Current Psychiatry

Dr. Sermsak Lolak’s well-written article on bedside psychotherapy (August 2004) stresses the importance of psychiatrists’ psychotherapeutic skills in medical settings. As our roles have shrunk to that of medication management in this era of managed care, this article is timely.

Regularly practicing psychotherapy in consultation and liaison psychiatry, however, is difficult because of several logistical issues:

• Length of stay. The 21-day hospital stay in Dr. Lolak’s example is an exception. On consult service, psychiatrists see up to 40% of patients for three follow-up visits or fewer.¹
• Inability to plan therapy. Often a patient is discharged without the psychiatric consultant’s knowledge or input.
• Reimbursement for follow-up psychotherapeutic visits. Most consultation and liaison psychiatrists generate income through billing; follow-up visits are not routinely reimbursed.

To address these issues, the psychiatrist needs to consider every visit as final and make every session complete in itself.

Although we clinically and intuitively recognize the benefits of ultra-brief psychotherapeutic interventions, clinical trials are needed to show their efficacy. These data would also provide rationale for payers to reimburse follow-up therapeutic visits.

Narsimha R. Pinninti MD, MBBS
Assistant professor of psychiatry
University of Medicine and Dentistry of New Jersey
Cherry Hill, NJ

References

1. Strain JJ, Ginsburg J, Fulop G, Strain JJ. Follow-up of psychiatric comorbidity in the general hospital. Int J Psychiatry Med 1990;20:227–34.

Dr. Lolak responds

Dr. Pinninti raises two commonly recognized major problems in consultation psychiatry:

• Visiting a hospitalized patient in a timely manner is often inconvenient.
• Reimbursement is limited.

These are among the many reasons that the American Board of Psychiatry and Neurology now recognizes psychosomatic medicine as a subspecialty. Hospital-based consultation psychiatrists who can see patients in a timely manner are needed. Often hospitals must subsidize a consultation psychiatrist’s position, but this expense offsets the risks of not seeing these patients.

Evidence for brief psychosocial treatment is another issue. Much of what we do is not supported by randomized controlled trials, but we cannot ignore our patients’ needs until more evidence is gathered. We hope a sufficient cadre of consultation psychiatrists do the research Dr. Pinninti suggests.

Sermsak Lolak, MD
Consultation-liaison psychiatry fellow

Thomas Wise, MD
Professor and vice chairman
Department of psychiatry Georgetown University,
Washington, DC

Dr. Sermsak Lolak’s well-written article on bedside psychotherapy (August 2004) stresses the importance of psychiatrists’ psychotherapeutic skills in medical settings. As our roles have shrunk to that of medication management in this era of managed care, this article is timely.

Regularly practicing psychotherapy in consultation and liaison psychiatry, however, is difficult because of several logistical issues:

• Length of stay. The 21-day hospital stay in Dr. Lolak’s example is an exception. On consult service, psychiatrists see up to 40% of patients for three follow-up visits or fewer.¹
• Inability to plan therapy. Often a patient is discharged without the psychiatric consultant’s knowledge or input.
• Reimbursement for follow-up psychotherapeutic visits. Most consultation and liaison psychiatrists generate income through billing; follow-up visits are not routinely reimbursed.

To address these issues, the psychiatrist needs to consider every visit as final and make every session complete in itself.

Although we clinically and intuitively recognize the benefits of ultra-brief psychotherapeutic interventions, clinical trials are needed to show their efficacy. These data would also provide rationale for payers to reimburse follow-up therapeutic visits.

Narsimha R. Pinninti MD, MBBS
Assistant professor of psychiatry
University of Medicine and Dentistry of New Jersey
Cherry Hill, NJ

References

1. Strain JJ, Ginsburg J, Fulop G, Strain JJ. Follow-up of psychiatric comorbidity in the general hospital. Int J Psychiatry Med 1990;20:227–34.

Dr. Lolak responds

Dr. Pinninti raises two commonly recognized major problems in consultation psychiatry:

• Visiting a hospitalized patient in a timely manner is often inconvenient.
• Reimbursement is limited.

These are among the many reasons that the American Board of Psychiatry and Neurology now recognizes psychosomatic medicine as a subspecialty. Hospital-based consultation psychiatrists who can see patients in a timely manner are needed. Often hospitals must subsidize a consultation psychiatrist’s position, but this expense offsets the risks of not seeing these patients.

Evidence for brief psychosocial treatment is another issue. Much of what we do is not supported by randomized controlled trials, but we cannot ignore our patients’ needs until more evidence is gathered. We hope a sufficient cadre of consultation psychiatrists do the research Dr. Pinninti suggests.

Sermsak Lolak, MD
Consultation-liaison psychiatry fellow

Thomas Wise, MD
Professor and vice chairman
Department of psychiatry Georgetown University,
Washington, DC

Dr. Sermsak Lolak’s well-written article on bedside psychotherapy (August 2004) stresses the importance of psychiatrists’ psychotherapeutic skills in medical settings. As our roles have shrunk to that of medication management in this era of managed care, this article is timely.

Regularly practicing psychotherapy in consultation and liaison psychiatry, however, is difficult because of several logistical issues:

• Length of stay. The 21-day hospital stay in Dr. Lolak’s example is an exception. On consult service, psychiatrists see up to 40% of patients for three follow-up visits or fewer.¹
• Inability to plan therapy. Often a patient is discharged without the psychiatric consultant’s knowledge or input.
• Reimbursement for follow-up psychotherapeutic visits. Most consultation and liaison psychiatrists generate income through billing; follow-up visits are not routinely reimbursed.

To address these issues, the psychiatrist needs to consider every visit as final and make every session complete in itself.

Although we clinically and intuitively recognize the benefits of ultra-brief psychotherapeutic interventions, clinical trials are needed to show their efficacy. These data would also provide rationale for payers to reimburse follow-up therapeutic visits.

Narsimha R. Pinninti MD, MBBS
Assistant professor of psychiatry
University of Medicine and Dentistry of New Jersey
Cherry Hill, NJ

References

1. Strain JJ, Ginsburg J, Fulop G, Strain JJ. Follow-up of psychiatric comorbidity in the general hospital. Int J Psychiatry Med 1990;20:227–34.

Dr. Lolak responds

Dr. Pinninti raises two commonly recognized major problems in consultation psychiatry:

• Visiting a hospitalized patient in a timely manner is often inconvenient.
• Reimbursement is limited.

These are among the many reasons that the American Board of Psychiatry and Neurology now recognizes psychosomatic medicine as a subspecialty. Hospital-based consultation psychiatrists who can see patients in a timely manner are needed. Often hospitals must subsidize a consultation psychiatrist’s position, but this expense offsets the risks of not seeing these patients.

Evidence for brief psychosocial treatment is another issue. Much of what we do is not supported by randomized controlled trials, but we cannot ignore our patients’ needs until more evidence is gathered. We hope a sufficient cadre of consultation psychiatrists do the research Dr. Pinninti suggests.

Sermsak Lolak, MD
Consultation-liaison psychiatry fellow

Thomas Wise, MD
Professor and vice chairman
Department of psychiatry Georgetown University,
Washington, DC

Reading “High-dose antipsychotics: Desperation or data-driven?” (Current Psychiatry, August 2004) brings back memories.

In 1980, I worked in a rural hospital. Our psychiatrist gave patients with acute mania a loading dosage of lithium, 900 mg tid to qid on day one, and drew a level in the morning. He reduced the dosage daily until target symptoms improved and/or serum concentration reached the high therapeutic range, or until side effects occurred.

During acute mania, the psychiatrist controlled unsafe agitation with short-acting barbiturates. Some extremely hostile, aggressive patients got IM sodium amobarbital. We observed the patient one-on-one until he or she awoke and could take fluids, food, and lithium. When manic symptoms began escalating, the patient again was medicated and the cycle continued. This cycle might continue for 1 to 3 days with decreasing frequency and sedation based on behavior.

Although I’m not advocating this regimen, it worked well from an empirical perspective. Our patients never acted out or had breathing problems.

Thomas Dowling, RN, MSN
St. Peter’s Hospital, Albany, NY

Reading “High-dose antipsychotics: Desperation or data-driven?” (Current Psychiatry, August 2004) brings back memories.

In 1980, I worked in a rural hospital. Our psychiatrist gave patients with acute mania a loading dosage of lithium, 900 mg tid to qid on day one, and drew a level in the morning. He reduced the dosage daily until target symptoms improved and/or serum concentration reached the high therapeutic range, or until side effects occurred.

During acute mania, the psychiatrist controlled unsafe agitation with short-acting barbiturates. Some extremely hostile, aggressive patients got IM sodium amobarbital. We observed the patient one-on-one until he or she awoke and could take fluids, food, and lithium. When manic symptoms began escalating, the patient again was medicated and the cycle continued. This cycle might continue for 1 to 3 days with decreasing frequency and sedation based on behavior.

Although I’m not advocating this regimen, it worked well from an empirical perspective. Our patients never acted out or had breathing problems.

Thomas Dowling, RN, MSN
St. Peter’s Hospital, Albany, NY

Reading “High-dose antipsychotics: Desperation or data-driven?” (Current Psychiatry, August 2004) brings back memories.

In 1980, I worked in a rural hospital. Our psychiatrist gave patients with acute mania a loading dosage of lithium, 900 mg tid to qid on day one, and drew a level in the morning. He reduced the dosage daily until target symptoms improved and/or serum concentration reached the high therapeutic range, or until side effects occurred.

During acute mania, the psychiatrist controlled unsafe agitation with short-acting barbiturates. Some extremely hostile, aggressive patients got IM sodium amobarbital. We observed the patient one-on-one until he or she awoke and could take fluids, food, and lithium. When manic symptoms began escalating, the patient again was medicated and the cycle continued. This cycle might continue for 1 to 3 days with decreasing frequency and sedation based on behavior.

Although I’m not advocating this regimen, it worked well from an empirical perspective. Our patients never acted out or had breathing problems.

Thomas Dowling, RN, MSN
St. Peter’s Hospital, Albany, NY

HISTORY: UNHAPPY NEW YEAR

On New Year’s Day Ms. M, age 43, begins experiencing persistent left-leg numbness, fatigue, and what she calls a “superallergic sensitivity to anything I put in my mouth.”

A few days later she sees her internist, who finds no medical cause and suspects that her symptoms are psychological. The internist prescribes fluoxetine, 10 mg/d. Fifteen minutes after taking the first dose, the patient reports “an anaphylactic episode,” which she describes as “screaming and shaking.”

Acting on the internist’s suggestion, Ms. M presents to me on Jan. 10. Her parents bring her to the appointment, as she feels too weak to drive.

A chemical engineer with a six-figure income, Ms. M has lived on her own most of her adult life but has stayed the past week with her elderly parents. With her vacation leave about to end, she says she is too weak and tired to return to work. She complains of extreme fatigue after eating most foods; after some meals, she says, welts surface throughout her body. Now living on bananas and homemade apple-sauce, she has lost 5 lbs in less than 2 weeks.

An only child, Ms. M is an award-winning athlete. She has enjoyed her career, which has taken her around the world. She has no significant psychiatric or medical history or family history of allergy or autoimmune disease. She says she is not depressed and is sleeping normally. Her Mini-Mental State Examination score of 30 indicates no cognitive impairment.

Ms. M denies feeling depressed. She mentions that her boyfriend broke off their relationship days before New Year’s Eve—the day on which she had expected they would become engaged. She sees no relationship between disappointment over this breakup and the symptoms that followed almost immediately. She has never had another intimate relationship and describes people she knows as “acquaintances” or “work buddies” rather than as friends.

Table 1

Diagnostic criteria for hypochondriasis

Ms. M refuses to try another psychotropic, fearing another “anaphylactic” episode like the one she described after the fluoxetine dose. She is willing to start psychotherapy, however.

Dr. Bernstein’s observations

Ms. M. complains of an array of food allergies and fatigue with no subjective feelings of depression. She has an athletic physique, is attractive without cosmetics, and is casually but neatly dressed, indicating good organization.

At this point, no physical or medical cause has been found for Ms. M’s symptoms, nor does she meet DSM-IV-TR criteria for hypochondriasis (Table 1). Her symptoms have persisted for 10 days—far short of the 6 months the diagnosis requires. Ms. M also believes that her medical problem is inconvenient but not serious.

Even though Ms. M denies feeling depressed, her symptoms most closely suggest depression with somatic complaints. She is not substantially distressed, but her symptoms are impairing her social and occupational functioning.

Antidepressants—particularly selective serotonin reuptake inhibitors—can help depressed patients with somatic symptoms, and low-dose atypical antipsychotics alternately are used to treat major depressive disorder with somatic delusions. Ms. M, however, will not try another medication, making psychotherapy my only treatment option.

TREATMENT: ‘SURFING’ FOR CLUES

For 6 months, Ms. M attends weekly psychodynamic psychotherapy sessions regularly and on time. She is courteous and pleasant, but her fatigue persists.

Early in treatment, Ms. M spends hours searching the Internet for doctors who specialize in malabsorption syndrome, allergy, and rare infectious diseases. Numerous internists, allergists, and immunologists perform blood work and other laboratory tests on her. She has the results—reams of clinical data—sent to me. I also order tests for HIV, syphilis, and gonorrhea. None of the results indicates a physical disorder. She refuses patch or intradermal testing for allergy, fearing anaphylaxis.

Ms. M also spends much of her day preparing her own meals. She introduces “new foods” one at a time, but reports that these trials often lead to fatigue and cause her to break out in welts. During psychotherapy, she points to bumps and rashes throughout her body that I cannot see.

Six months into psychotherapy, Ms. M is still staying with her parents and has not returned to work, citing disabling fatigue. Her parents, frustrated with her apparent unwillingness to get better, set a deadline for her to move out of their home. She finds an apartment nearby but about 2 miles from the train line she would use to commute to work. She refuses to take a taxi to the train station because of the expense, will not drive to the station because she cannot get up early, and will not drive directly to work for fear of tiring while driving. She refuses her company’s offer to let her work part time from home.

Ms. M’s company keeps her job open for her, but she is still not returning to work. After 1 year, the company finally fires her, then calls her a few months later asking if she’ll come back; she again says no. She collects disability benefits and taps into her savings and investment dividends to make ends meet. In discussing her lack of income during psychotherapy, Ms. M does not appear distressed.

Table 2

Undifferentiated somatoform disorder: diagnostic criteria

Dr. Bernstein’s observations

Although Ms. M meets criteria for undifferentiated somatoform disorder ( Table 2), her belief that she has a medical problem is tenacious and her disability persists despite lack of a medical diagnosis. To me, this suggests a delusional disorder (Table 3 ).

Table 3

Diagnostic criteria for delusional disorder

For 6 months, although she has cooperated with psychotherapy, Ms. M’s complaints have been unyielding. Despite our good relationship, she will not trust my recommendations to try a psychotropic. Nor does psychotherapy or the cooperation of her former employer enable her to resume her once-rewarding career, even part-time.

Ms. M is reclusive but not suspicious. She has no grandiose or paranoid delusions or hallucinations. She has had no depersonalization or derealization episodes, and no affective component exists. She is profoundly convinced that she suddenly developed severe, incapacitating food allergies. Her lifestyle has deteriorated—she feels unable to work and even her parents have virtually abandoned her—yet she seems oddly content.

How does Ms. M compare with other patients with:

• undifferentiated somatic disorder
• delusional disorder?

Dr. Bernstein’s observations

Somatoform disorder. Patients with undifferentiated somatoform disorder usually exhibit fluctuating symptoms, which often can be mitigated with psychodynamic therapy. In time, most accept that their problem is psychological rather than physical or that anxiety or depression are contributing to symptom fluctuation. Patients usually continue or resume social and vocational functioning.

By contrast, Ms. M believes immutably that her symptoms have an undiscovered physical cause. This belief has dramatically changed her life: She has sacrificed her career, social life, health insurance, even her financial security.

The depth and seeming permanence of Ms. M’s state does not distress her. She is not regressed nor affectively or cognitively impaired. She reports seeing and feeling welts and rashes that were not visible to me or to other medical/alternative medical specialists, suggesting reality testing impairment. I perceived no other break in reality testing during psychotherapy.

Delusional disorder can be treated with medication or cognitive-behavioral therapy. I once treated a young man who believed that his head was coming to a point, causing him tremendous emotional distress. An antipsychotic resulted in prompt remission.

By contrast, Ms. M has a delusional belief that food and medicine make her sick and could lead to anaphylaxis. She will not take medication, even in a hospital.

Perhaps someday we will find a neurobiological or biochemical cause for Ms. M’s behavior. Positron-emission tomography or augmented MRI could uncover such clues, but both tests require ingesting a foreign substance—something Ms. M will not do.

FOLLOW-UP: MS. M’S NEW LIFE

Having exhausted her savings and work disability benefits, Ms. M receives Social Security disability benefits. With her health insurance coverage having expired, she stops psychotherapy after 2 years and pursues no further medical workup.

Two years after presenting to me, she does not seem depressed but her presenting picture is unchanged. She sounds happy and cognitively intact. Her life revolves around her perceived disability.

Ms. M has spent much of the last 2 years alone in her apartment, content in her solitude. She has resumed playing tennis but only occasionally and has not resumed the sport for which she has won many awards. She says she feels slightly better but remains too tired to return to work. She has gradually expanded her menu to about a dozen foods. Despite her problems Ms. M, who is 5 feet 2 inches, has maintained her weight (114 lbs) and attractiveness.

All the while, Ms. M has refused medication. I repeatedly suggest hospitalization so that she can take psychotropics in a safe, supervised setting, but she declines.

Related resources

• Pilowsky I. Abnormal Illness Behaviour. New York: John Wiley & Sons, 1997.
• Isaac A, Wise T. A low-frustration strategy for treating somatization. Current Psychiatry 2003;2(8):32-50.

Drug brand names

• Fluoxetine • Prozac

Disclosure

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

HISTORY: UNHAPPY NEW YEAR

On New Year’s Day Ms. M, age 43, begins experiencing persistent left-leg numbness, fatigue, and what she calls a “superallergic sensitivity to anything I put in my mouth.”

A few days later she sees her internist, who finds no medical cause and suspects that her symptoms are psychological. The internist prescribes fluoxetine, 10 mg/d. Fifteen minutes after taking the first dose, the patient reports “an anaphylactic episode,” which she describes as “screaming and shaking.”

Acting on the internist’s suggestion, Ms. M presents to me on Jan. 10. Her parents bring her to the appointment, as she feels too weak to drive.

A chemical engineer with a six-figure income, Ms. M has lived on her own most of her adult life but has stayed the past week with her elderly parents. With her vacation leave about to end, she says she is too weak and tired to return to work. She complains of extreme fatigue after eating most foods; after some meals, she says, welts surface throughout her body. Now living on bananas and homemade apple-sauce, she has lost 5 lbs in less than 2 weeks.

An only child, Ms. M is an award-winning athlete. She has enjoyed her career, which has taken her around the world. She has no significant psychiatric or medical history or family history of allergy or autoimmune disease. She says she is not depressed and is sleeping normally. Her Mini-Mental State Examination score of 30 indicates no cognitive impairment.

Ms. M denies feeling depressed. She mentions that her boyfriend broke off their relationship days before New Year’s Eve—the day on which she had expected they would become engaged. She sees no relationship between disappointment over this breakup and the symptoms that followed almost immediately. She has never had another intimate relationship and describes people she knows as “acquaintances” or “work buddies” rather than as friends.

Table 1

Diagnostic criteria for hypochondriasis

Ms. M refuses to try another psychotropic, fearing another “anaphylactic” episode like the one she described after the fluoxetine dose. She is willing to start psychotherapy, however.

Dr. Bernstein’s observations

Ms. M. complains of an array of food allergies and fatigue with no subjective feelings of depression. She has an athletic physique, is attractive without cosmetics, and is casually but neatly dressed, indicating good organization.

At this point, no physical or medical cause has been found for Ms. M’s symptoms, nor does she meet DSM-IV-TR criteria for hypochondriasis (Table 1). Her symptoms have persisted for 10 days—far short of the 6 months the diagnosis requires. Ms. M also believes that her medical problem is inconvenient but not serious.

Even though Ms. M denies feeling depressed, her symptoms most closely suggest depression with somatic complaints. She is not substantially distressed, but her symptoms are impairing her social and occupational functioning.

Antidepressants—particularly selective serotonin reuptake inhibitors—can help depressed patients with somatic symptoms, and low-dose atypical antipsychotics alternately are used to treat major depressive disorder with somatic delusions. Ms. M, however, will not try another medication, making psychotherapy my only treatment option.

TREATMENT: ‘SURFING’ FOR CLUES

For 6 months, Ms. M attends weekly psychodynamic psychotherapy sessions regularly and on time. She is courteous and pleasant, but her fatigue persists.

Early in treatment, Ms. M spends hours searching the Internet for doctors who specialize in malabsorption syndrome, allergy, and rare infectious diseases. Numerous internists, allergists, and immunologists perform blood work and other laboratory tests on her. She has the results—reams of clinical data—sent to me. I also order tests for HIV, syphilis, and gonorrhea. None of the results indicates a physical disorder. She refuses patch or intradermal testing for allergy, fearing anaphylaxis.

Ms. M also spends much of her day preparing her own meals. She introduces “new foods” one at a time, but reports that these trials often lead to fatigue and cause her to break out in welts. During psychotherapy, she points to bumps and rashes throughout her body that I cannot see.

Six months into psychotherapy, Ms. M is still staying with her parents and has not returned to work, citing disabling fatigue. Her parents, frustrated with her apparent unwillingness to get better, set a deadline for her to move out of their home. She finds an apartment nearby but about 2 miles from the train line she would use to commute to work. She refuses to take a taxi to the train station because of the expense, will not drive to the station because she cannot get up early, and will not drive directly to work for fear of tiring while driving. She refuses her company’s offer to let her work part time from home.

Ms. M’s company keeps her job open for her, but she is still not returning to work. After 1 year, the company finally fires her, then calls her a few months later asking if she’ll come back; she again says no. She collects disability benefits and taps into her savings and investment dividends to make ends meet. In discussing her lack of income during psychotherapy, Ms. M does not appear distressed.

Table 2

Undifferentiated somatoform disorder: diagnostic criteria

Dr. Bernstein’s observations

Although Ms. M meets criteria for undifferentiated somatoform disorder ( Table 2), her belief that she has a medical problem is tenacious and her disability persists despite lack of a medical diagnosis. To me, this suggests a delusional disorder (Table 3 ).

Table 3

Diagnostic criteria for delusional disorder

For 6 months, although she has cooperated with psychotherapy, Ms. M’s complaints have been unyielding. Despite our good relationship, she will not trust my recommendations to try a psychotropic. Nor does psychotherapy or the cooperation of her former employer enable her to resume her once-rewarding career, even part-time.

Ms. M is reclusive but not suspicious. She has no grandiose or paranoid delusions or hallucinations. She has had no depersonalization or derealization episodes, and no affective component exists. She is profoundly convinced that she suddenly developed severe, incapacitating food allergies. Her lifestyle has deteriorated—she feels unable to work and even her parents have virtually abandoned her—yet she seems oddly content.

How does Ms. M compare with other patients with:

• undifferentiated somatic disorder
• delusional disorder?

Dr. Bernstein’s observations

Somatoform disorder. Patients with undifferentiated somatoform disorder usually exhibit fluctuating symptoms, which often can be mitigated with psychodynamic therapy. In time, most accept that their problem is psychological rather than physical or that anxiety or depression are contributing to symptom fluctuation. Patients usually continue or resume social and vocational functioning.

By contrast, Ms. M believes immutably that her symptoms have an undiscovered physical cause. This belief has dramatically changed her life: She has sacrificed her career, social life, health insurance, even her financial security.

The depth and seeming permanence of Ms. M’s state does not distress her. She is not regressed nor affectively or cognitively impaired. She reports seeing and feeling welts and rashes that were not visible to me or to other medical/alternative medical specialists, suggesting reality testing impairment. I perceived no other break in reality testing during psychotherapy.

Delusional disorder can be treated with medication or cognitive-behavioral therapy. I once treated a young man who believed that his head was coming to a point, causing him tremendous emotional distress. An antipsychotic resulted in prompt remission.

By contrast, Ms. M has a delusional belief that food and medicine make her sick and could lead to anaphylaxis. She will not take medication, even in a hospital.

Perhaps someday we will find a neurobiological or biochemical cause for Ms. M’s behavior. Positron-emission tomography or augmented MRI could uncover such clues, but both tests require ingesting a foreign substance—something Ms. M will not do.

FOLLOW-UP: MS. M’S NEW LIFE

Having exhausted her savings and work disability benefits, Ms. M receives Social Security disability benefits. With her health insurance coverage having expired, she stops psychotherapy after 2 years and pursues no further medical workup.

Two years after presenting to me, she does not seem depressed but her presenting picture is unchanged. She sounds happy and cognitively intact. Her life revolves around her perceived disability.

Ms. M has spent much of the last 2 years alone in her apartment, content in her solitude. She has resumed playing tennis but only occasionally and has not resumed the sport for which she has won many awards. She says she feels slightly better but remains too tired to return to work. She has gradually expanded her menu to about a dozen foods. Despite her problems Ms. M, who is 5 feet 2 inches, has maintained her weight (114 lbs) and attractiveness.

All the while, Ms. M has refused medication. I repeatedly suggest hospitalization so that she can take psychotropics in a safe, supervised setting, but she declines.

Related resources

• Pilowsky I. Abnormal Illness Behaviour. New York: John Wiley & Sons, 1997.
• Isaac A, Wise T. A low-frustration strategy for treating somatization. Current Psychiatry 2003;2(8):32-50.

Drug brand names

• Fluoxetine • Prozac

Disclosure

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

HISTORY: UNHAPPY NEW YEAR

On New Year’s Day Ms. M, age 43, begins experiencing persistent left-leg numbness, fatigue, and what she calls a “superallergic sensitivity to anything I put in my mouth.”

A few days later she sees her internist, who finds no medical cause and suspects that her symptoms are psychological. The internist prescribes fluoxetine, 10 mg/d. Fifteen minutes after taking the first dose, the patient reports “an anaphylactic episode,” which she describes as “screaming and shaking.”

Acting on the internist’s suggestion, Ms. M presents to me on Jan. 10. Her parents bring her to the appointment, as she feels too weak to drive.

A chemical engineer with a six-figure income, Ms. M has lived on her own most of her adult life but has stayed the past week with her elderly parents. With her vacation leave about to end, she says she is too weak and tired to return to work. She complains of extreme fatigue after eating most foods; after some meals, she says, welts surface throughout her body. Now living on bananas and homemade apple-sauce, she has lost 5 lbs in less than 2 weeks.

An only child, Ms. M is an award-winning athlete. She has enjoyed her career, which has taken her around the world. She has no significant psychiatric or medical history or family history of allergy or autoimmune disease. She says she is not depressed and is sleeping normally. Her Mini-Mental State Examination score of 30 indicates no cognitive impairment.

Ms. M denies feeling depressed. She mentions that her boyfriend broke off their relationship days before New Year’s Eve—the day on which she had expected they would become engaged. She sees no relationship between disappointment over this breakup and the symptoms that followed almost immediately. She has never had another intimate relationship and describes people she knows as “acquaintances” or “work buddies” rather than as friends.

Table 1

Diagnostic criteria for hypochondriasis

Ms. M refuses to try another psychotropic, fearing another “anaphylactic” episode like the one she described after the fluoxetine dose. She is willing to start psychotherapy, however.

Dr. Bernstein’s observations

Ms. M. complains of an array of food allergies and fatigue with no subjective feelings of depression. She has an athletic physique, is attractive without cosmetics, and is casually but neatly dressed, indicating good organization.

At this point, no physical or medical cause has been found for Ms. M’s symptoms, nor does she meet DSM-IV-TR criteria for hypochondriasis (Table 1). Her symptoms have persisted for 10 days—far short of the 6 months the diagnosis requires. Ms. M also believes that her medical problem is inconvenient but not serious.

Even though Ms. M denies feeling depressed, her symptoms most closely suggest depression with somatic complaints. She is not substantially distressed, but her symptoms are impairing her social and occupational functioning.

Antidepressants—particularly selective serotonin reuptake inhibitors—can help depressed patients with somatic symptoms, and low-dose atypical antipsychotics alternately are used to treat major depressive disorder with somatic delusions. Ms. M, however, will not try another medication, making psychotherapy my only treatment option.

TREATMENT: ‘SURFING’ FOR CLUES

For 6 months, Ms. M attends weekly psychodynamic psychotherapy sessions regularly and on time. She is courteous and pleasant, but her fatigue persists.

Early in treatment, Ms. M spends hours searching the Internet for doctors who specialize in malabsorption syndrome, allergy, and rare infectious diseases. Numerous internists, allergists, and immunologists perform blood work and other laboratory tests on her. She has the results—reams of clinical data—sent to me. I also order tests for HIV, syphilis, and gonorrhea. None of the results indicates a physical disorder. She refuses patch or intradermal testing for allergy, fearing anaphylaxis.

Ms. M also spends much of her day preparing her own meals. She introduces “new foods” one at a time, but reports that these trials often lead to fatigue and cause her to break out in welts. During psychotherapy, she points to bumps and rashes throughout her body that I cannot see.

Six months into psychotherapy, Ms. M is still staying with her parents and has not returned to work, citing disabling fatigue. Her parents, frustrated with her apparent unwillingness to get better, set a deadline for her to move out of their home. She finds an apartment nearby but about 2 miles from the train line she would use to commute to work. She refuses to take a taxi to the train station because of the expense, will not drive to the station because she cannot get up early, and will not drive directly to work for fear of tiring while driving. She refuses her company’s offer to let her work part time from home.

Ms. M’s company keeps her job open for her, but she is still not returning to work. After 1 year, the company finally fires her, then calls her a few months later asking if she’ll come back; she again says no. She collects disability benefits and taps into her savings and investment dividends to make ends meet. In discussing her lack of income during psychotherapy, Ms. M does not appear distressed.

Table 2

Undifferentiated somatoform disorder: diagnostic criteria

Dr. Bernstein’s observations

Although Ms. M meets criteria for undifferentiated somatoform disorder ( Table 2), her belief that she has a medical problem is tenacious and her disability persists despite lack of a medical diagnosis. To me, this suggests a delusional disorder (Table 3 ).

Table 3

Diagnostic criteria for delusional disorder

For 6 months, although she has cooperated with psychotherapy, Ms. M’s complaints have been unyielding. Despite our good relationship, she will not trust my recommendations to try a psychotropic. Nor does psychotherapy or the cooperation of her former employer enable her to resume her once-rewarding career, even part-time.

Ms. M is reclusive but not suspicious. She has no grandiose or paranoid delusions or hallucinations. She has had no depersonalization or derealization episodes, and no affective component exists. She is profoundly convinced that she suddenly developed severe, incapacitating food allergies. Her lifestyle has deteriorated—she feels unable to work and even her parents have virtually abandoned her—yet she seems oddly content.

How does Ms. M compare with other patients with:

• undifferentiated somatic disorder
• delusional disorder?

Dr. Bernstein’s observations

Somatoform disorder. Patients with undifferentiated somatoform disorder usually exhibit fluctuating symptoms, which often can be mitigated with psychodynamic therapy. In time, most accept that their problem is psychological rather than physical or that anxiety or depression are contributing to symptom fluctuation. Patients usually continue or resume social and vocational functioning.

By contrast, Ms. M believes immutably that her symptoms have an undiscovered physical cause. This belief has dramatically changed her life: She has sacrificed her career, social life, health insurance, even her financial security.

The depth and seeming permanence of Ms. M’s state does not distress her. She is not regressed nor affectively or cognitively impaired. She reports seeing and feeling welts and rashes that were not visible to me or to other medical/alternative medical specialists, suggesting reality testing impairment. I perceived no other break in reality testing during psychotherapy.

Delusional disorder can be treated with medication or cognitive-behavioral therapy. I once treated a young man who believed that his head was coming to a point, causing him tremendous emotional distress. An antipsychotic resulted in prompt remission.

By contrast, Ms. M has a delusional belief that food and medicine make her sick and could lead to anaphylaxis. She will not take medication, even in a hospital.

Perhaps someday we will find a neurobiological or biochemical cause for Ms. M’s behavior. Positron-emission tomography or augmented MRI could uncover such clues, but both tests require ingesting a foreign substance—something Ms. M will not do.

FOLLOW-UP: MS. M’S NEW LIFE

Having exhausted her savings and work disability benefits, Ms. M receives Social Security disability benefits. With her health insurance coverage having expired, she stops psychotherapy after 2 years and pursues no further medical workup.

Two years after presenting to me, she does not seem depressed but her presenting picture is unchanged. She sounds happy and cognitively intact. Her life revolves around her perceived disability.

Ms. M has spent much of the last 2 years alone in her apartment, content in her solitude. She has resumed playing tennis but only occasionally and has not resumed the sport for which she has won many awards. She says she feels slightly better but remains too tired to return to work. She has gradually expanded her menu to about a dozen foods. Despite her problems Ms. M, who is 5 feet 2 inches, has maintained her weight (114 lbs) and attractiveness.

All the while, Ms. M has refused medication. I repeatedly suggest hospitalization so that she can take psychotropics in a safe, supervised setting, but she declines.

Related resources

• Pilowsky I. Abnormal Illness Behaviour. New York: John Wiley & Sons, 1997.
• Isaac A, Wise T. A low-frustration strategy for treating somatization. Current Psychiatry 2003;2(8):32-50.

Drug brand names

• Fluoxetine • Prozac

Disclosure

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

When grappling with difficult cases, clinicians often wonder, “What does the evidence say, and how can I find it?” Thanks to evidence-based medicine (EBM), study results are easier to track down and apply to patient care.

From our experience, we tell how to use EBM’s 5 steps and offer tips and shortcuts to help you quickly find the evidence you need.

SPEEDING RESEARCH TO PRACTICE

Health care is rich in evidence-based innovations, but successful innovations are often disseminated slowly—if at all—to practitioners.¹ Studies that demonstrate effective treatments for mental disorders may take considerable time to find their way into clinical practice.² Similar delays have been noted in other specialties, depriving patients of the most up-to-date medical treatments. Academic medical center surveys show that up to 40% of clinical decisions are not supported by the literature.^3,4

Box

Two problems that daunt psychiatrists and other physicians—information overload and uncritical acceptance of information—contribute to less-than-optimal care.⁵

Information overload. With thousands of medical journals and millions of articles being published, no clinician can keep up with all developments in his or her field. Furthermore, study results often appear contradictory.

Review articles summarize the literature, but most are “journalistic” or “narrative”—not systematic reviews—and thus are subject to author bias in the studies cited and methods used to summarize conflicting results.⁴ Textbook chapters have the added problem of being almost immediately out-of-date.⁶

Uncritical acceptance of information occurs when clinicians rely too heavily on personal experiences, noncritiqued study results, expert opinion, and pharmaceutical industry influence.^4,6

Two approaches could narrow the gap between research and practice:

• clinical practice guidelines and pathways, a “top-down” approach favored by administrators
• evidence-based medicine, a “bottom-up” approach favored—and developed by—medical educators (Box 1).^5,7,8

EBM and patient care. EBM has been described as “partly a philosophy, partly a skill, and partly…a set of tools.”⁹ David Sackett—often considered the father of EBM—has defined it as “the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients.”¹⁰

Although some critics disagree, clinical expertise plays an important role in EBM, as the clinician must integrate research evidence, patient preferences, and the patient’s clinical condition when making decisions.¹¹

EBM’S 5 ACTION STEPS

With EBM, evidence from the medical literature is applied to patient care through 5 action steps:

• Formulate the question.
• Search for answers.
• Appraise the evidence.
• Apply the evidence to the patient.
• Assess the outcome.^5,6,12

Step 1: Formulate the question. EBM begins with a clinical question related to the diagnosis, treatment, prognosis, or cause of a patient’s illness. The question is formatted to include the patient’s problem or diagnosis, the intervention of interest and any comparison intervention, and the outcome of interest.

Take, for example, this question: “In patients with bipolar disorder, is lamotrigine as effective as lithium in preventing relapse?” In this format:

• bipolar disorder is the diagnosis
• lamotrigine is the intervention of interest
• lithium is the comparison intervention
• relapse is the outcome of interest.

Step 2: Search for answers. Because answering different types of questions requires different types of evidence, this step involves determining the most appropriate type of evidence and searching for it.¹³

Evidence hierarchies (Table 1) ^5,6,12,14 indicate that certain types of evidence are considered more credible than others. The higher the level of evidence, the more likely it is to provide valid, unbiased estimates of an intervention’s effect.

Table 1

Hierarchy for studies of therapy or harm

For example, observational studies may give misleading results, compared with randomized, controlled, clinical trials.¹⁵ Expert opinion—which does not necessarily reflect the best evidence in the literature—is considered the lowest evidence level.¹⁶

Beginning the literature search with Medline—as many clinicians do—is relatively inefficient. Searching the massive National Library of Medicine database often identifies a large number of articles, which you then must appraise for validity.

More efficient is Haynes’ “4S” strategy of:

• systems
• synopses
• syntheses
• studies.^12,17

It gives priority to sources of high-quality, pre-appraised information ( Table 2), so that you may omit Step 3—appraisal—in the EBM process.

Table 2

“4S” approach to answering clinical questions

Systems. A system is an information source that covers a variety of diagnoses, summarizes the results of high-quality systematic reviews, is frequently updated, and is linked to the original studies. Three examples are:

• Clinical Evidence, a journal published semi-annually in the United Kingdom that is also available online (www.clinicalevidence.com). Its ease of use, frequent updates, and clear links to the best evidence make it a good starting point.
• National electronic Library for Mental Health (NeLMH) (www.nelmh.org), created by the U.K.’s Centre for Evidence-Based Mental Health.¹⁸ It provides information only about depression, schizophrenia, and suicide but may become the model for evidence-based psychiatry.
• Evidence-based clinical practice guidelines developed by the American Psychiatric Association⁹ and other organizations. Many of these are included in the National Guideline Clearinghouse (www.guideline.gov), produced by the U.S. Department of Health and Human Service’s Agency for Healthcare Research and Quality.

Synopses. If a system does not answer a clinical question, look for synopses—structured abstracts of high-quality systematic reviews or original articles. Synopses are brief and pre-appraised for quality, allowing you to quickly get the point without reading a lengthy article. For psychiatry, Evidence-Based Mental Health and ACP Journal Club are the best synopses sources.

Syntheses. If you don’t find a synopsis to answer your question, then search for a synthesis—a high-quality systematic review. The best source is the Cochrane Database of Systematic Reviews;²⁰ others include:

• Database of Abstracts of Reviews of Effectiveness (DARE)
• Health Technology Assessment (HTA) database.

Table 3

Web sites for learning more about EBM

Studies. Only if the first three “S’s” fail to yield an answer would you then search Medline or similar databases for applicable studies.

The PubMed Clinical Queries interface lets you specify the type of question (therapy, diagnosis, etiology, or prognosis) and whether the search is to be sensitive or specific. You can also use filters built into this interface when searching for systematic reviews.

Step 3: Appraise the evidence. After you find an article, appraise its validity and importance. Checklists for appraising studies may be found in EBM texts^4,6,12 or downloaded from Web sites (Table 3). Specific appraisal questions to ask depend on the study design and your clinical question.

Step 4: Apply the results to your patient. Assuming the evidence is valid, important, and feasible in your setting, consider your patient’s preferences²¹ and apply the results.

Step 5: Assess the outcome. Evaluate your performance in searching the literature, and assess the patient’s response.

STREAMLINING THE EBM PROCESS

Studies in academic settings have shown that the full 5-step model can be incorporated into daily practice.^6,12 In nonacademic settings, however, practitioners complain about lack of time and information resources, as well as inadequate EBM skills.^11,12 Two shortcuts can streamline the process:

Use the 5-step process selectively; it is not required for every patient encounter.^6,12 After you research a question for one patient with a particular diagnosis, you can apply the answer to similar patients. Because most psychiatrists’ patients fall into relatively few diagnostic categories, only the exceptional patient would trigger the full 5-step process.

Use pre-appraised information sources. With online databases such as Clinical Evidence and Evidence-Based Mental Health, you can find information quickly, often in the time it takes to use textbooks.^5,12,17

Excellent EBM texts^4,6,12 and online resources may be useful (Table 3). Courses are listed on the EBM Education Center of Excellence Web site, and the Centre for Evidence-Based Mental Health at Oxford University offers an outstanding course for psychiatrists.

Related resources

• Gray GE. Concise guide to evidence-based psychiatry. Washington, DC: American Psychiatric Publishing, 2004.
• Greenhalgh T. How to read a paper: the basics of evidence-based medicine (2nd ed). London, BMJ Books, 2001.
• Guyatt G, Rennie D (eds). Users’ guides to the medical literature: a manual for evidence-based clinical care. Chicago: AMA Press, 2002.
• Sackett DL, Straus SE, Richardson WS, et al. Evidence-based medicine: how to practice and teach EBM (2nd ed). New York: Churchill Livingstone, 2000.

Acknowledgment

Supported in part by a grant from the National Institute of Mental Health (5-R24-MH61456-03).

When grappling with difficult cases, clinicians often wonder, “What does the evidence say, and how can I find it?” Thanks to evidence-based medicine (EBM), study results are easier to track down and apply to patient care.

From our experience, we tell how to use EBM’s 5 steps and offer tips and shortcuts to help you quickly find the evidence you need.

SPEEDING RESEARCH TO PRACTICE

Health care is rich in evidence-based innovations, but successful innovations are often disseminated slowly—if at all—to practitioners.¹ Studies that demonstrate effective treatments for mental disorders may take considerable time to find their way into clinical practice.² Similar delays have been noted in other specialties, depriving patients of the most up-to-date medical treatments. Academic medical center surveys show that up to 40% of clinical decisions are not supported by the literature.^3,4

Box

Two problems that daunt psychiatrists and other physicians—information overload and uncritical acceptance of information—contribute to less-than-optimal care.⁵

Information overload. With thousands of medical journals and millions of articles being published, no clinician can keep up with all developments in his or her field. Furthermore, study results often appear contradictory.

Review articles summarize the literature, but most are “journalistic” or “narrative”—not systematic reviews—and thus are subject to author bias in the studies cited and methods used to summarize conflicting results.⁴ Textbook chapters have the added problem of being almost immediately out-of-date.⁶

Uncritical acceptance of information occurs when clinicians rely too heavily on personal experiences, noncritiqued study results, expert opinion, and pharmaceutical industry influence.^4,6

Two approaches could narrow the gap between research and practice:

• clinical practice guidelines and pathways, a “top-down” approach favored by administrators
• evidence-based medicine, a “bottom-up” approach favored—and developed by—medical educators (Box 1).^5,7,8

EBM and patient care. EBM has been described as “partly a philosophy, partly a skill, and partly…a set of tools.”⁹ David Sackett—often considered the father of EBM—has defined it as “the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients.”¹⁰

Although some critics disagree, clinical expertise plays an important role in EBM, as the clinician must integrate research evidence, patient preferences, and the patient’s clinical condition when making decisions.¹¹

EBM’S 5 ACTION STEPS

With EBM, evidence from the medical literature is applied to patient care through 5 action steps:

• Formulate the question.
• Search for answers.
• Appraise the evidence.
• Apply the evidence to the patient.
• Assess the outcome.^5,6,12

Step 1: Formulate the question. EBM begins with a clinical question related to the diagnosis, treatment, prognosis, or cause of a patient’s illness. The question is formatted to include the patient’s problem or diagnosis, the intervention of interest and any comparison intervention, and the outcome of interest.

Take, for example, this question: “In patients with bipolar disorder, is lamotrigine as effective as lithium in preventing relapse?” In this format:

• bipolar disorder is the diagnosis
• lamotrigine is the intervention of interest
• lithium is the comparison intervention
• relapse is the outcome of interest.

Step 2: Search for answers. Because answering different types of questions requires different types of evidence, this step involves determining the most appropriate type of evidence and searching for it.¹³

Evidence hierarchies (Table 1) ^5,6,12,14 indicate that certain types of evidence are considered more credible than others. The higher the level of evidence, the more likely it is to provide valid, unbiased estimates of an intervention’s effect.

Table 1

Hierarchy for studies of therapy or harm

For example, observational studies may give misleading results, compared with randomized, controlled, clinical trials.¹⁵ Expert opinion—which does not necessarily reflect the best evidence in the literature—is considered the lowest evidence level.¹⁶

Beginning the literature search with Medline—as many clinicians do—is relatively inefficient. Searching the massive National Library of Medicine database often identifies a large number of articles, which you then must appraise for validity.

More efficient is Haynes’ “4S” strategy of:

• systems
• synopses
• syntheses
• studies.^12,17

It gives priority to sources of high-quality, pre-appraised information ( Table 2), so that you may omit Step 3—appraisal—in the EBM process.

Table 2

“4S” approach to answering clinical questions

Systems. A system is an information source that covers a variety of diagnoses, summarizes the results of high-quality systematic reviews, is frequently updated, and is linked to the original studies. Three examples are:

• Clinical Evidence, a journal published semi-annually in the United Kingdom that is also available online (www.clinicalevidence.com). Its ease of use, frequent updates, and clear links to the best evidence make it a good starting point.
• National electronic Library for Mental Health (NeLMH) (www.nelmh.org), created by the U.K.’s Centre for Evidence-Based Mental Health.¹⁸ It provides information only about depression, schizophrenia, and suicide but may become the model for evidence-based psychiatry.
• Evidence-based clinical practice guidelines developed by the American Psychiatric Association⁹ and other organizations. Many of these are included in the National Guideline Clearinghouse (www.guideline.gov), produced by the U.S. Department of Health and Human Service’s Agency for Healthcare Research and Quality.

Synopses. If a system does not answer a clinical question, look for synopses—structured abstracts of high-quality systematic reviews or original articles. Synopses are brief and pre-appraised for quality, allowing you to quickly get the point without reading a lengthy article. For psychiatry, Evidence-Based Mental Health and ACP Journal Club are the best synopses sources.

Syntheses. If you don’t find a synopsis to answer your question, then search for a synthesis—a high-quality systematic review. The best source is the Cochrane Database of Systematic Reviews;²⁰ others include:

• Database of Abstracts of Reviews of Effectiveness (DARE)
• Health Technology Assessment (HTA) database.

Table 3

Web sites for learning more about EBM

Studies. Only if the first three “S’s” fail to yield an answer would you then search Medline or similar databases for applicable studies.

The PubMed Clinical Queries interface lets you specify the type of question (therapy, diagnosis, etiology, or prognosis) and whether the search is to be sensitive or specific. You can also use filters built into this interface when searching for systematic reviews.

Step 3: Appraise the evidence. After you find an article, appraise its validity and importance. Checklists for appraising studies may be found in EBM texts^4,6,12 or downloaded from Web sites (Table 3). Specific appraisal questions to ask depend on the study design and your clinical question.

Step 4: Apply the results to your patient. Assuming the evidence is valid, important, and feasible in your setting, consider your patient’s preferences²¹ and apply the results.

Step 5: Assess the outcome. Evaluate your performance in searching the literature, and assess the patient’s response.

STREAMLINING THE EBM PROCESS

Studies in academic settings have shown that the full 5-step model can be incorporated into daily practice.^6,12 In nonacademic settings, however, practitioners complain about lack of time and information resources, as well as inadequate EBM skills.^11,12 Two shortcuts can streamline the process:

Use the 5-step process selectively; it is not required for every patient encounter.^6,12 After you research a question for one patient with a particular diagnosis, you can apply the answer to similar patients. Because most psychiatrists’ patients fall into relatively few diagnostic categories, only the exceptional patient would trigger the full 5-step process.

Use pre-appraised information sources. With online databases such as Clinical Evidence and Evidence-Based Mental Health, you can find information quickly, often in the time it takes to use textbooks.^5,12,17

Excellent EBM texts^4,6,12 and online resources may be useful (Table 3). Courses are listed on the EBM Education Center of Excellence Web site, and the Centre for Evidence-Based Mental Health at Oxford University offers an outstanding course for psychiatrists.

Related resources

• Gray GE. Concise guide to evidence-based psychiatry. Washington, DC: American Psychiatric Publishing, 2004.
• Greenhalgh T. How to read a paper: the basics of evidence-based medicine (2nd ed). London, BMJ Books, 2001.
• Guyatt G, Rennie D (eds). Users’ guides to the medical literature: a manual for evidence-based clinical care. Chicago: AMA Press, 2002.
• Sackett DL, Straus SE, Richardson WS, et al. Evidence-based medicine: how to practice and teach EBM (2nd ed). New York: Churchill Livingstone, 2000.

Acknowledgment

Supported in part by a grant from the National Institute of Mental Health (5-R24-MH61456-03).

When grappling with difficult cases, clinicians often wonder, “What does the evidence say, and how can I find it?” Thanks to evidence-based medicine (EBM), study results are easier to track down and apply to patient care.

From our experience, we tell how to use EBM’s 5 steps and offer tips and shortcuts to help you quickly find the evidence you need.

SPEEDING RESEARCH TO PRACTICE

Health care is rich in evidence-based innovations, but successful innovations are often disseminated slowly—if at all—to practitioners.¹ Studies that demonstrate effective treatments for mental disorders may take considerable time to find their way into clinical practice.² Similar delays have been noted in other specialties, depriving patients of the most up-to-date medical treatments. Academic medical center surveys show that up to 40% of clinical decisions are not supported by the literature.^3,4

Box

Two problems that daunt psychiatrists and other physicians—information overload and uncritical acceptance of information—contribute to less-than-optimal care.⁵

Information overload. With thousands of medical journals and millions of articles being published, no clinician can keep up with all developments in his or her field. Furthermore, study results often appear contradictory.

Review articles summarize the literature, but most are “journalistic” or “narrative”—not systematic reviews—and thus are subject to author bias in the studies cited and methods used to summarize conflicting results.⁴ Textbook chapters have the added problem of being almost immediately out-of-date.⁶

Uncritical acceptance of information occurs when clinicians rely too heavily on personal experiences, noncritiqued study results, expert opinion, and pharmaceutical industry influence.^4,6

Two approaches could narrow the gap between research and practice:

• clinical practice guidelines and pathways, a “top-down” approach favored by administrators
• evidence-based medicine, a “bottom-up” approach favored—and developed by—medical educators (Box 1).^5,7,8

EBM and patient care. EBM has been described as “partly a philosophy, partly a skill, and partly…a set of tools.”⁹ David Sackett—often considered the father of EBM—has defined it as “the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients.”¹⁰

Although some critics disagree, clinical expertise plays an important role in EBM, as the clinician must integrate research evidence, patient preferences, and the patient’s clinical condition when making decisions.¹¹

EBM’S 5 ACTION STEPS

With EBM, evidence from the medical literature is applied to patient care through 5 action steps:

• Formulate the question.
• Search for answers.
• Appraise the evidence.
• Apply the evidence to the patient.
• Assess the outcome.^5,6,12

Step 1: Formulate the question. EBM begins with a clinical question related to the diagnosis, treatment, prognosis, or cause of a patient’s illness. The question is formatted to include the patient’s problem or diagnosis, the intervention of interest and any comparison intervention, and the outcome of interest.

Take, for example, this question: “In patients with bipolar disorder, is lamotrigine as effective as lithium in preventing relapse?” In this format:

• bipolar disorder is the diagnosis
• lamotrigine is the intervention of interest
• lithium is the comparison intervention
• relapse is the outcome of interest.

Step 2: Search for answers. Because answering different types of questions requires different types of evidence, this step involves determining the most appropriate type of evidence and searching for it.¹³

Evidence hierarchies (Table 1) ^5,6,12,14 indicate that certain types of evidence are considered more credible than others. The higher the level of evidence, the more likely it is to provide valid, unbiased estimates of an intervention’s effect.

Table 1

Hierarchy for studies of therapy or harm

For example, observational studies may give misleading results, compared with randomized, controlled, clinical trials.¹⁵ Expert opinion—which does not necessarily reflect the best evidence in the literature—is considered the lowest evidence level.¹⁶

Beginning the literature search with Medline—as many clinicians do—is relatively inefficient. Searching the massive National Library of Medicine database often identifies a large number of articles, which you then must appraise for validity.

More efficient is Haynes’ “4S” strategy of:

• systems
• synopses
• syntheses
• studies.^12,17

It gives priority to sources of high-quality, pre-appraised information ( Table 2), so that you may omit Step 3—appraisal—in the EBM process.

Table 2

“4S” approach to answering clinical questions

Systems. A system is an information source that covers a variety of diagnoses, summarizes the results of high-quality systematic reviews, is frequently updated, and is linked to the original studies. Three examples are:

• Clinical Evidence, a journal published semi-annually in the United Kingdom that is also available online (www.clinicalevidence.com). Its ease of use, frequent updates, and clear links to the best evidence make it a good starting point.
• National electronic Library for Mental Health (NeLMH) (www.nelmh.org), created by the U.K.’s Centre for Evidence-Based Mental Health.¹⁸ It provides information only about depression, schizophrenia, and suicide but may become the model for evidence-based psychiatry.
• Evidence-based clinical practice guidelines developed by the American Psychiatric Association⁹ and other organizations. Many of these are included in the National Guideline Clearinghouse (www.guideline.gov), produced by the U.S. Department of Health and Human Service’s Agency for Healthcare Research and Quality.

Synopses. If a system does not answer a clinical question, look for synopses—structured abstracts of high-quality systematic reviews or original articles. Synopses are brief and pre-appraised for quality, allowing you to quickly get the point without reading a lengthy article. For psychiatry, Evidence-Based Mental Health and ACP Journal Club are the best synopses sources.

Syntheses. If you don’t find a synopsis to answer your question, then search for a synthesis—a high-quality systematic review. The best source is the Cochrane Database of Systematic Reviews;²⁰ others include:

• Database of Abstracts of Reviews of Effectiveness (DARE)
• Health Technology Assessment (HTA) database.

Table 3

Web sites for learning more about EBM

Studies. Only if the first three “S’s” fail to yield an answer would you then search Medline or similar databases for applicable studies.

The PubMed Clinical Queries interface lets you specify the type of question (therapy, diagnosis, etiology, or prognosis) and whether the search is to be sensitive or specific. You can also use filters built into this interface when searching for systematic reviews.

Step 3: Appraise the evidence. After you find an article, appraise its validity and importance. Checklists for appraising studies may be found in EBM texts^4,6,12 or downloaded from Web sites (Table 3). Specific appraisal questions to ask depend on the study design and your clinical question.

Step 4: Apply the results to your patient. Assuming the evidence is valid, important, and feasible in your setting, consider your patient’s preferences²¹ and apply the results.

Step 5: Assess the outcome. Evaluate your performance in searching the literature, and assess the patient’s response.

STREAMLINING THE EBM PROCESS

Studies in academic settings have shown that the full 5-step model can be incorporated into daily practice.^6,12 In nonacademic settings, however, practitioners complain about lack of time and information resources, as well as inadequate EBM skills.^11,12 Two shortcuts can streamline the process:

Use the 5-step process selectively; it is not required for every patient encounter.^6,12 After you research a question for one patient with a particular diagnosis, you can apply the answer to similar patients. Because most psychiatrists’ patients fall into relatively few diagnostic categories, only the exceptional patient would trigger the full 5-step process.

Use pre-appraised information sources. With online databases such as Clinical Evidence and Evidence-Based Mental Health, you can find information quickly, often in the time it takes to use textbooks.^5,12,17

Excellent EBM texts^4,6,12 and online resources may be useful (Table 3). Courses are listed on the EBM Education Center of Excellence Web site, and the Centre for Evidence-Based Mental Health at Oxford University offers an outstanding course for psychiatrists.

Related resources

• Gray GE. Concise guide to evidence-based psychiatry. Washington, DC: American Psychiatric Publishing, 2004.
• Greenhalgh T. How to read a paper: the basics of evidence-based medicine (2nd ed). London, BMJ Books, 2001.
• Guyatt G, Rennie D (eds). Users’ guides to the medical literature: a manual for evidence-based clinical care. Chicago: AMA Press, 2002.
• Sackett DL, Straus SE, Richardson WS, et al. Evidence-based medicine: how to practice and teach EBM (2nd ed). New York: Churchill Livingstone, 2000.

Acknowledgment

Supported in part by a grant from the National Institute of Mental Health (5-R24-MH61456-03).

When you evaluate patients whose seizures could be epileptic or psychogenic, the evidence points to three useful diagnostic steps:

• Step 1: Characterize the seizure by its triggers and presentation.
• Step 2: Identify psychiatric comorbidity that might be precipitating psychogenic nonepileptic seizures (PNES).¹
• Step 3: Obtain video EEG and blood tests for physiologic confirmation of epilepsy.

Using a case illustration, this article describes how to accomplish these steps by seeking clues in the patient’s seizure and psychiatric histories and choosing high-yield laboratory tests. When a PNES diagnosis becomes clear, we suggest an empathic approach that can help patients develop healthier responses to stress.

Box

CASE: A DIFFERENT KIND OF SEIZURE

Ms. X’s husband brought her to the emergency room after her third tonic-clonic convulsion within 1 week. He reported that her eyes suddenly rolled up and she became limp and fell down after they argued about money. She suffered a minor temple laceration, but this seizure—unlike past episodes—was not associated with mouth foaming or fecal or urinary incontinence.

Ms. X, age 35, has a history of seizure disorder and 5 years of unemployment when seizures were uncontrolled. Her seizures have been stabilized for 18 months with phenytoin, 300 mg bid. She has been hospitalized twice for major depressive disorder, most recently 2 years ago. Since then, her depression has been in remission with paroxetine, 20 mg once daily. She does not abuse drugs or alcohol. She has been married 8 years, has no children, and receives disability income.

Ms. X was stabilized and admitted for neurologic evaluation. CT and MRI were normal, EEG recordings were unremarkable, and blood workup revealed slightly elevated creatine kinase but normal prolactin. Her phenytoin serum level was 12 mcg/mL (therapeutic range, 5 to 20 mcg/mL). When video-EEG recording during one seizure revealed no abnormality, the neurologist requested psychiatric consultation.

PNES: NOT A ‘PSEUDO’ SEIZURE

Patients with PNES are a heterogeneous population (Box)^2-6 that appears repeatedly at emergency rooms, resulting in multiple investigations and treatment with antiepileptic medications. Those with both PNES and epileptic seizure disorders, such as Ms. X, present a particularly difficult clinical dilemma as:

• discontinuing anticonvulsants may exacerbate epileptic seizures
• increasing the medication in the mistaken belief that a seizure breakthrough has occurred could result in toxic serum levels.

Psychological stressors may precipitate PNES in a person who has never had a seizure or in someone with co-existing epilepsy. Patients with PNES frequently deny a correlation between emotional stress and their seizures, whereas the opposite usually holds for patients who have experienced an epileptic seizure.^7,8

PNES has been called “pseudoseizure” a term we believe is unfair to patients because the etiologic determinants are mostly subconscious.⁹ Although one author has defended the term,¹⁰ we agree with others that “pseudoseizure” can give a misleading impression that patients feign their symptoms.⁹

Psychiatric comorbidity. When PNES is suspected, a careful history is essential to identify precipitating psychiatric comorbidity, such as somatoform, somatization, conversion, or dissociative disorder. PNES may also be precipitated by or coexist with mood and anxiety disorders, schizophrenia, malingering and factitious disorders, diffuse organic brain disease, and developmental disorders (Table 1).^11,12

CASE CONTINUED: AN IMPORTANT CLUE

Ms. X was admitted to the psychiatric unit. Her psychiatric history showed recurrent depressive disorder and excluded head injury. She was mildly depressed but expressed minimal cognitive and biological depressive symptoms in the mental status examination. She denied suicidal or homicidal thoughts. Perception, thought process, and cognition were normal.

With her consent, the psychiatrist obtained collateral information from her brother. He reported that his sister had received a diagnosis of “pseudoseizures” about 5 years before this presentation. The brother was unsure of any emotional precipitants.

Table 1

Psychiatric disorders that may precipitate or coexist with PNES

PNES OR EPILEPSY?

Initial assessment of suspected PNES includes a medical, psychiatric, social, psychological, and substance abuse history, as well as a thorough physical examination.

In patients with suspected PNES, obtain collateral histories of seizure precipitants, abortants, childhood events, and family history. Ms. X, for example, has a history of depression and at least one past episode of probable PNES, as described by her brother. An argument with her husband apparently precipitated the most recent seizure episode.

Table 2

Psychogenic seizures PNES vs. epileptic seizures: Differences in presentation*

PNES features. Clinical features (Table 2)¹³—although not definitive—can help differentiate PNES from epileptic seizure.¹ PNES features to look for include:

• prolonged and bizarre prodrome
• prominent out-of-phase ictal or postictal activity
• clear-cut precipitants, especially in an emotionally charged atmosphere
• lack of falls or injuries
• fluctuating consciousness or vivid recall of details during ictal moments.

PNES’ physical symptoms are not voluntary. Patients often have out-of-phase upper- and lower-extremity movements and vocalization as the event starts, as opposed to about 20 seconds into the event when true tonic-clonic seizure makes the tonic-clonic transformation. Other common features are high-amplitude, forward pelvic thrusting, and lack of rigidity. Weeping during an apparent seizure strongly suggests a nonepileptic event.¹⁴

Ictal duration can be useful in assessment. Events that resemble tonic-clonic seizure but continue for >70 seconds or <20 seconds raise suspicion of nonepileptic seizures, although status epilepticus is possible.

Prolactin elevation. Epileptic tonic-clonic and partial complex seizures increase serum prolactin and are most reliable approximately 20 minutes after event onset. Nipple manipulation can spuriously increase serum prolactin, so observe female patients for this behavior if a seizure occurs in your presence. Psychotropics such as chlorpromazine and haloperidol may also elevate serum prolactin.

INVESTIGATIONS

Video EEG recording is available in most neurologic centers and is the investigation of choice. Epileptic seizure is characterized by recruitment of seizure activity in a physiologic distribution and postictal slowing, which would be difficult for a patient to imitate.

Unlike traditional EEG, video EEG shows evidence of electrographic paroxysmal changes immediately before, during, or after an epileptic seizure.¹⁵ Seizure presentations without paroxysmal electrographic changes would be considered PNES.

Traditional EEG is not recommended for a PNES workup because seizure activity is not recorded and myogenic infarcts may obscure readings. Moreover, because interictal EEG changes may occur even in patients with PNES, these changes in isolation cannot be interpreted as evidence of epilepsy.¹⁶

Laboratory testing includes full blood count, electrolytes, urea and creatinine, urine drug screen, and thyroid and liver function tests, as well as serum levels in patients taking anticonvulsants. These tests may exclude some seizure causes (such as hypokalemia or hypocalcemia with electrolyte disturbances) and provide baseline values for monitoring drug toxicity. Thyroid function testing will rule out hypo- or hyperthyroidism in patients with comorbid depressive or anxiety disorders. Urine drug screen reveals evidence of drug abuse—a possible organic seizure disorder precipitant.

Normal serum prolactin (men: 2 to 18 ng/mL; nonpregnant women: 3 to 30 ng/mL), cortisol (5 to 22 mcg/dL, morning blood specimen), and creatine kinase (50 to 200 U/L) rise substantially after an epileptic—but not psychogenic—seizure.¹⁷ Note, however, that creatine kinase and prolactin may be as elevated in PNES as in an epileptic seizure if PNES presents with vigorous muscular activity.¹⁸

Psychological testing may help distinguish PNES from epileptic seizure:

• Minnesota Multiphasic Personality Inventory (MMPI) is fairly sensitive and shows statistically significant differences between PNES and epileptic seizures in hypochondriasis, depression, hysteria, and schizophrenia.¹⁹
• Washington Psychosocial Seizure Inventory helps identify etiologic subgroups among PNES patients.⁹

Others. As in Ms. X’s case, CT and MRI for cerebral pathology may not help.²⁰ Provocative techniques²¹ based on suggestibility also may have little value because PNES can be provoked in patients with documented epileptic seizures. Hypnosis has been used in attempts to demonstrate the psychogenic component of nonepileptic seizures by reversing ictal and postictal amnesia. Patients with PNES seem to be more responsive to hypnosis than those with epileptic seizures.²²

CASE CONTINUED: EXPLAINING THE DIAGNOSIS

The psychiatric team concluded that even though Ms. X has a seizure disorder, this particular episode was psychogenic. This conclusion was based on the emotional precipitant, her brother’s collateral history, video EEG recordings, and her history of depression.

The psychiatrist explained to Ms. X that PNES and seizure disorders can coexist and that a PNES diagnosis does not imply that a patient is lying. It simply means that some seizures are precipitated by psychoemotional events.

Unlike some patients with PNES, Ms. X accepted the diagnosis without anger. Because her epilepsy and depression were stable, the psychiatrist did not change her phenytoin or paroxetine dosages. She was discharged, with follow-up care by a psychiatrist and neurologist.

TELLING AND TREATING THE PATIENT

It is important to present a PNES diagnosis with care because some patients react with aggressive denial and suicidal behavior. A patient who believes he or she has been perceived as a liar or “fake” may feel humiliated,⁹ which is why we prefer the term “PNES” rather than “pseudoseizure.”

PNES treatment calls for collaboration among the neurologist, psychiatrist, psychologist, therapists, support workers, and family. When PNES is clearly the only cause of seizures, avoid treating with anticonvulsants. If PNES coexists with a treated seizure disorder, no change in anticonvulsant dosage is necessary, especially if blood levels are normal.

Target the underlying psychiatric disorder, using medication and cognitive-behavioral therapy as appropriate. Advise patients and families that the seizure-like events may continue for some time but will disappear as the patient develops more-effective and appropriate coping strategies.

Prognosis. Favorable prognostic factors include being female, effective early intervention, normal premorbid psychological make-up, and good family support. Studies have shown that:

• up to 40% of treated PNES patients remained event-free for a median of 5 years.²³
• prognosis may depend on spell type (catatonic is more favorable than “thrashing”) and illness duration (<1 year is more favorable than >1 year).²⁴

Related resources

• Riley LT, Roy A. Pseudoseizures. Baltimore: Lippincott, Williams and Wilkins, 1982.
• Epilepsy.com. Psychogenic seizures. Resource for patients. www.epilepsy.com/epilepsy/seizure_psychogenic.html Accessed Oct. 14, 2004.
• Epilepsy Foundation. Information for patients and families. www.epilepsyfoundation.org. Search for nonepileptic seizures. Accessed Oct. 14, 2004.

Drug brand names

• Chlorpromazine • Thorazine
• Haloperidol • Haldol
• Phenytoin • Dilantin
• Paroxetine • Paxil

Disclosure

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

When you evaluate patients whose seizures could be epileptic or psychogenic, the evidence points to three useful diagnostic steps:

• Step 1: Characterize the seizure by its triggers and presentation.
• Step 2: Identify psychiatric comorbidity that might be precipitating psychogenic nonepileptic seizures (PNES).¹
• Step 3: Obtain video EEG and blood tests for physiologic confirmation of epilepsy.

Using a case illustration, this article describes how to accomplish these steps by seeking clues in the patient’s seizure and psychiatric histories and choosing high-yield laboratory tests. When a PNES diagnosis becomes clear, we suggest an empathic approach that can help patients develop healthier responses to stress.

Box

CASE: A DIFFERENT KIND OF SEIZURE

Ms. X’s husband brought her to the emergency room after her third tonic-clonic convulsion within 1 week. He reported that her eyes suddenly rolled up and she became limp and fell down after they argued about money. She suffered a minor temple laceration, but this seizure—unlike past episodes—was not associated with mouth foaming or fecal or urinary incontinence.

Ms. X, age 35, has a history of seizure disorder and 5 years of unemployment when seizures were uncontrolled. Her seizures have been stabilized for 18 months with phenytoin, 300 mg bid. She has been hospitalized twice for major depressive disorder, most recently 2 years ago. Since then, her depression has been in remission with paroxetine, 20 mg once daily. She does not abuse drugs or alcohol. She has been married 8 years, has no children, and receives disability income.

Ms. X was stabilized and admitted for neurologic evaluation. CT and MRI were normal, EEG recordings were unremarkable, and blood workup revealed slightly elevated creatine kinase but normal prolactin. Her phenytoin serum level was 12 mcg/mL (therapeutic range, 5 to 20 mcg/mL). When video-EEG recording during one seizure revealed no abnormality, the neurologist requested psychiatric consultation.

PNES: NOT A ‘PSEUDO’ SEIZURE

Patients with PNES are a heterogeneous population (Box)^2-6 that appears repeatedly at emergency rooms, resulting in multiple investigations and treatment with antiepileptic medications. Those with both PNES and epileptic seizure disorders, such as Ms. X, present a particularly difficult clinical dilemma as:

• discontinuing anticonvulsants may exacerbate epileptic seizures
• increasing the medication in the mistaken belief that a seizure breakthrough has occurred could result in toxic serum levels.

Psychological stressors may precipitate PNES in a person who has never had a seizure or in someone with co-existing epilepsy. Patients with PNES frequently deny a correlation between emotional stress and their seizures, whereas the opposite usually holds for patients who have experienced an epileptic seizure.^7,8

PNES has been called “pseudoseizure” a term we believe is unfair to patients because the etiologic determinants are mostly subconscious.⁹ Although one author has defended the term,¹⁰ we agree with others that “pseudoseizure” can give a misleading impression that patients feign their symptoms.⁹

Psychiatric comorbidity. When PNES is suspected, a careful history is essential to identify precipitating psychiatric comorbidity, such as somatoform, somatization, conversion, or dissociative disorder. PNES may also be precipitated by or coexist with mood and anxiety disorders, schizophrenia, malingering and factitious disorders, diffuse organic brain disease, and developmental disorders (Table 1).^11,12

CASE CONTINUED: AN IMPORTANT CLUE

Ms. X was admitted to the psychiatric unit. Her psychiatric history showed recurrent depressive disorder and excluded head injury. She was mildly depressed but expressed minimal cognitive and biological depressive symptoms in the mental status examination. She denied suicidal or homicidal thoughts. Perception, thought process, and cognition were normal.

With her consent, the psychiatrist obtained collateral information from her brother. He reported that his sister had received a diagnosis of “pseudoseizures” about 5 years before this presentation. The brother was unsure of any emotional precipitants.

Table 1

Psychiatric disorders that may precipitate or coexist with PNES

PNES OR EPILEPSY?

Initial assessment of suspected PNES includes a medical, psychiatric, social, psychological, and substance abuse history, as well as a thorough physical examination.

In patients with suspected PNES, obtain collateral histories of seizure precipitants, abortants, childhood events, and family history. Ms. X, for example, has a history of depression and at least one past episode of probable PNES, as described by her brother. An argument with her husband apparently precipitated the most recent seizure episode.

Table 2

Psychogenic seizures PNES vs. epileptic seizures: Differences in presentation*

PNES features. Clinical features (Table 2)¹³—although not definitive—can help differentiate PNES from epileptic seizure.¹ PNES features to look for include:

• prolonged and bizarre prodrome
• prominent out-of-phase ictal or postictal activity
• clear-cut precipitants, especially in an emotionally charged atmosphere
• lack of falls or injuries
• fluctuating consciousness or vivid recall of details during ictal moments.

PNES’ physical symptoms are not voluntary. Patients often have out-of-phase upper- and lower-extremity movements and vocalization as the event starts, as opposed to about 20 seconds into the event when true tonic-clonic seizure makes the tonic-clonic transformation. Other common features are high-amplitude, forward pelvic thrusting, and lack of rigidity. Weeping during an apparent seizure strongly suggests a nonepileptic event.¹⁴

Ictal duration can be useful in assessment. Events that resemble tonic-clonic seizure but continue for >70 seconds or <20 seconds raise suspicion of nonepileptic seizures, although status epilepticus is possible.

Prolactin elevation. Epileptic tonic-clonic and partial complex seizures increase serum prolactin and are most reliable approximately 20 minutes after event onset. Nipple manipulation can spuriously increase serum prolactin, so observe female patients for this behavior if a seizure occurs in your presence. Psychotropics such as chlorpromazine and haloperidol may also elevate serum prolactin.

INVESTIGATIONS

Video EEG recording is available in most neurologic centers and is the investigation of choice. Epileptic seizure is characterized by recruitment of seizure activity in a physiologic distribution and postictal slowing, which would be difficult for a patient to imitate.

Unlike traditional EEG, video EEG shows evidence of electrographic paroxysmal changes immediately before, during, or after an epileptic seizure.¹⁵ Seizure presentations without paroxysmal electrographic changes would be considered PNES.

Traditional EEG is not recommended for a PNES workup because seizure activity is not recorded and myogenic infarcts may obscure readings. Moreover, because interictal EEG changes may occur even in patients with PNES, these changes in isolation cannot be interpreted as evidence of epilepsy.¹⁶

Laboratory testing includes full blood count, electrolytes, urea and creatinine, urine drug screen, and thyroid and liver function tests, as well as serum levels in patients taking anticonvulsants. These tests may exclude some seizure causes (such as hypokalemia or hypocalcemia with electrolyte disturbances) and provide baseline values for monitoring drug toxicity. Thyroid function testing will rule out hypo- or hyperthyroidism in patients with comorbid depressive or anxiety disorders. Urine drug screen reveals evidence of drug abuse—a possible organic seizure disorder precipitant.

Normal serum prolactin (men: 2 to 18 ng/mL; nonpregnant women: 3 to 30 ng/mL), cortisol (5 to 22 mcg/dL, morning blood specimen), and creatine kinase (50 to 200 U/L) rise substantially after an epileptic—but not psychogenic—seizure.¹⁷ Note, however, that creatine kinase and prolactin may be as elevated in PNES as in an epileptic seizure if PNES presents with vigorous muscular activity.¹⁸

Psychological testing may help distinguish PNES from epileptic seizure:

• Minnesota Multiphasic Personality Inventory (MMPI) is fairly sensitive and shows statistically significant differences between PNES and epileptic seizures in hypochondriasis, depression, hysteria, and schizophrenia.¹⁹
• Washington Psychosocial Seizure Inventory helps identify etiologic subgroups among PNES patients.⁹

Others. As in Ms. X’s case, CT and MRI for cerebral pathology may not help.²⁰ Provocative techniques²¹ based on suggestibility also may have little value because PNES can be provoked in patients with documented epileptic seizures. Hypnosis has been used in attempts to demonstrate the psychogenic component of nonepileptic seizures by reversing ictal and postictal amnesia. Patients with PNES seem to be more responsive to hypnosis than those with epileptic seizures.²²

CASE CONTINUED: EXPLAINING THE DIAGNOSIS

The psychiatric team concluded that even though Ms. X has a seizure disorder, this particular episode was psychogenic. This conclusion was based on the emotional precipitant, her brother’s collateral history, video EEG recordings, and her history of depression.

The psychiatrist explained to Ms. X that PNES and seizure disorders can coexist and that a PNES diagnosis does not imply that a patient is lying. It simply means that some seizures are precipitated by psychoemotional events.

Unlike some patients with PNES, Ms. X accepted the diagnosis without anger. Because her epilepsy and depression were stable, the psychiatrist did not change her phenytoin or paroxetine dosages. She was discharged, with follow-up care by a psychiatrist and neurologist.

TELLING AND TREATING THE PATIENT

It is important to present a PNES diagnosis with care because some patients react with aggressive denial and suicidal behavior. A patient who believes he or she has been perceived as a liar or “fake” may feel humiliated,⁹ which is why we prefer the term “PNES” rather than “pseudoseizure.”

PNES treatment calls for collaboration among the neurologist, psychiatrist, psychologist, therapists, support workers, and family. When PNES is clearly the only cause of seizures, avoid treating with anticonvulsants. If PNES coexists with a treated seizure disorder, no change in anticonvulsant dosage is necessary, especially if blood levels are normal.

Target the underlying psychiatric disorder, using medication and cognitive-behavioral therapy as appropriate. Advise patients and families that the seizure-like events may continue for some time but will disappear as the patient develops more-effective and appropriate coping strategies.

Prognosis. Favorable prognostic factors include being female, effective early intervention, normal premorbid psychological make-up, and good family support. Studies have shown that:

• up to 40% of treated PNES patients remained event-free for a median of 5 years.²³
• prognosis may depend on spell type (catatonic is more favorable than “thrashing”) and illness duration (<1 year is more favorable than >1 year).²⁴

Related resources

• Riley LT, Roy A. Pseudoseizures. Baltimore: Lippincott, Williams and Wilkins, 1982.
• Epilepsy.com. Psychogenic seizures. Resource for patients. www.epilepsy.com/epilepsy/seizure_psychogenic.html Accessed Oct. 14, 2004.
• Epilepsy Foundation. Information for patients and families. www.epilepsyfoundation.org. Search for nonepileptic seizures. Accessed Oct. 14, 2004.

Drug brand names

• Chlorpromazine • Thorazine
• Haloperidol • Haldol
• Phenytoin • Dilantin
• Paroxetine • Paxil

Disclosure

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

When you evaluate patients whose seizures could be epileptic or psychogenic, the evidence points to three useful diagnostic steps:

• Step 1: Characterize the seizure by its triggers and presentation.
• Step 2: Identify psychiatric comorbidity that might be precipitating psychogenic nonepileptic seizures (PNES).¹
• Step 3: Obtain video EEG and blood tests for physiologic confirmation of epilepsy.

Using a case illustration, this article describes how to accomplish these steps by seeking clues in the patient’s seizure and psychiatric histories and choosing high-yield laboratory tests. When a PNES diagnosis becomes clear, we suggest an empathic approach that can help patients develop healthier responses to stress.

Box

CASE: A DIFFERENT KIND OF SEIZURE

Ms. X’s husband brought her to the emergency room after her third tonic-clonic convulsion within 1 week. He reported that her eyes suddenly rolled up and she became limp and fell down after they argued about money. She suffered a minor temple laceration, but this seizure—unlike past episodes—was not associated with mouth foaming or fecal or urinary incontinence.

Ms. X, age 35, has a history of seizure disorder and 5 years of unemployment when seizures were uncontrolled. Her seizures have been stabilized for 18 months with phenytoin, 300 mg bid. She has been hospitalized twice for major depressive disorder, most recently 2 years ago. Since then, her depression has been in remission with paroxetine, 20 mg once daily. She does not abuse drugs or alcohol. She has been married 8 years, has no children, and receives disability income.

Ms. X was stabilized and admitted for neurologic evaluation. CT and MRI were normal, EEG recordings were unremarkable, and blood workup revealed slightly elevated creatine kinase but normal prolactin. Her phenytoin serum level was 12 mcg/mL (therapeutic range, 5 to 20 mcg/mL). When video-EEG recording during one seizure revealed no abnormality, the neurologist requested psychiatric consultation.

PNES: NOT A ‘PSEUDO’ SEIZURE

Patients with PNES are a heterogeneous population (Box)^2-6 that appears repeatedly at emergency rooms, resulting in multiple investigations and treatment with antiepileptic medications. Those with both PNES and epileptic seizure disorders, such as Ms. X, present a particularly difficult clinical dilemma as:

• discontinuing anticonvulsants may exacerbate epileptic seizures
• increasing the medication in the mistaken belief that a seizure breakthrough has occurred could result in toxic serum levels.

Psychological stressors may precipitate PNES in a person who has never had a seizure or in someone with co-existing epilepsy. Patients with PNES frequently deny a correlation between emotional stress and their seizures, whereas the opposite usually holds for patients who have experienced an epileptic seizure.^7,8

PNES has been called “pseudoseizure” a term we believe is unfair to patients because the etiologic determinants are mostly subconscious.⁹ Although one author has defended the term,¹⁰ we agree with others that “pseudoseizure” can give a misleading impression that patients feign their symptoms.⁹

Psychiatric comorbidity. When PNES is suspected, a careful history is essential to identify precipitating psychiatric comorbidity, such as somatoform, somatization, conversion, or dissociative disorder. PNES may also be precipitated by or coexist with mood and anxiety disorders, schizophrenia, malingering and factitious disorders, diffuse organic brain disease, and developmental disorders (Table 1).^11,12

CASE CONTINUED: AN IMPORTANT CLUE

Ms. X was admitted to the psychiatric unit. Her psychiatric history showed recurrent depressive disorder and excluded head injury. She was mildly depressed but expressed minimal cognitive and biological depressive symptoms in the mental status examination. She denied suicidal or homicidal thoughts. Perception, thought process, and cognition were normal.

With her consent, the psychiatrist obtained collateral information from her brother. He reported that his sister had received a diagnosis of “pseudoseizures” about 5 years before this presentation. The brother was unsure of any emotional precipitants.

Table 1

Psychiatric disorders that may precipitate or coexist with PNES

PNES OR EPILEPSY?

Initial assessment of suspected PNES includes a medical, psychiatric, social, psychological, and substance abuse history, as well as a thorough physical examination.

In patients with suspected PNES, obtain collateral histories of seizure precipitants, abortants, childhood events, and family history. Ms. X, for example, has a history of depression and at least one past episode of probable PNES, as described by her brother. An argument with her husband apparently precipitated the most recent seizure episode.

Table 2

Psychogenic seizures PNES vs. epileptic seizures: Differences in presentation*

PNES features. Clinical features (Table 2)¹³—although not definitive—can help differentiate PNES from epileptic seizure.¹ PNES features to look for include:

• prolonged and bizarre prodrome
• prominent out-of-phase ictal or postictal activity
• clear-cut precipitants, especially in an emotionally charged atmosphere
• lack of falls or injuries
• fluctuating consciousness or vivid recall of details during ictal moments.

PNES’ physical symptoms are not voluntary. Patients often have out-of-phase upper- and lower-extremity movements and vocalization as the event starts, as opposed to about 20 seconds into the event when true tonic-clonic seizure makes the tonic-clonic transformation. Other common features are high-amplitude, forward pelvic thrusting, and lack of rigidity. Weeping during an apparent seizure strongly suggests a nonepileptic event.¹⁴

Ictal duration can be useful in assessment. Events that resemble tonic-clonic seizure but continue for >70 seconds or <20 seconds raise suspicion of nonepileptic seizures, although status epilepticus is possible.

Prolactin elevation. Epileptic tonic-clonic and partial complex seizures increase serum prolactin and are most reliable approximately 20 minutes after event onset. Nipple manipulation can spuriously increase serum prolactin, so observe female patients for this behavior if a seizure occurs in your presence. Psychotropics such as chlorpromazine and haloperidol may also elevate serum prolactin.

INVESTIGATIONS

Video EEG recording is available in most neurologic centers and is the investigation of choice. Epileptic seizure is characterized by recruitment of seizure activity in a physiologic distribution and postictal slowing, which would be difficult for a patient to imitate.

Unlike traditional EEG, video EEG shows evidence of electrographic paroxysmal changes immediately before, during, or after an epileptic seizure.¹⁵ Seizure presentations without paroxysmal electrographic changes would be considered PNES.

Traditional EEG is not recommended for a PNES workup because seizure activity is not recorded and myogenic infarcts may obscure readings. Moreover, because interictal EEG changes may occur even in patients with PNES, these changes in isolation cannot be interpreted as evidence of epilepsy.¹⁶

Laboratory testing includes full blood count, electrolytes, urea and creatinine, urine drug screen, and thyroid and liver function tests, as well as serum levels in patients taking anticonvulsants. These tests may exclude some seizure causes (such as hypokalemia or hypocalcemia with electrolyte disturbances) and provide baseline values for monitoring drug toxicity. Thyroid function testing will rule out hypo- or hyperthyroidism in patients with comorbid depressive or anxiety disorders. Urine drug screen reveals evidence of drug abuse—a possible organic seizure disorder precipitant.

Normal serum prolactin (men: 2 to 18 ng/mL; nonpregnant women: 3 to 30 ng/mL), cortisol (5 to 22 mcg/dL, morning blood specimen), and creatine kinase (50 to 200 U/L) rise substantially after an epileptic—but not psychogenic—seizure.¹⁷ Note, however, that creatine kinase and prolactin may be as elevated in PNES as in an epileptic seizure if PNES presents with vigorous muscular activity.¹⁸

Psychological testing may help distinguish PNES from epileptic seizure:

• Minnesota Multiphasic Personality Inventory (MMPI) is fairly sensitive and shows statistically significant differences between PNES and epileptic seizures in hypochondriasis, depression, hysteria, and schizophrenia.¹⁹
• Washington Psychosocial Seizure Inventory helps identify etiologic subgroups among PNES patients.⁹

Others. As in Ms. X’s case, CT and MRI for cerebral pathology may not help.²⁰ Provocative techniques²¹ based on suggestibility also may have little value because PNES can be provoked in patients with documented epileptic seizures. Hypnosis has been used in attempts to demonstrate the psychogenic component of nonepileptic seizures by reversing ictal and postictal amnesia. Patients with PNES seem to be more responsive to hypnosis than those with epileptic seizures.²²

CASE CONTINUED: EXPLAINING THE DIAGNOSIS

The psychiatric team concluded that even though Ms. X has a seizure disorder, this particular episode was psychogenic. This conclusion was based on the emotional precipitant, her brother’s collateral history, video EEG recordings, and her history of depression.

The psychiatrist explained to Ms. X that PNES and seizure disorders can coexist and that a PNES diagnosis does not imply that a patient is lying. It simply means that some seizures are precipitated by psychoemotional events.

Unlike some patients with PNES, Ms. X accepted the diagnosis without anger. Because her epilepsy and depression were stable, the psychiatrist did not change her phenytoin or paroxetine dosages. She was discharged, with follow-up care by a psychiatrist and neurologist.

TELLING AND TREATING THE PATIENT

It is important to present a PNES diagnosis with care because some patients react with aggressive denial and suicidal behavior. A patient who believes he or she has been perceived as a liar or “fake” may feel humiliated,⁹ which is why we prefer the term “PNES” rather than “pseudoseizure.”

PNES treatment calls for collaboration among the neurologist, psychiatrist, psychologist, therapists, support workers, and family. When PNES is clearly the only cause of seizures, avoid treating with anticonvulsants. If PNES coexists with a treated seizure disorder, no change in anticonvulsant dosage is necessary, especially if blood levels are normal.

Target the underlying psychiatric disorder, using medication and cognitive-behavioral therapy as appropriate. Advise patients and families that the seizure-like events may continue for some time but will disappear as the patient develops more-effective and appropriate coping strategies.

Prognosis. Favorable prognostic factors include being female, effective early intervention, normal premorbid psychological make-up, and good family support. Studies have shown that:

• up to 40% of treated PNES patients remained event-free for a median of 5 years.²³
• prognosis may depend on spell type (catatonic is more favorable than “thrashing”) and illness duration (<1 year is more favorable than >1 year).²⁴

Related resources

• Riley LT, Roy A. Pseudoseizures. Baltimore: Lippincott, Williams and Wilkins, 1982.
• Epilepsy.com. Psychogenic seizures. Resource for patients. www.epilepsy.com/epilepsy/seizure_psychogenic.html Accessed Oct. 14, 2004.
• Epilepsy Foundation. Information for patients and families. www.epilepsyfoundation.org. Search for nonepileptic seizures. Accessed Oct. 14, 2004.

Drug brand names

• Chlorpromazine • Thorazine
• Haloperidol • Haldol
• Phenytoin • Dilantin
• Paroxetine • Paxil

Disclosure

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

Rapid-cycling bipolar disorder is a moving target, with treatment-resistant depression recurring frequently and alternating with hypomanic/manic episodes (Box).^1,2 Can one medication adequately treat these complicated patients, or is combination therapy necessary? If more than one medication is needed, are some combinations more effective than others?

This article attempts to answer these questions by:

• discussing recent treatment trial results
• suggesting an algorithm for managing hypomanic/manic and depressive episodes in rapid-cycling patients with bipolar disorder types I or II.

CLINICAL CHARACTERISTICS

Rapid cycling is associated most consistently with female gender and bipolar II disorder² (Table); why these two groups are primarily affected is unknown. Results of studies linking rapid cycling with hypothyroidism, gonadal steroid effects, family history, and substance use have been inconsistent and contradictory.²

Age of onset. Recent studies examining bipolar disorder’s age of onset have contradicted earlier rapid-cycling literature. In two large studies, Schneck et al³ and Coryell et al⁴ found rapid cycling associated with early onset of bipolar illness. The authors note that high rates of rapid cycling in children and adolescents resemble adult rapid cycling and speculate that early-onset bipolar illness might lead to rapid cycling vulnerability.⁵

Box

Transient vs persistent state. Rapid cycling is thought to be either a transient state in long-term bipolar illness or a more chronic expression of the illness. Several studies^6,7 have described rapid cycling as a transient phenomenon, whereas others^8-11 have found a more persistent rapid cycling course during follow-up. Interestingly, a recent study¹¹ suggested the mood-cycle pattern may be the most important predictor of rapid cycling. Patients with a depression–hypomania/mania-euthymia course demonstrated more-persistent rapid cycling than did those with a hypomania/mania-depression-euthymia course.

Antidepressants. Antidepressants’ role in initiating or exacerbating rapid cycling also remains unclear. Wehr et al⁸ found that discontinuing antidepressants contributed to cycling cessation or slowing. However, two prospective studies by Coryell et al⁴ that controlled for major depression found no association between antidepressant use and rapid cycling.

More recently, Yildiz and Sachs¹² found a possible gender-specific relationship between antidepressants and rapid cycling. Women exposed to antidepressants before their first hypomanic/manic episode were more likely to develop rapid cycling than women who were not so exposed. This association was not evident in men.

NO DEFINITIVE CHOICES

Any discussion of treating rapid-cycling bipolar disorder is based on limited data, as few prospective studies of this exclusive cohort exist. Many studies report on mixed cohorts of refractory bipolar patients that include rapid cyclers, but separate analyses of rapid-cycling subgroups are not usually reported. Notable exceptions are recent studies by Calabrese et al, which are discussed below.

Lithium. Dunner and Fieve¹³ were the first to suggest that rapid-cycling bipolar patients respond poorly to lithium maintenance monotherapy. Later studies, however, suggested that lithium could benefit rapid cyclers, primarily in reducing hypomanic or manic episodes.

Baldessarini et al¹⁰ found that lithium was less effective for rapid than nonrapid cyclers only in reducing recurrence of depressive episodes. Kukopulos et al¹⁴ reported that lithium response in rapid cyclers increased from 16% to 78% after antidepressants were stopped, suggesting that a positive response to lithium may require more limited antidepressant use (or patients not having been exposed to antidepressants at all).

Thus, lithium prophylaxis has at least partial efficacy in many rapid cyclers, especially when antidepressants are avoided.

Divalproex. As with lithium, divalproex sodium appears more effective in treating and preventing hypomanic/manic episodes than depressive episodes in bipolar patients with rapid-cycling illness. Six open studies showed that patients who had not responded to lithium tended to do better with divalproex.¹⁵

Calabrese et al then tested the hypothesis that rapid cycling predicts nonresponse to lithium and positive response to divalproex.¹⁶ In a randomized controlled trial, they enrolled 254 recently hypomanic/manic rapid-cycling outpatients in an open-label stabilization phase involving combination lithium and divalproex therapy. Stabilized patients were then randomized to monotherapy with lithium, serum level ≥ 0.8 mEq/L, or divalproex, serum level ≥ 50 mcg/mL. Only 60 patients (24%) met stability criteria for randomization, achieving a persistent bimodal response as measured by continuous weeks of:

• Hamilton depression scale (24-item) score ≤ 20
• Young Mania Rating Scale score ≤ 12.5
• Global Assessment Scale score ≥ 51.

Most nonresponse was attributed to refractory depression.

After 20 months of maintenance therapy, about one-half of patients relapsed on either monotherapy. In the survival analysis, the median time to any mood episode was 45 weeks with divalproex monotherapy and 18 weeks with lithium monotherapy, although this difference was not statistically significant. The small sample size and high dropout rate may have created a false-negative error in this study.

Thus, these data did not show divalproex monotherapy to be more effective than lithium monotherapy in managing rapid-cycling bipolar disorder. The combination proved more effective in treating mania than depression and superior to monotherapy. This finding underscores combination therapy’s importance and the need to use mood stabilizers that also treat the depressed phase of bipolar disorder in rapid cyclers.

Table

Clinical characteristics of rapid cycling

Carbamazepine. Recent data refute earlier reports suggesting that rapid cycling predicted positive response to carbamazepine. Multiple open studies and four controlled studies suggest that carbamazepine—like lithium and divalproex—possesses moderate to marked efficacy in the hypomanic/manic phase but poor to moderate efficacy in the depressed phase of rapid-cycling bipolar disorder.¹⁷

Lamotrigine. Lamotrigine is the first mood-stabilizing agent that has shown efficacy in maintenance treatment of bipolar depression and rapid cycling. In a double-blind, prospective, placebo-controlled trial, Calabrese et al¹⁸ enrolled 324 rapid-cycling patients with bipolar disorder type I or II in an open-label stabilization phase with lamotrigine. The 182 stabilized patients were then randomly assigned to receive either lamotrigine (mean 288 +/- 94 mg/d) or placebo.

For 6 months, 41% of patients receiving lamotrigine and 26% of those receiving placebo remained stable without relapse (P = 0.03), although the difference was statistically significant only for the bipolar II subtype. Lamotrigine appeared most effective in patients with the biphasic pattern of depression-hypomania/mania-euthymia.

Topiramate. Most studies of topiramate in rapid cycling have been retrospective and/or small add-on studies to existing mood stabilizers, with topiramate use associated with moderately or markedly improved manic symptoms.¹⁹ Evidence supports further controlled investigations, particularly because topiramate’s weight-loss effects may help overweight or obese patients.

Gabapentin. Gabapentin’s efficacy in rapid cycling has not been established. Although open-label studies showed a 67% response rate when gabapentin was used as adjunctive therapy, two double-blind, placebocontrolled studies of bipolar patients failed to show efficacy.^20,21

Atypical antipsychotics. Five atypical antipsychotics—aripiprazole, olanzapine, quetiapine, risperidone, and ziprasidone—are FDA-approved for treating acute mania. Olanzapine is also indicated for bipolar maintenance treatment and has the most data showing efficacy in rapid cycling:

• In a 3-week, placebo-controlled study of 139 patients with bipolar I acute mania, olanzapine (median modal 15 mg/d) reduced manic symptoms to a statistically significantly extent in the 45 rapid cyclers.²²
• A long-term prospective study followed 23 patients—30% of whom were rapid cyclers—who used olanzapine (mean 8.2 mg/d) as an adjunct to mood stabilizers. Manic and depressive symptoms were reduced significantly in the cohort, which was followed for a mean of 303 days.²³
• An 8-week, double-blind, placebo-controlled study²⁴ compared olanzapine monotherapy with the olanzapine/fluoxetine combination (OFC) in 833 depressed bipolar I patients, of whom 315 (37%) had rapid cycling. Mean olanzapine dosage was 9.7 mg in monotherapy and 7.4 mg in combination therapy; mean fluoxetine dosage was 39.3 mg.

A follow-up analysis²⁵ showed that rapid cyclers’ depressive symptoms improved rapidly, and this improvement was sustained with OFC but not olanzapine monotherapy. Nonrapid-cycling patients responded to both treatments.

Other atypicals have shown partial efficacy in rapid-cycling bipolar disorder, although the studies have had methodologic limitations. Suppes et al²⁶ conducted the first controlled trial using clozapine as add-on therapy in a 1-year, randomized evaluation of 38 patients with treatment-refractory bipolar disorder. The 21 rapid cyclers received a mean peak of 234 mg/d. Brief Psychiatric Rating Scale and Clinical Global Improvement scores improved significantly overall, but data specific to the rapid-cycling patients were not reported.

Small, open-label studies using risperidone and quetiapine as adjuncts to mood stabilizers have shown modest efficacy in rapid cycling, usually in treating manic symptoms. A recent 8-week, double-blind, placebo-controlled trial of quetiapine in bipolar depression showed promising results, though its efficacy in rapid cycling was not reported.²⁷

RECOMMENDED TREATMENT

Because coincidental cycling may give the false appearance of efficacy in the short term, we recommend that you treat rapid cyclers methodically and judge outcomes over several months or cycle-lengths. A general approach includes:

• identify and treat underlying medical illnesses, such as hypothyroidism
• identify and treat comorbid alcohol/drug abuse
• taper or discontinue cycle-inducing agents such as antidepressants or sympathomimetics
• use standard mood stabilizers and/or atypical antipsychotics alone or in combination (Algorithm).

Algorithm Managing manic and depressive phases of rapid-cycling bipolar disorder

Treating the depressed phase in rapid cyclers is far more difficult than treating acute mania and may depend on bipolar subtype:

• Bipolar I patients likely will require one or more mood stabilizers (such as lithium, divalproex, olanzapine) plus add-on lamotrigine.
• Bipolar II patients may benefit from lamotrigine alone.
• Atypical antipsychotics that have putative antidepressant effects without apparent cycle-accelerating effects may also be considered. At this time, olanzapine has the most data.

Given depression’s refractory nature in rapid-cycling bipolar illness, you may need to combine any of the above medications, try electroconvulsive therapy, or use more-experimental strategies such as:

• omega-3 fatty acids
• donepezil
• pramipexole
• high-dose levothyroxine/T4.

Antidepressants. Before using antidepressants to treat bipolar depression, consider carefully the risk of initiating or exacerbating rapid cycling. No definitive evidence is available to guide your decision.

Likewise, the optimal duration of antidepressant treatment is unclear, although tapering the antidepressant as tolerated may be prudent after depressive symptoms are in remission.

Psychosocial interventions. Finally, don’t overlook psychosocial interventions. Bipolar-specific psychotherapies can enhance compliance, lessen depression, and improve treatment response.²⁸

CONCLUSION

Standard mood stabilizers appear to show partial efficacy in rapid cycling’s hypomanic/manic phase but only modest efficacy in the depressed phase. Lamotrigine appears more-promising in treating depressive than acute manic episodes and may be particularly effective for bipolar II patients. Evidence is growing that atypical antipsychotics also have partial efficacy in treating rapid cyclers, though whether this effect is phase-specific is unclear. As no single agent provides ideal bimodal treatment, combination therapy is recommended.

Related resources

• Bipolar Clinic and Research Program. Massachusetts General Hospital. Includes tools for clinicians and the clinical site for the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD). www.manicdepressive.org. Accessed Oct. 14, 2004.
• Goodwin FK, Jamison KR. Manic-depressive illness. New York: Oxford University Press, 1990.
• Marneros A, Goodwin FK (eds). Bipolar disorders: Mixed states, rapid cycling and atypical bipolar disorder. Cambridge, UK: Cambridge University Press (in press).

Drug brand names

• Aripiprazole • Abilify
• Carbamazepine • Carbatrol, others
• Clozapine • Clozaril
• Donepezil • Aricept
• Divalproex sodium • Depakote
• Gabapentin • Neurontin
• Lamotrigine • Lamictal
• Levothyroxine • Synthroid, others
• Lithium • Lithobid, others
• Olanzapine • Zyprexa
• Olanzapine/fluoxetine • Symbyax
• Oxcarbazepine • Trileptal
• Pramipexole • Mirapex
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Topiramate • Topamax
• Ziprasidone • Geodon

Disclosures

Dr. Altman is a speaker for Forest Pharmaceuticals, Janssen Pharmaceutica, AstraZeneca Pharmaceuticals, and Abbott Laboratories.

Dr. Schneck is a consultant to AstraZeneca Pharmaceuticals, UCB Pharma, and Bristol-Myers Squibb Co. and a speaker for AstraZeneca Pharmaceuticals.

Rapid-cycling bipolar disorder is a moving target, with treatment-resistant depression recurring frequently and alternating with hypomanic/manic episodes (Box).^1,2 Can one medication adequately treat these complicated patients, or is combination therapy necessary? If more than one medication is needed, are some combinations more effective than others?

This article attempts to answer these questions by:

• discussing recent treatment trial results
• suggesting an algorithm for managing hypomanic/manic and depressive episodes in rapid-cycling patients with bipolar disorder types I or II.

CLINICAL CHARACTERISTICS

Rapid cycling is associated most consistently with female gender and bipolar II disorder² (Table); why these two groups are primarily affected is unknown. Results of studies linking rapid cycling with hypothyroidism, gonadal steroid effects, family history, and substance use have been inconsistent and contradictory.²

Age of onset. Recent studies examining bipolar disorder’s age of onset have contradicted earlier rapid-cycling literature. In two large studies, Schneck et al³ and Coryell et al⁴ found rapid cycling associated with early onset of bipolar illness. The authors note that high rates of rapid cycling in children and adolescents resemble adult rapid cycling and speculate that early-onset bipolar illness might lead to rapid cycling vulnerability.⁵

Box

Transient vs persistent state. Rapid cycling is thought to be either a transient state in long-term bipolar illness or a more chronic expression of the illness. Several studies^6,7 have described rapid cycling as a transient phenomenon, whereas others^8-11 have found a more persistent rapid cycling course during follow-up. Interestingly, a recent study¹¹ suggested the mood-cycle pattern may be the most important predictor of rapid cycling. Patients with a depression–hypomania/mania-euthymia course demonstrated more-persistent rapid cycling than did those with a hypomania/mania-depression-euthymia course.

Antidepressants. Antidepressants’ role in initiating or exacerbating rapid cycling also remains unclear. Wehr et al⁸ found that discontinuing antidepressants contributed to cycling cessation or slowing. However, two prospective studies by Coryell et al⁴ that controlled for major depression found no association between antidepressant use and rapid cycling.

More recently, Yildiz and Sachs¹² found a possible gender-specific relationship between antidepressants and rapid cycling. Women exposed to antidepressants before their first hypomanic/manic episode were more likely to develop rapid cycling than women who were not so exposed. This association was not evident in men.

NO DEFINITIVE CHOICES

Any discussion of treating rapid-cycling bipolar disorder is based on limited data, as few prospective studies of this exclusive cohort exist. Many studies report on mixed cohorts of refractory bipolar patients that include rapid cyclers, but separate analyses of rapid-cycling subgroups are not usually reported. Notable exceptions are recent studies by Calabrese et al, which are discussed below.

Lithium. Dunner and Fieve¹³ were the first to suggest that rapid-cycling bipolar patients respond poorly to lithium maintenance monotherapy. Later studies, however, suggested that lithium could benefit rapid cyclers, primarily in reducing hypomanic or manic episodes.

Baldessarini et al¹⁰ found that lithium was less effective for rapid than nonrapid cyclers only in reducing recurrence of depressive episodes. Kukopulos et al¹⁴ reported that lithium response in rapid cyclers increased from 16% to 78% after antidepressants were stopped, suggesting that a positive response to lithium may require more limited antidepressant use (or patients not having been exposed to antidepressants at all).

Thus, lithium prophylaxis has at least partial efficacy in many rapid cyclers, especially when antidepressants are avoided.

Divalproex. As with lithium, divalproex sodium appears more effective in treating and preventing hypomanic/manic episodes than depressive episodes in bipolar patients with rapid-cycling illness. Six open studies showed that patients who had not responded to lithium tended to do better with divalproex.¹⁵

Calabrese et al then tested the hypothesis that rapid cycling predicts nonresponse to lithium and positive response to divalproex.¹⁶ In a randomized controlled trial, they enrolled 254 recently hypomanic/manic rapid-cycling outpatients in an open-label stabilization phase involving combination lithium and divalproex therapy. Stabilized patients were then randomized to monotherapy with lithium, serum level ≥ 0.8 mEq/L, or divalproex, serum level ≥ 50 mcg/mL. Only 60 patients (24%) met stability criteria for randomization, achieving a persistent bimodal response as measured by continuous weeks of:

• Hamilton depression scale (24-item) score ≤ 20
• Young Mania Rating Scale score ≤ 12.5
• Global Assessment Scale score ≥ 51.

Most nonresponse was attributed to refractory depression.

After 20 months of maintenance therapy, about one-half of patients relapsed on either monotherapy. In the survival analysis, the median time to any mood episode was 45 weeks with divalproex monotherapy and 18 weeks with lithium monotherapy, although this difference was not statistically significant. The small sample size and high dropout rate may have created a false-negative error in this study.

Thus, these data did not show divalproex monotherapy to be more effective than lithium monotherapy in managing rapid-cycling bipolar disorder. The combination proved more effective in treating mania than depression and superior to monotherapy. This finding underscores combination therapy’s importance and the need to use mood stabilizers that also treat the depressed phase of bipolar disorder in rapid cyclers.

Table

Clinical characteristics of rapid cycling

Carbamazepine. Recent data refute earlier reports suggesting that rapid cycling predicted positive response to carbamazepine. Multiple open studies and four controlled studies suggest that carbamazepine—like lithium and divalproex—possesses moderate to marked efficacy in the hypomanic/manic phase but poor to moderate efficacy in the depressed phase of rapid-cycling bipolar disorder.¹⁷

Lamotrigine. Lamotrigine is the first mood-stabilizing agent that has shown efficacy in maintenance treatment of bipolar depression and rapid cycling. In a double-blind, prospective, placebo-controlled trial, Calabrese et al¹⁸ enrolled 324 rapid-cycling patients with bipolar disorder type I or II in an open-label stabilization phase with lamotrigine. The 182 stabilized patients were then randomly assigned to receive either lamotrigine (mean 288 +/- 94 mg/d) or placebo.

For 6 months, 41% of patients receiving lamotrigine and 26% of those receiving placebo remained stable without relapse (P = 0.03), although the difference was statistically significant only for the bipolar II subtype. Lamotrigine appeared most effective in patients with the biphasic pattern of depression-hypomania/mania-euthymia.

Topiramate. Most studies of topiramate in rapid cycling have been retrospective and/or small add-on studies to existing mood stabilizers, with topiramate use associated with moderately or markedly improved manic symptoms.¹⁹ Evidence supports further controlled investigations, particularly because topiramate’s weight-loss effects may help overweight or obese patients.

Gabapentin. Gabapentin’s efficacy in rapid cycling has not been established. Although open-label studies showed a 67% response rate when gabapentin was used as adjunctive therapy, two double-blind, placebocontrolled studies of bipolar patients failed to show efficacy.^20,21

Atypical antipsychotics. Five atypical antipsychotics—aripiprazole, olanzapine, quetiapine, risperidone, and ziprasidone—are FDA-approved for treating acute mania. Olanzapine is also indicated for bipolar maintenance treatment and has the most data showing efficacy in rapid cycling:

• In a 3-week, placebo-controlled study of 139 patients with bipolar I acute mania, olanzapine (median modal 15 mg/d) reduced manic symptoms to a statistically significantly extent in the 45 rapid cyclers.²²
• A long-term prospective study followed 23 patients—30% of whom were rapid cyclers—who used olanzapine (mean 8.2 mg/d) as an adjunct to mood stabilizers. Manic and depressive symptoms were reduced significantly in the cohort, which was followed for a mean of 303 days.²³
• An 8-week, double-blind, placebo-controlled study²⁴ compared olanzapine monotherapy with the olanzapine/fluoxetine combination (OFC) in 833 depressed bipolar I patients, of whom 315 (37%) had rapid cycling. Mean olanzapine dosage was 9.7 mg in monotherapy and 7.4 mg in combination therapy; mean fluoxetine dosage was 39.3 mg.

A follow-up analysis²⁵ showed that rapid cyclers’ depressive symptoms improved rapidly, and this improvement was sustained with OFC but not olanzapine monotherapy. Nonrapid-cycling patients responded to both treatments.

Other atypicals have shown partial efficacy in rapid-cycling bipolar disorder, although the studies have had methodologic limitations. Suppes et al²⁶ conducted the first controlled trial using clozapine as add-on therapy in a 1-year, randomized evaluation of 38 patients with treatment-refractory bipolar disorder. The 21 rapid cyclers received a mean peak of 234 mg/d. Brief Psychiatric Rating Scale and Clinical Global Improvement scores improved significantly overall, but data specific to the rapid-cycling patients were not reported.

Small, open-label studies using risperidone and quetiapine as adjuncts to mood stabilizers have shown modest efficacy in rapid cycling, usually in treating manic symptoms. A recent 8-week, double-blind, placebo-controlled trial of quetiapine in bipolar depression showed promising results, though its efficacy in rapid cycling was not reported.²⁷

RECOMMENDED TREATMENT

Because coincidental cycling may give the false appearance of efficacy in the short term, we recommend that you treat rapid cyclers methodically and judge outcomes over several months or cycle-lengths. A general approach includes:

• identify and treat underlying medical illnesses, such as hypothyroidism
• identify and treat comorbid alcohol/drug abuse
• taper or discontinue cycle-inducing agents such as antidepressants or sympathomimetics
• use standard mood stabilizers and/or atypical antipsychotics alone or in combination (Algorithm).

Algorithm Managing manic and depressive phases of rapid-cycling bipolar disorder

Treating the depressed phase in rapid cyclers is far more difficult than treating acute mania and may depend on bipolar subtype:

• Bipolar I patients likely will require one or more mood stabilizers (such as lithium, divalproex, olanzapine) plus add-on lamotrigine.
• Bipolar II patients may benefit from lamotrigine alone.
• Atypical antipsychotics that have putative antidepressant effects without apparent cycle-accelerating effects may also be considered. At this time, olanzapine has the most data.

Given depression’s refractory nature in rapid-cycling bipolar illness, you may need to combine any of the above medications, try electroconvulsive therapy, or use more-experimental strategies such as:

• omega-3 fatty acids
• donepezil
• pramipexole
• high-dose levothyroxine/T4.

Antidepressants. Before using antidepressants to treat bipolar depression, consider carefully the risk of initiating or exacerbating rapid cycling. No definitive evidence is available to guide your decision.

Likewise, the optimal duration of antidepressant treatment is unclear, although tapering the antidepressant as tolerated may be prudent after depressive symptoms are in remission.

Psychosocial interventions. Finally, don’t overlook psychosocial interventions. Bipolar-specific psychotherapies can enhance compliance, lessen depression, and improve treatment response.²⁸

CONCLUSION

Standard mood stabilizers appear to show partial efficacy in rapid cycling’s hypomanic/manic phase but only modest efficacy in the depressed phase. Lamotrigine appears more-promising in treating depressive than acute manic episodes and may be particularly effective for bipolar II patients. Evidence is growing that atypical antipsychotics also have partial efficacy in treating rapid cyclers, though whether this effect is phase-specific is unclear. As no single agent provides ideal bimodal treatment, combination therapy is recommended.

Related resources

• Bipolar Clinic and Research Program. Massachusetts General Hospital. Includes tools for clinicians and the clinical site for the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD). www.manicdepressive.org. Accessed Oct. 14, 2004.
• Goodwin FK, Jamison KR. Manic-depressive illness. New York: Oxford University Press, 1990.
• Marneros A, Goodwin FK (eds). Bipolar disorders: Mixed states, rapid cycling and atypical bipolar disorder. Cambridge, UK: Cambridge University Press (in press).

Drug brand names

• Aripiprazole • Abilify
• Carbamazepine • Carbatrol, others
• Clozapine • Clozaril
• Donepezil • Aricept
• Divalproex sodium • Depakote
• Gabapentin • Neurontin
• Lamotrigine • Lamictal
• Levothyroxine • Synthroid, others
• Lithium • Lithobid, others
• Olanzapine • Zyprexa
• Olanzapine/fluoxetine • Symbyax
• Oxcarbazepine • Trileptal
• Pramipexole • Mirapex
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Topiramate • Topamax
• Ziprasidone • Geodon

Disclosures

Dr. Altman is a speaker for Forest Pharmaceuticals, Janssen Pharmaceutica, AstraZeneca Pharmaceuticals, and Abbott Laboratories.

Dr. Schneck is a consultant to AstraZeneca Pharmaceuticals, UCB Pharma, and Bristol-Myers Squibb Co. and a speaker for AstraZeneca Pharmaceuticals.

Rapid-cycling bipolar disorder is a moving target, with treatment-resistant depression recurring frequently and alternating with hypomanic/manic episodes (Box).^1,2 Can one medication adequately treat these complicated patients, or is combination therapy necessary? If more than one medication is needed, are some combinations more effective than others?

This article attempts to answer these questions by:

• discussing recent treatment trial results
• suggesting an algorithm for managing hypomanic/manic and depressive episodes in rapid-cycling patients with bipolar disorder types I or II.

CLINICAL CHARACTERISTICS

Rapid cycling is associated most consistently with female gender and bipolar II disorder² (Table); why these two groups are primarily affected is unknown. Results of studies linking rapid cycling with hypothyroidism, gonadal steroid effects, family history, and substance use have been inconsistent and contradictory.²

Age of onset. Recent studies examining bipolar disorder’s age of onset have contradicted earlier rapid-cycling literature. In two large studies, Schneck et al³ and Coryell et al⁴ found rapid cycling associated with early onset of bipolar illness. The authors note that high rates of rapid cycling in children and adolescents resemble adult rapid cycling and speculate that early-onset bipolar illness might lead to rapid cycling vulnerability.⁵

Box

Transient vs persistent state. Rapid cycling is thought to be either a transient state in long-term bipolar illness or a more chronic expression of the illness. Several studies^6,7 have described rapid cycling as a transient phenomenon, whereas others^8-11 have found a more persistent rapid cycling course during follow-up. Interestingly, a recent study¹¹ suggested the mood-cycle pattern may be the most important predictor of rapid cycling. Patients with a depression–hypomania/mania-euthymia course demonstrated more-persistent rapid cycling than did those with a hypomania/mania-depression-euthymia course.

Antidepressants. Antidepressants’ role in initiating or exacerbating rapid cycling also remains unclear. Wehr et al⁸ found that discontinuing antidepressants contributed to cycling cessation or slowing. However, two prospective studies by Coryell et al⁴ that controlled for major depression found no association between antidepressant use and rapid cycling.

More recently, Yildiz and Sachs¹² found a possible gender-specific relationship between antidepressants and rapid cycling. Women exposed to antidepressants before their first hypomanic/manic episode were more likely to develop rapid cycling than women who were not so exposed. This association was not evident in men.

NO DEFINITIVE CHOICES

Any discussion of treating rapid-cycling bipolar disorder is based on limited data, as few prospective studies of this exclusive cohort exist. Many studies report on mixed cohorts of refractory bipolar patients that include rapid cyclers, but separate analyses of rapid-cycling subgroups are not usually reported. Notable exceptions are recent studies by Calabrese et al, which are discussed below.

Lithium. Dunner and Fieve¹³ were the first to suggest that rapid-cycling bipolar patients respond poorly to lithium maintenance monotherapy. Later studies, however, suggested that lithium could benefit rapid cyclers, primarily in reducing hypomanic or manic episodes.

Baldessarini et al¹⁰ found that lithium was less effective for rapid than nonrapid cyclers only in reducing recurrence of depressive episodes. Kukopulos et al¹⁴ reported that lithium response in rapid cyclers increased from 16% to 78% after antidepressants were stopped, suggesting that a positive response to lithium may require more limited antidepressant use (or patients not having been exposed to antidepressants at all).

Thus, lithium prophylaxis has at least partial efficacy in many rapid cyclers, especially when antidepressants are avoided.

Divalproex. As with lithium, divalproex sodium appears more effective in treating and preventing hypomanic/manic episodes than depressive episodes in bipolar patients with rapid-cycling illness. Six open studies showed that patients who had not responded to lithium tended to do better with divalproex.¹⁵

Calabrese et al then tested the hypothesis that rapid cycling predicts nonresponse to lithium and positive response to divalproex.¹⁶ In a randomized controlled trial, they enrolled 254 recently hypomanic/manic rapid-cycling outpatients in an open-label stabilization phase involving combination lithium and divalproex therapy. Stabilized patients were then randomized to monotherapy with lithium, serum level ≥ 0.8 mEq/L, or divalproex, serum level ≥ 50 mcg/mL. Only 60 patients (24%) met stability criteria for randomization, achieving a persistent bimodal response as measured by continuous weeks of:

• Hamilton depression scale (24-item) score ≤ 20
• Young Mania Rating Scale score ≤ 12.5
• Global Assessment Scale score ≥ 51.

Most nonresponse was attributed to refractory depression.

After 20 months of maintenance therapy, about one-half of patients relapsed on either monotherapy. In the survival analysis, the median time to any mood episode was 45 weeks with divalproex monotherapy and 18 weeks with lithium monotherapy, although this difference was not statistically significant. The small sample size and high dropout rate may have created a false-negative error in this study.

Thus, these data did not show divalproex monotherapy to be more effective than lithium monotherapy in managing rapid-cycling bipolar disorder. The combination proved more effective in treating mania than depression and superior to monotherapy. This finding underscores combination therapy’s importance and the need to use mood stabilizers that also treat the depressed phase of bipolar disorder in rapid cyclers.

Table

Clinical characteristics of rapid cycling

Carbamazepine. Recent data refute earlier reports suggesting that rapid cycling predicted positive response to carbamazepine. Multiple open studies and four controlled studies suggest that carbamazepine—like lithium and divalproex—possesses moderate to marked efficacy in the hypomanic/manic phase but poor to moderate efficacy in the depressed phase of rapid-cycling bipolar disorder.¹⁷

Lamotrigine. Lamotrigine is the first mood-stabilizing agent that has shown efficacy in maintenance treatment of bipolar depression and rapid cycling. In a double-blind, prospective, placebo-controlled trial, Calabrese et al¹⁸ enrolled 324 rapid-cycling patients with bipolar disorder type I or II in an open-label stabilization phase with lamotrigine. The 182 stabilized patients were then randomly assigned to receive either lamotrigine (mean 288 +/- 94 mg/d) or placebo.

For 6 months, 41% of patients receiving lamotrigine and 26% of those receiving placebo remained stable without relapse (P = 0.03), although the difference was statistically significant only for the bipolar II subtype. Lamotrigine appeared most effective in patients with the biphasic pattern of depression-hypomania/mania-euthymia.

Topiramate. Most studies of topiramate in rapid cycling have been retrospective and/or small add-on studies to existing mood stabilizers, with topiramate use associated with moderately or markedly improved manic symptoms.¹⁹ Evidence supports further controlled investigations, particularly because topiramate’s weight-loss effects may help overweight or obese patients.

Gabapentin. Gabapentin’s efficacy in rapid cycling has not been established. Although open-label studies showed a 67% response rate when gabapentin was used as adjunctive therapy, two double-blind, placebocontrolled studies of bipolar patients failed to show efficacy.^20,21

Atypical antipsychotics. Five atypical antipsychotics—aripiprazole, olanzapine, quetiapine, risperidone, and ziprasidone—are FDA-approved for treating acute mania. Olanzapine is also indicated for bipolar maintenance treatment and has the most data showing efficacy in rapid cycling:

• In a 3-week, placebo-controlled study of 139 patients with bipolar I acute mania, olanzapine (median modal 15 mg/d) reduced manic symptoms to a statistically significantly extent in the 45 rapid cyclers.²²
• A long-term prospective study followed 23 patients—30% of whom were rapid cyclers—who used olanzapine (mean 8.2 mg/d) as an adjunct to mood stabilizers. Manic and depressive symptoms were reduced significantly in the cohort, which was followed for a mean of 303 days.²³
• An 8-week, double-blind, placebo-controlled study²⁴ compared olanzapine monotherapy with the olanzapine/fluoxetine combination (OFC) in 833 depressed bipolar I patients, of whom 315 (37%) had rapid cycling. Mean olanzapine dosage was 9.7 mg in monotherapy and 7.4 mg in combination therapy; mean fluoxetine dosage was 39.3 mg.

A follow-up analysis²⁵ showed that rapid cyclers’ depressive symptoms improved rapidly, and this improvement was sustained with OFC but not olanzapine monotherapy. Nonrapid-cycling patients responded to both treatments.

Other atypicals have shown partial efficacy in rapid-cycling bipolar disorder, although the studies have had methodologic limitations. Suppes et al²⁶ conducted the first controlled trial using clozapine as add-on therapy in a 1-year, randomized evaluation of 38 patients with treatment-refractory bipolar disorder. The 21 rapid cyclers received a mean peak of 234 mg/d. Brief Psychiatric Rating Scale and Clinical Global Improvement scores improved significantly overall, but data specific to the rapid-cycling patients were not reported.

Small, open-label studies using risperidone and quetiapine as adjuncts to mood stabilizers have shown modest efficacy in rapid cycling, usually in treating manic symptoms. A recent 8-week, double-blind, placebo-controlled trial of quetiapine in bipolar depression showed promising results, though its efficacy in rapid cycling was not reported.²⁷

RECOMMENDED TREATMENT

Because coincidental cycling may give the false appearance of efficacy in the short term, we recommend that you treat rapid cyclers methodically and judge outcomes over several months or cycle-lengths. A general approach includes:

• identify and treat underlying medical illnesses, such as hypothyroidism
• identify and treat comorbid alcohol/drug abuse
• taper or discontinue cycle-inducing agents such as antidepressants or sympathomimetics
• use standard mood stabilizers and/or atypical antipsychotics alone or in combination (Algorithm).

Algorithm Managing manic and depressive phases of rapid-cycling bipolar disorder

Treating the depressed phase in rapid cyclers is far more difficult than treating acute mania and may depend on bipolar subtype:

• Bipolar I patients likely will require one or more mood stabilizers (such as lithium, divalproex, olanzapine) plus add-on lamotrigine.
• Bipolar II patients may benefit from lamotrigine alone.
• Atypical antipsychotics that have putative antidepressant effects without apparent cycle-accelerating effects may also be considered. At this time, olanzapine has the most data.

Given depression’s refractory nature in rapid-cycling bipolar illness, you may need to combine any of the above medications, try electroconvulsive therapy, or use more-experimental strategies such as:

• omega-3 fatty acids
• donepezil
• pramipexole
• high-dose levothyroxine/T4.

Antidepressants. Before using antidepressants to treat bipolar depression, consider carefully the risk of initiating or exacerbating rapid cycling. No definitive evidence is available to guide your decision.

Likewise, the optimal duration of antidepressant treatment is unclear, although tapering the antidepressant as tolerated may be prudent after depressive symptoms are in remission.

Psychosocial interventions. Finally, don’t overlook psychosocial interventions. Bipolar-specific psychotherapies can enhance compliance, lessen depression, and improve treatment response.²⁸

CONCLUSION

Standard mood stabilizers appear to show partial efficacy in rapid cycling’s hypomanic/manic phase but only modest efficacy in the depressed phase. Lamotrigine appears more-promising in treating depressive than acute manic episodes and may be particularly effective for bipolar II patients. Evidence is growing that atypical antipsychotics also have partial efficacy in treating rapid cyclers, though whether this effect is phase-specific is unclear. As no single agent provides ideal bimodal treatment, combination therapy is recommended.

Related resources

• Bipolar Clinic and Research Program. Massachusetts General Hospital. Includes tools for clinicians and the clinical site for the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD). www.manicdepressive.org. Accessed Oct. 14, 2004.
• Goodwin FK, Jamison KR. Manic-depressive illness. New York: Oxford University Press, 1990.
• Marneros A, Goodwin FK (eds). Bipolar disorders: Mixed states, rapid cycling and atypical bipolar disorder. Cambridge, UK: Cambridge University Press (in press).

Drug brand names

• Aripiprazole • Abilify
• Carbamazepine • Carbatrol, others
• Clozapine • Clozaril
• Donepezil • Aricept
• Divalproex sodium • Depakote
• Gabapentin • Neurontin
• Lamotrigine • Lamictal
• Levothyroxine • Synthroid, others
• Lithium • Lithobid, others
• Olanzapine • Zyprexa
• Olanzapine/fluoxetine • Symbyax
• Oxcarbazepine • Trileptal
• Pramipexole • Mirapex
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Topiramate • Topamax
• Ziprasidone • Geodon

Disclosures

Dr. Altman is a speaker for Forest Pharmaceuticals, Janssen Pharmaceutica, AstraZeneca Pharmaceuticals, and Abbott Laboratories.

Dr. Schneck is a consultant to AstraZeneca Pharmaceuticals, UCB Pharma, and Bristol-Myers Squibb Co. and a speaker for AstraZeneca Pharmaceuticals.

Boys will be boys” and other platitudes may condone adolescent reckless driving, substance use, or sexual promiscuity—but to write off dangerous behavior as normal would be a mistake. Because adolescent impulsivity and sensation-seeking may have physiologic as well as emotional causes,^1,2 excessive risk taking may be treatable.

This article discusses the neurobiology of adolescent risk taking, suggests how to determine when such behavior may be pathologic, and offers a treatment approach for at-risk teens and their parents.

CASE: ‘WHAT’S WRONG WITH OUR SON?’

Josh, age 17, is brought to the adolescent psychiatry clinic by his distraught parents, who report their son has undergone a “personality change” over the past 2 years. They recall that he was respectful, studious, and soft-spoken until age 15. Since then, he has been skipping school, staying out late at night with his friends, and “obsessed” with TV poker games.

His parents recently discovered he has been gambling for money, which greatly upsets them. They also found a pack of cigarettes in their son’s car and are concerned that he might be using other substances. What finally prompted the psychiatric visit was Josh’s recent traffic citation for driving 25 miles over the speed limit.

CAUSES OF RISK TAKING

Normal development. In the absence of psychopathology, adolescent risk taking appears to be a normal development stage that is vital to successful transition to adulthood. This assumes that adolescents such as Josh learn to moderate their behavior and avoid long-term negative consequences.

Impulsivity and sensation seeking are recognized as key factors in adolescent risk taking Box 1.^1-4 Apparently, these traits result primarily from incomplete neural circuit maturation. Adolescent brain regions involved in impulsivity and risk taking are also involved in reward, and these centers exhibit an exaggerated response to stimuli.⁵ This amplified response may help explain an adolescent’s propensity for risky behavior.

Despite potential hazards, adolescent risk taking may confer benefits. In taking risks, adolescents:

• explore adult behavior
• learn to accomplish increasingly difficult developmental tasks
• reinforce their self-esteem.

Adolescent risk-takers have been found to be more self-confident, to feel more accepted, and to be better liked than their more-cautious peers.⁶

Psychiatric comorbidity. Excessive risk taking can be associated with psychiatric illness, including bipolar mania, psychosis, substance abuse, and impulse control disorders. Individuals with borderline personality and other cluster B disorders have marked impulsivity and thus are prone to risky behavior.

Teens with attention-deficit/hyperactivity disorder (ADHD), conduct disorder, and oppositional-defiant disorder (ODD) also tend to exhibit high impulsivity.

Box 1

CASE CONTINUED: JUST ‘HAVING FUN’

When interviewed alone, Josh admits to “occasional” truancy, which he attributes to being “bored” with school and wanting to spend time with his friends “doing fun stuff, like going to the beach.” He admits to gambling for money and smoking a half-pack of cigarettes daily, as well as drinking beer and smoking marijuana “a few times a week.”

Josh says he engages in these activities “because they’re fun,” and states he is annoyed by his parents’ concern. He blames the speeding ticket on “not paying attention.” He admits to drinking and driving but claims he always feels “in control.”

He also reports he has been sexually active since age 16 and often has had unprotected intercourse. When asked if he is concerned that he might contract a sexually transmitted disease or impregnate his partner, Josh appears ambivalent.

IMPULSIVITY IN ADOLESCENCE

Josh is engaging in numerous impulsive behaviors. Adolescents generally are more impulsive than adults, as demonstrated by their significantly higher impulsivity scores on standardized tests.⁷ Furthermore, as measured by improved response inhibition (go/no go tasks), the level of adolescent impulsivity is inversely related to age.⁸

Problem behavior syndrome. High impulsivity is predictive of problem gambling, drug use, and risky driving and sexual practices later in life.^1,2,9-11 Adolescents with what some authors describe as a “problem behavior syndrome” engage in behaviors—such as substance use, risky sexual behavior,¹² gambling,¹³ and reckless driving¹⁴—that share a common trend toward impulsivity.

Impaired data processing. Decision making has been proposed as a three-part cognitive process:

• accumulating sensory input
• processing this input and formulating a behavioral response appropriate to the situation
• planning and implementing the resultant motor output.²

Impulsivity is believed to result from impaired ability of the brain to process accumulated information or to formulate a response to it—or both. Impulsive individuals thus experience impaired data processing, in which they:

• misjudge the likely risk of a given action or overestimate their ability to accomplish a task
• show impaired response inhibition and thus find it difficult to resist an impulse to participate in a given activity.

Sensation seeking. Adolescents who exhibit risk-taking behavior may wish to experience the thrill of the behavior (sensation or novelty seeking). Alcoholic or drug-dependent individuals and those who engage in pathologic gambling or take chances while driving also demonstrate significantly impaired decision making.^15-17 Adolescents who engage in these and other problem behaviors have similarly scored high on sensation-seeking scales.^10,18

DECISION-MAKING BIOLOGY

At least four neural circuits process decisions, weighing the risks and benefits of a given situation and formulating a response. These circuits are:

• prefrontal cortices, including orbitofrontal, dorsolateral, and ventromedial
• ventral striatum, including the nucleus accumbens
• thalamus
• monoaminergic brainstem nuclei (ventral tegmental area [VTA] and raphe nuclei).¹⁹

Functional imaging studies—including MRI and PET, EEG, and electrophysiology—have confirmed that these four brain regions are integral to response inhibition and show abnormal activity in impulsive individuals.^20,21 Indeed, prefrontal cortex damage has been extensively documented to cause marked impulsivity, poor decision making, and an increased propensity for substance abuse and dependency.¹

Functional imaging studies also have shown that adolescents appear to use these neural regions inefficiently during decision making. Extensive areas of the involved brain regions are activated in individuals ages 8 to 20, whereas only focal activation occurs in adults.²²

Box 2

Activity within this network is modulated by excitatory glutamatergic transmission and inhibitory GABAergic transmission within the cortices and by dopaminergic and serotonergic transmission within the VTA and raphe nuclei, respectively.^2,20 Although all of these neurotransmitters have been implicated in impulsivity, dopamine and serotonin have been studied most extensively (box 2).^1,2,23-26

CASE CONTINUED: PSYCHIATRIC WORKUP

Josh clearly is engaged in worrisome behavior with potential long-term consequences. To evaluate him for underlying psychopathology, the psychiatrist used a structured psychiatric exam, Minnesota Multi-phasic Personality Inventory (MMPI), and SNAP-IV Rating Scale for ADHD (see Related resources). Josh endorsed some depressive symptoms—which were also evident on the MMPI—but did not meet DSM-IVTR criteria for major depressive disorder. Neither were his symptoms diagnostic for any other Axis I or Axis II disorder.

Given the risk of harm and likelihood of worsening behavior over time, the psychiatrist schedules Josh for weekly psychotherapy and possible medication.

Psychosocial interventions are discussed with Josh’s parents, including monitoring his activities, restricting access to peers who have been a poor influence, reinforcing good behavior, and enlisting help from teachers and his friends’ parents. The effect of these interventions is to be explored in follow-up visits.

Box 3

DEFINING DEGREES OF RISK

Although no criteria differentiate “normal” from “pathologic” risk taking, the definition of taking a risk implies potential adverse consequences. In evaluating the impulsive adolescent, it is important to determine which behaviors:

• can be instructive and promote maturation
• fall outside normal adolescent behavior and/or carry potentially severe outcomes.

Acceptable. Risk taking is acceptable if the potential adverse outcome is relatively benign and the adolescent is likely to learn from the experience. For example, driving 10 miles over the speed limit and receiving a ticket can lead to stricter observance of the speed limit.

Pathologic. Josh clearly exhibits risky behaviors that one would reasonably consider “pathologic,” as they carry potentially severe consequences that exceed any possible developmental gain. For example, drinking and driving can result in a DUI citation and/or a motor vehicle accident with physical injuries or death.

TREATMENT OPTIONS

Psychiatric comorbidity. When you evaluate an adolescent engaged in excessive risk taking, consider Axis I and II disorders characterized by marked impulsivity. If the patient meets diagnostic criteria for a psychopathology such as bipolar disorder or ADHD, treating the underlying condition will likely improve impulsivity.

Recommended approach. Even without an Axis I or Axis II disorder, adolescents who engage in pathologic risky behavior may benefit from psychosocial interventions (Box 3), psychotherapy, and perhaps medication.

Because very little evidence supports using psychotropics to treat pathologically impulsive adolescents, we recommend that you:

• first try psychosocial interventions and psychotherapy
• reserve medications for patients who do not respond adequately to nondrug approaches and engage in impulsive behaviors that pose a high risk for grave consequences.

Psychotherapy can be effective once the adolescent and clinician form a therapeutic alliance. Because Josh—like other such teens—will likely view his psychiatrist as “just another adult lecturing me on what to do,” focus first on establishing rapport by:

• getting to know him
• helping him feel at ease
• showing interest in his thoughts and empathy towards his concerns and complaints
• discussing anything but the reason his parents brought him to your office.

After you establish an alliance, focus therapy on helping the adolescent gain insight into his or her dangerous behaviors and their consequences. To illustrate to Josh the potential consequences of his behaviors, for example, you might introduce him to:

• someone disabled in a motor vehicle accident
• an HIV-positive activist
• a recovering alcoholic
• a long-time smoker with severe chronic obstructive pulmonary disease.

At-risk adolescents also could be encouraged to complete an educational program that teaches alternate activities for sensation seeking (such as skiing instead of high-speed driving).

Medication. Although the monoaminergic systems are known to modulate impulsive behavior, few studies have examined using medications to treat risk-taking adolescents, and no drugs are FDA-approved for this indication.

SSRIs. Selective serotonin reuptake inhibitors such as fluoxetine, sertraline, or escitalopram might be useful for treating excessive adolescent risk taking. A preliminary study with paroxetine—an SSRI not recommended for children and adolescents—suggests this class of antidepressants may help reduce impulsivity.²⁶ In the absence of data specific to risk-taking behavior, we recommend using SSRI dosages similar to those used to treat mood disorders in adolescents.

Clomipramine acts mainly on the serotonin receptor, preventing serotonin reuptake in a manner similar to an SSRI. Because it has the greatest serotonergic effect in its drug class, clomipramine is the only tricyclic proven effective in obsessive-compulsive disorder.²⁷ Although no data have shown that clomipramine affects impulsivity, it theoretically could be effective because of its effect on serotonin.

Divalproex sodium has been shown to effectively treat impulsivity, particularly in patients with autism spectrum disorders, intermittent explosive disorder, schizophrenia, borderline personality disorder, and bipolar disorder.^27-30 As an off-label use, one could consider trying this agent in an adolescent with pathologic risk-taking behavior. Use the same dosages and obtain routine labs as indicated for adolescents with other disorders.

Adherence. Like Josh, adolescents who engage in high-risk behaviors often do not recognize their pathology and resist psychiatric intervention. Getting them to take medication or participate in psychotherapy can be quite difficult.

Adolescents are far more likely to adhere to treatment if you develop a rapport with them and they trust you. As psychotherapy and psychosocial interventions progress, patients become more likely to gain insight into their conditions and become more adherent.

Other options to encourage adherence include having the parent:

• administer the medication and ensure that the patient swallows it
• use rewards to reinforce the adolescent’s good behavior and adherence to treatment.

Follow up weekly with patients such as Josh who exhibit high-risk behaviors and require psychotherapy and medication. Follow less-acute patients 2 weeks after the initial evaluation, then monthly if they are responding to medication.

Related resources

• Strauch B. The primal teen: what the new discoveries about the teenage brain tell us about our kids. New York: Doubleday, 2004.
• SNAP-IV Rating Scale to screen for attention-deficit/hyperactivity disorder. Is your child really ADD/ADHD? www.drbiofeedback.com. Accessed Sept. 8, 2004.
• Focus Adolescent Services. Resources and information for families with adolescent behavior problems, including high-risk behavior. http://www.focusas.com/BehavioralDisorders.html. Accessed Aug. 26, 2004.

Drug brand names

• Clomipramine • Anafranil
• Divalproex sodium • Depakote
• Escitalopram • Lexapro
• Fluoxetine • Prozac
• Paroxetine • Paxil
• Sertraline • Zoloft

Disclosure

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

Dr Shapira receives grant or research support from Abbott Laboratories, Janssen Pharmaceutica, Ortho-McNeil Pharmaceutical, Bristol-Myers Squibb Co., Eli Lilly and Co., and Pfizer Inc. He is a speaker for AstraZeneca Pharmaceuticals, Forest Laboratories, and Ortho-McNeil Pharmaceutical, Inc.

Boys will be boys” and other platitudes may condone adolescent reckless driving, substance use, or sexual promiscuity—but to write off dangerous behavior as normal would be a mistake. Because adolescent impulsivity and sensation-seeking may have physiologic as well as emotional causes,^1,2 excessive risk taking may be treatable.

This article discusses the neurobiology of adolescent risk taking, suggests how to determine when such behavior may be pathologic, and offers a treatment approach for at-risk teens and their parents.

CASE: ‘WHAT’S WRONG WITH OUR SON?’

Josh, age 17, is brought to the adolescent psychiatry clinic by his distraught parents, who report their son has undergone a “personality change” over the past 2 years. They recall that he was respectful, studious, and soft-spoken until age 15. Since then, he has been skipping school, staying out late at night with his friends, and “obsessed” with TV poker games.

His parents recently discovered he has been gambling for money, which greatly upsets them. They also found a pack of cigarettes in their son’s car and are concerned that he might be using other substances. What finally prompted the psychiatric visit was Josh’s recent traffic citation for driving 25 miles over the speed limit.

CAUSES OF RISK TAKING

Normal development. In the absence of psychopathology, adolescent risk taking appears to be a normal development stage that is vital to successful transition to adulthood. This assumes that adolescents such as Josh learn to moderate their behavior and avoid long-term negative consequences.

Impulsivity and sensation seeking are recognized as key factors in adolescent risk taking Box 1.^1-4 Apparently, these traits result primarily from incomplete neural circuit maturation. Adolescent brain regions involved in impulsivity and risk taking are also involved in reward, and these centers exhibit an exaggerated response to stimuli.⁵ This amplified response may help explain an adolescent’s propensity for risky behavior.

Despite potential hazards, adolescent risk taking may confer benefits. In taking risks, adolescents:

• explore adult behavior
• learn to accomplish increasingly difficult developmental tasks
• reinforce their self-esteem.

Adolescent risk-takers have been found to be more self-confident, to feel more accepted, and to be better liked than their more-cautious peers.⁶

Psychiatric comorbidity. Excessive risk taking can be associated with psychiatric illness, including bipolar mania, psychosis, substance abuse, and impulse control disorders. Individuals with borderline personality and other cluster B disorders have marked impulsivity and thus are prone to risky behavior.

Teens with attention-deficit/hyperactivity disorder (ADHD), conduct disorder, and oppositional-defiant disorder (ODD) also tend to exhibit high impulsivity.

Box 1

CASE CONTINUED: JUST ‘HAVING FUN’

When interviewed alone, Josh admits to “occasional” truancy, which he attributes to being “bored” with school and wanting to spend time with his friends “doing fun stuff, like going to the beach.” He admits to gambling for money and smoking a half-pack of cigarettes daily, as well as drinking beer and smoking marijuana “a few times a week.”

Josh says he engages in these activities “because they’re fun,” and states he is annoyed by his parents’ concern. He blames the speeding ticket on “not paying attention.” He admits to drinking and driving but claims he always feels “in control.”

He also reports he has been sexually active since age 16 and often has had unprotected intercourse. When asked if he is concerned that he might contract a sexually transmitted disease or impregnate his partner, Josh appears ambivalent.

IMPULSIVITY IN ADOLESCENCE

Josh is engaging in numerous impulsive behaviors. Adolescents generally are more impulsive than adults, as demonstrated by their significantly higher impulsivity scores on standardized tests.⁷ Furthermore, as measured by improved response inhibition (go/no go tasks), the level of adolescent impulsivity is inversely related to age.⁸

Problem behavior syndrome. High impulsivity is predictive of problem gambling, drug use, and risky driving and sexual practices later in life.^1,2,9-11 Adolescents with what some authors describe as a “problem behavior syndrome” engage in behaviors—such as substance use, risky sexual behavior,¹² gambling,¹³ and reckless driving¹⁴—that share a common trend toward impulsivity.

Impaired data processing. Decision making has been proposed as a three-part cognitive process:

• accumulating sensory input
• processing this input and formulating a behavioral response appropriate to the situation
• planning and implementing the resultant motor output.²

Impulsivity is believed to result from impaired ability of the brain to process accumulated information or to formulate a response to it—or both. Impulsive individuals thus experience impaired data processing, in which they:

• misjudge the likely risk of a given action or overestimate their ability to accomplish a task
• show impaired response inhibition and thus find it difficult to resist an impulse to participate in a given activity.

Sensation seeking. Adolescents who exhibit risk-taking behavior may wish to experience the thrill of the behavior (sensation or novelty seeking). Alcoholic or drug-dependent individuals and those who engage in pathologic gambling or take chances while driving also demonstrate significantly impaired decision making.^15-17 Adolescents who engage in these and other problem behaviors have similarly scored high on sensation-seeking scales.^10,18

DECISION-MAKING BIOLOGY

At least four neural circuits process decisions, weighing the risks and benefits of a given situation and formulating a response. These circuits are:

• prefrontal cortices, including orbitofrontal, dorsolateral, and ventromedial
• ventral striatum, including the nucleus accumbens
• thalamus
• monoaminergic brainstem nuclei (ventral tegmental area [VTA] and raphe nuclei).¹⁹

Functional imaging studies—including MRI and PET, EEG, and electrophysiology—have confirmed that these four brain regions are integral to response inhibition and show abnormal activity in impulsive individuals.^20,21 Indeed, prefrontal cortex damage has been extensively documented to cause marked impulsivity, poor decision making, and an increased propensity for substance abuse and dependency.¹

Functional imaging studies also have shown that adolescents appear to use these neural regions inefficiently during decision making. Extensive areas of the involved brain regions are activated in individuals ages 8 to 20, whereas only focal activation occurs in adults.²²

Box 2

Activity within this network is modulated by excitatory glutamatergic transmission and inhibitory GABAergic transmission within the cortices and by dopaminergic and serotonergic transmission within the VTA and raphe nuclei, respectively.^2,20 Although all of these neurotransmitters have been implicated in impulsivity, dopamine and serotonin have been studied most extensively (box 2).^1,2,23-26

CASE CONTINUED: PSYCHIATRIC WORKUP

Josh clearly is engaged in worrisome behavior with potential long-term consequences. To evaluate him for underlying psychopathology, the psychiatrist used a structured psychiatric exam, Minnesota Multi-phasic Personality Inventory (MMPI), and SNAP-IV Rating Scale for ADHD (see Related resources). Josh endorsed some depressive symptoms—which were also evident on the MMPI—but did not meet DSM-IVTR criteria for major depressive disorder. Neither were his symptoms diagnostic for any other Axis I or Axis II disorder.

Given the risk of harm and likelihood of worsening behavior over time, the psychiatrist schedules Josh for weekly psychotherapy and possible medication.

Psychosocial interventions are discussed with Josh’s parents, including monitoring his activities, restricting access to peers who have been a poor influence, reinforcing good behavior, and enlisting help from teachers and his friends’ parents. The effect of these interventions is to be explored in follow-up visits.

Box 3

DEFINING DEGREES OF RISK

Although no criteria differentiate “normal” from “pathologic” risk taking, the definition of taking a risk implies potential adverse consequences. In evaluating the impulsive adolescent, it is important to determine which behaviors:

• can be instructive and promote maturation
• fall outside normal adolescent behavior and/or carry potentially severe outcomes.

Acceptable. Risk taking is acceptable if the potential adverse outcome is relatively benign and the adolescent is likely to learn from the experience. For example, driving 10 miles over the speed limit and receiving a ticket can lead to stricter observance of the speed limit.

Pathologic. Josh clearly exhibits risky behaviors that one would reasonably consider “pathologic,” as they carry potentially severe consequences that exceed any possible developmental gain. For example, drinking and driving can result in a DUI citation and/or a motor vehicle accident with physical injuries or death.

TREATMENT OPTIONS

Psychiatric comorbidity. When you evaluate an adolescent engaged in excessive risk taking, consider Axis I and II disorders characterized by marked impulsivity. If the patient meets diagnostic criteria for a psychopathology such as bipolar disorder or ADHD, treating the underlying condition will likely improve impulsivity.

Recommended approach. Even without an Axis I or Axis II disorder, adolescents who engage in pathologic risky behavior may benefit from psychosocial interventions (Box 3), psychotherapy, and perhaps medication.

Because very little evidence supports using psychotropics to treat pathologically impulsive adolescents, we recommend that you:

• first try psychosocial interventions and psychotherapy
• reserve medications for patients who do not respond adequately to nondrug approaches and engage in impulsive behaviors that pose a high risk for grave consequences.

Psychotherapy can be effective once the adolescent and clinician form a therapeutic alliance. Because Josh—like other such teens—will likely view his psychiatrist as “just another adult lecturing me on what to do,” focus first on establishing rapport by:

• getting to know him
• helping him feel at ease
• showing interest in his thoughts and empathy towards his concerns and complaints
• discussing anything but the reason his parents brought him to your office.

After you establish an alliance, focus therapy on helping the adolescent gain insight into his or her dangerous behaviors and their consequences. To illustrate to Josh the potential consequences of his behaviors, for example, you might introduce him to:

• someone disabled in a motor vehicle accident
• an HIV-positive activist
• a recovering alcoholic
• a long-time smoker with severe chronic obstructive pulmonary disease.

At-risk adolescents also could be encouraged to complete an educational program that teaches alternate activities for sensation seeking (such as skiing instead of high-speed driving).

Medication. Although the monoaminergic systems are known to modulate impulsive behavior, few studies have examined using medications to treat risk-taking adolescents, and no drugs are FDA-approved for this indication.

SSRIs. Selective serotonin reuptake inhibitors such as fluoxetine, sertraline, or escitalopram might be useful for treating excessive adolescent risk taking. A preliminary study with paroxetine—an SSRI not recommended for children and adolescents—suggests this class of antidepressants may help reduce impulsivity.²⁶ In the absence of data specific to risk-taking behavior, we recommend using SSRI dosages similar to those used to treat mood disorders in adolescents.

Clomipramine acts mainly on the serotonin receptor, preventing serotonin reuptake in a manner similar to an SSRI. Because it has the greatest serotonergic effect in its drug class, clomipramine is the only tricyclic proven effective in obsessive-compulsive disorder.²⁷ Although no data have shown that clomipramine affects impulsivity, it theoretically could be effective because of its effect on serotonin.

Divalproex sodium has been shown to effectively treat impulsivity, particularly in patients with autism spectrum disorders, intermittent explosive disorder, schizophrenia, borderline personality disorder, and bipolar disorder.^27-30 As an off-label use, one could consider trying this agent in an adolescent with pathologic risk-taking behavior. Use the same dosages and obtain routine labs as indicated for adolescents with other disorders.

Adherence. Like Josh, adolescents who engage in high-risk behaviors often do not recognize their pathology and resist psychiatric intervention. Getting them to take medication or participate in psychotherapy can be quite difficult.

Adolescents are far more likely to adhere to treatment if you develop a rapport with them and they trust you. As psychotherapy and psychosocial interventions progress, patients become more likely to gain insight into their conditions and become more adherent.

Other options to encourage adherence include having the parent:

• administer the medication and ensure that the patient swallows it
• use rewards to reinforce the adolescent’s good behavior and adherence to treatment.

Follow up weekly with patients such as Josh who exhibit high-risk behaviors and require psychotherapy and medication. Follow less-acute patients 2 weeks after the initial evaluation, then monthly if they are responding to medication.

Related resources

• Strauch B. The primal teen: what the new discoveries about the teenage brain tell us about our kids. New York: Doubleday, 2004.
• SNAP-IV Rating Scale to screen for attention-deficit/hyperactivity disorder. Is your child really ADD/ADHD? www.drbiofeedback.com. Accessed Sept. 8, 2004.
• Focus Adolescent Services. Resources and information for families with adolescent behavior problems, including high-risk behavior. http://www.focusas.com/BehavioralDisorders.html. Accessed Aug. 26, 2004.

Drug brand names

• Clomipramine • Anafranil
• Divalproex sodium • Depakote
• Escitalopram • Lexapro
• Fluoxetine • Prozac
• Paroxetine • Paxil
• Sertraline • Zoloft

Disclosure

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

Dr Shapira receives grant or research support from Abbott Laboratories, Janssen Pharmaceutica, Ortho-McNeil Pharmaceutical, Bristol-Myers Squibb Co., Eli Lilly and Co., and Pfizer Inc. He is a speaker for AstraZeneca Pharmaceuticals, Forest Laboratories, and Ortho-McNeil Pharmaceutical, Inc.

Boys will be boys” and other platitudes may condone adolescent reckless driving, substance use, or sexual promiscuity—but to write off dangerous behavior as normal would be a mistake. Because adolescent impulsivity and sensation-seeking may have physiologic as well as emotional causes,^1,2 excessive risk taking may be treatable.

This article discusses the neurobiology of adolescent risk taking, suggests how to determine when such behavior may be pathologic, and offers a treatment approach for at-risk teens and their parents.

CASE: ‘WHAT’S WRONG WITH OUR SON?’

Josh, age 17, is brought to the adolescent psychiatry clinic by his distraught parents, who report their son has undergone a “personality change” over the past 2 years. They recall that he was respectful, studious, and soft-spoken until age 15. Since then, he has been skipping school, staying out late at night with his friends, and “obsessed” with TV poker games.

His parents recently discovered he has been gambling for money, which greatly upsets them. They also found a pack of cigarettes in their son’s car and are concerned that he might be using other substances. What finally prompted the psychiatric visit was Josh’s recent traffic citation for driving 25 miles over the speed limit.

CAUSES OF RISK TAKING

Normal development. In the absence of psychopathology, adolescent risk taking appears to be a normal development stage that is vital to successful transition to adulthood. This assumes that adolescents such as Josh learn to moderate their behavior and avoid long-term negative consequences.

Impulsivity and sensation seeking are recognized as key factors in adolescent risk taking Box 1.^1-4 Apparently, these traits result primarily from incomplete neural circuit maturation. Adolescent brain regions involved in impulsivity and risk taking are also involved in reward, and these centers exhibit an exaggerated response to stimuli.⁵ This amplified response may help explain an adolescent’s propensity for risky behavior.

Despite potential hazards, adolescent risk taking may confer benefits. In taking risks, adolescents:

• explore adult behavior
• learn to accomplish increasingly difficult developmental tasks
• reinforce their self-esteem.

Adolescent risk-takers have been found to be more self-confident, to feel more accepted, and to be better liked than their more-cautious peers.⁶

Psychiatric comorbidity. Excessive risk taking can be associated with psychiatric illness, including bipolar mania, psychosis, substance abuse, and impulse control disorders. Individuals with borderline personality and other cluster B disorders have marked impulsivity and thus are prone to risky behavior.

Teens with attention-deficit/hyperactivity disorder (ADHD), conduct disorder, and oppositional-defiant disorder (ODD) also tend to exhibit high impulsivity.

Box 1

CASE CONTINUED: JUST ‘HAVING FUN’

When interviewed alone, Josh admits to “occasional” truancy, which he attributes to being “bored” with school and wanting to spend time with his friends “doing fun stuff, like going to the beach.” He admits to gambling for money and smoking a half-pack of cigarettes daily, as well as drinking beer and smoking marijuana “a few times a week.”

Josh says he engages in these activities “because they’re fun,” and states he is annoyed by his parents’ concern. He blames the speeding ticket on “not paying attention.” He admits to drinking and driving but claims he always feels “in control.”

He also reports he has been sexually active since age 16 and often has had unprotected intercourse. When asked if he is concerned that he might contract a sexually transmitted disease or impregnate his partner, Josh appears ambivalent.

IMPULSIVITY IN ADOLESCENCE

Josh is engaging in numerous impulsive behaviors. Adolescents generally are more impulsive than adults, as demonstrated by their significantly higher impulsivity scores on standardized tests.⁷ Furthermore, as measured by improved response inhibition (go/no go tasks), the level of adolescent impulsivity is inversely related to age.⁸

Problem behavior syndrome. High impulsivity is predictive of problem gambling, drug use, and risky driving and sexual practices later in life.^1,2,9-11 Adolescents with what some authors describe as a “problem behavior syndrome” engage in behaviors—such as substance use, risky sexual behavior,¹² gambling,¹³ and reckless driving¹⁴—that share a common trend toward impulsivity.

Impaired data processing. Decision making has been proposed as a three-part cognitive process:

• accumulating sensory input
• processing this input and formulating a behavioral response appropriate to the situation
• planning and implementing the resultant motor output.²

Impulsivity is believed to result from impaired ability of the brain to process accumulated information or to formulate a response to it—or both. Impulsive individuals thus experience impaired data processing, in which they:

• misjudge the likely risk of a given action or overestimate their ability to accomplish a task
• show impaired response inhibition and thus find it difficult to resist an impulse to participate in a given activity.

Sensation seeking. Adolescents who exhibit risk-taking behavior may wish to experience the thrill of the behavior (sensation or novelty seeking). Alcoholic or drug-dependent individuals and those who engage in pathologic gambling or take chances while driving also demonstrate significantly impaired decision making.^15-17 Adolescents who engage in these and other problem behaviors have similarly scored high on sensation-seeking scales.^10,18

DECISION-MAKING BIOLOGY

At least four neural circuits process decisions, weighing the risks and benefits of a given situation and formulating a response. These circuits are:

• prefrontal cortices, including orbitofrontal, dorsolateral, and ventromedial
• ventral striatum, including the nucleus accumbens
• thalamus
• monoaminergic brainstem nuclei (ventral tegmental area [VTA] and raphe nuclei).¹⁹

Functional imaging studies—including MRI and PET, EEG, and electrophysiology—have confirmed that these four brain regions are integral to response inhibition and show abnormal activity in impulsive individuals.^20,21 Indeed, prefrontal cortex damage has been extensively documented to cause marked impulsivity, poor decision making, and an increased propensity for substance abuse and dependency.¹

Functional imaging studies also have shown that adolescents appear to use these neural regions inefficiently during decision making. Extensive areas of the involved brain regions are activated in individuals ages 8 to 20, whereas only focal activation occurs in adults.²²

Box 2

Activity within this network is modulated by excitatory glutamatergic transmission and inhibitory GABAergic transmission within the cortices and by dopaminergic and serotonergic transmission within the VTA and raphe nuclei, respectively.^2,20 Although all of these neurotransmitters have been implicated in impulsivity, dopamine and serotonin have been studied most extensively (box 2).^1,2,23-26

CASE CONTINUED: PSYCHIATRIC WORKUP

Josh clearly is engaged in worrisome behavior with potential long-term consequences. To evaluate him for underlying psychopathology, the psychiatrist used a structured psychiatric exam, Minnesota Multi-phasic Personality Inventory (MMPI), and SNAP-IV Rating Scale for ADHD (see Related resources). Josh endorsed some depressive symptoms—which were also evident on the MMPI—but did not meet DSM-IVTR criteria for major depressive disorder. Neither were his symptoms diagnostic for any other Axis I or Axis II disorder.

Given the risk of harm and likelihood of worsening behavior over time, the psychiatrist schedules Josh for weekly psychotherapy and possible medication.

Psychosocial interventions are discussed with Josh’s parents, including monitoring his activities, restricting access to peers who have been a poor influence, reinforcing good behavior, and enlisting help from teachers and his friends’ parents. The effect of these interventions is to be explored in follow-up visits.

Box 3

DEFINING DEGREES OF RISK

Although no criteria differentiate “normal” from “pathologic” risk taking, the definition of taking a risk implies potential adverse consequences. In evaluating the impulsive adolescent, it is important to determine which behaviors:

• can be instructive and promote maturation
• fall outside normal adolescent behavior and/or carry potentially severe outcomes.

Acceptable. Risk taking is acceptable if the potential adverse outcome is relatively benign and the adolescent is likely to learn from the experience. For example, driving 10 miles over the speed limit and receiving a ticket can lead to stricter observance of the speed limit.

Pathologic. Josh clearly exhibits risky behaviors that one would reasonably consider “pathologic,” as they carry potentially severe consequences that exceed any possible developmental gain. For example, drinking and driving can result in a DUI citation and/or a motor vehicle accident with physical injuries or death.

TREATMENT OPTIONS

Psychiatric comorbidity. When you evaluate an adolescent engaged in excessive risk taking, consider Axis I and II disorders characterized by marked impulsivity. If the patient meets diagnostic criteria for a psychopathology such as bipolar disorder or ADHD, treating the underlying condition will likely improve impulsivity.

Recommended approach. Even without an Axis I or Axis II disorder, adolescents who engage in pathologic risky behavior may benefit from psychosocial interventions (Box 3), psychotherapy, and perhaps medication.

Because very little evidence supports using psychotropics to treat pathologically impulsive adolescents, we recommend that you:

• first try psychosocial interventions and psychotherapy
• reserve medications for patients who do not respond adequately to nondrug approaches and engage in impulsive behaviors that pose a high risk for grave consequences.

Psychotherapy can be effective once the adolescent and clinician form a therapeutic alliance. Because Josh—like other such teens—will likely view his psychiatrist as “just another adult lecturing me on what to do,” focus first on establishing rapport by:

• getting to know him
• helping him feel at ease
• showing interest in his thoughts and empathy towards his concerns and complaints
• discussing anything but the reason his parents brought him to your office.

After you establish an alliance, focus therapy on helping the adolescent gain insight into his or her dangerous behaviors and their consequences. To illustrate to Josh the potential consequences of his behaviors, for example, you might introduce him to:

• someone disabled in a motor vehicle accident
• an HIV-positive activist
• a recovering alcoholic
• a long-time smoker with severe chronic obstructive pulmonary disease.

At-risk adolescents also could be encouraged to complete an educational program that teaches alternate activities for sensation seeking (such as skiing instead of high-speed driving).

Medication. Although the monoaminergic systems are known to modulate impulsive behavior, few studies have examined using medications to treat risk-taking adolescents, and no drugs are FDA-approved for this indication.

SSRIs. Selective serotonin reuptake inhibitors such as fluoxetine, sertraline, or escitalopram might be useful for treating excessive adolescent risk taking. A preliminary study with paroxetine—an SSRI not recommended for children and adolescents—suggests this class of antidepressants may help reduce impulsivity.²⁶ In the absence of data specific to risk-taking behavior, we recommend using SSRI dosages similar to those used to treat mood disorders in adolescents.

Clomipramine acts mainly on the serotonin receptor, preventing serotonin reuptake in a manner similar to an SSRI. Because it has the greatest serotonergic effect in its drug class, clomipramine is the only tricyclic proven effective in obsessive-compulsive disorder.²⁷ Although no data have shown that clomipramine affects impulsivity, it theoretically could be effective because of its effect on serotonin.

Divalproex sodium has been shown to effectively treat impulsivity, particularly in patients with autism spectrum disorders, intermittent explosive disorder, schizophrenia, borderline personality disorder, and bipolar disorder.^27-30 As an off-label use, one could consider trying this agent in an adolescent with pathologic risk-taking behavior. Use the same dosages and obtain routine labs as indicated for adolescents with other disorders.

Adherence. Like Josh, adolescents who engage in high-risk behaviors often do not recognize their pathology and resist psychiatric intervention. Getting them to take medication or participate in psychotherapy can be quite difficult.

Adolescents are far more likely to adhere to treatment if you develop a rapport with them and they trust you. As psychotherapy and psychosocial interventions progress, patients become more likely to gain insight into their conditions and become more adherent.

Other options to encourage adherence include having the parent:

• administer the medication and ensure that the patient swallows it
• use rewards to reinforce the adolescent’s good behavior and adherence to treatment.

Follow up weekly with patients such as Josh who exhibit high-risk behaviors and require psychotherapy and medication. Follow less-acute patients 2 weeks after the initial evaluation, then monthly if they are responding to medication.

Related resources

• Strauch B. The primal teen: what the new discoveries about the teenage brain tell us about our kids. New York: Doubleday, 2004.
• SNAP-IV Rating Scale to screen for attention-deficit/hyperactivity disorder. Is your child really ADD/ADHD? www.drbiofeedback.com. Accessed Sept. 8, 2004.
• Focus Adolescent Services. Resources and information for families with adolescent behavior problems, including high-risk behavior. http://www.focusas.com/BehavioralDisorders.html. Accessed Aug. 26, 2004.

Drug brand names

• Clomipramine • Anafranil
• Divalproex sodium • Depakote
• Escitalopram • Lexapro
• Fluoxetine • Prozac
• Paroxetine • Paxil
• Sertraline • Zoloft

Disclosure

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

Dr Shapira receives grant or research support from Abbott Laboratories, Janssen Pharmaceutica, Ortho-McNeil Pharmaceutical, Bristol-Myers Squibb Co., Eli Lilly and Co., and Pfizer Inc. He is a speaker for AstraZeneca Pharmaceuticals, Forest Laboratories, and Ortho-McNeil Pharmaceutical, Inc.

In last month’s Letters an editing error occurred in a letter by Dr. George Hilton and the reply by Dr. Edmund Higgins. The word “Oxytocin” was misread as “Oxycontin” and changed throughout to the incorrect “oxycodone” (Oxycontin’s generic name). Current Psychiatry regrets the error. See the corrected letter and reply at www.currentpsychiatry.com.

In last month’s Letters an editing error occurred in a letter by Dr. George Hilton and the reply by Dr. Edmund Higgins. The word “Oxytocin” was misread as “Oxycontin” and changed throughout to the incorrect “oxycodone” (Oxycontin’s generic name). Current Psychiatry regrets the error. See the corrected letter and reply at www.currentpsychiatry.com.

In last month’s Letters an editing error occurred in a letter by Dr. George Hilton and the reply by Dr. Edmund Higgins. The word “Oxytocin” was misread as “Oxycontin” and changed throughout to the incorrect “oxycodone” (Oxycontin’s generic name). Current Psychiatry regrets the error. See the corrected letter and reply at www.currentpsychiatry.com.