Evidence-Based Reviews

Discuss this article

Testifying in civil commitment proceedings sometimes is the only way to make sure dangerous patients get the hospital care they need. But for many psychiatrists, providing courtroom testimony can be a nerve-wracking experience because they:

• lack formal training about how to testify
• lack familiarity with laws and court procedures
• fear cross-examination.

Training programs are required to teach psychiatry residents about civil commitment but not about how to testify.^1,2 Residents who get to take the stand during training usually do not receive any instruction.² Knowing some fundamentals of testifying can reduce your anxiety and reluctance to take the stand³ and help you to perform better in court.

Court procedures

A doctor may not force a patient to stay in a hospital, no matter how much the patient needs treatment. Only courts have legal authority to order involuntary psychiatric hospitalization, and courts may do this only after receiving proof that civil commitment is legally justified. Statutory criteria for civil commitment vary across jurisdictions, but typically, the court must hear evidence proving that a person:

• exhibits clear signs of a mental illness
• and because of the mental illness recently did something that placed himself or others in physical danger.

Courts usually rely on testimony from patients’ caregivers for this evidence. Thus, testifying is a skill psychiatrists must exercise to care for seriously ill patients who need treatment but don’t want it.

Testifying and playing basketball have a lot in common. To score points in basketball, a player must put the ball through the hoop and stay in bounds.

To be effective in a civil commitment hearing, a psychiatrist needs a similar game plan. The ball is your testimony, the hoop is the law’s exact wording in your state, and the bounds are recent dangerous behavior.

Completing the Civil Commitment Checklist ( Figure ) will help you determine if you are ready to go to court. Download a PDF of this checklist and a worksheet to compile information you will need to provide accurate and relevant testimony.

Figure: Are you ready to testify in a civil commitment hearing?

*If you answer “yes” to all questions, you are ready to testify about the need for civil commitment. If you answer “no” to any bulleted items, civil commitment may be inappropriate. If you answer “no” to any of the other questions, you’re not ready to go to court

Shoot the ball through the hoop

As early as the mid-19th century, attorneys and physicians realized that “no physician or surgeon could be a satisfactory expert witness without some knowledge of the law.”⁴ You may have the best basketball shooting technique in the world, but it won’t help if you don’t know where the hoop is. Likewise, you’ll be shooting blind if you come to court without knowing your state’s requirements for civil commitment—which many psychiatrists don’t know.⁵

Your skills at diagnosis and verbal persuasiveness are critical to good testimony, but if you don’t directly address the requirements for involuntary hospitalization in your state, your testimony may be irrelevant. A court cannot authorize civil commitment unless your testimony clearly and convincingly shows that a patient is mentally ill and dangerous—as defined by the law in your state.

In most states, you can look up your state’s commitment statute on the Internet, and you can take a printed copy of the statute to the witness stand if you wish. Using the law’s actual wording, you can give the court examples of behavior that show why your patient needs hospitalization.

For example, Ohio law defines a “mental disorder” for purposes of involuntary hospitalization as “a substantial disorder of thought, mood, perception, orientation, or memory that grossly impairs judgment, behavior, capacity to recognize reality, or ability to meet the ordinary demands of life.”⁶ In Ohio and many other states, an official psychiatric diagnosis is neither necessary nor sufficient for civil commitment. Of course, psychiatrists should formulate diagnostic opinions using well-established criteria. But in court, the diagnosis is like the backboard—it is not the hoop that the ball must pass through. The court needs to know whether a patient’s recent actions are manifestations of impairments listed in the statute.

Here’s an example of testimony that makes the basketball hit the backboard but doesn’t put the ball through the hoop:

Doctor: “Your honor, my patient has schizophrenia, paranoid type, which he’s had for quite a number of years. Patients with paranoid schizophrenia have a hard time because they think people are after them. Based on my experience, I don’t see how my patient can survive outside the hospital right now. He’s too paranoid, and his thinking is messed up.”

 

Though this testimony may be medically sound, the psychiatrist has not told the court specifically how the patient’s illness impairs his present functioning. Here’s an example of a ball going through the hoop after bouncing off the backboard:

Doctor: “Your honor, my patient has schizophrenia, paranoid type. He hears voices saying that the government is trying to assassinate him by poisoning his food and medications. As a result, he stopped taking medications 5 days ago and left the safety of his group home. He has been living under an overpass and refusing to eat. He has become malnourished and dehydrated. This shows he has a substantial disorder of thought and judgment that keeps him from recognizing reality and meeting the ordinary demands of life.”

Staying in bounds

As a witness, your role is to tell the court the truth about your patient’s situation so that justice can be served—if necessary, by allowing the state to override your patient’s liberty interests through involuntary hospitalization.⁷ To justify involuntary hospitalization in most states, the court must find that your patient is both mentally ill and has recently done something dangerous.

In basketball, the referee stops the game when the ball goes out of bounds. Similarly, the court may stop you if your testimony goes out of bounds by invoking non-recent evidence to demonstrate current dangerousness. For example:

Doctor: “Your honor, this patient has been hospitalized twice in the last 10 months after intentionally overdosing on medications. He told me that last month he tried to kill himself with pills. I know this patient well, and I fear he’s going to overdose again.”

Patient’s attorney: “Objection, your honor. What my client said 1 month ago has nothing to do with his alleged dangerousness today.”

The judge may sustain the objection and disregard your testimony because the events you have recounted fall outside your jurisdiction’s time frame for a “recent” event. What your patient said last month may well make you think your patient is at risk now, but the law establishes sometimes-arbitrary boundaries to protect patients’ liberty. The legal justification is that placing time limits on dangerous behavior minimizes the risk of an erroneous civil commitment; evidence of actual, recent behavior increases the likelihood of real, current danger and reduces the risk of involuntarily hospitalizing someone who would not do harm.⁸

The definition of “recent” varies from state to state. Pennsylvania looks at behavior within the last 30 days;⁹ Utah uses 7 days.¹⁰ Often, the law is not clear; rather than set firm time restrictions, some states consider whether the actions in question are “material and relevant to the person’s present condition.”¹¹

Know what your jurisdiction considers “recent” behavior. If your state’s statute is not clear, ask a judge or attorney. As you think about your testimony, make sure that information describes dangerous behavior within the time frame that the court will accept as “recent.”

Cross-examination

Civil commitment hearings are supposed to be adversarial. The “proponent” of involuntary hospitalization (the State) puts on its best case, hoping to convince the judge (or occasionally, a jury) that commitment is justified. The “respondent” (the patient facing potential commitment) has many of the rights that criminal defendants have, including the right to an attorney and the right to challenge witnesses—including psychiatrists—through cross-examination.¹²

Psychiatrists aren’t accustomed to having their clinical reasoning forcefully challenged. When they disagree, psychiatrists usually seek to persuade each other and reach consensus rather than openly criticize colleagues and point out flaws in their opinions about patients. Even when insurance companies and patients disagree with you, they usually don’t try to discredit you.

Testifying is different.¹³ Expect to have your conclusions bluntly challenged in civil commitment hearings. But remember: as in a basketball game, attorneys’ cross-examination challenges are not personal—they’re intended only to win the game. Also, you can practice. Just as practicing against opponents improves skills needed to win basketball games, practicing against real or anticipated cross-examination can help you respond when you’re testifying. Be prepared to answer commonly asked questions ( Table 1 ), such as:

“Doctor, you’ve testified that Mr. Jones has bipolar disorder. Aren’t all psychiatric diagnoses just a guess?”

“Doctor, how can you be certain that Mr. Smith’s psychosis, as you call it, was the result of schizophrenia, not alcohol intoxication?”

 

“Well, doctor, if you’re saying that Mrs. Clark’s psychosis was caused by a urinary tract infection, isn’t that a medical problem and not a psychiatric problem that we can lock her up for?”

Other ways to prepare:

• Have a colleague play the part of an opposing attorney who is trying to find fault with your clinical reasoning.
• Imagine you are retained by an attorney who wants to find holes in your own testimony.
• Watch other psychiatrists testify, and learn from their triumphs or mistakes.

Table 1

Questions you’re likely to face when testifying

Question	Importance
What is your diagnosis?	In all states, a mental illness leading to dangerous behavior is required for involuntary hospitalization. Courts are less interested in the name of the disorder than in knowing how its symptoms affect the respondent and lead to danger
Why is the hospital the least restrictive environment?	The “least restrictive” legal standard14 is a safeguard against unwarranted hospitalization. Be ready to explain why other treatment options (outpatient, day hospitals, etc.) are not appropriate
What medications is the patient taking?	Be prepared to tell the court whether the patient was taking medications before and since admission. Know which medications are being prescribed for the patient in the hospital
Why does your diagnosis differ from the diagnosis of another clinician in the chart?	Cross-examining attorneys will often try to discredit your opinion by pointing out that other clinicians diagnosed something different. You can explain that different disorders belong within common diagnostic categories (for example, schizophrenia and schizoaffective disorder are both psychoses). Explaining your answer this way enhances its credibility by demonstrating agreement with the other clinician
What is the patient’s response to the allegations of dangerousness listed in the chart?	Don’t neglect patients’ views. Their answers allow you to testify about the patient from direct knowledge and often provide evidence of thought, mood, or judgment problems

Good and bad shots

Good basketball players avoid fouls and taking shots that opponents can block easily. A good psychiatric witness knows how to avoid committing legal fouls and having testimony blocked.

The hearsay block. In clinical practice, psychiatrists should use and often rely on information about their patients that they obtain from other persons. But in court, testifying about such information could be disallowed on grounds that it is hearsay—testimony about what someone else saw or heard.

In civil commitment hearings, rules against allowing hearsay protect patients from accusations that may be false or misleading and that they cannot challenge through cross-examination. Although many states have a “hearsay exception” for civil commitment hearings—meaning that doctors may testify about what others have told them—not all do. If you practice in a state without this exception, you’ll need to gather information and plan your testimony carefully to avoid having the basis for your opinion excluded.

To avoid this “block,” testify only about events you saw or heard. To be fair to patients, always ask them about their side of the story. You can then testify about your clinical findings—what you saw and heard—instead of what someone else said. For example:

Doctor: “Your wife told me you wanted to kill yourself. Is that true?”

Patient: “It wasn’t my wife. I told my brother I wanted to kill myself.”

Doctor: “How about now? Do you still want to kill yourself?”

Patient: “Yes, I do.”

The doctor now can testify about first-hand experience with the patient:

Doctor: “Your honor, I asked the patient whether he told anyone that he wanted to kill himself. He told me he had told his brother he wanted to kill himself, and that he still felt that way.”

Play only your position

Basketball teams get into trouble if players try to do things others are supposed to do. Players are not supposed to give orders to the coach. In court, your role is to provide expert testimony about your patient and psychiatry. Stay in that role:

• Don’t address the ultimate legal question, as in saying, “My patient meets this state’s commitment criteria.” That’s for the judge to decide.
• Don’t opine about the moral virtues or shortcomings of the courts or hearings: “My patient desperately needs treatment, but you’re just asking about whether he fits narrow legal rules.”
• Don’t testify about topics on which you’re not an expert: “I think the police used too much force when they handcuffed my patient.”

Table 2

Dos and don’ts of testifying in a civil commitment hearing

Do…	Don’t…
Wear conservative business attire, which shows that you take your work seriously	Dress casually. Though casual dress is OK in many workplaces, lawyers wear suits in court
Remain calm, professional, and respectful	Make jokes or sarcastic remarks; a cross-examining attorney will easily discredit you by pointing out that this is a ‘serious matter’
Use recent examples of the patient’s dangerousness	Testify about your patient’s childhood or remote events
Describe behaviors or statements you witnessed or obtained from the patient	Testify about information obtained only from other people
Pause before answering each question. Doing this allows time for you to think and for an attorney to object to the question	Lose your cool or argue with the attorneys or judge
Source: Adapted from Knoll JL, Resnick PJ. Deposition dos and don’ts: how to answer 8 tricky questions. Current Psychiatry. 2007;7(3):25-40

 

Related resources

• Melton GB, Petrila J, Poythress NG, et al. Psychological evaluations for the courts. New York, NY: Guilford Press; 2007.
• Perlin ML. Mental disability law—civil and criminal. 2nd ed. Charlottesville, VA: Lexis Law Publishing; 1998.

Disclosure

The authors report no financial relationship with any company whose products are mentioned in the article or with manufacturers of competing products.

Discuss this article

Testifying in civil commitment proceedings sometimes is the only way to make sure dangerous patients get the hospital care they need. But for many psychiatrists, providing courtroom testimony can be a nerve-wracking experience because they:

• lack formal training about how to testify
• lack familiarity with laws and court procedures
• fear cross-examination.

Training programs are required to teach psychiatry residents about civil commitment but not about how to testify.^1,2 Residents who get to take the stand during training usually do not receive any instruction.² Knowing some fundamentals of testifying can reduce your anxiety and reluctance to take the stand³ and help you to perform better in court.

Court procedures

A doctor may not force a patient to stay in a hospital, no matter how much the patient needs treatment. Only courts have legal authority to order involuntary psychiatric hospitalization, and courts may do this only after receiving proof that civil commitment is legally justified. Statutory criteria for civil commitment vary across jurisdictions, but typically, the court must hear evidence proving that a person:

• exhibits clear signs of a mental illness
• and because of the mental illness recently did something that placed himself or others in physical danger.

Courts usually rely on testimony from patients’ caregivers for this evidence. Thus, testifying is a skill psychiatrists must exercise to care for seriously ill patients who need treatment but don’t want it.

Testifying and playing basketball have a lot in common. To score points in basketball, a player must put the ball through the hoop and stay in bounds.

To be effective in a civil commitment hearing, a psychiatrist needs a similar game plan. The ball is your testimony, the hoop is the law’s exact wording in your state, and the bounds are recent dangerous behavior.

Completing the Civil Commitment Checklist ( Figure ) will help you determine if you are ready to go to court. Download a PDF of this checklist and a worksheet to compile information you will need to provide accurate and relevant testimony.

Figure: Are you ready to testify in a civil commitment hearing?

*If you answer “yes” to all questions, you are ready to testify about the need for civil commitment. If you answer “no” to any bulleted items, civil commitment may be inappropriate. If you answer “no” to any of the other questions, you’re not ready to go to court

Shoot the ball through the hoop

As early as the mid-19th century, attorneys and physicians realized that “no physician or surgeon could be a satisfactory expert witness without some knowledge of the law.”⁴ You may have the best basketball shooting technique in the world, but it won’t help if you don’t know where the hoop is. Likewise, you’ll be shooting blind if you come to court without knowing your state’s requirements for civil commitment—which many psychiatrists don’t know.⁵

Your skills at diagnosis and verbal persuasiveness are critical to good testimony, but if you don’t directly address the requirements for involuntary hospitalization in your state, your testimony may be irrelevant. A court cannot authorize civil commitment unless your testimony clearly and convincingly shows that a patient is mentally ill and dangerous—as defined by the law in your state.

In most states, you can look up your state’s commitment statute on the Internet, and you can take a printed copy of the statute to the witness stand if you wish. Using the law’s actual wording, you can give the court examples of behavior that show why your patient needs hospitalization.

For example, Ohio law defines a “mental disorder” for purposes of involuntary hospitalization as “a substantial disorder of thought, mood, perception, orientation, or memory that grossly impairs judgment, behavior, capacity to recognize reality, or ability to meet the ordinary demands of life.”⁶ In Ohio and many other states, an official psychiatric diagnosis is neither necessary nor sufficient for civil commitment. Of course, psychiatrists should formulate diagnostic opinions using well-established criteria. But in court, the diagnosis is like the backboard—it is not the hoop that the ball must pass through. The court needs to know whether a patient’s recent actions are manifestations of impairments listed in the statute.

Here’s an example of testimony that makes the basketball hit the backboard but doesn’t put the ball through the hoop:

Doctor: “Your honor, my patient has schizophrenia, paranoid type, which he’s had for quite a number of years. Patients with paranoid schizophrenia have a hard time because they think people are after them. Based on my experience, I don’t see how my patient can survive outside the hospital right now. He’s too paranoid, and his thinking is messed up.”

 

Though this testimony may be medically sound, the psychiatrist has not told the court specifically how the patient’s illness impairs his present functioning. Here’s an example of a ball going through the hoop after bouncing off the backboard:

Doctor: “Your honor, my patient has schizophrenia, paranoid type. He hears voices saying that the government is trying to assassinate him by poisoning his food and medications. As a result, he stopped taking medications 5 days ago and left the safety of his group home. He has been living under an overpass and refusing to eat. He has become malnourished and dehydrated. This shows he has a substantial disorder of thought and judgment that keeps him from recognizing reality and meeting the ordinary demands of life.”

Staying in bounds

As a witness, your role is to tell the court the truth about your patient’s situation so that justice can be served—if necessary, by allowing the state to override your patient’s liberty interests through involuntary hospitalization.⁷ To justify involuntary hospitalization in most states, the court must find that your patient is both mentally ill and has recently done something dangerous.

In basketball, the referee stops the game when the ball goes out of bounds. Similarly, the court may stop you if your testimony goes out of bounds by invoking non-recent evidence to demonstrate current dangerousness. For example:

Doctor: “Your honor, this patient has been hospitalized twice in the last 10 months after intentionally overdosing on medications. He told me that last month he tried to kill himself with pills. I know this patient well, and I fear he’s going to overdose again.”

Patient’s attorney: “Objection, your honor. What my client said 1 month ago has nothing to do with his alleged dangerousness today.”

The judge may sustain the objection and disregard your testimony because the events you have recounted fall outside your jurisdiction’s time frame for a “recent” event. What your patient said last month may well make you think your patient is at risk now, but the law establishes sometimes-arbitrary boundaries to protect patients’ liberty. The legal justification is that placing time limits on dangerous behavior minimizes the risk of an erroneous civil commitment; evidence of actual, recent behavior increases the likelihood of real, current danger and reduces the risk of involuntarily hospitalizing someone who would not do harm.⁸

The definition of “recent” varies from state to state. Pennsylvania looks at behavior within the last 30 days;⁹ Utah uses 7 days.¹⁰ Often, the law is not clear; rather than set firm time restrictions, some states consider whether the actions in question are “material and relevant to the person’s present condition.”¹¹

Know what your jurisdiction considers “recent” behavior. If your state’s statute is not clear, ask a judge or attorney. As you think about your testimony, make sure that information describes dangerous behavior within the time frame that the court will accept as “recent.”

Cross-examination

Civil commitment hearings are supposed to be adversarial. The “proponent” of involuntary hospitalization (the State) puts on its best case, hoping to convince the judge (or occasionally, a jury) that commitment is justified. The “respondent” (the patient facing potential commitment) has many of the rights that criminal defendants have, including the right to an attorney and the right to challenge witnesses—including psychiatrists—through cross-examination.¹²

Psychiatrists aren’t accustomed to having their clinical reasoning forcefully challenged. When they disagree, psychiatrists usually seek to persuade each other and reach consensus rather than openly criticize colleagues and point out flaws in their opinions about patients. Even when insurance companies and patients disagree with you, they usually don’t try to discredit you.

Testifying is different.¹³ Expect to have your conclusions bluntly challenged in civil commitment hearings. But remember: as in a basketball game, attorneys’ cross-examination challenges are not personal—they’re intended only to win the game. Also, you can practice. Just as practicing against opponents improves skills needed to win basketball games, practicing against real or anticipated cross-examination can help you respond when you’re testifying. Be prepared to answer commonly asked questions ( Table 1 ), such as:

“Doctor, you’ve testified that Mr. Jones has bipolar disorder. Aren’t all psychiatric diagnoses just a guess?”

“Doctor, how can you be certain that Mr. Smith’s psychosis, as you call it, was the result of schizophrenia, not alcohol intoxication?”

 

“Well, doctor, if you’re saying that Mrs. Clark’s psychosis was caused by a urinary tract infection, isn’t that a medical problem and not a psychiatric problem that we can lock her up for?”

Other ways to prepare:

• Have a colleague play the part of an opposing attorney who is trying to find fault with your clinical reasoning.
• Imagine you are retained by an attorney who wants to find holes in your own testimony.
• Watch other psychiatrists testify, and learn from their triumphs or mistakes.

Table 1

Questions you’re likely to face when testifying

Question	Importance
What is your diagnosis?	In all states, a mental illness leading to dangerous behavior is required for involuntary hospitalization. Courts are less interested in the name of the disorder than in knowing how its symptoms affect the respondent and lead to danger
Why is the hospital the least restrictive environment?	The “least restrictive” legal standard14 is a safeguard against unwarranted hospitalization. Be ready to explain why other treatment options (outpatient, day hospitals, etc.) are not appropriate
What medications is the patient taking?	Be prepared to tell the court whether the patient was taking medications before and since admission. Know which medications are being prescribed for the patient in the hospital
Why does your diagnosis differ from the diagnosis of another clinician in the chart?	Cross-examining attorneys will often try to discredit your opinion by pointing out that other clinicians diagnosed something different. You can explain that different disorders belong within common diagnostic categories (for example, schizophrenia and schizoaffective disorder are both psychoses). Explaining your answer this way enhances its credibility by demonstrating agreement with the other clinician
What is the patient’s response to the allegations of dangerousness listed in the chart?	Don’t neglect patients’ views. Their answers allow you to testify about the patient from direct knowledge and often provide evidence of thought, mood, or judgment problems

Good and bad shots

Good basketball players avoid fouls and taking shots that opponents can block easily. A good psychiatric witness knows how to avoid committing legal fouls and having testimony blocked.

The hearsay block. In clinical practice, psychiatrists should use and often rely on information about their patients that they obtain from other persons. But in court, testifying about such information could be disallowed on grounds that it is hearsay—testimony about what someone else saw or heard.

In civil commitment hearings, rules against allowing hearsay protect patients from accusations that may be false or misleading and that they cannot challenge through cross-examination. Although many states have a “hearsay exception” for civil commitment hearings—meaning that doctors may testify about what others have told them—not all do. If you practice in a state without this exception, you’ll need to gather information and plan your testimony carefully to avoid having the basis for your opinion excluded.

To avoid this “block,” testify only about events you saw or heard. To be fair to patients, always ask them about their side of the story. You can then testify about your clinical findings—what you saw and heard—instead of what someone else said. For example:

Doctor: “Your wife told me you wanted to kill yourself. Is that true?”

Patient: “It wasn’t my wife. I told my brother I wanted to kill myself.”

Doctor: “How about now? Do you still want to kill yourself?”

Patient: “Yes, I do.”

The doctor now can testify about first-hand experience with the patient:

Doctor: “Your honor, I asked the patient whether he told anyone that he wanted to kill himself. He told me he had told his brother he wanted to kill himself, and that he still felt that way.”

Play only your position

Basketball teams get into trouble if players try to do things others are supposed to do. Players are not supposed to give orders to the coach. In court, your role is to provide expert testimony about your patient and psychiatry. Stay in that role:

• Don’t address the ultimate legal question, as in saying, “My patient meets this state’s commitment criteria.” That’s for the judge to decide.
• Don’t opine about the moral virtues or shortcomings of the courts or hearings: “My patient desperately needs treatment, but you’re just asking about whether he fits narrow legal rules.”
• Don’t testify about topics on which you’re not an expert: “I think the police used too much force when they handcuffed my patient.”

Table 2

Dos and don’ts of testifying in a civil commitment hearing

Do…	Don’t…
Wear conservative business attire, which shows that you take your work seriously	Dress casually. Though casual dress is OK in many workplaces, lawyers wear suits in court
Remain calm, professional, and respectful	Make jokes or sarcastic remarks; a cross-examining attorney will easily discredit you by pointing out that this is a ‘serious matter’
Use recent examples of the patient’s dangerousness	Testify about your patient’s childhood or remote events
Describe behaviors or statements you witnessed or obtained from the patient	Testify about information obtained only from other people
Pause before answering each question. Doing this allows time for you to think and for an attorney to object to the question	Lose your cool or argue with the attorneys or judge
Source: Adapted from Knoll JL, Resnick PJ. Deposition dos and don’ts: how to answer 8 tricky questions. Current Psychiatry. 2007;7(3):25-40

 

Related resources

• Melton GB, Petrila J, Poythress NG, et al. Psychological evaluations for the courts. New York, NY: Guilford Press; 2007.
• Perlin ML. Mental disability law—civil and criminal. 2nd ed. Charlottesville, VA: Lexis Law Publishing; 1998.

Disclosure

The authors report no financial relationship with any company whose products are mentioned in the article or with manufacturers of competing products.

Discuss this article

Testifying in civil commitment proceedings sometimes is the only way to make sure dangerous patients get the hospital care they need. But for many psychiatrists, providing courtroom testimony can be a nerve-wracking experience because they:

• lack formal training about how to testify
• lack familiarity with laws and court procedures
• fear cross-examination.

Training programs are required to teach psychiatry residents about civil commitment but not about how to testify.^1,2 Residents who get to take the stand during training usually do not receive any instruction.² Knowing some fundamentals of testifying can reduce your anxiety and reluctance to take the stand³ and help you to perform better in court.

Court procedures

A doctor may not force a patient to stay in a hospital, no matter how much the patient needs treatment. Only courts have legal authority to order involuntary psychiatric hospitalization, and courts may do this only after receiving proof that civil commitment is legally justified. Statutory criteria for civil commitment vary across jurisdictions, but typically, the court must hear evidence proving that a person:

• exhibits clear signs of a mental illness
• and because of the mental illness recently did something that placed himself or others in physical danger.

Courts usually rely on testimony from patients’ caregivers for this evidence. Thus, testifying is a skill psychiatrists must exercise to care for seriously ill patients who need treatment but don’t want it.

Testifying and playing basketball have a lot in common. To score points in basketball, a player must put the ball through the hoop and stay in bounds.

To be effective in a civil commitment hearing, a psychiatrist needs a similar game plan. The ball is your testimony, the hoop is the law’s exact wording in your state, and the bounds are recent dangerous behavior.

Completing the Civil Commitment Checklist ( Figure ) will help you determine if you are ready to go to court. Download a PDF of this checklist and a worksheet to compile information you will need to provide accurate and relevant testimony.

Figure: Are you ready to testify in a civil commitment hearing?

*If you answer “yes” to all questions, you are ready to testify about the need for civil commitment. If you answer “no” to any bulleted items, civil commitment may be inappropriate. If you answer “no” to any of the other questions, you’re not ready to go to court

Shoot the ball through the hoop

As early as the mid-19th century, attorneys and physicians realized that “no physician or surgeon could be a satisfactory expert witness without some knowledge of the law.”⁴ You may have the best basketball shooting technique in the world, but it won’t help if you don’t know where the hoop is. Likewise, you’ll be shooting blind if you come to court without knowing your state’s requirements for civil commitment—which many psychiatrists don’t know.⁵

Your skills at diagnosis and verbal persuasiveness are critical to good testimony, but if you don’t directly address the requirements for involuntary hospitalization in your state, your testimony may be irrelevant. A court cannot authorize civil commitment unless your testimony clearly and convincingly shows that a patient is mentally ill and dangerous—as defined by the law in your state.

In most states, you can look up your state’s commitment statute on the Internet, and you can take a printed copy of the statute to the witness stand if you wish. Using the law’s actual wording, you can give the court examples of behavior that show why your patient needs hospitalization.

For example, Ohio law defines a “mental disorder” for purposes of involuntary hospitalization as “a substantial disorder of thought, mood, perception, orientation, or memory that grossly impairs judgment, behavior, capacity to recognize reality, or ability to meet the ordinary demands of life.”⁶ In Ohio and many other states, an official psychiatric diagnosis is neither necessary nor sufficient for civil commitment. Of course, psychiatrists should formulate diagnostic opinions using well-established criteria. But in court, the diagnosis is like the backboard—it is not the hoop that the ball must pass through. The court needs to know whether a patient’s recent actions are manifestations of impairments listed in the statute.

Here’s an example of testimony that makes the basketball hit the backboard but doesn’t put the ball through the hoop:

Doctor: “Your honor, my patient has schizophrenia, paranoid type, which he’s had for quite a number of years. Patients with paranoid schizophrenia have a hard time because they think people are after them. Based on my experience, I don’t see how my patient can survive outside the hospital right now. He’s too paranoid, and his thinking is messed up.”

 

Though this testimony may be medically sound, the psychiatrist has not told the court specifically how the patient’s illness impairs his present functioning. Here’s an example of a ball going through the hoop after bouncing off the backboard:

Doctor: “Your honor, my patient has schizophrenia, paranoid type. He hears voices saying that the government is trying to assassinate him by poisoning his food and medications. As a result, he stopped taking medications 5 days ago and left the safety of his group home. He has been living under an overpass and refusing to eat. He has become malnourished and dehydrated. This shows he has a substantial disorder of thought and judgment that keeps him from recognizing reality and meeting the ordinary demands of life.”

Staying in bounds

As a witness, your role is to tell the court the truth about your patient’s situation so that justice can be served—if necessary, by allowing the state to override your patient’s liberty interests through involuntary hospitalization.⁷ To justify involuntary hospitalization in most states, the court must find that your patient is both mentally ill and has recently done something dangerous.

In basketball, the referee stops the game when the ball goes out of bounds. Similarly, the court may stop you if your testimony goes out of bounds by invoking non-recent evidence to demonstrate current dangerousness. For example:

Doctor: “Your honor, this patient has been hospitalized twice in the last 10 months after intentionally overdosing on medications. He told me that last month he tried to kill himself with pills. I know this patient well, and I fear he’s going to overdose again.”

Patient’s attorney: “Objection, your honor. What my client said 1 month ago has nothing to do with his alleged dangerousness today.”

The judge may sustain the objection and disregard your testimony because the events you have recounted fall outside your jurisdiction’s time frame for a “recent” event. What your patient said last month may well make you think your patient is at risk now, but the law establishes sometimes-arbitrary boundaries to protect patients’ liberty. The legal justification is that placing time limits on dangerous behavior minimizes the risk of an erroneous civil commitment; evidence of actual, recent behavior increases the likelihood of real, current danger and reduces the risk of involuntarily hospitalizing someone who would not do harm.⁸

The definition of “recent” varies from state to state. Pennsylvania looks at behavior within the last 30 days;⁹ Utah uses 7 days.¹⁰ Often, the law is not clear; rather than set firm time restrictions, some states consider whether the actions in question are “material and relevant to the person’s present condition.”¹¹

Know what your jurisdiction considers “recent” behavior. If your state’s statute is not clear, ask a judge or attorney. As you think about your testimony, make sure that information describes dangerous behavior within the time frame that the court will accept as “recent.”

Cross-examination

Civil commitment hearings are supposed to be adversarial. The “proponent” of involuntary hospitalization (the State) puts on its best case, hoping to convince the judge (or occasionally, a jury) that commitment is justified. The “respondent” (the patient facing potential commitment) has many of the rights that criminal defendants have, including the right to an attorney and the right to challenge witnesses—including psychiatrists—through cross-examination.¹²

Psychiatrists aren’t accustomed to having their clinical reasoning forcefully challenged. When they disagree, psychiatrists usually seek to persuade each other and reach consensus rather than openly criticize colleagues and point out flaws in their opinions about patients. Even when insurance companies and patients disagree with you, they usually don’t try to discredit you.

Testifying is different.¹³ Expect to have your conclusions bluntly challenged in civil commitment hearings. But remember: as in a basketball game, attorneys’ cross-examination challenges are not personal—they’re intended only to win the game. Also, you can practice. Just as practicing against opponents improves skills needed to win basketball games, practicing against real or anticipated cross-examination can help you respond when you’re testifying. Be prepared to answer commonly asked questions ( Table 1 ), such as:

“Doctor, you’ve testified that Mr. Jones has bipolar disorder. Aren’t all psychiatric diagnoses just a guess?”

“Doctor, how can you be certain that Mr. Smith’s psychosis, as you call it, was the result of schizophrenia, not alcohol intoxication?”

 

“Well, doctor, if you’re saying that Mrs. Clark’s psychosis was caused by a urinary tract infection, isn’t that a medical problem and not a psychiatric problem that we can lock her up for?”

Other ways to prepare:

• Have a colleague play the part of an opposing attorney who is trying to find fault with your clinical reasoning.
• Imagine you are retained by an attorney who wants to find holes in your own testimony.
• Watch other psychiatrists testify, and learn from their triumphs or mistakes.

Table 1

Questions you’re likely to face when testifying

Question	Importance
What is your diagnosis?	In all states, a mental illness leading to dangerous behavior is required for involuntary hospitalization. Courts are less interested in the name of the disorder than in knowing how its symptoms affect the respondent and lead to danger
Why is the hospital the least restrictive environment?	The “least restrictive” legal standard14 is a safeguard against unwarranted hospitalization. Be ready to explain why other treatment options (outpatient, day hospitals, etc.) are not appropriate
What medications is the patient taking?	Be prepared to tell the court whether the patient was taking medications before and since admission. Know which medications are being prescribed for the patient in the hospital
Why does your diagnosis differ from the diagnosis of another clinician in the chart?	Cross-examining attorneys will often try to discredit your opinion by pointing out that other clinicians diagnosed something different. You can explain that different disorders belong within common diagnostic categories (for example, schizophrenia and schizoaffective disorder are both psychoses). Explaining your answer this way enhances its credibility by demonstrating agreement with the other clinician
What is the patient’s response to the allegations of dangerousness listed in the chart?	Don’t neglect patients’ views. Their answers allow you to testify about the patient from direct knowledge and often provide evidence of thought, mood, or judgment problems

Good and bad shots

Good basketball players avoid fouls and taking shots that opponents can block easily. A good psychiatric witness knows how to avoid committing legal fouls and having testimony blocked.

The hearsay block. In clinical practice, psychiatrists should use and often rely on information about their patients that they obtain from other persons. But in court, testifying about such information could be disallowed on grounds that it is hearsay—testimony about what someone else saw or heard.

In civil commitment hearings, rules against allowing hearsay protect patients from accusations that may be false or misleading and that they cannot challenge through cross-examination. Although many states have a “hearsay exception” for civil commitment hearings—meaning that doctors may testify about what others have told them—not all do. If you practice in a state without this exception, you’ll need to gather information and plan your testimony carefully to avoid having the basis for your opinion excluded.

To avoid this “block,” testify only about events you saw or heard. To be fair to patients, always ask them about their side of the story. You can then testify about your clinical findings—what you saw and heard—instead of what someone else said. For example:

Doctor: “Your wife told me you wanted to kill yourself. Is that true?”

Patient: “It wasn’t my wife. I told my brother I wanted to kill myself.”

Doctor: “How about now? Do you still want to kill yourself?”

Patient: “Yes, I do.”

The doctor now can testify about first-hand experience with the patient:

Doctor: “Your honor, I asked the patient whether he told anyone that he wanted to kill himself. He told me he had told his brother he wanted to kill himself, and that he still felt that way.”

Play only your position

Basketball teams get into trouble if players try to do things others are supposed to do. Players are not supposed to give orders to the coach. In court, your role is to provide expert testimony about your patient and psychiatry. Stay in that role:

• Don’t address the ultimate legal question, as in saying, “My patient meets this state’s commitment criteria.” That’s for the judge to decide.
• Don’t opine about the moral virtues or shortcomings of the courts or hearings: “My patient desperately needs treatment, but you’re just asking about whether he fits narrow legal rules.”
• Don’t testify about topics on which you’re not an expert: “I think the police used too much force when they handcuffed my patient.”

Table 2

Dos and don’ts of testifying in a civil commitment hearing

Do…	Don’t…
Wear conservative business attire, which shows that you take your work seriously	Dress casually. Though casual dress is OK in many workplaces, lawyers wear suits in court
Remain calm, professional, and respectful	Make jokes or sarcastic remarks; a cross-examining attorney will easily discredit you by pointing out that this is a ‘serious matter’
Use recent examples of the patient’s dangerousness	Testify about your patient’s childhood or remote events
Describe behaviors or statements you witnessed or obtained from the patient	Testify about information obtained only from other people
Pause before answering each question. Doing this allows time for you to think and for an attorney to object to the question	Lose your cool or argue with the attorneys or judge
Source: Adapted from Knoll JL, Resnick PJ. Deposition dos and don’ts: how to answer 8 tricky questions. Current Psychiatry. 2007;7(3):25-40

 

Related resources

• Melton GB, Petrila J, Poythress NG, et al. Psychological evaluations for the courts. New York, NY: Guilford Press; 2007.
• Perlin ML. Mental disability law—civil and criminal. 2nd ed. Charlottesville, VA: Lexis Law Publishing; 1998.

Disclosure

The authors report no financial relationship with any company whose products are mentioned in the article or with manufacturers of competing products.

Discuss this article

Five recent randomized controlled trials (RCTs) have demonstrated the efficacy of atypical antipsychotics for treating bipolar disorder in children and adolescents, but 4 of these 5 trials remain unpublished. The lag time between the completion of these trials and publication of their results—typically 4 to 5 years¹—leaves psychiatrists without important evidence to explain to families and critics² why they might recommend using these powerful medications in children with mental illness.

This article previews the preliminary results of these 5 RCTs of atypical antipsychotics, offers a treatment algorithm supported by this evidence, and discusses how to manage potentially serious risks when using antipsychotics to treat children and adolescents with bipolar disorder (BPD).

Where do atypical antipsychotics fit in?

Details of the 5 industry-sponsored RCTs of atypical antipsychotics in children and adolescents with bipolar I manic or mixed episodes are summarized in Table 1.^3-7 Only the olanzapine study⁴ has been published; data from the other 4 trials were presented at medical meetings in 2007 and 2008.

Change in Young Mania Rating Scale (YMRS) score was the primary outcome measure in these 5 trials, and each compound was more effective than placebo. The trials demonstrated statistically significant and clinically relevant differences between each antipsychotic and placebo. The number needed to treat (NNT)—how many patients need to be treated for 1 to benefit in a controlled clinical trial—ranged from 2 to 4. For comparison, the NNT for statins in the prevention of coronary events is 12 to 22,⁸ and the NNT in an analysis of trials of selective serotonin reuptake inhibitors for pediatric major depressive disorder was 9.⁹ Thus, an NNT of ≤4 represents a clinically significant effect.

Risperidone is FDA-approved for short-term treatment of acute bipolar I manic or mixed episodes in patients age 10 to 17. Aripiprazole is approved for acute and maintenance treatment of bipolar I manic or mixed episodes (with or without psychosis) as monotherapy or with lithium or valproate in patients age 10 to 17. In June, an FDA advisory committee recommended pediatric bipolar indications for olanzapine, quetiapine, and ziprasidone.

‘Mood stabilizers’ such as lithium, valproate, and carbamazepine have been used for years to treat bipolar mania in adults, adolescents, and children, despite limited supporting evidence. Preliminary results of a National Institute of Mental Health-funded double-blind RCT provide insights on their efficacy.¹⁰

The 153 outpatients age 7 to 17 in a bipolar I manic or mixed episode were randomly assigned to lithium, divalproex, or placebo for 8 weeks. Response rates—based on a Clinical Global Impressions-Improvement score of 1 or 2 (very much or much improved)—were divalproex, 54%; lithium, 42%; and placebo, 29%. Lithium showed a trend toward efficacy but did not clearly separate from placebo on the primary outcome measures. Effect sizes for lithium and divalproex were moderate.¹⁰

Only 1 study has compared a mood stabilizer with an atypical antipsychotic for treating mania in adolescents. In a double-blind trial, DelBello et al¹¹ randomly assigned 50 patients age 12 to 18 with a bipolar I manic or mixed episode to quetiapine, 400 to 600 mg/d, or divalproex, serum level 80 to 120 μg/mL, for 28 days. Manic symptoms resolved more rapidly, and remission rates measured by the YMRS were higher with quetiapine than with divalproex. Both medications were well tolerated.

Combination therapy. BPD as it presents in children and adolescents is often difficult to treat because of the disorder’s various phases (manic, depressed, mixed), frequent psychotic symptoms, and high rate of comorbidity. Pediatric BPD patients frequently require several psychotropics, including mood stabilizers and atypical antipsychotics.

In a double-blind, placebo-controlled study, 30 adolescents in a bipolar I manic or mixed episode initially received divalproex, 20 mg/kg/d, then were randomly assigned to 6 weeks of adjunctive quetiapine, titrated to 450 mg/d in 7 days (n=15), or placebo (n=15). Those receiving divalproex plus quetiapine showed a statistically significant greater reduction in manic symptoms (P=.03) and a higher response rate (87% vs 53%, P=.05), compared with those receiving divalproex and placebo. This suggests that a mood stabilizer plus an atypical antipsychotic is more effective than a mood stabilizer alone for adolescent mania. Quetiapine was well tolerated.¹²

Treatment. The American Psychiatric Association’s outdated 2002 practice guideline for acute bipolar I manic or mixed episodes in adults recommends lithium, valproate, and/or an antipsychotic.¹³ The more recent Texas Medication Algorithm Project (TMAP) guidelines recommend monotherapy with lithium, valproate, aripiprazole, quetiapine, risperidone, or ziprasidone for adults with euphoric or irritable manic or hypomanic symptoms.¹⁴

 

Based on the TMAP algorithm, recent clinical trial evidence, and our experience in treating pediatric BPD, we offer an approach for treating mania/hypomania in patients age 10 to 17 (see Proposed Algorithm). For dosing and precautions when using atypical antipsychotics in children and adolescents with BPD, see Table 2.^15-17

Comorbid psychiatric illnesses (such as anxiety disorders) are prevalent in adolescents with BPD. Evidence in adults and adolescents suggests that some atypical antipsychotics may provide additional benefit for these conditions as well. Thus, consider comorbid conditions and symptoms when choosing antimanic agents.

Attention-deficit/hyperactivity disorder (ADHD) is a common comorbidity in children with BPD, and stimulant medications are most often prescribed to treat inattentiveness and hyperactivity. Caution is imperative when treating bipolar youth with stimulants, which can exacerbate manic symptoms. Treat the patient’s mania before adding or reintroducing stimulant medication. Research and clinical experience suggest that if you first stabilize these patients on a mood stabilizer or atypical antipsychotic, adding a stimulant can be very helpful in treating comorbid ADHD symptoms. Start with low stimulant doses, and increase slowly.

Table 1

RCTs of atypical antipsychotics in patients age 10 to 17
with bipolar I disorder*

Antipsychotic and source	Bipolar I episode (# of subjects)	Trial duration (days)	Dosage (mg/d)	Response rate or YMRS score change	NNT	Mean weight gain (kg)
Risperidone Pandina et al3 AACAP 2007	Manic, mixed (169)	21	0.5 to 2.5 3 to 6	59% 63%	3.3 3.5	1.9 1.4
Olanzapine Tohen et al4	Manic, mixed (161)	21	10.4 ± 4.5	49%	4.1	3.7 ± 2.2
Quetiapine DelBello et al5 AACAP 2007	Manic (284)	21	400 600	64% 58%	4.4 4.2	1.7 1.7
Aripiprazole Wagner et al6 ACNP 2007	Manic, mixed (296)	28	10 30	45% 64%	4.1 2.4	0.9 0.54
Ziprasidone DelBello et al7 APA 2008	Manic, mixed (238)	28	80 to 160	–13.83 with ziprasidone, –8.61 with placebo	3.7	None
*Each trial included a 6-month open extension phase; results are pending
AACAP: American Academy of Child and Adolescent Psychiatry; ACNP: American College of Neuropsychopharmacology; APA: American Psychiatric Association; NNT: number needed to treat; RCT: randomized controlled trial; YMRS: Young Mania Rating Scale

Table 2

Recommended antipsychotic use in pediatric bipolar disorder

Drug	Starting dosage (mg)	Target dosage (mg/d)	Precautions
Aripiprazole	2.5 to 5 at bedtime	10 to 30	Monitor for CYP 3A4 and 2D6 interactions, weight, BMI, cholesterol, lipids, and glucose
Olanzapine	2.5 bid	10 to 20	Monitor for CYP 2D6 interactions, weight, BMI, cholesterol, lipids, glucose, and prolactin levels
Quetiapine	50 bid	400 to 1,200	Monitor for weight, BMI, cholesterol, lipids, and glucose
Risperidone	0.25 bid	1 to 2.5	Monitor for EPS, hyperprolactinemia (and associated sexual side effects, including galactorrhea), weight, BMI, cholesterol, lipids, glucose, and prolactin levels
Ziprasidone	20 bid	80 to 160	Check baseline ECG and as dose increases or with reason for high level of concern; monitor prolactin levels
BMI: body mass index; CYP: cytochrome P450; ECG: electrocardiography; EPS: extrapyramidal symptoms
Source: References 15-17

Proposed Algorithm: Treating a bipolar mixed/manic episode in patients age 10 to 17

Stage 1. Consider patient’s experience with antipsychotics, body weight, and family history when choosing first-line monotherapy (1A). Quetiapine poses low risk for extrapyramidal symptoms and tardive dyskinesia. Aripiprazole and ziprasidone pose relatively low risk of weight gain. Risperidone is potent at low doses but increases prolactin levels (long-term effect unknown).

Second-line choices (1B) are mood stabilizers lithium and valproate (because of lower potency than atypical antipsychotics), and olanzapine (which—although potent—causes substantial weight gain). In case of lack of response or intolerable side effects with initial agent, select an alternate from 1A or 1B. If this is not effective, move to Stage 2.

Stage 2. Consider augmentation for patients who show partial response to monotherapy (in your clinical judgment “mild to moderately improved” but not “much or very much improved”).

Stage 3. Combination therapy could include 2 mood stabilizers (such as lithium and valproate) plus an atypical antipsychotic; 2 atypical antipsychotics; or other combinations based on patient’s past responses. No research has shown these combinations to be efficacious in bipolar children and adolescents, but we find they sometimes help those with treatment-resistant symptoms.

Duration. Maintain psychotropics 12 to 18 months. When patient is euthymic, slowly taper 1 medication across several months. If symptoms recur, reintroduce the mood-stabilizing agent(s).

Source: Adapted and reprinted with permission from Kowatch RA, Fristad MA, Findling R, et al. Clinical manual for the management of bipolar disorder in children and adolescents. Arlington, VA: American Psychiatric Publishing, Inc.; 2008

Managing adverse effects

Although clinically effective, atypical antipsychotics may cause serious side effects that must be recognized and managed. These include extrapyramidal symptoms (EPS), tardive dyskinesia (TD), weight gain and obesity, hyperlipidemia, increased prolactin levels, and QTc changes. Counsel patients and families about the risks and benefits of antipsychotics when you consider them for children and adolescents with BPD (Table 3).

EPS. Drug-induced parkinsonism and akathisia are the most common EPS in children and adolescents with BPD treated with atypical antipsychotics.¹⁸

 

Correll et al¹⁹ reported a 10% rate of EPS in patients treated with aripiprazole. Treatment-emergent EPS also was observed in the RCT of risperidone.²⁰ EPS-related adverse events were associated with higher doses of risperidone, although none of the akathisia/EPS measures were thought to be “clinically significant.”

EPS frequency was relatively low and similar to placebo in the 3-week quetiapine trial,²¹ and no changes in movement disorder scale scores were observed during the olanzapine or ziprasidone RCTs.^4,7

Recommendations. If your pediatric patient develops EPS, first try an antipsychotic dose reduction. Because anticholinergics can contribute to antipsychotic-induced weight gain, reserve them until after a dosage reduction has been unsuccessful.

Benztropine (0.25 to 0.5 mg given 2 to 3 times daily, not to exceed 3 mg/d) or diphenhydramine (25 to 50 mg given 3 to 4 times daily; maximum dosage 5 mg/kg/d) can be effective in treating EPS. Avoid anticholinergics in children with narrow-angle glaucoma or age <3.

Akathisia may be managed with propranolol (20 to 120 mg/d in divided doses). Multiple doses (typically 3 times daily) are needed to prevent interdose withdrawal symptoms. Use this beta blocker with caution in children with asthma because of the possibility of bronchospasm.

TD. Short-term trials and a meta-analysis of atypical antipsychotic trials (>11 months’ duration, subject age <18) suggest a low annual risk for TD (0.4%).²² Large, prospective, long-term trials of atypical antipsychotics are necessary to more accurately define the risk of TD in the pediatric population, however. Retrospective analyses of adolescents treated with antipsychotics suggest 3 TD risk factors:

• early age of antipsychotic use
• medication nonadherence
• concomitant use of antiparkinsonian agents.²³

Kumra et al²⁴ identified lower premorbid functioning and greater positive symptoms at baseline as factors associated with “withdrawal dyskinesia/tardive dyskinesia” in children and adolescents with early-onset psychotic-spectrum disorders treated with typical or atypical antipsychotics.

Recommendations. To minimize TD risk, use the lowest effective antipsychotic dose, monitor for abnormal involuntary movements with standardized assessments (such as the Abnormal Involuntary Movement Scale), review risks and benefits with parents and patients, and regularly evaluate the indication and need for antipsychotic therapy. It is reasonable to attempt to lower the antipsychotic dose after the patient has attained remission and been stable for 1 year.

Neuroleptic malignant syndrome (NMS). This complication of dopamine-blocking medications:

• is among the most serious adverse effects of antipsychotic treatment
• continues to be associated with a mortality rate of 10%.²⁵

Recommendation. At least 1 recent review of pediatric NMS cases suggests that essential features (hyperthermia and severe muscular rigidity) are retained in children.²⁶ Nonetheless, monitor for variant presentations; hyper thermia or muscle rigidity may be absent or develop slowly over several days in patients treated with atypical antipsychotics.²⁷

Weight gain and glucose metabolism. A major adverse effect of most atypical antipsychotics is increased appetite, weight gain, and possible obesity.²⁸ In children, “obesity” refers to a body mass index (BMI) >95th percentile for age and sex; “over-weight” refers to BMI between the 85th and 95th percentile. Mean weight gain in the 5 atypical antipsychotic pediatric bipolar trials ranged from 0 to 8 lbs across 3 to 4 weeks of treatment (Figure).^3-7

Recommendations. Emphasize diet and exercise, with restriction of high-carbohydrate food, “fast foods,” and soft drinks. Another option is a trial of metformin, which decreases hepatic glucose production, decreases intestinal absorption of glucose, and improves insulin sensitivity by increasing peripheral glucose uptake and utilization.

Klein et al²⁹ studied 39 patients age 10 to 17 with mood and psychotic disorders whose weight increased by >10% during <1 year of olanzapine, risperidone, or quetiapine therapy. In this 16-week, double-blind, placebo-controlled trial, weight was stabilized in subjects receiving metformin, whereas those receiving placebo continued to gain weight (0.31 kg [0.68 lb]/week).

The usual starting metformin dose is 500 mg bid with meals. Increase in increments of 500 mg weekly, up to a maximum of 2,000 mg/d in divided doses. Potential side effects include diarrhea, nausea/vomiting, flatulence, and headache.

Hyperlipidemia. Patients who gain weight with atypical antipsychotics also may develop hyperlipidemia. Fasting serum triglycerides >150 mg/dL (1.70 mmol/L) in obese children are considered to be elevated and an early sign of metabolic syndrome.³⁰ Fasting total cholesterol >200 mg/dL (5.18 mmol/L) or low-density lipoprotein cholesterol >130 mg/dL (3.38 mmol/L) is consistent with hyperlipidemia.

 

Recommendation. Monitor and treat hyperlipidemia, which increases the risk of atherosclerosis as obese children grow older.³¹

Prolactin. Elevated prolactin concentrations may have deleterious effects in the developing child or adolescent, including gynecomastia, oligomenorrhea, and amenorrhea.¹⁷ Long-term effects on growth and sexual maturation have not been fully evaluated.

The relative tendency of atypical antipsychotics to cause hyperprolactinemia is roughly: risperidone/paliperidone > olanzapine > ziprasidone > quetiapine > clozapine > aripiprazole.¹⁸ In the risperidone RCT, mean changes in baseline prolactin levels were 41 ng/mL for boys and 59 ng/mL in girls.³ Results of the olanzapine RCT suggest a high incidence of hyperprolactinemia (26% of girls, 63% of boys).⁴ Decreases in serum prolactin were observed in bipolar children and adolescents treated with aripiprazole for 30 weeks.¹⁹

Recommendations. For any pediatric patient treated with an atypical antipsychotic that increases prolactin levels:

• Obtain a baseline prolactin level.
• Repeat after 6 months of treatment or with the emergence of elevated prolactin symptoms, such as gynecomastia in boys. Ask about increases in breast size, galactorrhea, changes in menstruation, sexual functioning, and pubertal development.

Switch patients who develop any of these side effects to another atypical agent that does not increase serum prolactin.³²

QTc interval prolongation. All atypical antipsychotics can cause QTc prolongation. Several cases of significant QTc prolongation have been reported in children and adolescents treated with ziprasidone.^33,34 In the RCT of ziprasidone, QTc prolongation was not clinically significant in most of the patients in which it was reported, and it did not lead to adverse events.³⁴ Mean QTc change was 8.1 msec at study termination.⁷

Patients enrolled in clinical trails are screened very carefully, however, and those with preexisting medical abnormalities typically are excluded. Thus, these findings may have limited usefulness for “real-world” patients.

Recommendations. Until additional information is known about the cardiac effects of atypical antipsychotics in children and adolescents:

• Perform a careful history, review of symptoms, and physical exam looking for any history of palpitations, shortness of breath, or syncope.
• Query specifically about any family history of sudden cardiac death.
• Perform a baseline resting ECG for patients starting ziprasidone or clozapine, or for other atypicals if indicated by history, review of systems, physical exam, etc.
• For patients treated with ziprasidone or clozapine, repeat ECG as the dose increases or if the patient has cardiac symptoms (unexplained shortness of breath, palpitations, skipped beats, etc.).

Table 3

Talking to families about using antipsychotics
in children with bipolar disorder

Effectiveness. Large, placebo-controlled studies have shown that atypical antipsychotics can significantly reduce manic symptoms in children and adolescents with bipolar disorder
Safety data. Additional 6-month safety data indicate that atypical antipsychotics continue to be effective in children and adolescents, without dramatic changes in side effects
Precautions. Antipsychotics are powerful medications and must be used carefully in pediatric patients
Potential side effects. All antipsychotics have serious potential side effects that must be recognized, monitored, and managed
Potential benefits from using atypical antipsychotics include mood stabilization, treatment of psychotic symptoms, and lower risk of extrapyramidal symptoms compared with typical antipsychotics
Risk vs benefit. On balance, the potential benefit of these agents outweighs the potential risk for children and adolescents with bipolar disorder

Figure: Mean weight gain with atypical antipsychotics in pediatric bipolar trials

Weight gain in children and adolescents with bipolar disorder varied among atypical antipsychotics used in 5 recent randomized controlled trials. Treatment duration was 3 weeks with olanzapine, risperidone, and quetiapine and 4 weeks with aripiprazole and ziprasidone. Dosages were olanzapine, 10.4 ± 4.5 mg/d; risperidone, 0.5 to 2.5 mg/d or 3 to 6 mg/d; aripiprazole, 10 or 30 mg/d; quetiapine, 400 or 600 mg/d; and ziprasidone, 80 to 160 mg/d.
Source: References 3-7Related resources

• Child and Adolescent Bipolar Foundation. www.bpkids.org.
• University of Illinois at Chicago Pediatric Mood Disorders Clinic. www.psych.uic.edu/pmdc.
• Ryan Licht Sang Bipolar Foundation. www.ryanlichtsangbipolarfoundation.org.

Drug brand names

• Aripiprazole • Abilify
• Benztropine • Cogentin
• Carbamazepine • Carbatrol
• Clozapine • Clozaril
• Diphenhydramine • Benadryl
• Divalproex sodium • Depakote
• Lithium • Lithobid, others
• Metformin • Glucophage
• Olanzapine • Zyprexa
• Paliperidone • Invega
• Propranolol • Inderal
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Valproate • Depacon
• Ziprasidone • Geodon

Disclosures

Dr. Kowatch is a consultant to and speaker for AstraZeneca and a consultant to Forest Pharmaceuticals. He receives research support from the National Alliance for Research on Schizophrenia and Depression, the National Institute of Child Health and Human Development, the National Institute of Mental Health, and the Stanley Foundation.

Dr. Strawn has received research support from the American Academy of Child and Adolescent Psychiatry (Lilly Pilot Research Award).

 

Dr. Sorter receives research support from the National Institute of Mental Health and the Health Foundation of Greater Cincinnati.

Discuss this article

Five recent randomized controlled trials (RCTs) have demonstrated the efficacy of atypical antipsychotics for treating bipolar disorder in children and adolescents, but 4 of these 5 trials remain unpublished. The lag time between the completion of these trials and publication of their results—typically 4 to 5 years¹—leaves psychiatrists without important evidence to explain to families and critics² why they might recommend using these powerful medications in children with mental illness.

This article previews the preliminary results of these 5 RCTs of atypical antipsychotics, offers a treatment algorithm supported by this evidence, and discusses how to manage potentially serious risks when using antipsychotics to treat children and adolescents with bipolar disorder (BPD).

Where do atypical antipsychotics fit in?

Details of the 5 industry-sponsored RCTs of atypical antipsychotics in children and adolescents with bipolar I manic or mixed episodes are summarized in Table 1.^3-7 Only the olanzapine study⁴ has been published; data from the other 4 trials were presented at medical meetings in 2007 and 2008.

Change in Young Mania Rating Scale (YMRS) score was the primary outcome measure in these 5 trials, and each compound was more effective than placebo. The trials demonstrated statistically significant and clinically relevant differences between each antipsychotic and placebo. The number needed to treat (NNT)—how many patients need to be treated for 1 to benefit in a controlled clinical trial—ranged from 2 to 4. For comparison, the NNT for statins in the prevention of coronary events is 12 to 22,⁸ and the NNT in an analysis of trials of selective serotonin reuptake inhibitors for pediatric major depressive disorder was 9.⁹ Thus, an NNT of ≤4 represents a clinically significant effect.

Risperidone is FDA-approved for short-term treatment of acute bipolar I manic or mixed episodes in patients age 10 to 17. Aripiprazole is approved for acute and maintenance treatment of bipolar I manic or mixed episodes (with or without psychosis) as monotherapy or with lithium or valproate in patients age 10 to 17. In June, an FDA advisory committee recommended pediatric bipolar indications for olanzapine, quetiapine, and ziprasidone.

‘Mood stabilizers’ such as lithium, valproate, and carbamazepine have been used for years to treat bipolar mania in adults, adolescents, and children, despite limited supporting evidence. Preliminary results of a National Institute of Mental Health-funded double-blind RCT provide insights on their efficacy.¹⁰

The 153 outpatients age 7 to 17 in a bipolar I manic or mixed episode were randomly assigned to lithium, divalproex, or placebo for 8 weeks. Response rates—based on a Clinical Global Impressions-Improvement score of 1 or 2 (very much or much improved)—were divalproex, 54%; lithium, 42%; and placebo, 29%. Lithium showed a trend toward efficacy but did not clearly separate from placebo on the primary outcome measures. Effect sizes for lithium and divalproex were moderate.¹⁰

Only 1 study has compared a mood stabilizer with an atypical antipsychotic for treating mania in adolescents. In a double-blind trial, DelBello et al¹¹ randomly assigned 50 patients age 12 to 18 with a bipolar I manic or mixed episode to quetiapine, 400 to 600 mg/d, or divalproex, serum level 80 to 120 μg/mL, for 28 days. Manic symptoms resolved more rapidly, and remission rates measured by the YMRS were higher with quetiapine than with divalproex. Both medications were well tolerated.

Combination therapy. BPD as it presents in children and adolescents is often difficult to treat because of the disorder’s various phases (manic, depressed, mixed), frequent psychotic symptoms, and high rate of comorbidity. Pediatric BPD patients frequently require several psychotropics, including mood stabilizers and atypical antipsychotics.

In a double-blind, placebo-controlled study, 30 adolescents in a bipolar I manic or mixed episode initially received divalproex, 20 mg/kg/d, then were randomly assigned to 6 weeks of adjunctive quetiapine, titrated to 450 mg/d in 7 days (n=15), or placebo (n=15). Those receiving divalproex plus quetiapine showed a statistically significant greater reduction in manic symptoms (P=.03) and a higher response rate (87% vs 53%, P=.05), compared with those receiving divalproex and placebo. This suggests that a mood stabilizer plus an atypical antipsychotic is more effective than a mood stabilizer alone for adolescent mania. Quetiapine was well tolerated.¹²

Treatment. The American Psychiatric Association’s outdated 2002 practice guideline for acute bipolar I manic or mixed episodes in adults recommends lithium, valproate, and/or an antipsychotic.¹³ The more recent Texas Medication Algorithm Project (TMAP) guidelines recommend monotherapy with lithium, valproate, aripiprazole, quetiapine, risperidone, or ziprasidone for adults with euphoric or irritable manic or hypomanic symptoms.¹⁴

 

Based on the TMAP algorithm, recent clinical trial evidence, and our experience in treating pediatric BPD, we offer an approach for treating mania/hypomania in patients age 10 to 17 (see Proposed Algorithm). For dosing and precautions when using atypical antipsychotics in children and adolescents with BPD, see Table 2.^15-17

Comorbid psychiatric illnesses (such as anxiety disorders) are prevalent in adolescents with BPD. Evidence in adults and adolescents suggests that some atypical antipsychotics may provide additional benefit for these conditions as well. Thus, consider comorbid conditions and symptoms when choosing antimanic agents.

Attention-deficit/hyperactivity disorder (ADHD) is a common comorbidity in children with BPD, and stimulant medications are most often prescribed to treat inattentiveness and hyperactivity. Caution is imperative when treating bipolar youth with stimulants, which can exacerbate manic symptoms. Treat the patient’s mania before adding or reintroducing stimulant medication. Research and clinical experience suggest that if you first stabilize these patients on a mood stabilizer or atypical antipsychotic, adding a stimulant can be very helpful in treating comorbid ADHD symptoms. Start with low stimulant doses, and increase slowly.

Table 1

RCTs of atypical antipsychotics in patients age 10 to 17
with bipolar I disorder*

Antipsychotic and source	Bipolar I episode (# of subjects)	Trial duration (days)	Dosage (mg/d)	Response rate or YMRS score change	NNT	Mean weight gain (kg)
Risperidone Pandina et al3 AACAP 2007	Manic, mixed (169)	21	0.5 to 2.5 3 to 6	59% 63%	3.3 3.5	1.9 1.4
Olanzapine Tohen et al4	Manic, mixed (161)	21	10.4 ± 4.5	49%	4.1	3.7 ± 2.2
Quetiapine DelBello et al5 AACAP 2007	Manic (284)	21	400 600	64% 58%	4.4 4.2	1.7 1.7
Aripiprazole Wagner et al6 ACNP 2007	Manic, mixed (296)	28	10 30	45% 64%	4.1 2.4	0.9 0.54
Ziprasidone DelBello et al7 APA 2008	Manic, mixed (238)	28	80 to 160	–13.83 with ziprasidone, –8.61 with placebo	3.7	None
*Each trial included a 6-month open extension phase; results are pending
AACAP: American Academy of Child and Adolescent Psychiatry; ACNP: American College of Neuropsychopharmacology; APA: American Psychiatric Association; NNT: number needed to treat; RCT: randomized controlled trial; YMRS: Young Mania Rating Scale

Table 2

Recommended antipsychotic use in pediatric bipolar disorder

Drug	Starting dosage (mg)	Target dosage (mg/d)	Precautions
Aripiprazole	2.5 to 5 at bedtime	10 to 30	Monitor for CYP 3A4 and 2D6 interactions, weight, BMI, cholesterol, lipids, and glucose
Olanzapine	2.5 bid	10 to 20	Monitor for CYP 2D6 interactions, weight, BMI, cholesterol, lipids, glucose, and prolactin levels
Quetiapine	50 bid	400 to 1,200	Monitor for weight, BMI, cholesterol, lipids, and glucose
Risperidone	0.25 bid	1 to 2.5	Monitor for EPS, hyperprolactinemia (and associated sexual side effects, including galactorrhea), weight, BMI, cholesterol, lipids, glucose, and prolactin levels
Ziprasidone	20 bid	80 to 160	Check baseline ECG and as dose increases or with reason for high level of concern; monitor prolactin levels
BMI: body mass index; CYP: cytochrome P450; ECG: electrocardiography; EPS: extrapyramidal symptoms
Source: References 15-17

Proposed Algorithm: Treating a bipolar mixed/manic episode in patients age 10 to 17

Stage 1. Consider patient’s experience with antipsychotics, body weight, and family history when choosing first-line monotherapy (1A). Quetiapine poses low risk for extrapyramidal symptoms and tardive dyskinesia. Aripiprazole and ziprasidone pose relatively low risk of weight gain. Risperidone is potent at low doses but increases prolactin levels (long-term effect unknown).

Second-line choices (1B) are mood stabilizers lithium and valproate (because of lower potency than atypical antipsychotics), and olanzapine (which—although potent—causes substantial weight gain). In case of lack of response or intolerable side effects with initial agent, select an alternate from 1A or 1B. If this is not effective, move to Stage 2.

Stage 2. Consider augmentation for patients who show partial response to monotherapy (in your clinical judgment “mild to moderately improved” but not “much or very much improved”).

Stage 3. Combination therapy could include 2 mood stabilizers (such as lithium and valproate) plus an atypical antipsychotic; 2 atypical antipsychotics; or other combinations based on patient’s past responses. No research has shown these combinations to be efficacious in bipolar children and adolescents, but we find they sometimes help those with treatment-resistant symptoms.

Duration. Maintain psychotropics 12 to 18 months. When patient is euthymic, slowly taper 1 medication across several months. If symptoms recur, reintroduce the mood-stabilizing agent(s).

Source: Adapted and reprinted with permission from Kowatch RA, Fristad MA, Findling R, et al. Clinical manual for the management of bipolar disorder in children and adolescents. Arlington, VA: American Psychiatric Publishing, Inc.; 2008

Managing adverse effects

Although clinically effective, atypical antipsychotics may cause serious side effects that must be recognized and managed. These include extrapyramidal symptoms (EPS), tardive dyskinesia (TD), weight gain and obesity, hyperlipidemia, increased prolactin levels, and QTc changes. Counsel patients and families about the risks and benefits of antipsychotics when you consider them for children and adolescents with BPD (Table 3).

EPS. Drug-induced parkinsonism and akathisia are the most common EPS in children and adolescents with BPD treated with atypical antipsychotics.¹⁸

 

Correll et al¹⁹ reported a 10% rate of EPS in patients treated with aripiprazole. Treatment-emergent EPS also was observed in the RCT of risperidone.²⁰ EPS-related adverse events were associated with higher doses of risperidone, although none of the akathisia/EPS measures were thought to be “clinically significant.”

EPS frequency was relatively low and similar to placebo in the 3-week quetiapine trial,²¹ and no changes in movement disorder scale scores were observed during the olanzapine or ziprasidone RCTs.^4,7

Recommendations. If your pediatric patient develops EPS, first try an antipsychotic dose reduction. Because anticholinergics can contribute to antipsychotic-induced weight gain, reserve them until after a dosage reduction has been unsuccessful.

Benztropine (0.25 to 0.5 mg given 2 to 3 times daily, not to exceed 3 mg/d) or diphenhydramine (25 to 50 mg given 3 to 4 times daily; maximum dosage 5 mg/kg/d) can be effective in treating EPS. Avoid anticholinergics in children with narrow-angle glaucoma or age <3.

Akathisia may be managed with propranolol (20 to 120 mg/d in divided doses). Multiple doses (typically 3 times daily) are needed to prevent interdose withdrawal symptoms. Use this beta blocker with caution in children with asthma because of the possibility of bronchospasm.

TD. Short-term trials and a meta-analysis of atypical antipsychotic trials (>11 months’ duration, subject age <18) suggest a low annual risk for TD (0.4%).²² Large, prospective, long-term trials of atypical antipsychotics are necessary to more accurately define the risk of TD in the pediatric population, however. Retrospective analyses of adolescents treated with antipsychotics suggest 3 TD risk factors:

• early age of antipsychotic use
• medication nonadherence
• concomitant use of antiparkinsonian agents.²³

Kumra et al²⁴ identified lower premorbid functioning and greater positive symptoms at baseline as factors associated with “withdrawal dyskinesia/tardive dyskinesia” in children and adolescents with early-onset psychotic-spectrum disorders treated with typical or atypical antipsychotics.

Recommendations. To minimize TD risk, use the lowest effective antipsychotic dose, monitor for abnormal involuntary movements with standardized assessments (such as the Abnormal Involuntary Movement Scale), review risks and benefits with parents and patients, and regularly evaluate the indication and need for antipsychotic therapy. It is reasonable to attempt to lower the antipsychotic dose after the patient has attained remission and been stable for 1 year.

Neuroleptic malignant syndrome (NMS). This complication of dopamine-blocking medications:

• is among the most serious adverse effects of antipsychotic treatment
• continues to be associated with a mortality rate of 10%.²⁵

Recommendation. At least 1 recent review of pediatric NMS cases suggests that essential features (hyperthermia and severe muscular rigidity) are retained in children.²⁶ Nonetheless, monitor for variant presentations; hyper thermia or muscle rigidity may be absent or develop slowly over several days in patients treated with atypical antipsychotics.²⁷

Weight gain and glucose metabolism. A major adverse effect of most atypical antipsychotics is increased appetite, weight gain, and possible obesity.²⁸ In children, “obesity” refers to a body mass index (BMI) >95th percentile for age and sex; “over-weight” refers to BMI between the 85th and 95th percentile. Mean weight gain in the 5 atypical antipsychotic pediatric bipolar trials ranged from 0 to 8 lbs across 3 to 4 weeks of treatment (Figure).^3-7

Recommendations. Emphasize diet and exercise, with restriction of high-carbohydrate food, “fast foods,” and soft drinks. Another option is a trial of metformin, which decreases hepatic glucose production, decreases intestinal absorption of glucose, and improves insulin sensitivity by increasing peripheral glucose uptake and utilization.

Klein et al²⁹ studied 39 patients age 10 to 17 with mood and psychotic disorders whose weight increased by >10% during <1 year of olanzapine, risperidone, or quetiapine therapy. In this 16-week, double-blind, placebo-controlled trial, weight was stabilized in subjects receiving metformin, whereas those receiving placebo continued to gain weight (0.31 kg [0.68 lb]/week).

The usual starting metformin dose is 500 mg bid with meals. Increase in increments of 500 mg weekly, up to a maximum of 2,000 mg/d in divided doses. Potential side effects include diarrhea, nausea/vomiting, flatulence, and headache.

Hyperlipidemia. Patients who gain weight with atypical antipsychotics also may develop hyperlipidemia. Fasting serum triglycerides >150 mg/dL (1.70 mmol/L) in obese children are considered to be elevated and an early sign of metabolic syndrome.³⁰ Fasting total cholesterol >200 mg/dL (5.18 mmol/L) or low-density lipoprotein cholesterol >130 mg/dL (3.38 mmol/L) is consistent with hyperlipidemia.

 

Recommendation. Monitor and treat hyperlipidemia, which increases the risk of atherosclerosis as obese children grow older.³¹

Prolactin. Elevated prolactin concentrations may have deleterious effects in the developing child or adolescent, including gynecomastia, oligomenorrhea, and amenorrhea.¹⁷ Long-term effects on growth and sexual maturation have not been fully evaluated.

The relative tendency of atypical antipsychotics to cause hyperprolactinemia is roughly: risperidone/paliperidone > olanzapine > ziprasidone > quetiapine > clozapine > aripiprazole.¹⁸ In the risperidone RCT, mean changes in baseline prolactin levels were 41 ng/mL for boys and 59 ng/mL in girls.³ Results of the olanzapine RCT suggest a high incidence of hyperprolactinemia (26% of girls, 63% of boys).⁴ Decreases in serum prolactin were observed in bipolar children and adolescents treated with aripiprazole for 30 weeks.¹⁹

Recommendations. For any pediatric patient treated with an atypical antipsychotic that increases prolactin levels:

• Obtain a baseline prolactin level.
• Repeat after 6 months of treatment or with the emergence of elevated prolactin symptoms, such as gynecomastia in boys. Ask about increases in breast size, galactorrhea, changes in menstruation, sexual functioning, and pubertal development.

Switch patients who develop any of these side effects to another atypical agent that does not increase serum prolactin.³²

QTc interval prolongation. All atypical antipsychotics can cause QTc prolongation. Several cases of significant QTc prolongation have been reported in children and adolescents treated with ziprasidone.^33,34 In the RCT of ziprasidone, QTc prolongation was not clinically significant in most of the patients in which it was reported, and it did not lead to adverse events.³⁴ Mean QTc change was 8.1 msec at study termination.⁷

Patients enrolled in clinical trails are screened very carefully, however, and those with preexisting medical abnormalities typically are excluded. Thus, these findings may have limited usefulness for “real-world” patients.

Recommendations. Until additional information is known about the cardiac effects of atypical antipsychotics in children and adolescents:

• Perform a careful history, review of symptoms, and physical exam looking for any history of palpitations, shortness of breath, or syncope.
• Query specifically about any family history of sudden cardiac death.
• Perform a baseline resting ECG for patients starting ziprasidone or clozapine, or for other atypicals if indicated by history, review of systems, physical exam, etc.
• For patients treated with ziprasidone or clozapine, repeat ECG as the dose increases or if the patient has cardiac symptoms (unexplained shortness of breath, palpitations, skipped beats, etc.).

Table 3

Talking to families about using antipsychotics
in children with bipolar disorder

Effectiveness. Large, placebo-controlled studies have shown that atypical antipsychotics can significantly reduce manic symptoms in children and adolescents with bipolar disorder
Safety data. Additional 6-month safety data indicate that atypical antipsychotics continue to be effective in children and adolescents, without dramatic changes in side effects
Precautions. Antipsychotics are powerful medications and must be used carefully in pediatric patients
Potential side effects. All antipsychotics have serious potential side effects that must be recognized, monitored, and managed
Potential benefits from using atypical antipsychotics include mood stabilization, treatment of psychotic symptoms, and lower risk of extrapyramidal symptoms compared with typical antipsychotics
Risk vs benefit. On balance, the potential benefit of these agents outweighs the potential risk for children and adolescents with bipolar disorder

Figure: Mean weight gain with atypical antipsychotics in pediatric bipolar trials

Weight gain in children and adolescents with bipolar disorder varied among atypical antipsychotics used in 5 recent randomized controlled trials. Treatment duration was 3 weeks with olanzapine, risperidone, and quetiapine and 4 weeks with aripiprazole and ziprasidone. Dosages were olanzapine, 10.4 ± 4.5 mg/d; risperidone, 0.5 to 2.5 mg/d or 3 to 6 mg/d; aripiprazole, 10 or 30 mg/d; quetiapine, 400 or 600 mg/d; and ziprasidone, 80 to 160 mg/d.
Source: References 3-7Related resources

• Child and Adolescent Bipolar Foundation. www.bpkids.org.
• University of Illinois at Chicago Pediatric Mood Disorders Clinic. www.psych.uic.edu/pmdc.
• Ryan Licht Sang Bipolar Foundation. www.ryanlichtsangbipolarfoundation.org.

Drug brand names

• Aripiprazole • Abilify
• Benztropine • Cogentin
• Carbamazepine • Carbatrol
• Clozapine • Clozaril
• Diphenhydramine • Benadryl
• Divalproex sodium • Depakote
• Lithium • Lithobid, others
• Metformin • Glucophage
• Olanzapine • Zyprexa
• Paliperidone • Invega
• Propranolol • Inderal
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Valproate • Depacon
• Ziprasidone • Geodon

Disclosures

Dr. Kowatch is a consultant to and speaker for AstraZeneca and a consultant to Forest Pharmaceuticals. He receives research support from the National Alliance for Research on Schizophrenia and Depression, the National Institute of Child Health and Human Development, the National Institute of Mental Health, and the Stanley Foundation.

Dr. Strawn has received research support from the American Academy of Child and Adolescent Psychiatry (Lilly Pilot Research Award).

 

Dr. Sorter receives research support from the National Institute of Mental Health and the Health Foundation of Greater Cincinnati.

Discuss this article

Five recent randomized controlled trials (RCTs) have demonstrated the efficacy of atypical antipsychotics for treating bipolar disorder in children and adolescents, but 4 of these 5 trials remain unpublished. The lag time between the completion of these trials and publication of their results—typically 4 to 5 years¹—leaves psychiatrists without important evidence to explain to families and critics² why they might recommend using these powerful medications in children with mental illness.

This article previews the preliminary results of these 5 RCTs of atypical antipsychotics, offers a treatment algorithm supported by this evidence, and discusses how to manage potentially serious risks when using antipsychotics to treat children and adolescents with bipolar disorder (BPD).

Where do atypical antipsychotics fit in?

Details of the 5 industry-sponsored RCTs of atypical antipsychotics in children and adolescents with bipolar I manic or mixed episodes are summarized in Table 1.^3-7 Only the olanzapine study⁴ has been published; data from the other 4 trials were presented at medical meetings in 2007 and 2008.

Change in Young Mania Rating Scale (YMRS) score was the primary outcome measure in these 5 trials, and each compound was more effective than placebo. The trials demonstrated statistically significant and clinically relevant differences between each antipsychotic and placebo. The number needed to treat (NNT)—how many patients need to be treated for 1 to benefit in a controlled clinical trial—ranged from 2 to 4. For comparison, the NNT for statins in the prevention of coronary events is 12 to 22,⁸ and the NNT in an analysis of trials of selective serotonin reuptake inhibitors for pediatric major depressive disorder was 9.⁹ Thus, an NNT of ≤4 represents a clinically significant effect.

Risperidone is FDA-approved for short-term treatment of acute bipolar I manic or mixed episodes in patients age 10 to 17. Aripiprazole is approved for acute and maintenance treatment of bipolar I manic or mixed episodes (with or without psychosis) as monotherapy or with lithium or valproate in patients age 10 to 17. In June, an FDA advisory committee recommended pediatric bipolar indications for olanzapine, quetiapine, and ziprasidone.

‘Mood stabilizers’ such as lithium, valproate, and carbamazepine have been used for years to treat bipolar mania in adults, adolescents, and children, despite limited supporting evidence. Preliminary results of a National Institute of Mental Health-funded double-blind RCT provide insights on their efficacy.¹⁰

The 153 outpatients age 7 to 17 in a bipolar I manic or mixed episode were randomly assigned to lithium, divalproex, or placebo for 8 weeks. Response rates—based on a Clinical Global Impressions-Improvement score of 1 or 2 (very much or much improved)—were divalproex, 54%; lithium, 42%; and placebo, 29%. Lithium showed a trend toward efficacy but did not clearly separate from placebo on the primary outcome measures. Effect sizes for lithium and divalproex were moderate.¹⁰

Only 1 study has compared a mood stabilizer with an atypical antipsychotic for treating mania in adolescents. In a double-blind trial, DelBello et al¹¹ randomly assigned 50 patients age 12 to 18 with a bipolar I manic or mixed episode to quetiapine, 400 to 600 mg/d, or divalproex, serum level 80 to 120 μg/mL, for 28 days. Manic symptoms resolved more rapidly, and remission rates measured by the YMRS were higher with quetiapine than with divalproex. Both medications were well tolerated.

Combination therapy. BPD as it presents in children and adolescents is often difficult to treat because of the disorder’s various phases (manic, depressed, mixed), frequent psychotic symptoms, and high rate of comorbidity. Pediatric BPD patients frequently require several psychotropics, including mood stabilizers and atypical antipsychotics.

In a double-blind, placebo-controlled study, 30 adolescents in a bipolar I manic or mixed episode initially received divalproex, 20 mg/kg/d, then were randomly assigned to 6 weeks of adjunctive quetiapine, titrated to 450 mg/d in 7 days (n=15), or placebo (n=15). Those receiving divalproex plus quetiapine showed a statistically significant greater reduction in manic symptoms (P=.03) and a higher response rate (87% vs 53%, P=.05), compared with those receiving divalproex and placebo. This suggests that a mood stabilizer plus an atypical antipsychotic is more effective than a mood stabilizer alone for adolescent mania. Quetiapine was well tolerated.¹²

Treatment. The American Psychiatric Association’s outdated 2002 practice guideline for acute bipolar I manic or mixed episodes in adults recommends lithium, valproate, and/or an antipsychotic.¹³ The more recent Texas Medication Algorithm Project (TMAP) guidelines recommend monotherapy with lithium, valproate, aripiprazole, quetiapine, risperidone, or ziprasidone for adults with euphoric or irritable manic or hypomanic symptoms.¹⁴

 

Based on the TMAP algorithm, recent clinical trial evidence, and our experience in treating pediatric BPD, we offer an approach for treating mania/hypomania in patients age 10 to 17 (see Proposed Algorithm). For dosing and precautions when using atypical antipsychotics in children and adolescents with BPD, see Table 2.^15-17

Comorbid psychiatric illnesses (such as anxiety disorders) are prevalent in adolescents with BPD. Evidence in adults and adolescents suggests that some atypical antipsychotics may provide additional benefit for these conditions as well. Thus, consider comorbid conditions and symptoms when choosing antimanic agents.

Attention-deficit/hyperactivity disorder (ADHD) is a common comorbidity in children with BPD, and stimulant medications are most often prescribed to treat inattentiveness and hyperactivity. Caution is imperative when treating bipolar youth with stimulants, which can exacerbate manic symptoms. Treat the patient’s mania before adding or reintroducing stimulant medication. Research and clinical experience suggest that if you first stabilize these patients on a mood stabilizer or atypical antipsychotic, adding a stimulant can be very helpful in treating comorbid ADHD symptoms. Start with low stimulant doses, and increase slowly.

Table 1

RCTs of atypical antipsychotics in patients age 10 to 17
with bipolar I disorder*

Antipsychotic and source	Bipolar I episode (# of subjects)	Trial duration (days)	Dosage (mg/d)	Response rate or YMRS score change	NNT	Mean weight gain (kg)
Risperidone Pandina et al3 AACAP 2007	Manic, mixed (169)	21	0.5 to 2.5 3 to 6	59% 63%	3.3 3.5	1.9 1.4
Olanzapine Tohen et al4	Manic, mixed (161)	21	10.4 ± 4.5	49%	4.1	3.7 ± 2.2
Quetiapine DelBello et al5 AACAP 2007	Manic (284)	21	400 600	64% 58%	4.4 4.2	1.7 1.7
Aripiprazole Wagner et al6 ACNP 2007	Manic, mixed (296)	28	10 30	45% 64%	4.1 2.4	0.9 0.54
Ziprasidone DelBello et al7 APA 2008	Manic, mixed (238)	28	80 to 160	–13.83 with ziprasidone, –8.61 with placebo	3.7	None
*Each trial included a 6-month open extension phase; results are pending
AACAP: American Academy of Child and Adolescent Psychiatry; ACNP: American College of Neuropsychopharmacology; APA: American Psychiatric Association; NNT: number needed to treat; RCT: randomized controlled trial; YMRS: Young Mania Rating Scale

Table 2

Recommended antipsychotic use in pediatric bipolar disorder

Drug	Starting dosage (mg)	Target dosage (mg/d)	Precautions
Aripiprazole	2.5 to 5 at bedtime	10 to 30	Monitor for CYP 3A4 and 2D6 interactions, weight, BMI, cholesterol, lipids, and glucose
Olanzapine	2.5 bid	10 to 20	Monitor for CYP 2D6 interactions, weight, BMI, cholesterol, lipids, glucose, and prolactin levels
Quetiapine	50 bid	400 to 1,200	Monitor for weight, BMI, cholesterol, lipids, and glucose
Risperidone	0.25 bid	1 to 2.5	Monitor for EPS, hyperprolactinemia (and associated sexual side effects, including galactorrhea), weight, BMI, cholesterol, lipids, glucose, and prolactin levels
Ziprasidone	20 bid	80 to 160	Check baseline ECG and as dose increases or with reason for high level of concern; monitor prolactin levels
BMI: body mass index; CYP: cytochrome P450; ECG: electrocardiography; EPS: extrapyramidal symptoms
Source: References 15-17

Proposed Algorithm: Treating a bipolar mixed/manic episode in patients age 10 to 17

Stage 1. Consider patient’s experience with antipsychotics, body weight, and family history when choosing first-line monotherapy (1A). Quetiapine poses low risk for extrapyramidal symptoms and tardive dyskinesia. Aripiprazole and ziprasidone pose relatively low risk of weight gain. Risperidone is potent at low doses but increases prolactin levels (long-term effect unknown).

Second-line choices (1B) are mood stabilizers lithium and valproate (because of lower potency than atypical antipsychotics), and olanzapine (which—although potent—causes substantial weight gain). In case of lack of response or intolerable side effects with initial agent, select an alternate from 1A or 1B. If this is not effective, move to Stage 2.

Stage 2. Consider augmentation for patients who show partial response to monotherapy (in your clinical judgment “mild to moderately improved” but not “much or very much improved”).

Stage 3. Combination therapy could include 2 mood stabilizers (such as lithium and valproate) plus an atypical antipsychotic; 2 atypical antipsychotics; or other combinations based on patient’s past responses. No research has shown these combinations to be efficacious in bipolar children and adolescents, but we find they sometimes help those with treatment-resistant symptoms.

Duration. Maintain psychotropics 12 to 18 months. When patient is euthymic, slowly taper 1 medication across several months. If symptoms recur, reintroduce the mood-stabilizing agent(s).

Source: Adapted and reprinted with permission from Kowatch RA, Fristad MA, Findling R, et al. Clinical manual for the management of bipolar disorder in children and adolescents. Arlington, VA: American Psychiatric Publishing, Inc.; 2008

Managing adverse effects

Although clinically effective, atypical antipsychotics may cause serious side effects that must be recognized and managed. These include extrapyramidal symptoms (EPS), tardive dyskinesia (TD), weight gain and obesity, hyperlipidemia, increased prolactin levels, and QTc changes. Counsel patients and families about the risks and benefits of antipsychotics when you consider them for children and adolescents with BPD (Table 3).

EPS. Drug-induced parkinsonism and akathisia are the most common EPS in children and adolescents with BPD treated with atypical antipsychotics.¹⁸

 

Correll et al¹⁹ reported a 10% rate of EPS in patients treated with aripiprazole. Treatment-emergent EPS also was observed in the RCT of risperidone.²⁰ EPS-related adverse events were associated with higher doses of risperidone, although none of the akathisia/EPS measures were thought to be “clinically significant.”

EPS frequency was relatively low and similar to placebo in the 3-week quetiapine trial,²¹ and no changes in movement disorder scale scores were observed during the olanzapine or ziprasidone RCTs.^4,7

Recommendations. If your pediatric patient develops EPS, first try an antipsychotic dose reduction. Because anticholinergics can contribute to antipsychotic-induced weight gain, reserve them until after a dosage reduction has been unsuccessful.

Benztropine (0.25 to 0.5 mg given 2 to 3 times daily, not to exceed 3 mg/d) or diphenhydramine (25 to 50 mg given 3 to 4 times daily; maximum dosage 5 mg/kg/d) can be effective in treating EPS. Avoid anticholinergics in children with narrow-angle glaucoma or age <3.

Akathisia may be managed with propranolol (20 to 120 mg/d in divided doses). Multiple doses (typically 3 times daily) are needed to prevent interdose withdrawal symptoms. Use this beta blocker with caution in children with asthma because of the possibility of bronchospasm.

TD. Short-term trials and a meta-analysis of atypical antipsychotic trials (>11 months’ duration, subject age <18) suggest a low annual risk for TD (0.4%).²² Large, prospective, long-term trials of atypical antipsychotics are necessary to more accurately define the risk of TD in the pediatric population, however. Retrospective analyses of adolescents treated with antipsychotics suggest 3 TD risk factors:

• early age of antipsychotic use
• medication nonadherence
• concomitant use of antiparkinsonian agents.²³

Kumra et al²⁴ identified lower premorbid functioning and greater positive symptoms at baseline as factors associated with “withdrawal dyskinesia/tardive dyskinesia” in children and adolescents with early-onset psychotic-spectrum disorders treated with typical or atypical antipsychotics.

Recommendations. To minimize TD risk, use the lowest effective antipsychotic dose, monitor for abnormal involuntary movements with standardized assessments (such as the Abnormal Involuntary Movement Scale), review risks and benefits with parents and patients, and regularly evaluate the indication and need for antipsychotic therapy. It is reasonable to attempt to lower the antipsychotic dose after the patient has attained remission and been stable for 1 year.

Neuroleptic malignant syndrome (NMS). This complication of dopamine-blocking medications:

• is among the most serious adverse effects of antipsychotic treatment
• continues to be associated with a mortality rate of 10%.²⁵

Recommendation. At least 1 recent review of pediatric NMS cases suggests that essential features (hyperthermia and severe muscular rigidity) are retained in children.²⁶ Nonetheless, monitor for variant presentations; hyper thermia or muscle rigidity may be absent or develop slowly over several days in patients treated with atypical antipsychotics.²⁷

Weight gain and glucose metabolism. A major adverse effect of most atypical antipsychotics is increased appetite, weight gain, and possible obesity.²⁸ In children, “obesity” refers to a body mass index (BMI) >95th percentile for age and sex; “over-weight” refers to BMI between the 85th and 95th percentile. Mean weight gain in the 5 atypical antipsychotic pediatric bipolar trials ranged from 0 to 8 lbs across 3 to 4 weeks of treatment (Figure).^3-7

Recommendations. Emphasize diet and exercise, with restriction of high-carbohydrate food, “fast foods,” and soft drinks. Another option is a trial of metformin, which decreases hepatic glucose production, decreases intestinal absorption of glucose, and improves insulin sensitivity by increasing peripheral glucose uptake and utilization.

Klein et al²⁹ studied 39 patients age 10 to 17 with mood and psychotic disorders whose weight increased by >10% during <1 year of olanzapine, risperidone, or quetiapine therapy. In this 16-week, double-blind, placebo-controlled trial, weight was stabilized in subjects receiving metformin, whereas those receiving placebo continued to gain weight (0.31 kg [0.68 lb]/week).

The usual starting metformin dose is 500 mg bid with meals. Increase in increments of 500 mg weekly, up to a maximum of 2,000 mg/d in divided doses. Potential side effects include diarrhea, nausea/vomiting, flatulence, and headache.

Hyperlipidemia. Patients who gain weight with atypical antipsychotics also may develop hyperlipidemia. Fasting serum triglycerides >150 mg/dL (1.70 mmol/L) in obese children are considered to be elevated and an early sign of metabolic syndrome.³⁰ Fasting total cholesterol >200 mg/dL (5.18 mmol/L) or low-density lipoprotein cholesterol >130 mg/dL (3.38 mmol/L) is consistent with hyperlipidemia.

 

Recommendation. Monitor and treat hyperlipidemia, which increases the risk of atherosclerosis as obese children grow older.³¹

Prolactin. Elevated prolactin concentrations may have deleterious effects in the developing child or adolescent, including gynecomastia, oligomenorrhea, and amenorrhea.¹⁷ Long-term effects on growth and sexual maturation have not been fully evaluated.

The relative tendency of atypical antipsychotics to cause hyperprolactinemia is roughly: risperidone/paliperidone > olanzapine > ziprasidone > quetiapine > clozapine > aripiprazole.¹⁸ In the risperidone RCT, mean changes in baseline prolactin levels were 41 ng/mL for boys and 59 ng/mL in girls.³ Results of the olanzapine RCT suggest a high incidence of hyperprolactinemia (26% of girls, 63% of boys).⁴ Decreases in serum prolactin were observed in bipolar children and adolescents treated with aripiprazole for 30 weeks.¹⁹

Recommendations. For any pediatric patient treated with an atypical antipsychotic that increases prolactin levels:

• Obtain a baseline prolactin level.
• Repeat after 6 months of treatment or with the emergence of elevated prolactin symptoms, such as gynecomastia in boys. Ask about increases in breast size, galactorrhea, changes in menstruation, sexual functioning, and pubertal development.

Switch patients who develop any of these side effects to another atypical agent that does not increase serum prolactin.³²

QTc interval prolongation. All atypical antipsychotics can cause QTc prolongation. Several cases of significant QTc prolongation have been reported in children and adolescents treated with ziprasidone.^33,34 In the RCT of ziprasidone, QTc prolongation was not clinically significant in most of the patients in which it was reported, and it did not lead to adverse events.³⁴ Mean QTc change was 8.1 msec at study termination.⁷

Patients enrolled in clinical trails are screened very carefully, however, and those with preexisting medical abnormalities typically are excluded. Thus, these findings may have limited usefulness for “real-world” patients.

Recommendations. Until additional information is known about the cardiac effects of atypical antipsychotics in children and adolescents:

• Perform a careful history, review of symptoms, and physical exam looking for any history of palpitations, shortness of breath, or syncope.
• Query specifically about any family history of sudden cardiac death.
• Perform a baseline resting ECG for patients starting ziprasidone or clozapine, or for other atypicals if indicated by history, review of systems, physical exam, etc.
• For patients treated with ziprasidone or clozapine, repeat ECG as the dose increases or if the patient has cardiac symptoms (unexplained shortness of breath, palpitations, skipped beats, etc.).

Table 3

Talking to families about using antipsychotics
in children with bipolar disorder

Effectiveness. Large, placebo-controlled studies have shown that atypical antipsychotics can significantly reduce manic symptoms in children and adolescents with bipolar disorder
Safety data. Additional 6-month safety data indicate that atypical antipsychotics continue to be effective in children and adolescents, without dramatic changes in side effects
Precautions. Antipsychotics are powerful medications and must be used carefully in pediatric patients
Potential side effects. All antipsychotics have serious potential side effects that must be recognized, monitored, and managed
Potential benefits from using atypical antipsychotics include mood stabilization, treatment of psychotic symptoms, and lower risk of extrapyramidal symptoms compared with typical antipsychotics
Risk vs benefit. On balance, the potential benefit of these agents outweighs the potential risk for children and adolescents with bipolar disorder

Figure: Mean weight gain with atypical antipsychotics in pediatric bipolar trials

Weight gain in children and adolescents with bipolar disorder varied among atypical antipsychotics used in 5 recent randomized controlled trials. Treatment duration was 3 weeks with olanzapine, risperidone, and quetiapine and 4 weeks with aripiprazole and ziprasidone. Dosages were olanzapine, 10.4 ± 4.5 mg/d; risperidone, 0.5 to 2.5 mg/d or 3 to 6 mg/d; aripiprazole, 10 or 30 mg/d; quetiapine, 400 or 600 mg/d; and ziprasidone, 80 to 160 mg/d.
Source: References 3-7Related resources

• Child and Adolescent Bipolar Foundation. www.bpkids.org.
• University of Illinois at Chicago Pediatric Mood Disorders Clinic. www.psych.uic.edu/pmdc.
• Ryan Licht Sang Bipolar Foundation. www.ryanlichtsangbipolarfoundation.org.

Drug brand names

• Aripiprazole • Abilify
• Benztropine • Cogentin
• Carbamazepine • Carbatrol
• Clozapine • Clozaril
• Diphenhydramine • Benadryl
• Divalproex sodium • Depakote
• Lithium • Lithobid, others
• Metformin • Glucophage
• Olanzapine • Zyprexa
• Paliperidone • Invega
• Propranolol • Inderal
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Valproate • Depacon
• Ziprasidone • Geodon

Disclosures

Dr. Kowatch is a consultant to and speaker for AstraZeneca and a consultant to Forest Pharmaceuticals. He receives research support from the National Alliance for Research on Schizophrenia and Depression, the National Institute of Child Health and Human Development, the National Institute of Mental Health, and the Stanley Foundation.

Dr. Strawn has received research support from the American Academy of Child and Adolescent Psychiatry (Lilly Pilot Research Award).

 

Dr. Sorter receives research support from the National Institute of Mental Health and the Health Foundation of Greater Cincinnati.

Comment on this article

Americans with depression turn to complementary and alternative medicine (CAM) more often than conventional psychotherapy or FDA-approved medication. In a nationally representative sample, 54% of respondents with self-reported “severe depression”—including two-thirds of those receiving conventional therapies—reported using CAM during the previous 12 months.¹

Unfortunately, popular acceptance of CAM for depression is disproportionate to the evidence base, which—although growing—remains limited. As a result, your patients may be self-medicating with poorly supported treatments that are unlikely to help them recover from depression.

In reviewing CAM treatments for depression, we found some with enough evidence of positive effect that we feel comfortable recommending them as evidence-based options. These promising, short-term treatments are supported by level 1a or 1b evidence and at least 1 study that demonstrates an ability to induce remission ( Table 1 ).²

For patients seeking “natural” or nonprescription treatments—or when you wish to augment standard treatments that are not working adequately—you might recommend fatty acids, St. John’s wort, or S-adenosyl-L-methionine (SAMe). Similar recommendations can be made for yoga, exercise, and bibliotherapy, as we discuss here.

Table 1

Evidence these authors required to recommend a CAM treatment

Minimum requirements	Level of evidence	Recommendation
Systematic review showing superiority to placebo or standard treatment Plus 1 study showing CAM treatment can induce remission	1a + 1b or 2b	A
At least 2 RCTs showing superiority to placebo or standard treatment Plus 1 study showing CAM treatment can induce remission	1b 1b or 2b	A–
CAM: complementary and alternative medicine; RCT: randomized controlled trial
Source: Reference 2

Reviewing CAM evidence

This article refers to as “alternative” any treatment other than a form of psychotherapy or an FDA-approved medication that substitutes for a standard psychiatric treatment. When used to augment standard psychiatric treatments, these approaches are considered “complementary.”

Our search for evidence on CAM treatments for depressive disorders raised questions about what constitutes acceptable and convincing methodology:

• Studies often had problems with blinding and suitable placebos. Many were small, with short duration and no long-term follow-up.
  
• Comparisons with active treatments that showed no differences were assumed to imply comparability, even though the studies were powered to detect only large differences.
  

On the other hand, multiple randomized controlled trials (RCTs) have shown consistent superiority of some CAM treatments over comparison conditions.

Applying the evidence. Because CAM use is widespread, be sure to ask psychiatric patients if they are using CAM treatments. If the answer is “yes,” a risk-benefit assessment is needed. Inform patients who are using poorly supported CAM approaches that they could consider better-supported CAM options as well as standard effective antidepressants.

Monitor patients for an adequately prompt positive response to an evidence-based CAM approach that has shown efficacy for their level of depression. As with any treatment, consider other evidence-based options when CAM treatments are inadequate or unsuccessful in achieving remission of depressive symptoms.

Sufficient evidence of efficacy

Yoga. In their systematic review of yoga’s effectiveness for depression, Pilkington et al³ analyzed 5 RCTs that met 3 criteria:

• participants had mild to severe depression or depressive disorders
  
• yoga or yoga-based exercises alone were used for treatment
  
• depression rating scales were used as outcome measures.
  

They found evidence that yoga can reduce depressive symptoms and induce remission ( Table 2 ). The studies were generally small and of short duration, and depression severity and interventions varied widely. Most participants were young and relatively fit, raising questions about yoga’s applicability to older or less fit patients. Reporting of adverse events was limited, but no safety issues or adverse effects were identified.

Conclusion. Yoga has been studied primarily as an alternative treatment, but its physical health and group participation benefits and lack of side effects make it a suitable complementary treatment as well.

Exercise. Extensive literature has examined the relationship between exercise and depression. We identified 7 reviews published between 1993 and 2008 ( Table 3 ). All supported positive effects of exercise except for patients age

• 45% with supervised exercise
  
• 40% with home-based exercise
  
• 47% with sertraline, 50 to 200 mg/d
  
• 31% with placebo.⁴
  

Table 2

5 RCTs of yoga’s effectiveness in treating depression

RCT	Interventions	Conclusion
Broota and Dhir, 1990	Yoga and PMR vs control	Yoga and PMR were more effective than control, with yoga more effective than PMR
Khumar et al, 1993	Shavasana yoga vs no intervention	College students with severe depression improved during and after yoga treatment
Janakiramaiah et al, 2000	SKY vs ECT vs imipramine	Reductions in BDI scores for all 3 groups; ECT > SKY or imipramine, SKY=imipramine
Rohini et al, 2000	Full SKY vs partial SKY	30 individuals with MDD improved with either therapy, but results were not statistically significant
Woolery, 2004	Iyengar yoga vs wait list	28 mildly depressed individuals benefitted from yoga, as measured by BDI scores at midpoint and throughout treatment
BDI: Beck Depression Inventory; ECT: electroconvulsive therapy; MDD: major depressive disorder; PMR: progressive muscle relaxation; RCT: randomized controlled trial; SKY: Sudarshan Kriya yoga
Source: Broota A, Dhir R. Efficacy of two relaxation techniques in depression. Journal of Personality and Clinical Studies. 1990;6(1):83-90. Khumar SS, Kaur P, Kaur S. Effectiveness of Shavasana on depression among university students. Indian J Clin Psychol. 1993;20(2):82-87. Janakiramaiah N, Gangadhar BN, Naga Venkatesha Murthy PJ, et al. Antidepressant efficacy of Sudarshan Kriya yoga (SKY) in melancholia: a randomized comparison with electroconvulsive therapy (ECT) and imipramine. J Affect Disord. 2000;57(1-3):255-259. Rohini V, Pandey RS, Janakiramaiah N, et al. A comparative study of full and partial Sudarshan Kriya yoga (SKY) in major depressive disorder. NIMHANS Journal. 2000;18(1):53-57. Woolery A, Myers H, Sternlieb B, et al. A yoga intervention for young adults with elevated symptoms of depression. Altern Ther Health Med. 2004;10(2):60-63.

 

Table 3

Evidence of the antidepressant effect of exercise

Literature review	Methodology	Conclusion
Byrne and Byrne, 1993	13 studies, clinical samples, measured changes in depressed mood	90% of studies reported beneficial effects, especially in clinical populations
Scully et al, 1998	4 literature reviews, 1 monograph, 1 study	Positive relationship of physical activity and depression in healthy and clinical samples, increased over time
Lawlor and Hopker, 2001	14 RCTs from 1966 to 1999 with depression as an outcome	Significant methodologic weaknesses, but exercise effect > no treatment and=cognitive therapy
Dunn et al, 2001	Examined dose effect in 37 studies; subjects diagnosed with depressive disorders, mild-to-moderate symptoms, and no medical comorbidity	Only level B and C evidence; positive effects with exercise from light to heavy intensity; aerobic=nonaerobic; improvement may or may not be related to improved fitness
Brosse et al, 2002	12 RCTs from 1979 to 1999	Significant methodologic limitations, but authors concluded evidence supports a positive effect of exercise in healthy and clinical populations
Larun et al, 2006	4 RCTs in children and youth age	Exercise effect same as no intervention, low-intensity relaxation, or psychosocial intervention
Barbour et al, 2007	2 meta-analyses, 1 RCT, 2 studies	Positive effect; high-energy was optimal dose; aerobic=nonaerobic; improvement may or may not be related to improved fitness
RCT: randomized controlled trial
Source: Byrne AE, Byrne DG. The effect of exercise on depression, anxiety and other mood states: A review. J Psychosom Res. 1993;37(6):565-574. Scully D, Kremer J, Meade MM, et al. Physical exercise and psychological well being: a critical review. Br J Sports Med. 1998;32(2):111-120. Lawlor DA, Hopker SW. The effectiveness of exercise as an intervention in the management of depression: systematic review and meta-regression analysis of randomised controlled trials. BMJ. 2001;322(7289):763-767. Dunn AL, Trivedi MH, O’Neal HA. Physical activity dose-response effects on outcomes of depression and anxiety. Med Sci Sports Exerc. 2001;33(6):S587. Brosse AL, Sheets ES, Lett HS, et al. Exercise and the treatment of clinical depression in adults: recent findings and future directions. Sports Med. 2002;32(12):741-760. Larun L, Nordheim LV, Ekeland E, et al. Exercise in prevention and treatment of anxiety and depression among children and young people. Cochrane Database Syst Rev. 2006;3:CD004691. Barbour KA, Edenfield TM, Blumenthal JA. Exercise as a treatment for depression and other psychiatric disorders: a review. J Cardiopulm Rehabil Prev. 2007;27(6):359-367.

Conclusion. Evidence supports exercise for short-term treatment of mild or moderate depression in adults. Studies tend to be small and brief, to enroll young physically -sound patients, and to include little follow-up. Studies of subjects age

At least 2 studies suggest that high-energy exercise and aerobic or resistance training provide greater reductions in depressive symptoms than exercises such as walking.^5,6 Yoga’s positive effects suggest, however, that an aerobic effect is not necessary for an antidepressant benefit.

Exercise has not been adequately tested as a complementary treatment but likely is safe for most psychiatric patients. Perspiration and dehydration might alter therapeutic blood levels of lithium or other medications. Advise patients to drink water before, during, and after exercise and to avoid outdoor exercise in extreme temperatures. More vigorous monitoring might be indicated in specific cases.

Tailor exercise programs to individual needs, considering the patient’s age and health status. Refer a patient with a known heart problem or increased cardiovascular risk to his or her physician for selective exercise testing.

Bibliotherapy—reading self-help books, usually about cognitive-behavioral approaches to depressive disorders—has been relatively well studied. A recent meta-analysis examined 29 studies with pre-post designs. Group differences in the 17 controlled studies yielded a large effect size of 0.77. Participants who read the materials benefitted similarly whether they met in groups or applied the information on their own. Older adults tended to be less depressed at baseline and made smaller treatment gains.⁷

A study of 31 patients age >60 with mild-to-moderate depression⁸ compared 16 sessions of professionally administered cognitive-behavioral therapy (CBT) with self-administered cognitive therapy learned from reading a book.⁹ Both groups showed greater improvement in depressive symptoms compared with a control group. Subjects in the CBT group did somewhat better during the trial, but at 3-month follow-up most patients in both treatment groups no longer met diagnostic criteria for MDD.

Conclusion. Evidence supports bibliotherapy as an effective treatment for mild-to-moderate depression. No convincing data support its use as a complementary treatment, but it poses virtually no risk.

St. John’s wort (Hypericum perforatum) has been extensively studied for depressive disorders, with 29 RCTs in a meta-analysis of MDD trials through July 2008.¹⁰ Another meta-analysis compared St. John’s wort with selective serotonin reuptake inhibitors (SSRIs) in 13 studies through June 2008.¹¹ These and most RCTs have found St. John’s wort significantly more effective than placebo in reducing depressive symptoms.

 

Data selected from double-blind RCTs totaling 217 patients with mild depression [Hamilton Depression Rating Scale (HDRS) scores 12 Studies routinely show that treating MDD with St. John’s wort is comparable to using tricyclic or SSRI antidepressants.

Side effects with St. John’s wort generally are no different than with placebo and significantly less than with comparison treatments. Even so, using St. John’s wort instead of SSRIs for MDD remains controversial.

Studies vs SSRIs. Many of the favorable St. John’s wort trials were conducted in Europe, particularly in Germany. Two large RCTs conducted in the United States reported that the St. John’s wort standardized extract LI-160 was not more effective than placebo, but neither could be clearly interpreted as negative for St. John’s wort:

• In an 8-week trial, St. John’s wort and placebo groups improved significantly but at unusually low rates. The remission rate with St. John’s wort was small but significantly higher than with placebo.¹³
  
• A study sponsored by the National Institute of Mental Health compared St. John’s wort, 900 to 1,500 mg/d; sertraline, 50 to 100 mg/d; and placebo in 340 adults with MDD. No positive effects were found for St. John’s wort or sertraline.¹⁴
  

Side effects. St. John’s wort can affect blood levels of circulating medications metabolized by the cytochrome P450 liver enzyme system, including tricyclic antidepressants. Case studies have reported pregnancy from oral contraceptive failure, skin rashes, headache, and mania with St. John’s wort use. Although these reports are disturbing, St. John’s wort’s side effects when compared with SSRIs have been less frequent (40% vs 49%) and milder (clinical trial dropout rate 2% vs 7%).¹¹

Conclusion. Standardized extracts of St. John’s wort—particularly WS5570, 300 mg tid, and ZE117, 250 mg bid—appear to be effective treatments, especially for mild-to-moderate MDD. Because St. John’s wort is available without prescription and can interact with SSRIs or other antidepressants:

• care is required for its complementary use
  
• it is important to ask if patients are using St. John’s wort on their own.
  

St. John’s wort is used as a first-line depression treatment in Europe, but U.S. physicians may be less familiar with its potential interactions with other medications. We recommend that you consider St. John’s wort:

• for first-line use only when you can adequately gauge its effects on your patient’s other medications
  
• especially for depressed patients who cannot tolerate SSRIs.
  

SAMe is a metabolite involved in bio-synthesis of norepinephrine, serotonin, and dopamine.¹⁵ SAMe salts (such as 1,4-butanedisulfonate) are used as an over-the-counter supplement for depression treatment. Dozens of RCTs show SAMe has greater efficacy than placebo and positive effects comparable to those of standard antidepressants. In a meta-analysis of 28 RCTs by the Agency for Healthcare Quality and Research, SAMe produced significantly greater symptom improvement compared with placebo.¹⁶

SAMe has become a popular alternative treatment for depression since its introduction to the United States in the late 1990s, but it has been studied in only 2 U.S. open trials. One showed SAMe to be very effective in reducing depressive symptoms in patients with HIV/AIDS.¹⁷ The other found a 50% response rate and 43% remission rate with adjunctive SAMe, 800 to 1,600 mg/d for 6 weeks, in 30 adults with MDD who failed to respond adequately to SSRIs or the serotonin-norepinephrine reuptake inhibitor (SNRI) venlafaxine. The most common side effects were gastrointestinal (GI) symptoms and headaches.¹⁸ This open trial led to an ongoing National Institutes of Health-sponsored RCT on the safety and efficacy of SAMe for patients with treatment-resistant depression.

Conclusion. SAMe appears to have a faster onset of antidepressant effect than standard SSRIs or SNRIs and a favorable side-effect profile, which make the lack of rigorous trials in the United States striking. We recommend that you consider SAMe:

• as an adjunct in patients with incomplete response to standard treatments
  
• as a complementary treatment to speed onset of antidepressant effects.
  

Polyunsaturated fatty acids (PUFAs), usually from fish oils, have long been popular nutritional supplements because of their beneficial effects on cholesterol and cardiovascular health. Omega-3 and omega-6 fatty acids are the most common PUFAs. The omega-3 PUFAs include eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA).

Four meta-analyses independently looked at largely the same dozen RCTs through 2006 and found that 1 to 2 grams daily of omega-3 PUFAs was significantly more effective at reducing depressive symptoms than placebo.^19-22 Other data suggest that omega-3 PUFAs can induce depression remission in depressed Parkinson’s disease patients²³ and depressed pregnant women.²⁴ Since 2006, however, findings have been inconsistent. Several trials have found PUFAs no more effective than placebo.^25-27

 

An 8-week double-blind study compared EPA, 1 gram daily; fluoxetine, 20 mg/d; or both agents in 60 outpatients with MDD. Response rates—as measured by ≥50% reduction in baseline HDRS scores—were 50% with fluoxetine, 56% with EPA, and 81% with combination therapy.²⁸

Conclusion. Questions remain about dosing, ratio of EPA to DHA, patient selection, and baseline blood levels of PUFAs compared with response. PUFAs have a benign side-effect profile, with occasional reports of diarrhea or GI upset. Although their therapeutic effects are being clarified, PUFAs appear safe to recommend as an adjunct treatment if standard care is not satisfactory.

Insufficient evidence

L-tryptophan. It seems reasonable to expect a serotonin precursor to increase serotonin in the CNS and improve depressive symptoms. Of 111 trials on L-tryptophan for depression, however, only 2 met the quality criteria for inclusion in a recent meta-analysis.²⁹ Combining the 2 trials showed L-tryptophan alone and in combination with a tricyclic antidepressant was more effective than placebo for treating depressive disorders in adults.

Conclusion. Very little research continues to test L-tryptophan as a viable CAM for depressive disorder. Its serious side effect of eosinophilia-myalgia syndrome makes clinical use of this agent unlikely.

Acupuncture. Numerous small studies with questionable controls, different needling placements, and poor allocation concealment and blinding limit the ability to draw conclusions about acupuncture for treating depression ( Table 4 ). A recent meta-analysis by Wang et al³⁰ added 2 Chinese trials not included in an earlier review³¹ and found acupuncture significantly reduced depressive symptoms. No consistent differences were detected in response or remission rates, however.

Conclusion. Evidence is methodologically weak, and the use of acupuncture as an alternative or complementary treatment of depression is questionable.

Table 4

Acupuncture: Insufficient evidence of antidepressant effect

Literature review	Methodology	Conclusion
Mukaino et al, 2005	Systematic review of 7 RCTs including 509 patients; compared either manual or electroacupuncture with any control procedure	Inconsistent evidence of manual acupuncture’s effectiveness vs sham; electroacupuncture’s effect may be similar to that of antidepressant medication and merits further study
Leo and Ligot, 2007	Systematic review of 9 RCTs, 5 considered low quality; some focused on very specific populations (ie, hospitalized stroke patients or pregnant depressed patients)	Evidence inconclusive because of study designs and methodologies
Smith and Hay, 2005	Meta-analysis of 7 trials including 517 adults with mild-to-moderate depression; 5 trials (409 participants) compared acupuncture with medication; 2 trials compared acupuncture with wait list or sham acupuncture	No difference between acupuncture and medication; study quality too poor to support acupuncture’s efficacy
Wang et al, 2008	Meta-analysis of 8 small RCTs totalling 477 subjects (256 received active acupuncture, remainder received sham acupuncture); sham acupuncture design, number of acupuncture sessions, and duration varied among studies	Significant reduction in HRSD or BDI scores for acupuncture vs sham, but no significant effect of acupuncture on response or remission rates
BDI: Beck Depression Inventory; HRSD: Hamilton Rating Scale for Depression; RCT: randomized controlled trial
Source: Mukaino Y, Park J, White A, et al. The effectiveness of acupuncture for depression—a systematic review of randomised controlled trials. Acupunct Med. 2005;23(2):70-76. Leo RJ, Ligot JS Jr. A systematic review of randomized controlled trials of acupuncture in the treatment of depression. J Affect Disord. 2007;97(1-3):13-22. Smith CA, Hay PPJ. Acupuncture for depression. Cochrane Database Syst Rev. 2005;(2):CD004046. Wang H, Qi H, Wang BS, et al. Is acupuncture beneficial in depression? A meta-analysis of 8 randomized controlled trials. J Affect Disord. 2008;111(2-3):125-134.

Related resources

• National Center for Complementary and Alternative Medicine, National Institutes of Health. http://nccam.nih.gov.
  
• Journal of Alternative and Complementary Medicine. www.liebertpub.com/products/product.aspx?pid=26.
  
• Complementary and alternative medicine. www.nlm.nih.gov/medlineplus/complementaryandalternativemedicine.html.
  

Drug brand names

• Fluoxetine • Prozac
  
• Imipramine • Tofranil
  
• Lithium • Eskalith, Lithobid
  
• Sertraline • Zoloft
  
• Venlafaxine • Effexor
  

Disclosure

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Comment on this article

Americans with depression turn to complementary and alternative medicine (CAM) more often than conventional psychotherapy or FDA-approved medication. In a nationally representative sample, 54% of respondents with self-reported “severe depression”—including two-thirds of those receiving conventional therapies—reported using CAM during the previous 12 months.¹

Unfortunately, popular acceptance of CAM for depression is disproportionate to the evidence base, which—although growing—remains limited. As a result, your patients may be self-medicating with poorly supported treatments that are unlikely to help them recover from depression.

In reviewing CAM treatments for depression, we found some with enough evidence of positive effect that we feel comfortable recommending them as evidence-based options. These promising, short-term treatments are supported by level 1a or 1b evidence and at least 1 study that demonstrates an ability to induce remission ( Table 1 ).²

For patients seeking “natural” or nonprescription treatments—or when you wish to augment standard treatments that are not working adequately—you might recommend fatty acids, St. John’s wort, or S-adenosyl-L-methionine (SAMe). Similar recommendations can be made for yoga, exercise, and bibliotherapy, as we discuss here.

Table 1

Evidence these authors required to recommend a CAM treatment

Minimum requirements	Level of evidence	Recommendation
Systematic review showing superiority to placebo or standard treatment Plus 1 study showing CAM treatment can induce remission	1a + 1b or 2b	A
At least 2 RCTs showing superiority to placebo or standard treatment Plus 1 study showing CAM treatment can induce remission	1b 1b or 2b	A–
CAM: complementary and alternative medicine; RCT: randomized controlled trial
Source: Reference 2

Reviewing CAM evidence

This article refers to as “alternative” any treatment other than a form of psychotherapy or an FDA-approved medication that substitutes for a standard psychiatric treatment. When used to augment standard psychiatric treatments, these approaches are considered “complementary.”

Our search for evidence on CAM treatments for depressive disorders raised questions about what constitutes acceptable and convincing methodology:

• Studies often had problems with blinding and suitable placebos. Many were small, with short duration and no long-term follow-up.
  
• Comparisons with active treatments that showed no differences were assumed to imply comparability, even though the studies were powered to detect only large differences.
  

On the other hand, multiple randomized controlled trials (RCTs) have shown consistent superiority of some CAM treatments over comparison conditions.

Applying the evidence. Because CAM use is widespread, be sure to ask psychiatric patients if they are using CAM treatments. If the answer is “yes,” a risk-benefit assessment is needed. Inform patients who are using poorly supported CAM approaches that they could consider better-supported CAM options as well as standard effective antidepressants.

Monitor patients for an adequately prompt positive response to an evidence-based CAM approach that has shown efficacy for their level of depression. As with any treatment, consider other evidence-based options when CAM treatments are inadequate or unsuccessful in achieving remission of depressive symptoms.

Sufficient evidence of efficacy

Yoga. In their systematic review of yoga’s effectiveness for depression, Pilkington et al³ analyzed 5 RCTs that met 3 criteria:

• participants had mild to severe depression or depressive disorders
  
• yoga or yoga-based exercises alone were used for treatment
  
• depression rating scales were used as outcome measures.
  

They found evidence that yoga can reduce depressive symptoms and induce remission ( Table 2 ). The studies were generally small and of short duration, and depression severity and interventions varied widely. Most participants were young and relatively fit, raising questions about yoga’s applicability to older or less fit patients. Reporting of adverse events was limited, but no safety issues or adverse effects were identified.

Conclusion. Yoga has been studied primarily as an alternative treatment, but its physical health and group participation benefits and lack of side effects make it a suitable complementary treatment as well.

Exercise. Extensive literature has examined the relationship between exercise and depression. We identified 7 reviews published between 1993 and 2008 ( Table 3 ). All supported positive effects of exercise except for patients age

• 45% with supervised exercise
  
• 40% with home-based exercise
  
• 47% with sertraline, 50 to 200 mg/d
  
• 31% with placebo.⁴
  

Table 2

5 RCTs of yoga’s effectiveness in treating depression

RCT	Interventions	Conclusion
Broota and Dhir, 1990	Yoga and PMR vs control	Yoga and PMR were more effective than control, with yoga more effective than PMR
Khumar et al, 1993	Shavasana yoga vs no intervention	College students with severe depression improved during and after yoga treatment
Janakiramaiah et al, 2000	SKY vs ECT vs imipramine	Reductions in BDI scores for all 3 groups; ECT > SKY or imipramine, SKY=imipramine
Rohini et al, 2000	Full SKY vs partial SKY	30 individuals with MDD improved with either therapy, but results were not statistically significant
Woolery, 2004	Iyengar yoga vs wait list	28 mildly depressed individuals benefitted from yoga, as measured by BDI scores at midpoint and throughout treatment
BDI: Beck Depression Inventory; ECT: electroconvulsive therapy; MDD: major depressive disorder; PMR: progressive muscle relaxation; RCT: randomized controlled trial; SKY: Sudarshan Kriya yoga
Source: Broota A, Dhir R. Efficacy of two relaxation techniques in depression. Journal of Personality and Clinical Studies. 1990;6(1):83-90. Khumar SS, Kaur P, Kaur S. Effectiveness of Shavasana on depression among university students. Indian J Clin Psychol. 1993;20(2):82-87. Janakiramaiah N, Gangadhar BN, Naga Venkatesha Murthy PJ, et al. Antidepressant efficacy of Sudarshan Kriya yoga (SKY) in melancholia: a randomized comparison with electroconvulsive therapy (ECT) and imipramine. J Affect Disord. 2000;57(1-3):255-259. Rohini V, Pandey RS, Janakiramaiah N, et al. A comparative study of full and partial Sudarshan Kriya yoga (SKY) in major depressive disorder. NIMHANS Journal. 2000;18(1):53-57. Woolery A, Myers H, Sternlieb B, et al. A yoga intervention for young adults with elevated symptoms of depression. Altern Ther Health Med. 2004;10(2):60-63.

 

Table 3

Evidence of the antidepressant effect of exercise

Literature review	Methodology	Conclusion
Byrne and Byrne, 1993	13 studies, clinical samples, measured changes in depressed mood	90% of studies reported beneficial effects, especially in clinical populations
Scully et al, 1998	4 literature reviews, 1 monograph, 1 study	Positive relationship of physical activity and depression in healthy and clinical samples, increased over time
Lawlor and Hopker, 2001	14 RCTs from 1966 to 1999 with depression as an outcome	Significant methodologic weaknesses, but exercise effect > no treatment and=cognitive therapy
Dunn et al, 2001	Examined dose effect in 37 studies; subjects diagnosed with depressive disorders, mild-to-moderate symptoms, and no medical comorbidity	Only level B and C evidence; positive effects with exercise from light to heavy intensity; aerobic=nonaerobic; improvement may or may not be related to improved fitness
Brosse et al, 2002	12 RCTs from 1979 to 1999	Significant methodologic limitations, but authors concluded evidence supports a positive effect of exercise in healthy and clinical populations
Larun et al, 2006	4 RCTs in children and youth age	Exercise effect same as no intervention, low-intensity relaxation, or psychosocial intervention
Barbour et al, 2007	2 meta-analyses, 1 RCT, 2 studies	Positive effect; high-energy was optimal dose; aerobic=nonaerobic; improvement may or may not be related to improved fitness
RCT: randomized controlled trial
Source: Byrne AE, Byrne DG. The effect of exercise on depression, anxiety and other mood states: A review. J Psychosom Res. 1993;37(6):565-574. Scully D, Kremer J, Meade MM, et al. Physical exercise and psychological well being: a critical review. Br J Sports Med. 1998;32(2):111-120. Lawlor DA, Hopker SW. The effectiveness of exercise as an intervention in the management of depression: systematic review and meta-regression analysis of randomised controlled trials. BMJ. 2001;322(7289):763-767. Dunn AL, Trivedi MH, O’Neal HA. Physical activity dose-response effects on outcomes of depression and anxiety. Med Sci Sports Exerc. 2001;33(6):S587. Brosse AL, Sheets ES, Lett HS, et al. Exercise and the treatment of clinical depression in adults: recent findings and future directions. Sports Med. 2002;32(12):741-760. Larun L, Nordheim LV, Ekeland E, et al. Exercise in prevention and treatment of anxiety and depression among children and young people. Cochrane Database Syst Rev. 2006;3:CD004691. Barbour KA, Edenfield TM, Blumenthal JA. Exercise as a treatment for depression and other psychiatric disorders: a review. J Cardiopulm Rehabil Prev. 2007;27(6):359-367.

Conclusion. Evidence supports exercise for short-term treatment of mild or moderate depression in adults. Studies tend to be small and brief, to enroll young physically -sound patients, and to include little follow-up. Studies of subjects age

At least 2 studies suggest that high-energy exercise and aerobic or resistance training provide greater reductions in depressive symptoms than exercises such as walking.^5,6 Yoga’s positive effects suggest, however, that an aerobic effect is not necessary for an antidepressant benefit.

Exercise has not been adequately tested as a complementary treatment but likely is safe for most psychiatric patients. Perspiration and dehydration might alter therapeutic blood levels of lithium or other medications. Advise patients to drink water before, during, and after exercise and to avoid outdoor exercise in extreme temperatures. More vigorous monitoring might be indicated in specific cases.

Tailor exercise programs to individual needs, considering the patient’s age and health status. Refer a patient with a known heart problem or increased cardiovascular risk to his or her physician for selective exercise testing.

Bibliotherapy—reading self-help books, usually about cognitive-behavioral approaches to depressive disorders—has been relatively well studied. A recent meta-analysis examined 29 studies with pre-post designs. Group differences in the 17 controlled studies yielded a large effect size of 0.77. Participants who read the materials benefitted similarly whether they met in groups or applied the information on their own. Older adults tended to be less depressed at baseline and made smaller treatment gains.⁷

A study of 31 patients age >60 with mild-to-moderate depression⁸ compared 16 sessions of professionally administered cognitive-behavioral therapy (CBT) with self-administered cognitive therapy learned from reading a book.⁹ Both groups showed greater improvement in depressive symptoms compared with a control group. Subjects in the CBT group did somewhat better during the trial, but at 3-month follow-up most patients in both treatment groups no longer met diagnostic criteria for MDD.

Conclusion. Evidence supports bibliotherapy as an effective treatment for mild-to-moderate depression. No convincing data support its use as a complementary treatment, but it poses virtually no risk.

St. John’s wort (Hypericum perforatum) has been extensively studied for depressive disorders, with 29 RCTs in a meta-analysis of MDD trials through July 2008.¹⁰ Another meta-analysis compared St. John’s wort with selective serotonin reuptake inhibitors (SSRIs) in 13 studies through June 2008.¹¹ These and most RCTs have found St. John’s wort significantly more effective than placebo in reducing depressive symptoms.

 

Data selected from double-blind RCTs totaling 217 patients with mild depression [Hamilton Depression Rating Scale (HDRS) scores 12 Studies routinely show that treating MDD with St. John’s wort is comparable to using tricyclic or SSRI antidepressants.

Side effects with St. John’s wort generally are no different than with placebo and significantly less than with comparison treatments. Even so, using St. John’s wort instead of SSRIs for MDD remains controversial.

Studies vs SSRIs. Many of the favorable St. John’s wort trials were conducted in Europe, particularly in Germany. Two large RCTs conducted in the United States reported that the St. John’s wort standardized extract LI-160 was not more effective than placebo, but neither could be clearly interpreted as negative for St. John’s wort:

• In an 8-week trial, St. John’s wort and placebo groups improved significantly but at unusually low rates. The remission rate with St. John’s wort was small but significantly higher than with placebo.¹³
  
• A study sponsored by the National Institute of Mental Health compared St. John’s wort, 900 to 1,500 mg/d; sertraline, 50 to 100 mg/d; and placebo in 340 adults with MDD. No positive effects were found for St. John’s wort or sertraline.¹⁴
  

Side effects. St. John’s wort can affect blood levels of circulating medications metabolized by the cytochrome P450 liver enzyme system, including tricyclic antidepressants. Case studies have reported pregnancy from oral contraceptive failure, skin rashes, headache, and mania with St. John’s wort use. Although these reports are disturbing, St. John’s wort’s side effects when compared with SSRIs have been less frequent (40% vs 49%) and milder (clinical trial dropout rate 2% vs 7%).¹¹

Conclusion. Standardized extracts of St. John’s wort—particularly WS5570, 300 mg tid, and ZE117, 250 mg bid—appear to be effective treatments, especially for mild-to-moderate MDD. Because St. John’s wort is available without prescription and can interact with SSRIs or other antidepressants:

• care is required for its complementary use
  
• it is important to ask if patients are using St. John’s wort on their own.
  

St. John’s wort is used as a first-line depression treatment in Europe, but U.S. physicians may be less familiar with its potential interactions with other medications. We recommend that you consider St. John’s wort:

• for first-line use only when you can adequately gauge its effects on your patient’s other medications
  
• especially for depressed patients who cannot tolerate SSRIs.
  

SAMe is a metabolite involved in bio-synthesis of norepinephrine, serotonin, and dopamine.¹⁵ SAMe salts (such as 1,4-butanedisulfonate) are used as an over-the-counter supplement for depression treatment. Dozens of RCTs show SAMe has greater efficacy than placebo and positive effects comparable to those of standard antidepressants. In a meta-analysis of 28 RCTs by the Agency for Healthcare Quality and Research, SAMe produced significantly greater symptom improvement compared with placebo.¹⁶

SAMe has become a popular alternative treatment for depression since its introduction to the United States in the late 1990s, but it has been studied in only 2 U.S. open trials. One showed SAMe to be very effective in reducing depressive symptoms in patients with HIV/AIDS.¹⁷ The other found a 50% response rate and 43% remission rate with adjunctive SAMe, 800 to 1,600 mg/d for 6 weeks, in 30 adults with MDD who failed to respond adequately to SSRIs or the serotonin-norepinephrine reuptake inhibitor (SNRI) venlafaxine. The most common side effects were gastrointestinal (GI) symptoms and headaches.¹⁸ This open trial led to an ongoing National Institutes of Health-sponsored RCT on the safety and efficacy of SAMe for patients with treatment-resistant depression.

Conclusion. SAMe appears to have a faster onset of antidepressant effect than standard SSRIs or SNRIs and a favorable side-effect profile, which make the lack of rigorous trials in the United States striking. We recommend that you consider SAMe:

• as an adjunct in patients with incomplete response to standard treatments
  
• as a complementary treatment to speed onset of antidepressant effects.
  

Polyunsaturated fatty acids (PUFAs), usually from fish oils, have long been popular nutritional supplements because of their beneficial effects on cholesterol and cardiovascular health. Omega-3 and omega-6 fatty acids are the most common PUFAs. The omega-3 PUFAs include eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA).

Four meta-analyses independently looked at largely the same dozen RCTs through 2006 and found that 1 to 2 grams daily of omega-3 PUFAs was significantly more effective at reducing depressive symptoms than placebo.^19-22 Other data suggest that omega-3 PUFAs can induce depression remission in depressed Parkinson’s disease patients²³ and depressed pregnant women.²⁴ Since 2006, however, findings have been inconsistent. Several trials have found PUFAs no more effective than placebo.^25-27

 

An 8-week double-blind study compared EPA, 1 gram daily; fluoxetine, 20 mg/d; or both agents in 60 outpatients with MDD. Response rates—as measured by ≥50% reduction in baseline HDRS scores—were 50% with fluoxetine, 56% with EPA, and 81% with combination therapy.²⁸

Conclusion. Questions remain about dosing, ratio of EPA to DHA, patient selection, and baseline blood levels of PUFAs compared with response. PUFAs have a benign side-effect profile, with occasional reports of diarrhea or GI upset. Although their therapeutic effects are being clarified, PUFAs appear safe to recommend as an adjunct treatment if standard care is not satisfactory.

Insufficient evidence

L-tryptophan. It seems reasonable to expect a serotonin precursor to increase serotonin in the CNS and improve depressive symptoms. Of 111 trials on L-tryptophan for depression, however, only 2 met the quality criteria for inclusion in a recent meta-analysis.²⁹ Combining the 2 trials showed L-tryptophan alone and in combination with a tricyclic antidepressant was more effective than placebo for treating depressive disorders in adults.

Conclusion. Very little research continues to test L-tryptophan as a viable CAM for depressive disorder. Its serious side effect of eosinophilia-myalgia syndrome makes clinical use of this agent unlikely.

Acupuncture. Numerous small studies with questionable controls, different needling placements, and poor allocation concealment and blinding limit the ability to draw conclusions about acupuncture for treating depression ( Table 4 ). A recent meta-analysis by Wang et al³⁰ added 2 Chinese trials not included in an earlier review³¹ and found acupuncture significantly reduced depressive symptoms. No consistent differences were detected in response or remission rates, however.

Conclusion. Evidence is methodologically weak, and the use of acupuncture as an alternative or complementary treatment of depression is questionable.

Table 4

Acupuncture: Insufficient evidence of antidepressant effect

Literature review	Methodology	Conclusion
Mukaino et al, 2005	Systematic review of 7 RCTs including 509 patients; compared either manual or electroacupuncture with any control procedure	Inconsistent evidence of manual acupuncture’s effectiveness vs sham; electroacupuncture’s effect may be similar to that of antidepressant medication and merits further study
Leo and Ligot, 2007	Systematic review of 9 RCTs, 5 considered low quality; some focused on very specific populations (ie, hospitalized stroke patients or pregnant depressed patients)	Evidence inconclusive because of study designs and methodologies
Smith and Hay, 2005	Meta-analysis of 7 trials including 517 adults with mild-to-moderate depression; 5 trials (409 participants) compared acupuncture with medication; 2 trials compared acupuncture with wait list or sham acupuncture	No difference between acupuncture and medication; study quality too poor to support acupuncture’s efficacy
Wang et al, 2008	Meta-analysis of 8 small RCTs totalling 477 subjects (256 received active acupuncture, remainder received sham acupuncture); sham acupuncture design, number of acupuncture sessions, and duration varied among studies	Significant reduction in HRSD or BDI scores for acupuncture vs sham, but no significant effect of acupuncture on response or remission rates
BDI: Beck Depression Inventory; HRSD: Hamilton Rating Scale for Depression; RCT: randomized controlled trial
Source: Mukaino Y, Park J, White A, et al. The effectiveness of acupuncture for depression—a systematic review of randomised controlled trials. Acupunct Med. 2005;23(2):70-76. Leo RJ, Ligot JS Jr. A systematic review of randomized controlled trials of acupuncture in the treatment of depression. J Affect Disord. 2007;97(1-3):13-22. Smith CA, Hay PPJ. Acupuncture for depression. Cochrane Database Syst Rev. 2005;(2):CD004046. Wang H, Qi H, Wang BS, et al. Is acupuncture beneficial in depression? A meta-analysis of 8 randomized controlled trials. J Affect Disord. 2008;111(2-3):125-134.

Related resources

• National Center for Complementary and Alternative Medicine, National Institutes of Health. http://nccam.nih.gov.
  
• Journal of Alternative and Complementary Medicine. www.liebertpub.com/products/product.aspx?pid=26.
  
• Complementary and alternative medicine. www.nlm.nih.gov/medlineplus/complementaryandalternativemedicine.html.
  

Drug brand names

• Fluoxetine • Prozac
  
• Imipramine • Tofranil
  
• Lithium • Eskalith, Lithobid
  
• Sertraline • Zoloft
  
• Venlafaxine • Effexor
  

Disclosure

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Comment on this article

Americans with depression turn to complementary and alternative medicine (CAM) more often than conventional psychotherapy or FDA-approved medication. In a nationally representative sample, 54% of respondents with self-reported “severe depression”—including two-thirds of those receiving conventional therapies—reported using CAM during the previous 12 months.¹

Unfortunately, popular acceptance of CAM for depression is disproportionate to the evidence base, which—although growing—remains limited. As a result, your patients may be self-medicating with poorly supported treatments that are unlikely to help them recover from depression.

In reviewing CAM treatments for depression, we found some with enough evidence of positive effect that we feel comfortable recommending them as evidence-based options. These promising, short-term treatments are supported by level 1a or 1b evidence and at least 1 study that demonstrates an ability to induce remission ( Table 1 ).²

For patients seeking “natural” or nonprescription treatments—or when you wish to augment standard treatments that are not working adequately—you might recommend fatty acids, St. John’s wort, or S-adenosyl-L-methionine (SAMe). Similar recommendations can be made for yoga, exercise, and bibliotherapy, as we discuss here.

Table 1

Evidence these authors required to recommend a CAM treatment

Minimum requirements	Level of evidence	Recommendation
Systematic review showing superiority to placebo or standard treatment Plus 1 study showing CAM treatment can induce remission	1a + 1b or 2b	A
At least 2 RCTs showing superiority to placebo or standard treatment Plus 1 study showing CAM treatment can induce remission	1b 1b or 2b	A–
CAM: complementary and alternative medicine; RCT: randomized controlled trial
Source: Reference 2

Reviewing CAM evidence

This article refers to as “alternative” any treatment other than a form of psychotherapy or an FDA-approved medication that substitutes for a standard psychiatric treatment. When used to augment standard psychiatric treatments, these approaches are considered “complementary.”

Our search for evidence on CAM treatments for depressive disorders raised questions about what constitutes acceptable and convincing methodology:

• Studies often had problems with blinding and suitable placebos. Many were small, with short duration and no long-term follow-up.
  
• Comparisons with active treatments that showed no differences were assumed to imply comparability, even though the studies were powered to detect only large differences.
  

On the other hand, multiple randomized controlled trials (RCTs) have shown consistent superiority of some CAM treatments over comparison conditions.

Applying the evidence. Because CAM use is widespread, be sure to ask psychiatric patients if they are using CAM treatments. If the answer is “yes,” a risk-benefit assessment is needed. Inform patients who are using poorly supported CAM approaches that they could consider better-supported CAM options as well as standard effective antidepressants.

Monitor patients for an adequately prompt positive response to an evidence-based CAM approach that has shown efficacy for their level of depression. As with any treatment, consider other evidence-based options when CAM treatments are inadequate or unsuccessful in achieving remission of depressive symptoms.

Sufficient evidence of efficacy

Yoga. In their systematic review of yoga’s effectiveness for depression, Pilkington et al³ analyzed 5 RCTs that met 3 criteria:

• participants had mild to severe depression or depressive disorders
  
• yoga or yoga-based exercises alone were used for treatment
  
• depression rating scales were used as outcome measures.
  

They found evidence that yoga can reduce depressive symptoms and induce remission ( Table 2 ). The studies were generally small and of short duration, and depression severity and interventions varied widely. Most participants were young and relatively fit, raising questions about yoga’s applicability to older or less fit patients. Reporting of adverse events was limited, but no safety issues or adverse effects were identified.

Conclusion. Yoga has been studied primarily as an alternative treatment, but its physical health and group participation benefits and lack of side effects make it a suitable complementary treatment as well.

Exercise. Extensive literature has examined the relationship between exercise and depression. We identified 7 reviews published between 1993 and 2008 ( Table 3 ). All supported positive effects of exercise except for patients age

• 45% with supervised exercise
  
• 40% with home-based exercise
  
• 47% with sertraline, 50 to 200 mg/d
  
• 31% with placebo.⁴
  

Table 2

5 RCTs of yoga’s effectiveness in treating depression

RCT	Interventions	Conclusion
Broota and Dhir, 1990	Yoga and PMR vs control	Yoga and PMR were more effective than control, with yoga more effective than PMR
Khumar et al, 1993	Shavasana yoga vs no intervention	College students with severe depression improved during and after yoga treatment
Janakiramaiah et al, 2000	SKY vs ECT vs imipramine	Reductions in BDI scores for all 3 groups; ECT > SKY or imipramine, SKY=imipramine
Rohini et al, 2000	Full SKY vs partial SKY	30 individuals with MDD improved with either therapy, but results were not statistically significant
Woolery, 2004	Iyengar yoga vs wait list	28 mildly depressed individuals benefitted from yoga, as measured by BDI scores at midpoint and throughout treatment
BDI: Beck Depression Inventory; ECT: electroconvulsive therapy; MDD: major depressive disorder; PMR: progressive muscle relaxation; RCT: randomized controlled trial; SKY: Sudarshan Kriya yoga
Source: Broota A, Dhir R. Efficacy of two relaxation techniques in depression. Journal of Personality and Clinical Studies. 1990;6(1):83-90. Khumar SS, Kaur P, Kaur S. Effectiveness of Shavasana on depression among university students. Indian J Clin Psychol. 1993;20(2):82-87. Janakiramaiah N, Gangadhar BN, Naga Venkatesha Murthy PJ, et al. Antidepressant efficacy of Sudarshan Kriya yoga (SKY) in melancholia: a randomized comparison with electroconvulsive therapy (ECT) and imipramine. J Affect Disord. 2000;57(1-3):255-259. Rohini V, Pandey RS, Janakiramaiah N, et al. A comparative study of full and partial Sudarshan Kriya yoga (SKY) in major depressive disorder. NIMHANS Journal. 2000;18(1):53-57. Woolery A, Myers H, Sternlieb B, et al. A yoga intervention for young adults with elevated symptoms of depression. Altern Ther Health Med. 2004;10(2):60-63.

 

Table 3

Evidence of the antidepressant effect of exercise

Literature review	Methodology	Conclusion
Byrne and Byrne, 1993	13 studies, clinical samples, measured changes in depressed mood	90% of studies reported beneficial effects, especially in clinical populations
Scully et al, 1998	4 literature reviews, 1 monograph, 1 study	Positive relationship of physical activity and depression in healthy and clinical samples, increased over time
Lawlor and Hopker, 2001	14 RCTs from 1966 to 1999 with depression as an outcome	Significant methodologic weaknesses, but exercise effect > no treatment and=cognitive therapy
Dunn et al, 2001	Examined dose effect in 37 studies; subjects diagnosed with depressive disorders, mild-to-moderate symptoms, and no medical comorbidity	Only level B and C evidence; positive effects with exercise from light to heavy intensity; aerobic=nonaerobic; improvement may or may not be related to improved fitness
Brosse et al, 2002	12 RCTs from 1979 to 1999	Significant methodologic limitations, but authors concluded evidence supports a positive effect of exercise in healthy and clinical populations
Larun et al, 2006	4 RCTs in children and youth age	Exercise effect same as no intervention, low-intensity relaxation, or psychosocial intervention
Barbour et al, 2007	2 meta-analyses, 1 RCT, 2 studies	Positive effect; high-energy was optimal dose; aerobic=nonaerobic; improvement may or may not be related to improved fitness
RCT: randomized controlled trial
Source: Byrne AE, Byrne DG. The effect of exercise on depression, anxiety and other mood states: A review. J Psychosom Res. 1993;37(6):565-574. Scully D, Kremer J, Meade MM, et al. Physical exercise and psychological well being: a critical review. Br J Sports Med. 1998;32(2):111-120. Lawlor DA, Hopker SW. The effectiveness of exercise as an intervention in the management of depression: systematic review and meta-regression analysis of randomised controlled trials. BMJ. 2001;322(7289):763-767. Dunn AL, Trivedi MH, O’Neal HA. Physical activity dose-response effects on outcomes of depression and anxiety. Med Sci Sports Exerc. 2001;33(6):S587. Brosse AL, Sheets ES, Lett HS, et al. Exercise and the treatment of clinical depression in adults: recent findings and future directions. Sports Med. 2002;32(12):741-760. Larun L, Nordheim LV, Ekeland E, et al. Exercise in prevention and treatment of anxiety and depression among children and young people. Cochrane Database Syst Rev. 2006;3:CD004691. Barbour KA, Edenfield TM, Blumenthal JA. Exercise as a treatment for depression and other psychiatric disorders: a review. J Cardiopulm Rehabil Prev. 2007;27(6):359-367.

Conclusion. Evidence supports exercise for short-term treatment of mild or moderate depression in adults. Studies tend to be small and brief, to enroll young physically -sound patients, and to include little follow-up. Studies of subjects age

At least 2 studies suggest that high-energy exercise and aerobic or resistance training provide greater reductions in depressive symptoms than exercises such as walking.^5,6 Yoga’s positive effects suggest, however, that an aerobic effect is not necessary for an antidepressant benefit.

Exercise has not been adequately tested as a complementary treatment but likely is safe for most psychiatric patients. Perspiration and dehydration might alter therapeutic blood levels of lithium or other medications. Advise patients to drink water before, during, and after exercise and to avoid outdoor exercise in extreme temperatures. More vigorous monitoring might be indicated in specific cases.

Tailor exercise programs to individual needs, considering the patient’s age and health status. Refer a patient with a known heart problem or increased cardiovascular risk to his or her physician for selective exercise testing.

Bibliotherapy—reading self-help books, usually about cognitive-behavioral approaches to depressive disorders—has been relatively well studied. A recent meta-analysis examined 29 studies with pre-post designs. Group differences in the 17 controlled studies yielded a large effect size of 0.77. Participants who read the materials benefitted similarly whether they met in groups or applied the information on their own. Older adults tended to be less depressed at baseline and made smaller treatment gains.⁷

A study of 31 patients age >60 with mild-to-moderate depression⁸ compared 16 sessions of professionally administered cognitive-behavioral therapy (CBT) with self-administered cognitive therapy learned from reading a book.⁹ Both groups showed greater improvement in depressive symptoms compared with a control group. Subjects in the CBT group did somewhat better during the trial, but at 3-month follow-up most patients in both treatment groups no longer met diagnostic criteria for MDD.

Conclusion. Evidence supports bibliotherapy as an effective treatment for mild-to-moderate depression. No convincing data support its use as a complementary treatment, but it poses virtually no risk.

St. John’s wort (Hypericum perforatum) has been extensively studied for depressive disorders, with 29 RCTs in a meta-analysis of MDD trials through July 2008.¹⁰ Another meta-analysis compared St. John’s wort with selective serotonin reuptake inhibitors (SSRIs) in 13 studies through June 2008.¹¹ These and most RCTs have found St. John’s wort significantly more effective than placebo in reducing depressive symptoms.

 

Data selected from double-blind RCTs totaling 217 patients with mild depression [Hamilton Depression Rating Scale (HDRS) scores 12 Studies routinely show that treating MDD with St. John’s wort is comparable to using tricyclic or SSRI antidepressants.

Side effects with St. John’s wort generally are no different than with placebo and significantly less than with comparison treatments. Even so, using St. John’s wort instead of SSRIs for MDD remains controversial.

Studies vs SSRIs. Many of the favorable St. John’s wort trials were conducted in Europe, particularly in Germany. Two large RCTs conducted in the United States reported that the St. John’s wort standardized extract LI-160 was not more effective than placebo, but neither could be clearly interpreted as negative for St. John’s wort:

• In an 8-week trial, St. John’s wort and placebo groups improved significantly but at unusually low rates. The remission rate with St. John’s wort was small but significantly higher than with placebo.¹³
  
• A study sponsored by the National Institute of Mental Health compared St. John’s wort, 900 to 1,500 mg/d; sertraline, 50 to 100 mg/d; and placebo in 340 adults with MDD. No positive effects were found for St. John’s wort or sertraline.¹⁴
  

Side effects. St. John’s wort can affect blood levels of circulating medications metabolized by the cytochrome P450 liver enzyme system, including tricyclic antidepressants. Case studies have reported pregnancy from oral contraceptive failure, skin rashes, headache, and mania with St. John’s wort use. Although these reports are disturbing, St. John’s wort’s side effects when compared with SSRIs have been less frequent (40% vs 49%) and milder (clinical trial dropout rate 2% vs 7%).¹¹

Conclusion. Standardized extracts of St. John’s wort—particularly WS5570, 300 mg tid, and ZE117, 250 mg bid—appear to be effective treatments, especially for mild-to-moderate MDD. Because St. John’s wort is available without prescription and can interact with SSRIs or other antidepressants:

• care is required for its complementary use
  
• it is important to ask if patients are using St. John’s wort on their own.
  

St. John’s wort is used as a first-line depression treatment in Europe, but U.S. physicians may be less familiar with its potential interactions with other medications. We recommend that you consider St. John’s wort:

• for first-line use only when you can adequately gauge its effects on your patient’s other medications
  
• especially for depressed patients who cannot tolerate SSRIs.
  

SAMe is a metabolite involved in bio-synthesis of norepinephrine, serotonin, and dopamine.¹⁵ SAMe salts (such as 1,4-butanedisulfonate) are used as an over-the-counter supplement for depression treatment. Dozens of RCTs show SAMe has greater efficacy than placebo and positive effects comparable to those of standard antidepressants. In a meta-analysis of 28 RCTs by the Agency for Healthcare Quality and Research, SAMe produced significantly greater symptom improvement compared with placebo.¹⁶

SAMe has become a popular alternative treatment for depression since its introduction to the United States in the late 1990s, but it has been studied in only 2 U.S. open trials. One showed SAMe to be very effective in reducing depressive symptoms in patients with HIV/AIDS.¹⁷ The other found a 50% response rate and 43% remission rate with adjunctive SAMe, 800 to 1,600 mg/d for 6 weeks, in 30 adults with MDD who failed to respond adequately to SSRIs or the serotonin-norepinephrine reuptake inhibitor (SNRI) venlafaxine. The most common side effects were gastrointestinal (GI) symptoms and headaches.¹⁸ This open trial led to an ongoing National Institutes of Health-sponsored RCT on the safety and efficacy of SAMe for patients with treatment-resistant depression.

Conclusion. SAMe appears to have a faster onset of antidepressant effect than standard SSRIs or SNRIs and a favorable side-effect profile, which make the lack of rigorous trials in the United States striking. We recommend that you consider SAMe:

• as an adjunct in patients with incomplete response to standard treatments
  
• as a complementary treatment to speed onset of antidepressant effects.
  

Polyunsaturated fatty acids (PUFAs), usually from fish oils, have long been popular nutritional supplements because of their beneficial effects on cholesterol and cardiovascular health. Omega-3 and omega-6 fatty acids are the most common PUFAs. The omega-3 PUFAs include eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA).

Four meta-analyses independently looked at largely the same dozen RCTs through 2006 and found that 1 to 2 grams daily of omega-3 PUFAs was significantly more effective at reducing depressive symptoms than placebo.^19-22 Other data suggest that omega-3 PUFAs can induce depression remission in depressed Parkinson’s disease patients²³ and depressed pregnant women.²⁴ Since 2006, however, findings have been inconsistent. Several trials have found PUFAs no more effective than placebo.^25-27

 

An 8-week double-blind study compared EPA, 1 gram daily; fluoxetine, 20 mg/d; or both agents in 60 outpatients with MDD. Response rates—as measured by ≥50% reduction in baseline HDRS scores—were 50% with fluoxetine, 56% with EPA, and 81% with combination therapy.²⁸

Conclusion. Questions remain about dosing, ratio of EPA to DHA, patient selection, and baseline blood levels of PUFAs compared with response. PUFAs have a benign side-effect profile, with occasional reports of diarrhea or GI upset. Although their therapeutic effects are being clarified, PUFAs appear safe to recommend as an adjunct treatment if standard care is not satisfactory.

Insufficient evidence

L-tryptophan. It seems reasonable to expect a serotonin precursor to increase serotonin in the CNS and improve depressive symptoms. Of 111 trials on L-tryptophan for depression, however, only 2 met the quality criteria for inclusion in a recent meta-analysis.²⁹ Combining the 2 trials showed L-tryptophan alone and in combination with a tricyclic antidepressant was more effective than placebo for treating depressive disorders in adults.

Conclusion. Very little research continues to test L-tryptophan as a viable CAM for depressive disorder. Its serious side effect of eosinophilia-myalgia syndrome makes clinical use of this agent unlikely.

Acupuncture. Numerous small studies with questionable controls, different needling placements, and poor allocation concealment and blinding limit the ability to draw conclusions about acupuncture for treating depression ( Table 4 ). A recent meta-analysis by Wang et al³⁰ added 2 Chinese trials not included in an earlier review³¹ and found acupuncture significantly reduced depressive symptoms. No consistent differences were detected in response or remission rates, however.

Conclusion. Evidence is methodologically weak, and the use of acupuncture as an alternative or complementary treatment of depression is questionable.

Table 4

Acupuncture: Insufficient evidence of antidepressant effect

Literature review	Methodology	Conclusion
Mukaino et al, 2005	Systematic review of 7 RCTs including 509 patients; compared either manual or electroacupuncture with any control procedure	Inconsistent evidence of manual acupuncture’s effectiveness vs sham; electroacupuncture’s effect may be similar to that of antidepressant medication and merits further study
Leo and Ligot, 2007	Systematic review of 9 RCTs, 5 considered low quality; some focused on very specific populations (ie, hospitalized stroke patients or pregnant depressed patients)	Evidence inconclusive because of study designs and methodologies
Smith and Hay, 2005	Meta-analysis of 7 trials including 517 adults with mild-to-moderate depression; 5 trials (409 participants) compared acupuncture with medication; 2 trials compared acupuncture with wait list or sham acupuncture	No difference between acupuncture and medication; study quality too poor to support acupuncture’s efficacy
Wang et al, 2008	Meta-analysis of 8 small RCTs totalling 477 subjects (256 received active acupuncture, remainder received sham acupuncture); sham acupuncture design, number of acupuncture sessions, and duration varied among studies	Significant reduction in HRSD or BDI scores for acupuncture vs sham, but no significant effect of acupuncture on response or remission rates
BDI: Beck Depression Inventory; HRSD: Hamilton Rating Scale for Depression; RCT: randomized controlled trial
Source: Mukaino Y, Park J, White A, et al. The effectiveness of acupuncture for depression—a systematic review of randomised controlled trials. Acupunct Med. 2005;23(2):70-76. Leo RJ, Ligot JS Jr. A systematic review of randomized controlled trials of acupuncture in the treatment of depression. J Affect Disord. 2007;97(1-3):13-22. Smith CA, Hay PPJ. Acupuncture for depression. Cochrane Database Syst Rev. 2005;(2):CD004046. Wang H, Qi H, Wang BS, et al. Is acupuncture beneficial in depression? A meta-analysis of 8 randomized controlled trials. J Affect Disord. 2008;111(2-3):125-134.

Related resources

• National Center for Complementary and Alternative Medicine, National Institutes of Health. http://nccam.nih.gov.
  
• Journal of Alternative and Complementary Medicine. www.liebertpub.com/products/product.aspx?pid=26.
  
• Complementary and alternative medicine. www.nlm.nih.gov/medlineplus/complementaryandalternativemedicine.html.
  

Drug brand names

• Fluoxetine • Prozac
  
• Imipramine • Tofranil
  
• Lithium • Eskalith, Lithobid
  
• Sertraline • Zoloft
  
• Venlafaxine • Effexor
  

Disclosure

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Comment on this article

Sex offenders traditionally are managed by the criminal justice system, but psychiatrists are frequently called on to assess and treat these individuals. Part of the reason is the overlap of paraphilias (disorders of sexual preference) and sexual offending. Many sexual offenders do not meet DSM criteria for paraphilias,¹ however, and individuals with paraphilias do not necessarily commit offenses or come into contact with the legal system.

As clinicians, we may need to assess and treat a wide range of sexual issues, from persons with paraphilias who are self-referred and have no legal involvement, to recurrent sexual offenders who are at a high risk of repeat offending. Successfully managing sex offenders includes psychological and pharmacologic interventions and possibly incarceration and post-incarceration surveillance. This article focuses on pharmacologic interventions for male sexual offenders.

Reducing sexual drive

Sex offending likely is the result of a complex interplay of environment and psychological and biologic factors. The biology of sexual function provides numerous targets for pharmacologic intervention, including:²

• endocrine factors, such as testosterone
• neurotransmitters, such as serotonin.

The use of pharmacologic treatments for sex offenders is off-label, and evidence is limited. In general, pharmacologic treatments are geared toward reducing sexual drive through nonhormonal or hormonal means (Table 1).^3-5

Table 1: Pharmacologic treatment of male sex offenders: A risk-based approach

*Not available in the United States
^†Some authors suggest administering this dosage once every 2 weeks
CPA: cyproterone acetate; GNRH: gonadotropin-releasing hormone; IM: intramuscular; MPA: medroxyprogesterone acetate; SSRI: selective serotonin reuptake inhibitor
Source: References 3-5

Nonhormonal treatments

SSRIs. Selective serotonin reuptake inhibitors act by blocking serotonin reuptake in the synaptic cleft. Soon after the first SSRIs were approved in 1988, reports appeared of SSRIs interfering with sexual functioning.⁶ This side effect quickly was exploited to assist the treatment of sexual offenders.⁷

The mechanism of action may include:⁸

• direct effects, such as general inhibition of sexual activity, reduced impulsiveness, and an effect on the hypothesized “obsessive-compulsive” nature of paraphilias⁹
• indirect reduction of testosterone.

A growing body of literature supports SSRIs’ effectiveness in treating paraphilias and sexual offenders. Greenberg⁷ reviewed case studies and open drug trials of nearly 200 patients receiving fluoxetine, fluvoxamine, or sertraline. Most studies showed response rates of 50% to 90%.¹⁰ Positive effects included decreases in:

• paraphiliac fantasies, urges, and sexual acts
• masturbation
• hypersexual activity.

Some studies reported a preferential decrease in paraphiliac interests with an increase in conventional sexual interests, although this may be related to placebo or halo effects—patients may have reported an increase in conventional interests because they noticed a decrease in paraphiliac interests. Negative side effects included decreased sexual desires, delayed ejaculation, decreased libido, and anorgasmia.

Adi et al¹¹ completed a more rigorous literature evaluation that included 9 studies with a total of 225 patients receiving fluoxetine, fluvoxamine, sertraline, or paroxetine. Eight studies showed benefits; however, Adi noted that this preliminary evidence was “far from conclusive.”

SSRIs generally are well tolerated and may be more appealing to patients than the “chemical castration” of hormonal treatments. Dosing is similar to that used in depression or obsessive-compulsive disorder. Although most patients notice beneficial effects in 2 to 4 weeks, some notice the effect nearly immediately.

Naltrexone. An opioid antagonist thought to affect the CNS processes of pleasure and pain, naltrexone has been used to treat alcohol dependence and pathologic gambling. A few case studies^12-14 and 1 study of 21 adolescent sex offenders¹⁵ have shown benefits in treating sexual offenders or paraphiliacs. Benefits were seen at 50 mg/d, with suggested dosing of 100 to 200 mg/d. Because data are very limited, consider naltrexone only on an individual basis or as a possible adjunctive treatment.

Psychostimulants. Methylphenidate was added to augment SSRI treatment in a study of 26 men with paraphilias or paraphilia-related disorders.¹⁶ Results included further significant decreases in total sexual outlets (orgasms per week) and average time spent per day in paraphilia and paraphilia-related behavior. These gains appeared to be independent of the presence of attention-deficit/hyperactivity disorder.

Again, because data are very limited, consider this strategy only on an individual basis or as a possible adjunctive treatment. Because sexual offenders have high rates of substance abuse,¹⁷ consider the potential for stimulant abuse.

Hormonal treatments

Because testosterone is required for healthy bone metabolism, the antiandrogen medications used in hormonal treatment can cause osteoporosis.^18,19 Therefore, long-term antiandrogen treatment should include bone scans to monitor for osteopenia and osteoporosis. Some authors have suggested that monthly doses of 25 to 50 mg of testosterone could minimize this risk.²⁰ Bisphosphonates, vitamin D, and calcium supplements at osteoporosis treatment levels might be helpful.¹⁸ Other common side effects of antiandrogen medications are listed in Table 2.

 

Finasteride is approved for treating benign prostatic hyperplasia and androgenetic alopecia. It works by preventing conversion of testosterone to dihydrotestosterone (DHT) by the type II isoenzyme.²¹ Serum DHT contributes to male sexual behavior and predicts frequency of orgasms in healthy volunteers.²² Although there have been no studies of finasteride in sex offenders, it may be more acceptable to patients than other hormonal treatments and have a theoretical benefit in reducing sexual drive. Clinically, some patients describe increased control over urges without substantial side effects. Because there is no evidence supporting finasteride use in sex offenders, consider this medication only on an individual basis or as a possible adjunctive treatment.

Cyproterone acetate (CPA) is a synthetic steroid that blocks androgen receptors.^3,23 Some evidence supports its use in treating sex offenders,²⁴ although this agent is not available in the United States. For more information about CPA, see Box.

Medroxyprogesterone acetate (MPA), a derivative of progesterone, lowers serum testosterone by inhibiting its production through reducing pituitary luteinizing hormone (LH).²⁴ The typical dose range for use in sex offenders is 100 to 600 mg/d orally or 100 to 700 mg IM every week,³ although some authors suggest giving similar doses every 2 weeks.⁴

Side effects of MPA include hypersomnia; neurasthenia; weight gain; hot flashes; gynecomastia; increased scalp hair; and decreased erections, ejaculate volume, spermatogenesis, and body and facial hair. The drug decreases testosterone levels by about 50%. Positive effects include reduced interest in and energy spent on pursuing sexual goals, but preservation of nondeviant sexual arousal.⁴

MPA has been shown to effectively decrease deviant sexual arousal and recidivism. In a study of 100 patients receiving MPA (average 250 mg IM every 2 weeks) for an average of 3 years, only 1 re-offended while taking MPA.⁴

In a 5-year follow-up study²⁵ of 275 men, subjects were classified into high risk/treatment with MPA, 200 to 400 mg IM every 2 weeks, and low risk/nontreatment groups. A portion of the high risk/treatment group did not receive MPA. No sexual re-offenses occurred among high-risk subjects who received MPA, whereas the recidivism rate was 18% among high-risk subjects who did not receive MPA. Subjects in the low risk/nontreatment group had a recidivism rate of 15%, which suggests the need for more liberal use of antiandrogens. One major confounding factor was that subjects in the high risk/treatment group had to report every 2 weeks for injections; this may have resulted in closer follow-up, monitoring, and support, which may have contributed to lower recidivism.

Gonadotropin-releasing hormone (GNRH) agonists. The terms gonadotropin-releasing hormone and luteinizing-releasing hormone are used interchangeably. Most body testosterone is produced and released by Leydig cells in the testes in response to stimulation by LH released by the pituitary gland. LH release is controlled by the pulsatile release of GNRH from the hypothalamus. GNRH agonists are high-potency analogs of GNRH that work by causing an initial surge of LH followed by down-regulation of gonadotroph cells, a drop in LH, and a drop in testosterone to castration levels.

The GNRH analogs leuprolide, goserelin, and triptorelin are used to treat paraphiliacs and sexual offenders.²⁰ Leuprolide typically is dosed at 7.5 mg IM every month, 22.5 mg IM every 3 months, or 30 mg IM every 4 months. Goserelin is provided as a subcutaneous implant/depot injected as 3.6 mg every month or 10.8 mg every 3 months.

Triptorelin is FDA-approved as treatment for advanced prostate cancer. Triptorelin is given in depot formulation as 3.75 mg IM every month or in a long-acting form as 11.25 mg IM every 3 months.

When starting these medications, an initial surge of LH and testosterone can last up to 3 weeks.²⁶ Theoretically, this could worsen paraphiliac interests. Many practitioners will use a testosterone blocker such as flutamide, 250 mg tid, for the initial weeks of treatment.

Side effects of the GNRH agonists are similar. Most patients initially experience hot flashes. A systemic literature review²⁷ reported:

• weight gain
• perspiration
• gynecomastia
• urinary incontinence
• hot flashes
• decreased growth of facial and body hair
• asthenia
• erectile failure
• muscle tenderness
• frequent bone demineralization.

Rare cases of pituitary apoplexy (a clinical syndrome secondary to infarction of the pituitary gland) have been reported, possibly related to an underlying pituitary adenoma.²⁸

In a literature review that totalled 118 patients,²⁷ GNRH agonists significantly decreased erections, ejaculations, paraphiliac fantasies, and paraphiliac behavior. Patients also reported feeling more relaxed, and recidivism rates were low. Some patients who failed to respond to CPA and MPA responded to GNRH agonists. Subsequent studies found similar results.^29,30

 

Box

Table 2

Common side effects of antiandrogen therapy

Depression
Erectile dysfunction
Fatigue
Gynecomastia
Hot flashes
Hypertension
Low libido
Myalgia
Osteopenia
Osteoporosis
Sweating
Thromboembolism
Weight gain

Monitoring

Laboratory investigations are recommended to monitor for side effects of antiandrogen medications (Table 3).^19,27,31 Medical contraindications to rule out before initiating antiandrogen medications include:

• thromboembolic diseases
• liver disease
• bone demineralization disorders
• hypersensitivity to the drug.

Measure prolactin to rule out pituitary adenomas. Monitor serum testosterone because some patients will not experience testosterone suppression from GNRH agonists or other antiandrogens. Noncompliant patients could potentially reverse the effects of MPA and GNRH agonists by taking exogenous testosterone.

Table 3

Monitoring patients receiving antiandrogen medications

Pre-therapy workup	Periodic monitoring
Endocrinology or internist consultation Bone scan Weight Blood pressure Electrocardiogram CBC, renal function, liver function, fasting glucose, and lipids LH, FSH, testosterone, prolactin	Monthly: testosterone for the first 6 months Every 6 months: testosterone, LH, FSH, prolactin, CBC, renal function, liver function, fasting glucose and lipids, weight, blood pressure Yearly: bone scan
CBC: complete blood count; FSH: follicle-stimulating hormone; LH: luteinizing hormone
Source: References 19,27,31

Medication selection

The goals of pharmacologic treatment of sex offenders are to:

• reduce sexual offending and victimization
• suppress sexual drive to a controllable level
• possibly preferentially eliminate deviant arousal/thoughts
• allow normal sexual relationships.

Gauging risk. In determining which pharmacologic treatment to offer a patient, first evaluate the individual’s risk for recidivism. Actuarial scales^32,33 suggest that recidivism risk can be categorized, based on clinical factors (Table 4).^4,25,34

In addition to statistical risk factors, several other factors affect medication selection. Self-referred individuals may be more reliable in taking oral medications than those referred by the courts. A developmentally delayed individual may be a poor candidate for oral medication, unless he resides in a group home setting where compliance can be assured. Efficacy also guides medication choice. Finally, some patients will be legally required to provide proof of compliance, which only IM medications provide.

Treatment. Based on clinical experience and available literature, Bradford⁵ created an algorithm to help clinicians select appropriate pharmacologic interventions. Although it has not been validated, this algorithm provides a reasonable starting place.

In general, start treatment with an SSRI for low-risk individuals (Table 1).^3-5 If this strategy is insufficient, consider augmentation with methylphenidate, naltrexone, or finasteride.

The next step would be to add oral MPA or CPA, 50 mg/d, which would partially inhibit testosterone and may allow some normal sexual functioning.^4,23 Higher-dose oral MPA or CPA would be tried next. For higher-risk individuals or treatment failures, IM MPA or CPA would be offered next, followed by a GNRH agonist. For individuals at highest risk for re-offending, combinations of agents may be indicated.

This simple strategy is appealing, but in reality, treatment should be individualized. Choose medications based on the patient’s risk, wishes, and the previously mentioned clinical factors.

Table 4: Will my patient commit another sexual offense? Evaluating risk

Source: References 4,25,34Related resources

• Krueger RB, Kaplan MS. The paraphilic and hypersexual disorders: an overview. J Psychiatr Pract. 2001;7:391-403.
• Krueger RB, Kaplan MS. Behavioral and psychopharmacological treatment of the paraphilic and hypersexual disorders. J Psychiatr Pract. 2002;8:21-32.
• Association for the Treatment of Sexual Abusers. www.atsa.com.

 

Drug brand names

• Cyproterone acetate • Androcur
• Finasteride • Propecia, Proscar
• Fluoxetine • Prozac
• Flutamide • Eulexin
• Fluvoxamine • Luvox
• Goserelin • Zoladex
• Leuprolide • Eligard, Lupron
• Medroxyprogesterone acetate • Depo-Provera, Provera
• Methylphenidate • Ritalin
• Naltrexone • ReVia
• Paroxetine • Paxil
• Sertraline • Zoloft
• Triptorelin • Trelstar Depot

Disclosure

Dr. Booth reports no financial relationship with any company whose products are mentioned in the article or with manufacturers of competing products.

Comment on this article

Sex offenders traditionally are managed by the criminal justice system, but psychiatrists are frequently called on to assess and treat these individuals. Part of the reason is the overlap of paraphilias (disorders of sexual preference) and sexual offending. Many sexual offenders do not meet DSM criteria for paraphilias,¹ however, and individuals with paraphilias do not necessarily commit offenses or come into contact with the legal system.

As clinicians, we may need to assess and treat a wide range of sexual issues, from persons with paraphilias who are self-referred and have no legal involvement, to recurrent sexual offenders who are at a high risk of repeat offending. Successfully managing sex offenders includes psychological and pharmacologic interventions and possibly incarceration and post-incarceration surveillance. This article focuses on pharmacologic interventions for male sexual offenders.

Reducing sexual drive

Sex offending likely is the result of a complex interplay of environment and psychological and biologic factors. The biology of sexual function provides numerous targets for pharmacologic intervention, including:²

• endocrine factors, such as testosterone
• neurotransmitters, such as serotonin.

The use of pharmacologic treatments for sex offenders is off-label, and evidence is limited. In general, pharmacologic treatments are geared toward reducing sexual drive through nonhormonal or hormonal means (Table 1).^3-5

Table 1: Pharmacologic treatment of male sex offenders: A risk-based approach

*Not available in the United States
^†Some authors suggest administering this dosage once every 2 weeks
CPA: cyproterone acetate; GNRH: gonadotropin-releasing hormone; IM: intramuscular; MPA: medroxyprogesterone acetate; SSRI: selective serotonin reuptake inhibitor
Source: References 3-5

Nonhormonal treatments

SSRIs. Selective serotonin reuptake inhibitors act by blocking serotonin reuptake in the synaptic cleft. Soon after the first SSRIs were approved in 1988, reports appeared of SSRIs interfering with sexual functioning.⁶ This side effect quickly was exploited to assist the treatment of sexual offenders.⁷

The mechanism of action may include:⁸

• direct effects, such as general inhibition of sexual activity, reduced impulsiveness, and an effect on the hypothesized “obsessive-compulsive” nature of paraphilias⁹
• indirect reduction of testosterone.

A growing body of literature supports SSRIs’ effectiveness in treating paraphilias and sexual offenders. Greenberg⁷ reviewed case studies and open drug trials of nearly 200 patients receiving fluoxetine, fluvoxamine, or sertraline. Most studies showed response rates of 50% to 90%.¹⁰ Positive effects included decreases in:

• paraphiliac fantasies, urges, and sexual acts
• masturbation
• hypersexual activity.

Some studies reported a preferential decrease in paraphiliac interests with an increase in conventional sexual interests, although this may be related to placebo or halo effects—patients may have reported an increase in conventional interests because they noticed a decrease in paraphiliac interests. Negative side effects included decreased sexual desires, delayed ejaculation, decreased libido, and anorgasmia.

Adi et al¹¹ completed a more rigorous literature evaluation that included 9 studies with a total of 225 patients receiving fluoxetine, fluvoxamine, sertraline, or paroxetine. Eight studies showed benefits; however, Adi noted that this preliminary evidence was “far from conclusive.”

SSRIs generally are well tolerated and may be more appealing to patients than the “chemical castration” of hormonal treatments. Dosing is similar to that used in depression or obsessive-compulsive disorder. Although most patients notice beneficial effects in 2 to 4 weeks, some notice the effect nearly immediately.

Naltrexone. An opioid antagonist thought to affect the CNS processes of pleasure and pain, naltrexone has been used to treat alcohol dependence and pathologic gambling. A few case studies^12-14 and 1 study of 21 adolescent sex offenders¹⁵ have shown benefits in treating sexual offenders or paraphiliacs. Benefits were seen at 50 mg/d, with suggested dosing of 100 to 200 mg/d. Because data are very limited, consider naltrexone only on an individual basis or as a possible adjunctive treatment.

Psychostimulants. Methylphenidate was added to augment SSRI treatment in a study of 26 men with paraphilias or paraphilia-related disorders.¹⁶ Results included further significant decreases in total sexual outlets (orgasms per week) and average time spent per day in paraphilia and paraphilia-related behavior. These gains appeared to be independent of the presence of attention-deficit/hyperactivity disorder.

Again, because data are very limited, consider this strategy only on an individual basis or as a possible adjunctive treatment. Because sexual offenders have high rates of substance abuse,¹⁷ consider the potential for stimulant abuse.

Hormonal treatments

Because testosterone is required for healthy bone metabolism, the antiandrogen medications used in hormonal treatment can cause osteoporosis.^18,19 Therefore, long-term antiandrogen treatment should include bone scans to monitor for osteopenia and osteoporosis. Some authors have suggested that monthly doses of 25 to 50 mg of testosterone could minimize this risk.²⁰ Bisphosphonates, vitamin D, and calcium supplements at osteoporosis treatment levels might be helpful.¹⁸ Other common side effects of antiandrogen medications are listed in Table 2.

 

Finasteride is approved for treating benign prostatic hyperplasia and androgenetic alopecia. It works by preventing conversion of testosterone to dihydrotestosterone (DHT) by the type II isoenzyme.²¹ Serum DHT contributes to male sexual behavior and predicts frequency of orgasms in healthy volunteers.²² Although there have been no studies of finasteride in sex offenders, it may be more acceptable to patients than other hormonal treatments and have a theoretical benefit in reducing sexual drive. Clinically, some patients describe increased control over urges without substantial side effects. Because there is no evidence supporting finasteride use in sex offenders, consider this medication only on an individual basis or as a possible adjunctive treatment.

Cyproterone acetate (CPA) is a synthetic steroid that blocks androgen receptors.^3,23 Some evidence supports its use in treating sex offenders,²⁴ although this agent is not available in the United States. For more information about CPA, see Box.

Medroxyprogesterone acetate (MPA), a derivative of progesterone, lowers serum testosterone by inhibiting its production through reducing pituitary luteinizing hormone (LH).²⁴ The typical dose range for use in sex offenders is 100 to 600 mg/d orally or 100 to 700 mg IM every week,³ although some authors suggest giving similar doses every 2 weeks.⁴

Side effects of MPA include hypersomnia; neurasthenia; weight gain; hot flashes; gynecomastia; increased scalp hair; and decreased erections, ejaculate volume, spermatogenesis, and body and facial hair. The drug decreases testosterone levels by about 50%. Positive effects include reduced interest in and energy spent on pursuing sexual goals, but preservation of nondeviant sexual arousal.⁴

MPA has been shown to effectively decrease deviant sexual arousal and recidivism. In a study of 100 patients receiving MPA (average 250 mg IM every 2 weeks) for an average of 3 years, only 1 re-offended while taking MPA.⁴

In a 5-year follow-up study²⁵ of 275 men, subjects were classified into high risk/treatment with MPA, 200 to 400 mg IM every 2 weeks, and low risk/nontreatment groups. A portion of the high risk/treatment group did not receive MPA. No sexual re-offenses occurred among high-risk subjects who received MPA, whereas the recidivism rate was 18% among high-risk subjects who did not receive MPA. Subjects in the low risk/nontreatment group had a recidivism rate of 15%, which suggests the need for more liberal use of antiandrogens. One major confounding factor was that subjects in the high risk/treatment group had to report every 2 weeks for injections; this may have resulted in closer follow-up, monitoring, and support, which may have contributed to lower recidivism.

Gonadotropin-releasing hormone (GNRH) agonists. The terms gonadotropin-releasing hormone and luteinizing-releasing hormone are used interchangeably. Most body testosterone is produced and released by Leydig cells in the testes in response to stimulation by LH released by the pituitary gland. LH release is controlled by the pulsatile release of GNRH from the hypothalamus. GNRH agonists are high-potency analogs of GNRH that work by causing an initial surge of LH followed by down-regulation of gonadotroph cells, a drop in LH, and a drop in testosterone to castration levels.

The GNRH analogs leuprolide, goserelin, and triptorelin are used to treat paraphiliacs and sexual offenders.²⁰ Leuprolide typically is dosed at 7.5 mg IM every month, 22.5 mg IM every 3 months, or 30 mg IM every 4 months. Goserelin is provided as a subcutaneous implant/depot injected as 3.6 mg every month or 10.8 mg every 3 months.

Triptorelin is FDA-approved as treatment for advanced prostate cancer. Triptorelin is given in depot formulation as 3.75 mg IM every month or in a long-acting form as 11.25 mg IM every 3 months.

When starting these medications, an initial surge of LH and testosterone can last up to 3 weeks.²⁶ Theoretically, this could worsen paraphiliac interests. Many practitioners will use a testosterone blocker such as flutamide, 250 mg tid, for the initial weeks of treatment.

Side effects of the GNRH agonists are similar. Most patients initially experience hot flashes. A systemic literature review²⁷ reported:

• weight gain
• perspiration
• gynecomastia
• urinary incontinence
• hot flashes
• decreased growth of facial and body hair
• asthenia
• erectile failure
• muscle tenderness
• frequent bone demineralization.

Rare cases of pituitary apoplexy (a clinical syndrome secondary to infarction of the pituitary gland) have been reported, possibly related to an underlying pituitary adenoma.²⁸

In a literature review that totalled 118 patients,²⁷ GNRH agonists significantly decreased erections, ejaculations, paraphiliac fantasies, and paraphiliac behavior. Patients also reported feeling more relaxed, and recidivism rates were low. Some patients who failed to respond to CPA and MPA responded to GNRH agonists. Subsequent studies found similar results.^29,30

 

Box

Table 2

Common side effects of antiandrogen therapy

Depression
Erectile dysfunction
Fatigue
Gynecomastia
Hot flashes
Hypertension
Low libido
Myalgia
Osteopenia
Osteoporosis
Sweating
Thromboembolism
Weight gain

Monitoring

Laboratory investigations are recommended to monitor for side effects of antiandrogen medications (Table 3).^19,27,31 Medical contraindications to rule out before initiating antiandrogen medications include:

• thromboembolic diseases
• liver disease
• bone demineralization disorders
• hypersensitivity to the drug.

Measure prolactin to rule out pituitary adenomas. Monitor serum testosterone because some patients will not experience testosterone suppression from GNRH agonists or other antiandrogens. Noncompliant patients could potentially reverse the effects of MPA and GNRH agonists by taking exogenous testosterone.

Table 3

Monitoring patients receiving antiandrogen medications

Pre-therapy workup	Periodic monitoring
Endocrinology or internist consultation Bone scan Weight Blood pressure Electrocardiogram CBC, renal function, liver function, fasting glucose, and lipids LH, FSH, testosterone, prolactin	Monthly: testosterone for the first 6 months Every 6 months: testosterone, LH, FSH, prolactin, CBC, renal function, liver function, fasting glucose and lipids, weight, blood pressure Yearly: bone scan
CBC: complete blood count; FSH: follicle-stimulating hormone; LH: luteinizing hormone
Source: References 19,27,31

Medication selection

The goals of pharmacologic treatment of sex offenders are to:

• reduce sexual offending and victimization
• suppress sexual drive to a controllable level
• possibly preferentially eliminate deviant arousal/thoughts
• allow normal sexual relationships.

Gauging risk. In determining which pharmacologic treatment to offer a patient, first evaluate the individual’s risk for recidivism. Actuarial scales^32,33 suggest that recidivism risk can be categorized, based on clinical factors (Table 4).^4,25,34

In addition to statistical risk factors, several other factors affect medication selection. Self-referred individuals may be more reliable in taking oral medications than those referred by the courts. A developmentally delayed individual may be a poor candidate for oral medication, unless he resides in a group home setting where compliance can be assured. Efficacy also guides medication choice. Finally, some patients will be legally required to provide proof of compliance, which only IM medications provide.

Treatment. Based on clinical experience and available literature, Bradford⁵ created an algorithm to help clinicians select appropriate pharmacologic interventions. Although it has not been validated, this algorithm provides a reasonable starting place.

In general, start treatment with an SSRI for low-risk individuals (Table 1).^3-5 If this strategy is insufficient, consider augmentation with methylphenidate, naltrexone, or finasteride.

The next step would be to add oral MPA or CPA, 50 mg/d, which would partially inhibit testosterone and may allow some normal sexual functioning.^4,23 Higher-dose oral MPA or CPA would be tried next. For higher-risk individuals or treatment failures, IM MPA or CPA would be offered next, followed by a GNRH agonist. For individuals at highest risk for re-offending, combinations of agents may be indicated.

This simple strategy is appealing, but in reality, treatment should be individualized. Choose medications based on the patient’s risk, wishes, and the previously mentioned clinical factors.

Table 4: Will my patient commit another sexual offense? Evaluating risk

Source: References 4,25,34Related resources

• Krueger RB, Kaplan MS. The paraphilic and hypersexual disorders: an overview. J Psychiatr Pract. 2001;7:391-403.
• Krueger RB, Kaplan MS. Behavioral and psychopharmacological treatment of the paraphilic and hypersexual disorders. J Psychiatr Pract. 2002;8:21-32.
• Association for the Treatment of Sexual Abusers. www.atsa.com.

 

Drug brand names

• Cyproterone acetate • Androcur
• Finasteride • Propecia, Proscar
• Fluoxetine • Prozac
• Flutamide • Eulexin
• Fluvoxamine • Luvox
• Goserelin • Zoladex
• Leuprolide • Eligard, Lupron
• Medroxyprogesterone acetate • Depo-Provera, Provera
• Methylphenidate • Ritalin
• Naltrexone • ReVia
• Paroxetine • Paxil
• Sertraline • Zoloft
• Triptorelin • Trelstar Depot

Disclosure

Dr. Booth reports no financial relationship with any company whose products are mentioned in the article or with manufacturers of competing products.

Comment on this article

Sex offenders traditionally are managed by the criminal justice system, but psychiatrists are frequently called on to assess and treat these individuals. Part of the reason is the overlap of paraphilias (disorders of sexual preference) and sexual offending. Many sexual offenders do not meet DSM criteria for paraphilias,¹ however, and individuals with paraphilias do not necessarily commit offenses or come into contact with the legal system.

As clinicians, we may need to assess and treat a wide range of sexual issues, from persons with paraphilias who are self-referred and have no legal involvement, to recurrent sexual offenders who are at a high risk of repeat offending. Successfully managing sex offenders includes psychological and pharmacologic interventions and possibly incarceration and post-incarceration surveillance. This article focuses on pharmacologic interventions for male sexual offenders.

Reducing sexual drive

Sex offending likely is the result of a complex interplay of environment and psychological and biologic factors. The biology of sexual function provides numerous targets for pharmacologic intervention, including:²

• endocrine factors, such as testosterone
• neurotransmitters, such as serotonin.

The use of pharmacologic treatments for sex offenders is off-label, and evidence is limited. In general, pharmacologic treatments are geared toward reducing sexual drive through nonhormonal or hormonal means (Table 1).^3-5

Table 1: Pharmacologic treatment of male sex offenders: A risk-based approach

*Not available in the United States
^†Some authors suggest administering this dosage once every 2 weeks
CPA: cyproterone acetate; GNRH: gonadotropin-releasing hormone; IM: intramuscular; MPA: medroxyprogesterone acetate; SSRI: selective serotonin reuptake inhibitor
Source: References 3-5

Nonhormonal treatments

SSRIs. Selective serotonin reuptake inhibitors act by blocking serotonin reuptake in the synaptic cleft. Soon after the first SSRIs were approved in 1988, reports appeared of SSRIs interfering with sexual functioning.⁶ This side effect quickly was exploited to assist the treatment of sexual offenders.⁷

The mechanism of action may include:⁸

• direct effects, such as general inhibition of sexual activity, reduced impulsiveness, and an effect on the hypothesized “obsessive-compulsive” nature of paraphilias⁹
• indirect reduction of testosterone.

A growing body of literature supports SSRIs’ effectiveness in treating paraphilias and sexual offenders. Greenberg⁷ reviewed case studies and open drug trials of nearly 200 patients receiving fluoxetine, fluvoxamine, or sertraline. Most studies showed response rates of 50% to 90%.¹⁰ Positive effects included decreases in:

• paraphiliac fantasies, urges, and sexual acts
• masturbation
• hypersexual activity.

Some studies reported a preferential decrease in paraphiliac interests with an increase in conventional sexual interests, although this may be related to placebo or halo effects—patients may have reported an increase in conventional interests because they noticed a decrease in paraphiliac interests. Negative side effects included decreased sexual desires, delayed ejaculation, decreased libido, and anorgasmia.

Adi et al¹¹ completed a more rigorous literature evaluation that included 9 studies with a total of 225 patients receiving fluoxetine, fluvoxamine, sertraline, or paroxetine. Eight studies showed benefits; however, Adi noted that this preliminary evidence was “far from conclusive.”

SSRIs generally are well tolerated and may be more appealing to patients than the “chemical castration” of hormonal treatments. Dosing is similar to that used in depression or obsessive-compulsive disorder. Although most patients notice beneficial effects in 2 to 4 weeks, some notice the effect nearly immediately.

Naltrexone. An opioid antagonist thought to affect the CNS processes of pleasure and pain, naltrexone has been used to treat alcohol dependence and pathologic gambling. A few case studies^12-14 and 1 study of 21 adolescent sex offenders¹⁵ have shown benefits in treating sexual offenders or paraphiliacs. Benefits were seen at 50 mg/d, with suggested dosing of 100 to 200 mg/d. Because data are very limited, consider naltrexone only on an individual basis or as a possible adjunctive treatment.

Psychostimulants. Methylphenidate was added to augment SSRI treatment in a study of 26 men with paraphilias or paraphilia-related disorders.¹⁶ Results included further significant decreases in total sexual outlets (orgasms per week) and average time spent per day in paraphilia and paraphilia-related behavior. These gains appeared to be independent of the presence of attention-deficit/hyperactivity disorder.

Again, because data are very limited, consider this strategy only on an individual basis or as a possible adjunctive treatment. Because sexual offenders have high rates of substance abuse,¹⁷ consider the potential for stimulant abuse.

Hormonal treatments

Because testosterone is required for healthy bone metabolism, the antiandrogen medications used in hormonal treatment can cause osteoporosis.^18,19 Therefore, long-term antiandrogen treatment should include bone scans to monitor for osteopenia and osteoporosis. Some authors have suggested that monthly doses of 25 to 50 mg of testosterone could minimize this risk.²⁰ Bisphosphonates, vitamin D, and calcium supplements at osteoporosis treatment levels might be helpful.¹⁸ Other common side effects of antiandrogen medications are listed in Table 2.

 

Finasteride is approved for treating benign prostatic hyperplasia and androgenetic alopecia. It works by preventing conversion of testosterone to dihydrotestosterone (DHT) by the type II isoenzyme.²¹ Serum DHT contributes to male sexual behavior and predicts frequency of orgasms in healthy volunteers.²² Although there have been no studies of finasteride in sex offenders, it may be more acceptable to patients than other hormonal treatments and have a theoretical benefit in reducing sexual drive. Clinically, some patients describe increased control over urges without substantial side effects. Because there is no evidence supporting finasteride use in sex offenders, consider this medication only on an individual basis or as a possible adjunctive treatment.

Cyproterone acetate (CPA) is a synthetic steroid that blocks androgen receptors.^3,23 Some evidence supports its use in treating sex offenders,²⁴ although this agent is not available in the United States. For more information about CPA, see Box.

Medroxyprogesterone acetate (MPA), a derivative of progesterone, lowers serum testosterone by inhibiting its production through reducing pituitary luteinizing hormone (LH).²⁴ The typical dose range for use in sex offenders is 100 to 600 mg/d orally or 100 to 700 mg IM every week,³ although some authors suggest giving similar doses every 2 weeks.⁴

Side effects of MPA include hypersomnia; neurasthenia; weight gain; hot flashes; gynecomastia; increased scalp hair; and decreased erections, ejaculate volume, spermatogenesis, and body and facial hair. The drug decreases testosterone levels by about 50%. Positive effects include reduced interest in and energy spent on pursuing sexual goals, but preservation of nondeviant sexual arousal.⁴

MPA has been shown to effectively decrease deviant sexual arousal and recidivism. In a study of 100 patients receiving MPA (average 250 mg IM every 2 weeks) for an average of 3 years, only 1 re-offended while taking MPA.⁴

In a 5-year follow-up study²⁵ of 275 men, subjects were classified into high risk/treatment with MPA, 200 to 400 mg IM every 2 weeks, and low risk/nontreatment groups. A portion of the high risk/treatment group did not receive MPA. No sexual re-offenses occurred among high-risk subjects who received MPA, whereas the recidivism rate was 18% among high-risk subjects who did not receive MPA. Subjects in the low risk/nontreatment group had a recidivism rate of 15%, which suggests the need for more liberal use of antiandrogens. One major confounding factor was that subjects in the high risk/treatment group had to report every 2 weeks for injections; this may have resulted in closer follow-up, monitoring, and support, which may have contributed to lower recidivism.

Gonadotropin-releasing hormone (GNRH) agonists. The terms gonadotropin-releasing hormone and luteinizing-releasing hormone are used interchangeably. Most body testosterone is produced and released by Leydig cells in the testes in response to stimulation by LH released by the pituitary gland. LH release is controlled by the pulsatile release of GNRH from the hypothalamus. GNRH agonists are high-potency analogs of GNRH that work by causing an initial surge of LH followed by down-regulation of gonadotroph cells, a drop in LH, and a drop in testosterone to castration levels.

The GNRH analogs leuprolide, goserelin, and triptorelin are used to treat paraphiliacs and sexual offenders.²⁰ Leuprolide typically is dosed at 7.5 mg IM every month, 22.5 mg IM every 3 months, or 30 mg IM every 4 months. Goserelin is provided as a subcutaneous implant/depot injected as 3.6 mg every month or 10.8 mg every 3 months.

Triptorelin is FDA-approved as treatment for advanced prostate cancer. Triptorelin is given in depot formulation as 3.75 mg IM every month or in a long-acting form as 11.25 mg IM every 3 months.

When starting these medications, an initial surge of LH and testosterone can last up to 3 weeks.²⁶ Theoretically, this could worsen paraphiliac interests. Many practitioners will use a testosterone blocker such as flutamide, 250 mg tid, for the initial weeks of treatment.

Side effects of the GNRH agonists are similar. Most patients initially experience hot flashes. A systemic literature review²⁷ reported:

• weight gain
• perspiration
• gynecomastia
• urinary incontinence
• hot flashes
• decreased growth of facial and body hair
• asthenia
• erectile failure
• muscle tenderness
• frequent bone demineralization.

Rare cases of pituitary apoplexy (a clinical syndrome secondary to infarction of the pituitary gland) have been reported, possibly related to an underlying pituitary adenoma.²⁸

In a literature review that totalled 118 patients,²⁷ GNRH agonists significantly decreased erections, ejaculations, paraphiliac fantasies, and paraphiliac behavior. Patients also reported feeling more relaxed, and recidivism rates were low. Some patients who failed to respond to CPA and MPA responded to GNRH agonists. Subsequent studies found similar results.^29,30

 

Box

Table 2

Common side effects of antiandrogen therapy

Depression
Erectile dysfunction
Fatigue
Gynecomastia
Hot flashes
Hypertension
Low libido
Myalgia
Osteopenia
Osteoporosis
Sweating
Thromboembolism
Weight gain

Monitoring

Laboratory investigations are recommended to monitor for side effects of antiandrogen medications (Table 3).^19,27,31 Medical contraindications to rule out before initiating antiandrogen medications include:

• thromboembolic diseases
• liver disease
• bone demineralization disorders
• hypersensitivity to the drug.

Measure prolactin to rule out pituitary adenomas. Monitor serum testosterone because some patients will not experience testosterone suppression from GNRH agonists or other antiandrogens. Noncompliant patients could potentially reverse the effects of MPA and GNRH agonists by taking exogenous testosterone.

Table 3

Monitoring patients receiving antiandrogen medications

Pre-therapy workup	Periodic monitoring
Endocrinology or internist consultation Bone scan Weight Blood pressure Electrocardiogram CBC, renal function, liver function, fasting glucose, and lipids LH, FSH, testosterone, prolactin	Monthly: testosterone for the first 6 months Every 6 months: testosterone, LH, FSH, prolactin, CBC, renal function, liver function, fasting glucose and lipids, weight, blood pressure Yearly: bone scan
CBC: complete blood count; FSH: follicle-stimulating hormone; LH: luteinizing hormone
Source: References 19,27,31

Medication selection

The goals of pharmacologic treatment of sex offenders are to:

• reduce sexual offending and victimization
• suppress sexual drive to a controllable level
• possibly preferentially eliminate deviant arousal/thoughts
• allow normal sexual relationships.

Gauging risk. In determining which pharmacologic treatment to offer a patient, first evaluate the individual’s risk for recidivism. Actuarial scales^32,33 suggest that recidivism risk can be categorized, based on clinical factors (Table 4).^4,25,34

In addition to statistical risk factors, several other factors affect medication selection. Self-referred individuals may be more reliable in taking oral medications than those referred by the courts. A developmentally delayed individual may be a poor candidate for oral medication, unless he resides in a group home setting where compliance can be assured. Efficacy also guides medication choice. Finally, some patients will be legally required to provide proof of compliance, which only IM medications provide.

Treatment. Based on clinical experience and available literature, Bradford⁵ created an algorithm to help clinicians select appropriate pharmacologic interventions. Although it has not been validated, this algorithm provides a reasonable starting place.

In general, start treatment with an SSRI for low-risk individuals (Table 1).^3-5 If this strategy is insufficient, consider augmentation with methylphenidate, naltrexone, or finasteride.

The next step would be to add oral MPA or CPA, 50 mg/d, which would partially inhibit testosterone and may allow some normal sexual functioning.^4,23 Higher-dose oral MPA or CPA would be tried next. For higher-risk individuals or treatment failures, IM MPA or CPA would be offered next, followed by a GNRH agonist. For individuals at highest risk for re-offending, combinations of agents may be indicated.

This simple strategy is appealing, but in reality, treatment should be individualized. Choose medications based on the patient’s risk, wishes, and the previously mentioned clinical factors.

Table 4: Will my patient commit another sexual offense? Evaluating risk

Source: References 4,25,34Related resources

• Krueger RB, Kaplan MS. The paraphilic and hypersexual disorders: an overview. J Psychiatr Pract. 2001;7:391-403.
• Krueger RB, Kaplan MS. Behavioral and psychopharmacological treatment of the paraphilic and hypersexual disorders. J Psychiatr Pract. 2002;8:21-32.
• Association for the Treatment of Sexual Abusers. www.atsa.com.

 

Drug brand names

• Cyproterone acetate • Androcur
• Finasteride • Propecia, Proscar
• Fluoxetine • Prozac
• Flutamide • Eulexin
• Fluvoxamine • Luvox
• Goserelin • Zoladex
• Leuprolide • Eligard, Lupron
• Medroxyprogesterone acetate • Depo-Provera, Provera
• Methylphenidate • Ritalin
• Naltrexone • ReVia
• Paroxetine • Paxil
• Sertraline • Zoloft
• Triptorelin • Trelstar Depot

Disclosure

Dr. Booth reports no financial relationship with any company whose products are mentioned in the article or with manufacturers of competing products.

Comment on this article

Treatment of borderline personality disorder (BPD) often is viewed as challenging and the results so discouraging that some clinicians avoid referrals of BPD patients.^1-3 Psychotherapy has been the treatment mainstay for decades, and supportive approaches are probably the most widely employed.⁴ Psychodynamic therapy often has been recommended.

This article introduces a new evidence-based group treatment program that we developed for BPD patients. Systems Training for Emotional Predictability and Problem Solving (STEPPS) is founded on the successes of better known psychoeducational models but is easier for practicing psychiatrists to implement.

A different approach to BPD

Linehan⁵ introduced dialectical behavior therapy (DBT)—a manualized, time-limited, cognitive-behavioral approach in which patients learn to regulate their emotions and behaviors rather than change their personality structure. Other evidence-based BPD treatments include transference-focused psychotherapy,⁶ schema-focused psychotherapy,⁷ and Bateman and Fonagy’s mentalization program.⁸ For a description of the unique challenges presented by BPD patients, see Box.

In the mid-1990s, we set out to create a treatment program for our BPD patients in response to managed care directives to lower the cost of care, decrease length of inpatient treatment, and reduce rehospitalization rates. Despite DBT’s many appealing features, we felt this model was too lengthy and labor-intensive for our treatment setting. We concluded that modifying a program developed by Bartels and Crotty⁹ would serve our needs. This 12-week psychoeducational program:

• employs established cognitive-behavioral techniques in group treatment intended to supplement but not replace patients’ ongoing treatment
  
• incorporates a “systems” component that recognizes the importance of the patient’s family and friends.
  

We adapted Bartels and Crotty’s manual (with permission), lengthened the program to 20 weeks, and developed specific session agendas with explicit facilitator guidelines.

We eventually renamed the program Systems Training for Emotional Predictability and Problem Solving (STEPPS)¹⁰ and created a new manual (see Related Resources) to simplify group leader training and ensure fidelity to the model. Data from 5 studies, including 2 randomized controlled trials (Table 1), show that STEPPS has a robust antidepressant effect and leads to broad-based improvements in the affective, cognitive, impulsive, and disturbed relationship domains of BPD.^11-15

Box

 

Table 1

STEPPS: Trials show improvement across BPD domains

Study	Patients	Results
Uncontrolled trials
Blum et al, 200211	52 outpatients; 94% female; mean age 33	Significant improvement in BEST score; significant drop in BDI score and the PANAS negative affect scale
Black et al, 200812	12 incarcerated women; mean age 35	Significant improvement in BEST score; significant drop in BDI score and the PANAS negative affect scale
Freije et al, 200213	85 patients; 91% female; mean age 32	Significant improvement in score on a Dutch version of BEST; significant improvement on SCL-90 subscales, especially those rating anxiety, depression, and interpersonal sensitivity
Randomized controlled trials
Blum et al, 200814	165 adults with BPD assigned to STEPPS plus treatment as usual or only treatment as usual	Patients receiving STEPPS plus treatment as usual experienced greater improvements in ZAN-BPD total score, impulsivity, negative affect, mood, and global functioning
van Wel, 200715	79 adults with BPD assigned to STEPPS plus treatment as usual or only treatment as usuals	Patients receiving STEPPS plus treatment as usual had greater improvements in global psychiatric symptoms using the SCL-90, BPD symptoms, and quality of life measures at the end of treatment and at 6-month follow-up
BDI: Beck Depression Inventory; BEST: Borderline Evaluation of Severity Over Time; BPD: borderline personality disorder; PANAS: Positive and Negative Affect Scale; SCL-90: Symptoms Checklist-90; STEPPS: Systems Training for Emotional Predictability and Problem Solving; ZAN-BPD: Zanarini Rating Scale for Borderline Personality Disorder

STEPPS’ theoretical foundation

Because STEPPS employs general psychotherapy principles and techniques commonly taught in graduate-level psychotherapy training programs, it requires little additional training for mental health workers.¹⁶ Further, because it supplements ongoing treatment, STEPPS:

• does not disrupt the patient’s present regimen, and
  
• potentially enhances relationship skills by encouraging the patient to remain in longer relationships with professional and non-professional support.
  

STEPPS also integrates the patient’s ongoing social and professional support system, thereby avoiding the perception of abandonment common among patients with BPD.

STEPPS employs cognitive-behavioral methods, including identifying and challenging distorted thoughts and specific behavioral change, combined with psycho-education and skills training.^11,12 The addition of a systems component that enlists the help of the patient’s family and friends is unique to STEPPS (Box 1).

Emotional intensity disorder. Many clinicians assume that the core deficit in BPD is inability to manage emotional intensity. In STEPPS, therapists reframe BPD as emotional intensity disorder (EID), a term patients find easier to understand and accept. Patients tend to “see themselves as driven by the disorder to seek relief from a painful illness through desperate behaviors that are reinforced by negative and distorted thinking.”¹⁶ Starting with the first session, STEPPS therapists validate the patients’ experience of BPD and provide hope by teaching that patients can acquire skills to manage the disorder.

Box 1

 

Group format

STEPPS consists of 20 consecutive weekly, 2-hour sessions led by 2 therapists (we prefer the term “facilitators”). Sessions take place in a classroom setting and are highly structured, with specific facilitator guidelines for each session.

When they arrive at each session, patients fill out the Borderline Evaluation of Severity Over Time (BEST) scale (Box 2)^11,17 and record the results on a graph to measure their progress. Each session has a specific handout, including an agenda, followed by the homework assignment for the next week. Participants read the handout material aloud during the group session and start the homework assignment to be sure they understand it.

Handouts also include poems, essays, drawings, and examples created by previous STEPPS participants; these provide a sense of ownership among participants past, present, and future. Participants are encouraged to share their own writings and drawings, as well as other resources they have found helpful to illustrate the skills being taught.

Skills training. Facilitators introduce a new skill at each session, and each skill builds on previously taught skills. A recurring theme in STEPPS is that “most of the work is done between sessions”—during the week, patients are expected to practice the skill taught at the previous session. Using the STEPPS skills is framed as “change from the outside in.” As patients challenge maladaptive filters and distorted cognitions, they find that negative feelings and dysfunctional behaviors change.

Patients identify their use of specific skills by completing a 5-point Emotional Intensity Continuum (EIC) scale. This abstract concept is made concrete with drawings of pots on a burner. At level 1 (baseline), there is no heat under the burner; at level 5, the pot is boiling over.

Box 2

3 components of STEPPS

Awareness of illness. The first step is for patients to replace misconceptions about BPD with awareness of the behaviors and feelings that define the disorder. They are provided with a printed handout listing DSM-IV criteria for BPD and given time to acknowledge examples in their own behavior. This is called “owning” the illness.

The second step is to introduce the concept of schemas, referred to as cognitive filters. With the author’s permission, we extracted 64 items from Young’s Schema Questionnaire,¹⁸ which helps patients identify their early maladaptive schemas. We encourage patients to understand the relationship among these filters, DSM-IV criteria, and their subsequent pattern of feelings, thoughts, and behaviors.

Emotion management skills taught in STEPPS are distancing, communicating, challenging, distracting, and managing problems (Table 2). Using these skills, participants learn to:

• predict the course of emotional states
  
• anticipate stressful situations
  
• develop functional coping strategies.
  

Behavior management skills include goal-setting, sleep hygiene, diet and nutrition, exercise and physical health, relaxation and leisure, abuse avoidance, and interpersonal relationship management. Participants learn that following a daily routine and managing behaviors such as sleep and diet yields the energy needed to manage the disorder.

Table 2

Patient skills taught in STEPPS

Awareness of illness
Understand what BPD is
Reframe BPD as ‘emotional intensity disorder’
‘Own’ the illness
Identify and challenge dysfunctional schemas
Emotion management skills
Distancing: Provide distance from emotional intensity
Communicating: Describe and define feelings, physical sensations, thoughts, filters, action urges, and behaviors
Challenging: Identify distorted thinking and develop alternate ways of thinking
Distracting: Identify and engage in behaviors that lower emotional intensity or assist in getting through an episode without resorting to damaging behaviors
Managing problems: Identify and define problems, then plan and carry out action steps
Behavior management skills
Setting goals: Identify specific goals and develop strategies to manage specific problematic behaviors
Eating: Balanced diet
Sleeping: Good sleep hygiene
Exercising: Regular and balanced exercise
Leisure: Regular leisure activities
Physical health: Manage medical problems
Abuse avoidance: Develop strategies to replace abusive behaviors (self-harm, substance abuse, gambling, etc.)
Relationship management: Identify and determine strategies to develop healthy relationships. Understand and implement healthy boundaries
BPD: borderline personality disorder; STEPPS: Systems Training for Emotional Predictability and Problem Solving

 

Patient characteristics

Although some patients learn of STEPPS from previous participants, at our facility we usually request a formal professional referral. We then send potential participants a letter inviting them to attend the group, along with a brochure describing STEPPS and a dated syllabus. We generally begin with 12 to 15 patients but typically have 7 to 10 by the fifth session.

Patients with strong narcissistic or antisocial traits may have difficulty in group settings, probably because they prefer to be the center of attention. That said, we have successfully implemented STEPPS in Iowa prisons and have not experienced difficulties.¹² Patients who are abusing substances or have active eating disorders (primarily anorexia nervosa) may not be cognitively able to benefit from STEPPS until these behaviors are better controlled. We recommend that patients seek treatment for these behaviors before—or concurrent with—STEPPS participation.

Persons who deal with conflict by physical threats or intimidation are potential threats to group integrity and are removed immediately. We avoid forming groups with a lone male participant because:

• he may come to represent all men to the rest of the group
  
• he may have difficulty identifying with problems unique to women with BPD.
  

Patients who do well in STEPPS are able to share time with others, limit discussion of their own problems, have some capacity for empathy, and demonstrate an ability to consider that another’s perception may be different from their own. We encourage referring clinicians to discuss with the patient his or her readiness to enter STEPPS, as well as the requirements for and expectations of the program (such as capacity to listen, compliance with homework assignments, etc.). Some patients may need to wait until they are psychologically ready to participate.

In a recent study we found that patients who were rated as more symptomatic at baseline experienced the greatest improvement. Apart from this finding, there were few response predictors, but it was reassuring that both men and women improved.¹⁹ Members are cautiously encouraged to use each other as reinforcement team members between sessions, once they feel safe in the group. They are instructed to follow the reinforcement team guidelines.

The facilitators’ role

STEPPS groups are led by 2 facilitators with graduate level training in social sciences and psychotherapy experience. Therapists may be trained in STEPPS during a 1- to 2-day on-site workshop or by attending a 20-week group. These trainees are identified as professionals and do not participate in the sessions.

Using 1 male and 1 female facilitator for a STEPPS group allows modeling of relationship behaviors between genders, projects a healthy male role, and provides support for male participants, who in most groups are in the minority. Initially the facilitators’ stance is active and directive, although this tends to decrease as patients gradually are given increasing leadership responsibilities (such as leading brief reviews of homework assignments).

The therapists’ main tasks include:

• maintaining the psychoeducational format
  
• adhering to the guidelines
  
• avoiding involvement in individual issues and past traumas (providing individual psychotherapy in a group setting)
  
• maintaining focus on skills acquisition
  
• encouraging group cohesion through identification
  
• facilitating participants’ change of perspective from victims of EID to experts on managing EID.
  

Crises are common among patients with BPD and if not attended to appropriately can easily derail the group process. Crises are acknowledged, then managed in the group by careful attention to the use of skills (such as using a crisis as an example of applying the skill to be learned that session). Facilitators direct patients to deal with long-standing personal issues with their individual therapists. Patients who appear in imminent danger of self-harm or suicide are removed from the group and immediately referred to emergency personnel. On these rare occasions, the referral is done swiftly to avoid disrupting the group and creating a perception of special treatment.

Follow-up: STAIRWAYS

STAIRWAYS is a 1-year follow-up group that meets twice a month after the 20-week STEPPS program and consists of stand-alone modules addressing:

• Setting goals
  
• Trying new things (oriented toward long-term goals, such as obtaining a degree, employment, etc.)
  
• Anger management
  
• Impulsivity control
  
• Relationship management (emphasis on conflict management)
  
• Writing a script (identifying and preparing for future stressors)
  
• Assertiveness training
  
• Your choices (making healthy choices)
  
• Staying on track (relapse prevention).
  

STAIRWAYS follows a format similar to STEPPS, with a classroom-like setting and homework assignments. It maintains participants’ contact with the STEPPS model by emphasizing ongoing use of newly learned skills and reinforcing STEPPS skills.

 

Related resources

• The STEPPS Model for Borderline Personality Disorder Manual. www.steppsforbpd.com.
  
• van Wel B, Kockmann I, Blum N, et al. STEPPS group treatment for borderline personality disorder in The Netherlands. Ann Clin Psychiatry. 2006;18(1):63-67.
  

Disclosures

Dr. Black receives research/grant support from AstraZeneca and Forest Laboratories and is a consultant to Jazz Pharmaceuticals.

Ms. Blum and Mr. St. John receive royalties from Blums’ Books LLC, publisher of the Systems Training for Emotional Predictability and Problem Solving (STEPPS) manual.

Comment on this article

Treatment of borderline personality disorder (BPD) often is viewed as challenging and the results so discouraging that some clinicians avoid referrals of BPD patients.^1-3 Psychotherapy has been the treatment mainstay for decades, and supportive approaches are probably the most widely employed.⁴ Psychodynamic therapy often has been recommended.

This article introduces a new evidence-based group treatment program that we developed for BPD patients. Systems Training for Emotional Predictability and Problem Solving (STEPPS) is founded on the successes of better known psychoeducational models but is easier for practicing psychiatrists to implement.

A different approach to BPD

Linehan⁵ introduced dialectical behavior therapy (DBT)—a manualized, time-limited, cognitive-behavioral approach in which patients learn to regulate their emotions and behaviors rather than change their personality structure. Other evidence-based BPD treatments include transference-focused psychotherapy,⁶ schema-focused psychotherapy,⁷ and Bateman and Fonagy’s mentalization program.⁸ For a description of the unique challenges presented by BPD patients, see Box.

In the mid-1990s, we set out to create a treatment program for our BPD patients in response to managed care directives to lower the cost of care, decrease length of inpatient treatment, and reduce rehospitalization rates. Despite DBT’s many appealing features, we felt this model was too lengthy and labor-intensive for our treatment setting. We concluded that modifying a program developed by Bartels and Crotty⁹ would serve our needs. This 12-week psychoeducational program:

• employs established cognitive-behavioral techniques in group treatment intended to supplement but not replace patients’ ongoing treatment
  
• incorporates a “systems” component that recognizes the importance of the patient’s family and friends.
  

We adapted Bartels and Crotty’s manual (with permission), lengthened the program to 20 weeks, and developed specific session agendas with explicit facilitator guidelines.

We eventually renamed the program Systems Training for Emotional Predictability and Problem Solving (STEPPS)¹⁰ and created a new manual (see Related Resources) to simplify group leader training and ensure fidelity to the model. Data from 5 studies, including 2 randomized controlled trials (Table 1), show that STEPPS has a robust antidepressant effect and leads to broad-based improvements in the affective, cognitive, impulsive, and disturbed relationship domains of BPD.^11-15

Box

 

Table 1

STEPPS: Trials show improvement across BPD domains

Study	Patients	Results
Uncontrolled trials
Blum et al, 200211	52 outpatients; 94% female; mean age 33	Significant improvement in BEST score; significant drop in BDI score and the PANAS negative affect scale
Black et al, 200812	12 incarcerated women; mean age 35	Significant improvement in BEST score; significant drop in BDI score and the PANAS negative affect scale
Freije et al, 200213	85 patients; 91% female; mean age 32	Significant improvement in score on a Dutch version of BEST; significant improvement on SCL-90 subscales, especially those rating anxiety, depression, and interpersonal sensitivity
Randomized controlled trials
Blum et al, 200814	165 adults with BPD assigned to STEPPS plus treatment as usual or only treatment as usual	Patients receiving STEPPS plus treatment as usual experienced greater improvements in ZAN-BPD total score, impulsivity, negative affect, mood, and global functioning
van Wel, 200715	79 adults with BPD assigned to STEPPS plus treatment as usual or only treatment as usuals	Patients receiving STEPPS plus treatment as usual had greater improvements in global psychiatric symptoms using the SCL-90, BPD symptoms, and quality of life measures at the end of treatment and at 6-month follow-up
BDI: Beck Depression Inventory; BEST: Borderline Evaluation of Severity Over Time; BPD: borderline personality disorder; PANAS: Positive and Negative Affect Scale; SCL-90: Symptoms Checklist-90; STEPPS: Systems Training for Emotional Predictability and Problem Solving; ZAN-BPD: Zanarini Rating Scale for Borderline Personality Disorder

STEPPS’ theoretical foundation

Because STEPPS employs general psychotherapy principles and techniques commonly taught in graduate-level psychotherapy training programs, it requires little additional training for mental health workers.¹⁶ Further, because it supplements ongoing treatment, STEPPS:

• does not disrupt the patient’s present regimen, and
  
• potentially enhances relationship skills by encouraging the patient to remain in longer relationships with professional and non-professional support.
  

STEPPS also integrates the patient’s ongoing social and professional support system, thereby avoiding the perception of abandonment common among patients with BPD.

STEPPS employs cognitive-behavioral methods, including identifying and challenging distorted thoughts and specific behavioral change, combined with psycho-education and skills training.^11,12 The addition of a systems component that enlists the help of the patient’s family and friends is unique to STEPPS (Box 1).

Emotional intensity disorder. Many clinicians assume that the core deficit in BPD is inability to manage emotional intensity. In STEPPS, therapists reframe BPD as emotional intensity disorder (EID), a term patients find easier to understand and accept. Patients tend to “see themselves as driven by the disorder to seek relief from a painful illness through desperate behaviors that are reinforced by negative and distorted thinking.”¹⁶ Starting with the first session, STEPPS therapists validate the patients’ experience of BPD and provide hope by teaching that patients can acquire skills to manage the disorder.

Box 1

 

Group format

STEPPS consists of 20 consecutive weekly, 2-hour sessions led by 2 therapists (we prefer the term “facilitators”). Sessions take place in a classroom setting and are highly structured, with specific facilitator guidelines for each session.

When they arrive at each session, patients fill out the Borderline Evaluation of Severity Over Time (BEST) scale (Box 2)^11,17 and record the results on a graph to measure their progress. Each session has a specific handout, including an agenda, followed by the homework assignment for the next week. Participants read the handout material aloud during the group session and start the homework assignment to be sure they understand it.

Handouts also include poems, essays, drawings, and examples created by previous STEPPS participants; these provide a sense of ownership among participants past, present, and future. Participants are encouraged to share their own writings and drawings, as well as other resources they have found helpful to illustrate the skills being taught.

Skills training. Facilitators introduce a new skill at each session, and each skill builds on previously taught skills. A recurring theme in STEPPS is that “most of the work is done between sessions”—during the week, patients are expected to practice the skill taught at the previous session. Using the STEPPS skills is framed as “change from the outside in.” As patients challenge maladaptive filters and distorted cognitions, they find that negative feelings and dysfunctional behaviors change.

Patients identify their use of specific skills by completing a 5-point Emotional Intensity Continuum (EIC) scale. This abstract concept is made concrete with drawings of pots on a burner. At level 1 (baseline), there is no heat under the burner; at level 5, the pot is boiling over.

Box 2

3 components of STEPPS

Awareness of illness. The first step is for patients to replace misconceptions about BPD with awareness of the behaviors and feelings that define the disorder. They are provided with a printed handout listing DSM-IV criteria for BPD and given time to acknowledge examples in their own behavior. This is called “owning” the illness.

The second step is to introduce the concept of schemas, referred to as cognitive filters. With the author’s permission, we extracted 64 items from Young’s Schema Questionnaire,¹⁸ which helps patients identify their early maladaptive schemas. We encourage patients to understand the relationship among these filters, DSM-IV criteria, and their subsequent pattern of feelings, thoughts, and behaviors.

Emotion management skills taught in STEPPS are distancing, communicating, challenging, distracting, and managing problems (Table 2). Using these skills, participants learn to:

• predict the course of emotional states
  
• anticipate stressful situations
  
• develop functional coping strategies.
  

Behavior management skills include goal-setting, sleep hygiene, diet and nutrition, exercise and physical health, relaxation and leisure, abuse avoidance, and interpersonal relationship management. Participants learn that following a daily routine and managing behaviors such as sleep and diet yields the energy needed to manage the disorder.

Table 2

Patient skills taught in STEPPS

Awareness of illness
Understand what BPD is
Reframe BPD as ‘emotional intensity disorder’
‘Own’ the illness
Identify and challenge dysfunctional schemas
Emotion management skills
Distancing: Provide distance from emotional intensity
Communicating: Describe and define feelings, physical sensations, thoughts, filters, action urges, and behaviors
Challenging: Identify distorted thinking and develop alternate ways of thinking
Distracting: Identify and engage in behaviors that lower emotional intensity or assist in getting through an episode without resorting to damaging behaviors
Managing problems: Identify and define problems, then plan and carry out action steps
Behavior management skills
Setting goals: Identify specific goals and develop strategies to manage specific problematic behaviors
Eating: Balanced diet
Sleeping: Good sleep hygiene
Exercising: Regular and balanced exercise
Leisure: Regular leisure activities
Physical health: Manage medical problems
Abuse avoidance: Develop strategies to replace abusive behaviors (self-harm, substance abuse, gambling, etc.)
Relationship management: Identify and determine strategies to develop healthy relationships. Understand and implement healthy boundaries
BPD: borderline personality disorder; STEPPS: Systems Training for Emotional Predictability and Problem Solving

 

Patient characteristics

Although some patients learn of STEPPS from previous participants, at our facility we usually request a formal professional referral. We then send potential participants a letter inviting them to attend the group, along with a brochure describing STEPPS and a dated syllabus. We generally begin with 12 to 15 patients but typically have 7 to 10 by the fifth session.

Patients with strong narcissistic or antisocial traits may have difficulty in group settings, probably because they prefer to be the center of attention. That said, we have successfully implemented STEPPS in Iowa prisons and have not experienced difficulties.¹² Patients who are abusing substances or have active eating disorders (primarily anorexia nervosa) may not be cognitively able to benefit from STEPPS until these behaviors are better controlled. We recommend that patients seek treatment for these behaviors before—or concurrent with—STEPPS participation.

Persons who deal with conflict by physical threats or intimidation are potential threats to group integrity and are removed immediately. We avoid forming groups with a lone male participant because:

• he may come to represent all men to the rest of the group
  
• he may have difficulty identifying with problems unique to women with BPD.
  

Patients who do well in STEPPS are able to share time with others, limit discussion of their own problems, have some capacity for empathy, and demonstrate an ability to consider that another’s perception may be different from their own. We encourage referring clinicians to discuss with the patient his or her readiness to enter STEPPS, as well as the requirements for and expectations of the program (such as capacity to listen, compliance with homework assignments, etc.). Some patients may need to wait until they are psychologically ready to participate.

In a recent study we found that patients who were rated as more symptomatic at baseline experienced the greatest improvement. Apart from this finding, there were few response predictors, but it was reassuring that both men and women improved.¹⁹ Members are cautiously encouraged to use each other as reinforcement team members between sessions, once they feel safe in the group. They are instructed to follow the reinforcement team guidelines.

The facilitators’ role

STEPPS groups are led by 2 facilitators with graduate level training in social sciences and psychotherapy experience. Therapists may be trained in STEPPS during a 1- to 2-day on-site workshop or by attending a 20-week group. These trainees are identified as professionals and do not participate in the sessions.

Using 1 male and 1 female facilitator for a STEPPS group allows modeling of relationship behaviors between genders, projects a healthy male role, and provides support for male participants, who in most groups are in the minority. Initially the facilitators’ stance is active and directive, although this tends to decrease as patients gradually are given increasing leadership responsibilities (such as leading brief reviews of homework assignments).

The therapists’ main tasks include:

• maintaining the psychoeducational format
  
• adhering to the guidelines
  
• avoiding involvement in individual issues and past traumas (providing individual psychotherapy in a group setting)
  
• maintaining focus on skills acquisition
  
• encouraging group cohesion through identification
  
• facilitating participants’ change of perspective from victims of EID to experts on managing EID.
  

Crises are common among patients with BPD and if not attended to appropriately can easily derail the group process. Crises are acknowledged, then managed in the group by careful attention to the use of skills (such as using a crisis as an example of applying the skill to be learned that session). Facilitators direct patients to deal with long-standing personal issues with their individual therapists. Patients who appear in imminent danger of self-harm or suicide are removed from the group and immediately referred to emergency personnel. On these rare occasions, the referral is done swiftly to avoid disrupting the group and creating a perception of special treatment.

Follow-up: STAIRWAYS

STAIRWAYS is a 1-year follow-up group that meets twice a month after the 20-week STEPPS program and consists of stand-alone modules addressing:

• Setting goals
  
• Trying new things (oriented toward long-term goals, such as obtaining a degree, employment, etc.)
  
• Anger management
  
• Impulsivity control
  
• Relationship management (emphasis on conflict management)
  
• Writing a script (identifying and preparing for future stressors)
  
• Assertiveness training
  
• Your choices (making healthy choices)
  
• Staying on track (relapse prevention).
  

STAIRWAYS follows a format similar to STEPPS, with a classroom-like setting and homework assignments. It maintains participants’ contact with the STEPPS model by emphasizing ongoing use of newly learned skills and reinforcing STEPPS skills.

 

Related resources

• The STEPPS Model for Borderline Personality Disorder Manual. www.steppsforbpd.com.
  
• van Wel B, Kockmann I, Blum N, et al. STEPPS group treatment for borderline personality disorder in The Netherlands. Ann Clin Psychiatry. 2006;18(1):63-67.
  

Disclosures

Dr. Black receives research/grant support from AstraZeneca and Forest Laboratories and is a consultant to Jazz Pharmaceuticals.

Ms. Blum and Mr. St. John receive royalties from Blums’ Books LLC, publisher of the Systems Training for Emotional Predictability and Problem Solving (STEPPS) manual.

Comment on this article

Treatment of borderline personality disorder (BPD) often is viewed as challenging and the results so discouraging that some clinicians avoid referrals of BPD patients.^1-3 Psychotherapy has been the treatment mainstay for decades, and supportive approaches are probably the most widely employed.⁴ Psychodynamic therapy often has been recommended.

This article introduces a new evidence-based group treatment program that we developed for BPD patients. Systems Training for Emotional Predictability and Problem Solving (STEPPS) is founded on the successes of better known psychoeducational models but is easier for practicing psychiatrists to implement.

A different approach to BPD

Linehan⁵ introduced dialectical behavior therapy (DBT)—a manualized, time-limited, cognitive-behavioral approach in which patients learn to regulate their emotions and behaviors rather than change their personality structure. Other evidence-based BPD treatments include transference-focused psychotherapy,⁶ schema-focused psychotherapy,⁷ and Bateman and Fonagy’s mentalization program.⁸ For a description of the unique challenges presented by BPD patients, see Box.

In the mid-1990s, we set out to create a treatment program for our BPD patients in response to managed care directives to lower the cost of care, decrease length of inpatient treatment, and reduce rehospitalization rates. Despite DBT’s many appealing features, we felt this model was too lengthy and labor-intensive for our treatment setting. We concluded that modifying a program developed by Bartels and Crotty⁹ would serve our needs. This 12-week psychoeducational program:

• employs established cognitive-behavioral techniques in group treatment intended to supplement but not replace patients’ ongoing treatment
  
• incorporates a “systems” component that recognizes the importance of the patient’s family and friends.
  

We adapted Bartels and Crotty’s manual (with permission), lengthened the program to 20 weeks, and developed specific session agendas with explicit facilitator guidelines.

We eventually renamed the program Systems Training for Emotional Predictability and Problem Solving (STEPPS)¹⁰ and created a new manual (see Related Resources) to simplify group leader training and ensure fidelity to the model. Data from 5 studies, including 2 randomized controlled trials (Table 1), show that STEPPS has a robust antidepressant effect and leads to broad-based improvements in the affective, cognitive, impulsive, and disturbed relationship domains of BPD.^11-15

Box

 

Table 1

STEPPS: Trials show improvement across BPD domains

Study	Patients	Results
Uncontrolled trials
Blum et al, 200211	52 outpatients; 94% female; mean age 33	Significant improvement in BEST score; significant drop in BDI score and the PANAS negative affect scale
Black et al, 200812	12 incarcerated women; mean age 35	Significant improvement in BEST score; significant drop in BDI score and the PANAS negative affect scale
Freije et al, 200213	85 patients; 91% female; mean age 32	Significant improvement in score on a Dutch version of BEST; significant improvement on SCL-90 subscales, especially those rating anxiety, depression, and interpersonal sensitivity
Randomized controlled trials
Blum et al, 200814	165 adults with BPD assigned to STEPPS plus treatment as usual or only treatment as usual	Patients receiving STEPPS plus treatment as usual experienced greater improvements in ZAN-BPD total score, impulsivity, negative affect, mood, and global functioning
van Wel, 200715	79 adults with BPD assigned to STEPPS plus treatment as usual or only treatment as usuals	Patients receiving STEPPS plus treatment as usual had greater improvements in global psychiatric symptoms using the SCL-90, BPD symptoms, and quality of life measures at the end of treatment and at 6-month follow-up
BDI: Beck Depression Inventory; BEST: Borderline Evaluation of Severity Over Time; BPD: borderline personality disorder; PANAS: Positive and Negative Affect Scale; SCL-90: Symptoms Checklist-90; STEPPS: Systems Training for Emotional Predictability and Problem Solving; ZAN-BPD: Zanarini Rating Scale for Borderline Personality Disorder

STEPPS’ theoretical foundation

Because STEPPS employs general psychotherapy principles and techniques commonly taught in graduate-level psychotherapy training programs, it requires little additional training for mental health workers.¹⁶ Further, because it supplements ongoing treatment, STEPPS:

• does not disrupt the patient’s present regimen, and
  
• potentially enhances relationship skills by encouraging the patient to remain in longer relationships with professional and non-professional support.
  

STEPPS also integrates the patient’s ongoing social and professional support system, thereby avoiding the perception of abandonment common among patients with BPD.

STEPPS employs cognitive-behavioral methods, including identifying and challenging distorted thoughts and specific behavioral change, combined with psycho-education and skills training.^11,12 The addition of a systems component that enlists the help of the patient’s family and friends is unique to STEPPS (Box 1).

Emotional intensity disorder. Many clinicians assume that the core deficit in BPD is inability to manage emotional intensity. In STEPPS, therapists reframe BPD as emotional intensity disorder (EID), a term patients find easier to understand and accept. Patients tend to “see themselves as driven by the disorder to seek relief from a painful illness through desperate behaviors that are reinforced by negative and distorted thinking.”¹⁶ Starting with the first session, STEPPS therapists validate the patients’ experience of BPD and provide hope by teaching that patients can acquire skills to manage the disorder.

Box 1

 

Group format

STEPPS consists of 20 consecutive weekly, 2-hour sessions led by 2 therapists (we prefer the term “facilitators”). Sessions take place in a classroom setting and are highly structured, with specific facilitator guidelines for each session.

When they arrive at each session, patients fill out the Borderline Evaluation of Severity Over Time (BEST) scale (Box 2)^11,17 and record the results on a graph to measure their progress. Each session has a specific handout, including an agenda, followed by the homework assignment for the next week. Participants read the handout material aloud during the group session and start the homework assignment to be sure they understand it.

Handouts also include poems, essays, drawings, and examples created by previous STEPPS participants; these provide a sense of ownership among participants past, present, and future. Participants are encouraged to share their own writings and drawings, as well as other resources they have found helpful to illustrate the skills being taught.

Skills training. Facilitators introduce a new skill at each session, and each skill builds on previously taught skills. A recurring theme in STEPPS is that “most of the work is done between sessions”—during the week, patients are expected to practice the skill taught at the previous session. Using the STEPPS skills is framed as “change from the outside in.” As patients challenge maladaptive filters and distorted cognitions, they find that negative feelings and dysfunctional behaviors change.

Patients identify their use of specific skills by completing a 5-point Emotional Intensity Continuum (EIC) scale. This abstract concept is made concrete with drawings of pots on a burner. At level 1 (baseline), there is no heat under the burner; at level 5, the pot is boiling over.

Box 2

3 components of STEPPS

Awareness of illness. The first step is for patients to replace misconceptions about BPD with awareness of the behaviors and feelings that define the disorder. They are provided with a printed handout listing DSM-IV criteria for BPD and given time to acknowledge examples in their own behavior. This is called “owning” the illness.

The second step is to introduce the concept of schemas, referred to as cognitive filters. With the author’s permission, we extracted 64 items from Young’s Schema Questionnaire,¹⁸ which helps patients identify their early maladaptive schemas. We encourage patients to understand the relationship among these filters, DSM-IV criteria, and their subsequent pattern of feelings, thoughts, and behaviors.

Emotion management skills taught in STEPPS are distancing, communicating, challenging, distracting, and managing problems (Table 2). Using these skills, participants learn to:

• predict the course of emotional states
  
• anticipate stressful situations
  
• develop functional coping strategies.
  

Behavior management skills include goal-setting, sleep hygiene, diet and nutrition, exercise and physical health, relaxation and leisure, abuse avoidance, and interpersonal relationship management. Participants learn that following a daily routine and managing behaviors such as sleep and diet yields the energy needed to manage the disorder.

Table 2

Patient skills taught in STEPPS

Awareness of illness
Understand what BPD is
Reframe BPD as ‘emotional intensity disorder’
‘Own’ the illness
Identify and challenge dysfunctional schemas
Emotion management skills
Distancing: Provide distance from emotional intensity
Communicating: Describe and define feelings, physical sensations, thoughts, filters, action urges, and behaviors
Challenging: Identify distorted thinking and develop alternate ways of thinking
Distracting: Identify and engage in behaviors that lower emotional intensity or assist in getting through an episode without resorting to damaging behaviors
Managing problems: Identify and define problems, then plan and carry out action steps
Behavior management skills
Setting goals: Identify specific goals and develop strategies to manage specific problematic behaviors
Eating: Balanced diet
Sleeping: Good sleep hygiene
Exercising: Regular and balanced exercise
Leisure: Regular leisure activities
Physical health: Manage medical problems
Abuse avoidance: Develop strategies to replace abusive behaviors (self-harm, substance abuse, gambling, etc.)
Relationship management: Identify and determine strategies to develop healthy relationships. Understand and implement healthy boundaries
BPD: borderline personality disorder; STEPPS: Systems Training for Emotional Predictability and Problem Solving

 

Patient characteristics

Although some patients learn of STEPPS from previous participants, at our facility we usually request a formal professional referral. We then send potential participants a letter inviting them to attend the group, along with a brochure describing STEPPS and a dated syllabus. We generally begin with 12 to 15 patients but typically have 7 to 10 by the fifth session.

Patients with strong narcissistic or antisocial traits may have difficulty in group settings, probably because they prefer to be the center of attention. That said, we have successfully implemented STEPPS in Iowa prisons and have not experienced difficulties.¹² Patients who are abusing substances or have active eating disorders (primarily anorexia nervosa) may not be cognitively able to benefit from STEPPS until these behaviors are better controlled. We recommend that patients seek treatment for these behaviors before—or concurrent with—STEPPS participation.

Persons who deal with conflict by physical threats or intimidation are potential threats to group integrity and are removed immediately. We avoid forming groups with a lone male participant because:

• he may come to represent all men to the rest of the group
  
• he may have difficulty identifying with problems unique to women with BPD.
  

Patients who do well in STEPPS are able to share time with others, limit discussion of their own problems, have some capacity for empathy, and demonstrate an ability to consider that another’s perception may be different from their own. We encourage referring clinicians to discuss with the patient his or her readiness to enter STEPPS, as well as the requirements for and expectations of the program (such as capacity to listen, compliance with homework assignments, etc.). Some patients may need to wait until they are psychologically ready to participate.

In a recent study we found that patients who were rated as more symptomatic at baseline experienced the greatest improvement. Apart from this finding, there were few response predictors, but it was reassuring that both men and women improved.¹⁹ Members are cautiously encouraged to use each other as reinforcement team members between sessions, once they feel safe in the group. They are instructed to follow the reinforcement team guidelines.

The facilitators’ role

STEPPS groups are led by 2 facilitators with graduate level training in social sciences and psychotherapy experience. Therapists may be trained in STEPPS during a 1- to 2-day on-site workshop or by attending a 20-week group. These trainees are identified as professionals and do not participate in the sessions.

Using 1 male and 1 female facilitator for a STEPPS group allows modeling of relationship behaviors between genders, projects a healthy male role, and provides support for male participants, who in most groups are in the minority. Initially the facilitators’ stance is active and directive, although this tends to decrease as patients gradually are given increasing leadership responsibilities (such as leading brief reviews of homework assignments).

The therapists’ main tasks include:

• maintaining the psychoeducational format
  
• adhering to the guidelines
  
• avoiding involvement in individual issues and past traumas (providing individual psychotherapy in a group setting)
  
• maintaining focus on skills acquisition
  
• encouraging group cohesion through identification
  
• facilitating participants’ change of perspective from victims of EID to experts on managing EID.
  

Crises are common among patients with BPD and if not attended to appropriately can easily derail the group process. Crises are acknowledged, then managed in the group by careful attention to the use of skills (such as using a crisis as an example of applying the skill to be learned that session). Facilitators direct patients to deal with long-standing personal issues with their individual therapists. Patients who appear in imminent danger of self-harm or suicide are removed from the group and immediately referred to emergency personnel. On these rare occasions, the referral is done swiftly to avoid disrupting the group and creating a perception of special treatment.

Follow-up: STAIRWAYS

STAIRWAYS is a 1-year follow-up group that meets twice a month after the 20-week STEPPS program and consists of stand-alone modules addressing:

• Setting goals
  
• Trying new things (oriented toward long-term goals, such as obtaining a degree, employment, etc.)
  
• Anger management
  
• Impulsivity control
  
• Relationship management (emphasis on conflict management)
  
• Writing a script (identifying and preparing for future stressors)
  
• Assertiveness training
  
• Your choices (making healthy choices)
  
• Staying on track (relapse prevention).
  

STAIRWAYS follows a format similar to STEPPS, with a classroom-like setting and homework assignments. It maintains participants’ contact with the STEPPS model by emphasizing ongoing use of newly learned skills and reinforcing STEPPS skills.

 

Related resources

• The STEPPS Model for Borderline Personality Disorder Manual. www.steppsforbpd.com.
  
• van Wel B, Kockmann I, Blum N, et al. STEPPS group treatment for borderline personality disorder in The Netherlands. Ann Clin Psychiatry. 2006;18(1):63-67.
  

Disclosures

Dr. Black receives research/grant support from AstraZeneca and Forest Laboratories and is a consultant to Jazz Pharmaceuticals.

Ms. Blum and Mr. St. John receive royalties from Blums’ Books LLC, publisher of the Systems Training for Emotional Predictability and Problem Solving (STEPPS) manual.