Law & Medicine

Postpartum Shortness of Breath

A 35-year-old woman presented to the ED after referral by her obstetrician. Six days prior, the patient had given birth to twins without incident. On postpartum hospital day 2, however, she developed mild shortness of breath, and a chest X-ray was ordered. Since there was concern for possible pneumonia, the patient was prescribed oral antibiotics and discharged home on hospital day 4. She continued to complain of shortness of breath following discharge; at that time, the obstetrician referred the patient to the ED for further evaluation.

The patient was evaluated by the EP, who ordered a chest X-ray. He was also concerned that patient had pneumonia and prescribed a different class of antibiotic before discharging the patient home.

One week later, the patient presented back to the same ED with continued shortness of breath. On this visit, she was seen by a physician assistant (PA). Following the history taking and physical examination, a chest X-ray and rapid flu and rapid strep tests were ordered. Both the PA and supervising EP reviewed the chest X-ray and were concerned for pneumonia; however, both the flu and strep tests were negative. A third class of antibiotic was prescribed, and the patient was discharged home.

The chest X-ray on the second ED visit was not interpreted by a radiologist until 3 days later (The patient was seen on a Friday evening and the films were not read until the following Monday morning). The radiologist’s interpretation was “worsening congestive heart failure” (CHF). Two days later (5 days following the second ED visit), the EP was notified of the interpretation discrepancy and made multiple attempts to contact the patient, which included leaving a voice-mail message on her home phone.

The patient returned the call the following day and spoke with one of the ED nurses, who encouraged her to return immediately to the ED. The patient returned to the ED the next day (1 week after her second ED visit) and was admitted to the hospital for CHF secondary to postpartum cardiomyopathy. Unfortunately, she developed an embolus to her kidney, followed by an ischemic cerebrovascular accident, and died 3 weeks after admission.

The patient’s family filed a malpractice lawsuit against the hospital, the EPs, and the PA for negligent delay in making the correct diagnosis, stating that patient’s subsequent stroke and death were a direct result of this delay. Following deliberations, all of the EPs involved in the case were found free of negligence; the PA and the hospital ED, however, were found guilty.

Discussion

Interestingly, the majority of successful malpractice suits against physicians involve cognitive errors and system issues; this case is no exception. Making the correct diagnosis in a patient is a complex process, involving data gathering and synthesis, intuition, clinical experience, and logical thinking. Unfortunately, biases can occur during this process and result in misdiagnosis or delayed diagnosis. These biases include anchoring bias, confirmation bias, premature closure, and diagnosis momentum.¹

Anchoring Bias. This occurs when a physician relies too heavily on the first piece of information or one’s initial impression.¹ Despite evidence to the contrary, the physician keeps returning to the initial diagnosis (ie, he or she is “anchored” to it).

Confirmation Bias. Related to anchoring bias, in confirmation bias, the physician ignores or discounts evidence that contradicts one’s initial impression and focuses solely on the evidence supporting it.¹

Both anchoring bias and confirmation bias appear to have played a role in this case. The differential diagnosis was never broadened beyond pneumonia, despite the fact that one must also consider pulmonary embolism (ie, a hypercoagulable state) and CHF (ie, postpartum cardiomyopathy) in a patient complaining of dyspnea in the postpartum period.

Premature Closure. This occurs when the physician finds a cause that fits the clinical picture and ceases to search for other diagnostic possibilities.^1,2

Diagnosis Momentum. A bias that occurs when the diagnosis considered a possibility by one physician becomes a definitive diagnosis as it is passed from one physician to the next; it then becomes accepted without question by physicians down the line.¹ This type of bias also seems to have played a role in this case.

There are a few strategies to help prevent or minimize these types of errors. First, as new data are gathered, one should reconsider and reprioritize the differential diagnosis. When certain data points do not fit neatly with an earlier diagnosis, careful attention must be paid to them. This is especially true for the patient receiving appropriate treatment but not showing clinical improvement. While it is usually helpful to know previous working diagnoses, the clinician must try to keep an open mind and consider alternative diagnoses. At the end of the day, the lesson is the need to develop a broad differential diagnosis.²

The system error in this case involves the lack of timely overread of radiology studies. Emergency medicine is a 24-hour, 7-day-a-week specialty; therefore, support services to the ED need to be similarly available. With today’s teleradiology and dedicated night readers, it is difficult to justify not providing such a service. This situation explains, in part, the negligent verdict against the hospital.

Chest Pain in a Man With Type 2 Diabetes

A 54-year-old man presented to the ED with a several-hour history of chest pain and shortness of breath. His medical history was remarkable for type 2 diabetes mellitus. He denied any associated nausea, vomiting, or diaphoresis. He smoked one pack of cigarettes per day and drank alcohol on occasion. The patient further denied having experienced similar symptoms in the past. All vital signs were normal, and he appeared comfortable and in no acute distress. His physical examination was completely unremarkable.

An electrocardiogram (ECG) revealed Q-waves in the anterior leads, but was otherwise nondiagnostic. The results of a portable chest X-ray were interpreted as normal. Cardiac enzyme testing revealed a positive troponin I, but a normal creatine phosphokinase-MB (CPK-MB). The patient’s complete blood count and coagulation studies were normal, and a basic metabolic profile was remarkable only for a blood sugar of 540 mg/dL; his serum bicarbonate value was normal.

The patient was placed on oxygen via nasal cannula, administered an aspirin (325 mg), and given subcutaneous regular insulin for hyperglycemia. The EP contacted a hospitalist to admit the patient. The hospitalist accepted the patient and admitted him to the telemetry floor.

Upon arrival at telemetry, the patient was examined by a nurse who noted rhonchi bilaterally on lung auscultation. Shortly afterward, he became anxious, and the nurse consulted the hospitalist, who ordered administration of lorazepam. Approximately 20 minutes later, the patient went into cardiac arrest and died.

This patient was never examined by the hospitalist prior to coding. An autopsy revealed evidence of severe coronary artery disease, a previous infarction that was at least a few months old with scarring of the left ventricle, and a recent infarction that had begun at least 12 hours prior to death.

The family sued all of the providers involved in the care of the patient. At trial, there was a factual dispute regarding whether the EP informed the hospitalist of the elevated troponin I level. The defendants argued that the patient had sustained irreversible heart damage prior to his arrival at the hospital, and that nothing any of the defendants could have done would have saved his life. The jury deliberated for approximately 2 hours before delivering a verdict in favor of the defense.

Discussion

Clearly, this verdict could have gone the other way. This patient was experiencing a non-ST segment elevation myocardial infarction (MI). The nondiagnostic ECG, coupled with the elevated troponin I, indicated damage of heart muscle from an acute interruption of coronary blood flow.

There are several problems with the management of this case. First, this patient required a cardiology consult for risk stratification. Several scoring systems could have been used to determine whether this patient was a candidate for early (ie, within 4-48 hours) invasive treatments such as percutaneous intervention or a more conservative approach.

Second, having sustained an MI, this patient was at high risk for complications such as ventricular arrhythmias, heart failure, cardiogenic shock, and other serious adverse events. Patients with acute MI should be admitted to the critical care or intensive care unit. In addition to aspirin, this patient should have received nitroglycerin and anticoagulation therapy. Either heparin or a low molecular weight heparin, such as enoxaparin, would have been appropriate if no contraindications existed. Finally, additional therapy including glycoprotein IIB/IIIA inhibitors, clopidogrel, etc, may have been indicated depending upon the timing of percutaneous intervention.¹

It appears that both the EP and the hospitalist either failed to appreciate the significance of the elevated troponin I or overlooked it. This patient had normal renal function, so the only explanation—especially in the setting of a middle-aged man complaining of chest pain—was that myocardial damage had occurred.

Alcohol Detoxification in a Young Man

A young man was brought to the ED by a friend. The patient’s sole complaint was the need for help with his alcohol dependency. In addition to alcohol abuse, his medical history was remarkable only for an admission 1 month prior for suicidal ideation. The patient denied suicidal or homicidal ideations on this presentation. The physical examination revealed stable vital signs, but conjunctival injection, slurred speech, and a strong odor of alcohol. A blood alcohol test showed a concentration level of 0.36 g/dL. The patient, however, was alert and able to ambulate without assistance. He was medically cleared by the EP and arrangements were made to admit him to a local detoxification center.

Approximately 4 hours after his arrival, while waiting for transport to the detoxification center, the patient became impatient with the delay, removed his intravenous line, and told the nurse that he was going home via taxi. The nurse encouraged him to call a friend to take him home, to which to the patient agreed. The nurse left the patient to inform the EP of his desire to go home; when she returned, she discovered that the patient had already left the hospital. The physician notified hospital security but not the police. Approximately 2 hours later, the patient was struck by a car and seriously injured.

The patient sued the EP and the hospital for negligence and medical malpractice. The suit alleged that the physician and the hospital should have prevented the patient from leaving the ED. The physician and the hospital requested a dismissal, arguing that the patient did not exhibit any suicidal or homicidal ideation, presented on his own volition, and, though intoxicated, could still make decisions for himself. An appellate court granted the motion, holding that the defendants “lacked authority to confine the plaintiff upon his departure” from the ED.

Discussion

When a patient attempts to leave the ED against medical advice, the treating physician should make an attempt to convince him or her to remain for treatment. Often, something as simple as offering a cup of coffee will change a patient’s mind. In other instances, the use of chemical or physical restraint may be required. The handling of the case ultimately becomes a question of whether the patient was competent to make decisions and whether he presented a danger to himself or others. The extent of intoxication varies by the degree and does not of itself constitute incapability to make decisions. All practicing EPs have cared for patients with blood alcohol levels above the legal limit for driving, but who were functionally sober and able to make decisions. If a patient is competent and does not present a danger to self or others, he or she can decide to leave the ED without further management. However, it is best to release the patient in the company of friends or family, as was urged by the nurse in this case. Obviously, such a patient should not be allowed to drive himself home.

When a patient does leave against medical advice, the physician and nurse should document in the ED record their conversations urging him or her to stay. Alternatively, when a patient is found incompetent of making decisions or is a danger to self or others, he or she must be prevented from leaving the ED. This includes use of the minimal amount of physical or chemical restraint needed to keep the patient from leaving. When there is doubt that the patient is able to make a competent decision, it is better to err on the side of caution and keep him or her in the ED for his own safety and the safety of others.

Postpartum Shortness of Breath

A 35-year-old woman presented to the ED after referral by her obstetrician. Six days prior, the patient had given birth to twins without incident. On postpartum hospital day 2, however, she developed mild shortness of breath, and a chest X-ray was ordered. Since there was concern for possible pneumonia, the patient was prescribed oral antibiotics and discharged home on hospital day 4. She continued to complain of shortness of breath following discharge; at that time, the obstetrician referred the patient to the ED for further evaluation.

The patient was evaluated by the EP, who ordered a chest X-ray. He was also concerned that patient had pneumonia and prescribed a different class of antibiotic before discharging the patient home.

One week later, the patient presented back to the same ED with continued shortness of breath. On this visit, she was seen by a physician assistant (PA). Following the history taking and physical examination, a chest X-ray and rapid flu and rapid strep tests were ordered. Both the PA and supervising EP reviewed the chest X-ray and were concerned for pneumonia; however, both the flu and strep tests were negative. A third class of antibiotic was prescribed, and the patient was discharged home.

The chest X-ray on the second ED visit was not interpreted by a radiologist until 3 days later (The patient was seen on a Friday evening and the films were not read until the following Monday morning). The radiologist’s interpretation was “worsening congestive heart failure” (CHF). Two days later (5 days following the second ED visit), the EP was notified of the interpretation discrepancy and made multiple attempts to contact the patient, which included leaving a voice-mail message on her home phone.

The patient returned the call the following day and spoke with one of the ED nurses, who encouraged her to return immediately to the ED. The patient returned to the ED the next day (1 week after her second ED visit) and was admitted to the hospital for CHF secondary to postpartum cardiomyopathy. Unfortunately, she developed an embolus to her kidney, followed by an ischemic cerebrovascular accident, and died 3 weeks after admission.

The patient’s family filed a malpractice lawsuit against the hospital, the EPs, and the PA for negligent delay in making the correct diagnosis, stating that patient’s subsequent stroke and death were a direct result of this delay. Following deliberations, all of the EPs involved in the case were found free of negligence; the PA and the hospital ED, however, were found guilty.

Discussion

Interestingly, the majority of successful malpractice suits against physicians involve cognitive errors and system issues; this case is no exception. Making the correct diagnosis in a patient is a complex process, involving data gathering and synthesis, intuition, clinical experience, and logical thinking. Unfortunately, biases can occur during this process and result in misdiagnosis or delayed diagnosis. These biases include anchoring bias, confirmation bias, premature closure, and diagnosis momentum.¹

Anchoring Bias. This occurs when a physician relies too heavily on the first piece of information or one’s initial impression.¹ Despite evidence to the contrary, the physician keeps returning to the initial diagnosis (ie, he or she is “anchored” to it).

Confirmation Bias. Related to anchoring bias, in confirmation bias, the physician ignores or discounts evidence that contradicts one’s initial impression and focuses solely on the evidence supporting it.¹

Both anchoring bias and confirmation bias appear to have played a role in this case. The differential diagnosis was never broadened beyond pneumonia, despite the fact that one must also consider pulmonary embolism (ie, a hypercoagulable state) and CHF (ie, postpartum cardiomyopathy) in a patient complaining of dyspnea in the postpartum period.

Premature Closure. This occurs when the physician finds a cause that fits the clinical picture and ceases to search for other diagnostic possibilities.^1,2

Diagnosis Momentum. A bias that occurs when the diagnosis considered a possibility by one physician becomes a definitive diagnosis as it is passed from one physician to the next; it then becomes accepted without question by physicians down the line.¹ This type of bias also seems to have played a role in this case.

There are a few strategies to help prevent or minimize these types of errors. First, as new data are gathered, one should reconsider and reprioritize the differential diagnosis. When certain data points do not fit neatly with an earlier diagnosis, careful attention must be paid to them. This is especially true for the patient receiving appropriate treatment but not showing clinical improvement. While it is usually helpful to know previous working diagnoses, the clinician must try to keep an open mind and consider alternative diagnoses. At the end of the day, the lesson is the need to develop a broad differential diagnosis.²

The system error in this case involves the lack of timely overread of radiology studies. Emergency medicine is a 24-hour, 7-day-a-week specialty; therefore, support services to the ED need to be similarly available. With today’s teleradiology and dedicated night readers, it is difficult to justify not providing such a service. This situation explains, in part, the negligent verdict against the hospital.

Chest Pain in a Man With Type 2 Diabetes

A 54-year-old man presented to the ED with a several-hour history of chest pain and shortness of breath. His medical history was remarkable for type 2 diabetes mellitus. He denied any associated nausea, vomiting, or diaphoresis. He smoked one pack of cigarettes per day and drank alcohol on occasion. The patient further denied having experienced similar symptoms in the past. All vital signs were normal, and he appeared comfortable and in no acute distress. His physical examination was completely unremarkable.

An electrocardiogram (ECG) revealed Q-waves in the anterior leads, but was otherwise nondiagnostic. The results of a portable chest X-ray were interpreted as normal. Cardiac enzyme testing revealed a positive troponin I, but a normal creatine phosphokinase-MB (CPK-MB). The patient’s complete blood count and coagulation studies were normal, and a basic metabolic profile was remarkable only for a blood sugar of 540 mg/dL; his serum bicarbonate value was normal.

The patient was placed on oxygen via nasal cannula, administered an aspirin (325 mg), and given subcutaneous regular insulin for hyperglycemia. The EP contacted a hospitalist to admit the patient. The hospitalist accepted the patient and admitted him to the telemetry floor.

Upon arrival at telemetry, the patient was examined by a nurse who noted rhonchi bilaterally on lung auscultation. Shortly afterward, he became anxious, and the nurse consulted the hospitalist, who ordered administration of lorazepam. Approximately 20 minutes later, the patient went into cardiac arrest and died.

This patient was never examined by the hospitalist prior to coding. An autopsy revealed evidence of severe coronary artery disease, a previous infarction that was at least a few months old with scarring of the left ventricle, and a recent infarction that had begun at least 12 hours prior to death.

The family sued all of the providers involved in the care of the patient. At trial, there was a factual dispute regarding whether the EP informed the hospitalist of the elevated troponin I level. The defendants argued that the patient had sustained irreversible heart damage prior to his arrival at the hospital, and that nothing any of the defendants could have done would have saved his life. The jury deliberated for approximately 2 hours before delivering a verdict in favor of the defense.

Discussion

Clearly, this verdict could have gone the other way. This patient was experiencing a non-ST segment elevation myocardial infarction (MI). The nondiagnostic ECG, coupled with the elevated troponin I, indicated damage of heart muscle from an acute interruption of coronary blood flow.

There are several problems with the management of this case. First, this patient required a cardiology consult for risk stratification. Several scoring systems could have been used to determine whether this patient was a candidate for early (ie, within 4-48 hours) invasive treatments such as percutaneous intervention or a more conservative approach.

Second, having sustained an MI, this patient was at high risk for complications such as ventricular arrhythmias, heart failure, cardiogenic shock, and other serious adverse events. Patients with acute MI should be admitted to the critical care or intensive care unit. In addition to aspirin, this patient should have received nitroglycerin and anticoagulation therapy. Either heparin or a low molecular weight heparin, such as enoxaparin, would have been appropriate if no contraindications existed. Finally, additional therapy including glycoprotein IIB/IIIA inhibitors, clopidogrel, etc, may have been indicated depending upon the timing of percutaneous intervention.¹

It appears that both the EP and the hospitalist either failed to appreciate the significance of the elevated troponin I or overlooked it. This patient had normal renal function, so the only explanation—especially in the setting of a middle-aged man complaining of chest pain—was that myocardial damage had occurred.

Alcohol Detoxification in a Young Man

A young man was brought to the ED by a friend. The patient’s sole complaint was the need for help with his alcohol dependency. In addition to alcohol abuse, his medical history was remarkable only for an admission 1 month prior for suicidal ideation. The patient denied suicidal or homicidal ideations on this presentation. The physical examination revealed stable vital signs, but conjunctival injection, slurred speech, and a strong odor of alcohol. A blood alcohol test showed a concentration level of 0.36 g/dL. The patient, however, was alert and able to ambulate without assistance. He was medically cleared by the EP and arrangements were made to admit him to a local detoxification center.

Approximately 4 hours after his arrival, while waiting for transport to the detoxification center, the patient became impatient with the delay, removed his intravenous line, and told the nurse that he was going home via taxi. The nurse encouraged him to call a friend to take him home, to which to the patient agreed. The nurse left the patient to inform the EP of his desire to go home; when she returned, she discovered that the patient had already left the hospital. The physician notified hospital security but not the police. Approximately 2 hours later, the patient was struck by a car and seriously injured.

The patient sued the EP and the hospital for negligence and medical malpractice. The suit alleged that the physician and the hospital should have prevented the patient from leaving the ED. The physician and the hospital requested a dismissal, arguing that the patient did not exhibit any suicidal or homicidal ideation, presented on his own volition, and, though intoxicated, could still make decisions for himself. An appellate court granted the motion, holding that the defendants “lacked authority to confine the plaintiff upon his departure” from the ED.

Discussion

When a patient attempts to leave the ED against medical advice, the treating physician should make an attempt to convince him or her to remain for treatment. Often, something as simple as offering a cup of coffee will change a patient’s mind. In other instances, the use of chemical or physical restraint may be required. The handling of the case ultimately becomes a question of whether the patient was competent to make decisions and whether he presented a danger to himself or others. The extent of intoxication varies by the degree and does not of itself constitute incapability to make decisions. All practicing EPs have cared for patients with blood alcohol levels above the legal limit for driving, but who were functionally sober and able to make decisions. If a patient is competent and does not present a danger to self or others, he or she can decide to leave the ED without further management. However, it is best to release the patient in the company of friends or family, as was urged by the nurse in this case. Obviously, such a patient should not be allowed to drive himself home.

When a patient does leave against medical advice, the physician and nurse should document in the ED record their conversations urging him or her to stay. Alternatively, when a patient is found incompetent of making decisions or is a danger to self or others, he or she must be prevented from leaving the ED. This includes use of the minimal amount of physical or chemical restraint needed to keep the patient from leaving. When there is doubt that the patient is able to make a competent decision, it is better to err on the side of caution and keep him or her in the ED for his own safety and the safety of others.

Postpartum Shortness of Breath

A 35-year-old woman presented to the ED after referral by her obstetrician. Six days prior, the patient had given birth to twins without incident. On postpartum hospital day 2, however, she developed mild shortness of breath, and a chest X-ray was ordered. Since there was concern for possible pneumonia, the patient was prescribed oral antibiotics and discharged home on hospital day 4. She continued to complain of shortness of breath following discharge; at that time, the obstetrician referred the patient to the ED for further evaluation.

The patient was evaluated by the EP, who ordered a chest X-ray. He was also concerned that patient had pneumonia and prescribed a different class of antibiotic before discharging the patient home.

One week later, the patient presented back to the same ED with continued shortness of breath. On this visit, she was seen by a physician assistant (PA). Following the history taking and physical examination, a chest X-ray and rapid flu and rapid strep tests were ordered. Both the PA and supervising EP reviewed the chest X-ray and were concerned for pneumonia; however, both the flu and strep tests were negative. A third class of antibiotic was prescribed, and the patient was discharged home.

The chest X-ray on the second ED visit was not interpreted by a radiologist until 3 days later (The patient was seen on a Friday evening and the films were not read until the following Monday morning). The radiologist’s interpretation was “worsening congestive heart failure” (CHF). Two days later (5 days following the second ED visit), the EP was notified of the interpretation discrepancy and made multiple attempts to contact the patient, which included leaving a voice-mail message on her home phone.

The patient returned the call the following day and spoke with one of the ED nurses, who encouraged her to return immediately to the ED. The patient returned to the ED the next day (1 week after her second ED visit) and was admitted to the hospital for CHF secondary to postpartum cardiomyopathy. Unfortunately, she developed an embolus to her kidney, followed by an ischemic cerebrovascular accident, and died 3 weeks after admission.

The patient’s family filed a malpractice lawsuit against the hospital, the EPs, and the PA for negligent delay in making the correct diagnosis, stating that patient’s subsequent stroke and death were a direct result of this delay. Following deliberations, all of the EPs involved in the case were found free of negligence; the PA and the hospital ED, however, were found guilty.

Discussion

Interestingly, the majority of successful malpractice suits against physicians involve cognitive errors and system issues; this case is no exception. Making the correct diagnosis in a patient is a complex process, involving data gathering and synthesis, intuition, clinical experience, and logical thinking. Unfortunately, biases can occur during this process and result in misdiagnosis or delayed diagnosis. These biases include anchoring bias, confirmation bias, premature closure, and diagnosis momentum.¹

Anchoring Bias. This occurs when a physician relies too heavily on the first piece of information or one’s initial impression.¹ Despite evidence to the contrary, the physician keeps returning to the initial diagnosis (ie, he or she is “anchored” to it).

Confirmation Bias. Related to anchoring bias, in confirmation bias, the physician ignores or discounts evidence that contradicts one’s initial impression and focuses solely on the evidence supporting it.¹

Both anchoring bias and confirmation bias appear to have played a role in this case. The differential diagnosis was never broadened beyond pneumonia, despite the fact that one must also consider pulmonary embolism (ie, a hypercoagulable state) and CHF (ie, postpartum cardiomyopathy) in a patient complaining of dyspnea in the postpartum period.

Premature Closure. This occurs when the physician finds a cause that fits the clinical picture and ceases to search for other diagnostic possibilities.^1,2

Diagnosis Momentum. A bias that occurs when the diagnosis considered a possibility by one physician becomes a definitive diagnosis as it is passed from one physician to the next; it then becomes accepted without question by physicians down the line.¹ This type of bias also seems to have played a role in this case.

There are a few strategies to help prevent or minimize these types of errors. First, as new data are gathered, one should reconsider and reprioritize the differential diagnosis. When certain data points do not fit neatly with an earlier diagnosis, careful attention must be paid to them. This is especially true for the patient receiving appropriate treatment but not showing clinical improvement. While it is usually helpful to know previous working diagnoses, the clinician must try to keep an open mind and consider alternative diagnoses. At the end of the day, the lesson is the need to develop a broad differential diagnosis.²

The system error in this case involves the lack of timely overread of radiology studies. Emergency medicine is a 24-hour, 7-day-a-week specialty; therefore, support services to the ED need to be similarly available. With today’s teleradiology and dedicated night readers, it is difficult to justify not providing such a service. This situation explains, in part, the negligent verdict against the hospital.

Chest Pain in a Man With Type 2 Diabetes

A 54-year-old man presented to the ED with a several-hour history of chest pain and shortness of breath. His medical history was remarkable for type 2 diabetes mellitus. He denied any associated nausea, vomiting, or diaphoresis. He smoked one pack of cigarettes per day and drank alcohol on occasion. The patient further denied having experienced similar symptoms in the past. All vital signs were normal, and he appeared comfortable and in no acute distress. His physical examination was completely unremarkable.

An electrocardiogram (ECG) revealed Q-waves in the anterior leads, but was otherwise nondiagnostic. The results of a portable chest X-ray were interpreted as normal. Cardiac enzyme testing revealed a positive troponin I, but a normal creatine phosphokinase-MB (CPK-MB). The patient’s complete blood count and coagulation studies were normal, and a basic metabolic profile was remarkable only for a blood sugar of 540 mg/dL; his serum bicarbonate value was normal.

The patient was placed on oxygen via nasal cannula, administered an aspirin (325 mg), and given subcutaneous regular insulin for hyperglycemia. The EP contacted a hospitalist to admit the patient. The hospitalist accepted the patient and admitted him to the telemetry floor.

Upon arrival at telemetry, the patient was examined by a nurse who noted rhonchi bilaterally on lung auscultation. Shortly afterward, he became anxious, and the nurse consulted the hospitalist, who ordered administration of lorazepam. Approximately 20 minutes later, the patient went into cardiac arrest and died.

This patient was never examined by the hospitalist prior to coding. An autopsy revealed evidence of severe coronary artery disease, a previous infarction that was at least a few months old with scarring of the left ventricle, and a recent infarction that had begun at least 12 hours prior to death.

The family sued all of the providers involved in the care of the patient. At trial, there was a factual dispute regarding whether the EP informed the hospitalist of the elevated troponin I level. The defendants argued that the patient had sustained irreversible heart damage prior to his arrival at the hospital, and that nothing any of the defendants could have done would have saved his life. The jury deliberated for approximately 2 hours before delivering a verdict in favor of the defense.

Discussion

Clearly, this verdict could have gone the other way. This patient was experiencing a non-ST segment elevation myocardial infarction (MI). The nondiagnostic ECG, coupled with the elevated troponin I, indicated damage of heart muscle from an acute interruption of coronary blood flow.

There are several problems with the management of this case. First, this patient required a cardiology consult for risk stratification. Several scoring systems could have been used to determine whether this patient was a candidate for early (ie, within 4-48 hours) invasive treatments such as percutaneous intervention or a more conservative approach.

Second, having sustained an MI, this patient was at high risk for complications such as ventricular arrhythmias, heart failure, cardiogenic shock, and other serious adverse events. Patients with acute MI should be admitted to the critical care or intensive care unit. In addition to aspirin, this patient should have received nitroglycerin and anticoagulation therapy. Either heparin or a low molecular weight heparin, such as enoxaparin, would have been appropriate if no contraindications existed. Finally, additional therapy including glycoprotein IIB/IIIA inhibitors, clopidogrel, etc, may have been indicated depending upon the timing of percutaneous intervention.¹

It appears that both the EP and the hospitalist either failed to appreciate the significance of the elevated troponin I or overlooked it. This patient had normal renal function, so the only explanation—especially in the setting of a middle-aged man complaining of chest pain—was that myocardial damage had occurred.

Alcohol Detoxification in a Young Man

A young man was brought to the ED by a friend. The patient’s sole complaint was the need for help with his alcohol dependency. In addition to alcohol abuse, his medical history was remarkable only for an admission 1 month prior for suicidal ideation. The patient denied suicidal or homicidal ideations on this presentation. The physical examination revealed stable vital signs, but conjunctival injection, slurred speech, and a strong odor of alcohol. A blood alcohol test showed a concentration level of 0.36 g/dL. The patient, however, was alert and able to ambulate without assistance. He was medically cleared by the EP and arrangements were made to admit him to a local detoxification center.

Approximately 4 hours after his arrival, while waiting for transport to the detoxification center, the patient became impatient with the delay, removed his intravenous line, and told the nurse that he was going home via taxi. The nurse encouraged him to call a friend to take him home, to which to the patient agreed. The nurse left the patient to inform the EP of his desire to go home; when she returned, she discovered that the patient had already left the hospital. The physician notified hospital security but not the police. Approximately 2 hours later, the patient was struck by a car and seriously injured.

The patient sued the EP and the hospital for negligence and medical malpractice. The suit alleged that the physician and the hospital should have prevented the patient from leaving the ED. The physician and the hospital requested a dismissal, arguing that the patient did not exhibit any suicidal or homicidal ideation, presented on his own volition, and, though intoxicated, could still make decisions for himself. An appellate court granted the motion, holding that the defendants “lacked authority to confine the plaintiff upon his departure” from the ED.

Discussion

When a patient attempts to leave the ED against medical advice, the treating physician should make an attempt to convince him or her to remain for treatment. Often, something as simple as offering a cup of coffee will change a patient’s mind. In other instances, the use of chemical or physical restraint may be required. The handling of the case ultimately becomes a question of whether the patient was competent to make decisions and whether he presented a danger to himself or others. The extent of intoxication varies by the degree and does not of itself constitute incapability to make decisions. All practicing EPs have cared for patients with blood alcohol levels above the legal limit for driving, but who were functionally sober and able to make decisions. If a patient is competent and does not present a danger to self or others, he or she can decide to leave the ED without further management. However, it is best to release the patient in the company of friends or family, as was urged by the nurse in this case. Obviously, such a patient should not be allowed to drive himself home.

When a patient does leave against medical advice, the physician and nurse should document in the ED record their conversations urging him or her to stay. Alternatively, when a patient is found incompetent of making decisions or is a danger to self or others, he or she must be prevented from leaving the ED. This includes use of the minimal amount of physical or chemical restraint needed to keep the patient from leaving. When there is doubt that the patient is able to make a competent decision, it is better to err on the side of caution and keep him or her in the ED for his own safety and the safety of others.

Question: A 71-year-old woman with heart disease and breast cancer was hospitalized for uncontrolled diabetes and a hip fracture. There, she suffered two grand mal seizures that could not be controlled with anticonvulsants, and the patient lapsed into coma. Her daughter became the surrogate decision maker, and she made it clear that her mother always said she wanted everything done.

After several weeks, the physicians decided that further care would be futile. The chair of the ethics committee took the view that the family’s opinion was not relevant, because cardiopulmonary resuscitation (CPR) was not a genuine therapeutic option and would be "medically contraindicated, inhumane, and unethical." Accordingly, the attending physician entered a do-not-resuscitate (DNR) order despite strong protest from the daughter. The patient died shortly thereafter without receiving CPR.

In this actual case where the daughter filed a negligence lawsuit against the hospital, which of the following statements is incorrect?

A. The defendant’s expert relied upon the position paper of the American Thoracic Society, which states that life support "can be limited without the consent of patient or surrogate when the intervention is judged to be futile."

B. Futile intervention may be defined as treatment that would be highly unlikely to result in a meaningful survival of the patient.

C. The jury found that if competent, the patient would have wanted CPR and would have wanted ventilation until death.

D. The jury found such treatment would be futile.

E. The jury entered a verdict of negligence.

Answer: E. The above narrative is based on Gilgunn v. Massachusetts General Hospital (verdict issued April 21, 1995) and adapted from an article by the prominent ethicist Alexander M. Capron.¹ All of the options listed were evident at trial, except that it was a defense verdict, i.e., no negligence. The case remains the best-known litigated example of medical futility. In earlier cases such as Wanglie² and Baby K,³ the courts had avoided addressing the issue directly.

Gilgunn supports the notion that futility of CPR can trump a patient’s family insistence on having such intervention. But being a trial court verdict, it lacks the precedential authority that an appellate decision would confer.

The verdict also was not without its critics. As Mr. Capron wrote:

"But to allow Mrs. Gilgunn’s physicians to impose this view, however widely held, on their patient is the equivalent of allowing them to abandon the patient. We still need means ... to reach a social consensus on whether health professionals should have authority to decide, among the interventions patients (or surrogates) will accept, which will actually be provided and which they may withhold based upon their evaluation of the worth of the outcome. When we come to adopt such policies, we would do well to ponder long and hard before adopting a utilitarian measure that affords the waning lives of the most vulnerable in our society less protection from unilateral decisions by powerful professionals."¹

In 1999, the District of Columbia Court of Appeals agreed with a trial court’s order to issue a DNR order for a neglected 2-year-old child who was born prematurely with serious medical problems and virtually nonexistent cognition, but who could still experience pain.⁴ The biological mother and putative father appealed, but the Court of Appeals affirmed. The court held, among other things, that the standard of proof required for issuance of a DNR order is clear and convincing evidence, and the applicable standard is the best interests of the child test rather than a substituted judgment standard. However, this case pitted the biological parents against a court-appointed guardian, rather than the medical providers.

The American Medical Association’s current Code of Ethics urges that when neither a patient nor surrogate is able or available to make a decision regarding CPR, an attending physician contemplating a DNR order should consult another physician or a hospital ethics committee if one is available.⁵ If the physician determines that a request for resuscitation would not be medically effective, "the physician should seek to resolve the conflict through a fair decision-making process, when time permits." In an earlier version,⁶ the AMA stated: "CPR may be withheld if, in the judgment of the treating physician, an attempt to resuscitate the patient would be futile."

DNR orders are at the heart of the futility conundrum, especially because CPR is a highly invasive, low-success procedure (notable exceptions exist, however). Medical futility denotes treatment that cannot confer an overall benefit on the whole person even if it can restore some physiologic variable.⁷ The Latin word "futilis" means leaky, and in Greek mythology, the daughters of Danaus were condemned in the underworld to draw water in leaky sieves, conveying the full meaning of futility.

The debate is over whether an intervention, however hopeless and ineffective, can ever be characterized as completely futile. Moreover, without absolute certainty, will this suffice to override a patient’s insistence on having that treatment?

Others have countered that absoluteness is an unrealistic standard, and that inhumane suffering and costs are relevant factors to ponder. The AMA has clearly stated that "physicians are not ethically obligated to deliver care that, in their best professional judgment, will not have a reasonable chance of benefiting their patients," and "patients should not be given treatments simply because they demand them."⁸

One legal consequence of discontinuing medical treatment that ends with a patient’s demise is the specter of criminal prosecution, although charges of homicide are unlikely to prevail.

In a landmark California case, Barber v. Superior Court of Los Angeles County,⁹ the court recognized that "a physician is authorized under the standards of medical practice to discontinue a form of therapy which in his medical judgment is useless. ... If the treating physicians have determined that continued use of a respirator is useless, then they may decide to discontinue it without fear of civil or criminal liability."

In Kansas v. Naramore,¹⁰ the state appeals court reversed and entered a verdict of acquittal despite a jury finding Dr. Naramore guilty over his provision of palliative treatment. The court noted that the burden of proof to establish the criminal guilt of a physician for acts arising out of providing medical treatment is higher than that necessary to find medical malpractice or to impose medical licensure discipline. It held that "with no direct evidence of criminal intent, it is highly disturbing that testimony by such an impressive array of apparently objective medical experts, who found the defendant’s actions to be not only noncriminal, but medically appropriate, can be dismissed as ‘unbelievable’ and not even capable of generating reasonable doubt."

Notwithstanding widely accepted ethical notions of medical futility, there are state and federal statutes touching on DNR orders that warrant careful attention.

For example, New York Public Health Law Section 2962, paragraph 1, states: "Every person admitted to a hospital shall be presumed to consent to the administration of cardiopulmonary resuscitation in the event of cardiac or respiratory arrest, unless there is consent to the issuance of an order not to resuscitate as provided in this article." This raises the question as to whether it is ever legally permissible to enter a unilateral DNR order against the wishes of the patient.

On the other hand, Hawaii Revised Statutes 327E-13(d) notes: "This chapter shall not authorize or require a health care provider or institution to provide health care contrary to generally accepted health care standards." This has been interpreted as allowing the prevailing standard of care to serve as the yardstick of propriety.

Finally, the federal "antidumping" law governing emergency treatment in hospitals, widely known as EMTALA (Emergency Medical Treatment & Labor Act), requires all emergency departments to provide treatment necessary to prevent the material deterioration of the individual’s condition. This would always include the use of CPR unless specifically rejected by the patient or surrogate, because the law does not contain a "standard of care" or futility exception.

References

1. Capron, A.M., Abandoning a Waning Life. Hastings Center Report 1995;25:24-6.

2. In re the conservatorship of Helga M. Wanglie, No. PX-91-283, District Probate Division, 4th Judicial District of the County of Hennepin, State of Minnesota.

3. In the Matter of Baby "K", 16 F.3d F. Supp. 590 (E.D. VA 1993). WL 38674 (4th Cir. 1994).

4. In re K.I., 735 A.2d 448 (D.C. Ct. App. 1999).

5. Code of Ethics of the AMA, section 2.22, 2012-2013 edition.

6. JAMA 1991;265:1868-71.

7. Ann. Int. Med. 1990;112:949-54.

8. Code of Ethics of the AMA, section 2.035, 2012-2013 edition.

9. Barber v. Superior Court of Los Angeles County, 147 Cal. App. 3d 1006 (1983).

10. Kansas v. Naramore, 965 P.2d 211 (Kan. 1998).

Dr. Tan is professor emeritus of medicine and former adjunct professor of law at the University of Hawaii. This article is meant to be educational and does not constitute medical, ethical, or legal advice. It is adapted from the author’s book, "Medical Malpractice: Understanding the Law, Managing the Risk" (2006). For additional information, readers may contact the author at [email protected].

Question: A 71-year-old woman with heart disease and breast cancer was hospitalized for uncontrolled diabetes and a hip fracture. There, she suffered two grand mal seizures that could not be controlled with anticonvulsants, and the patient lapsed into coma. Her daughter became the surrogate decision maker, and she made it clear that her mother always said she wanted everything done.

After several weeks, the physicians decided that further care would be futile. The chair of the ethics committee took the view that the family’s opinion was not relevant, because cardiopulmonary resuscitation (CPR) was not a genuine therapeutic option and would be "medically contraindicated, inhumane, and unethical." Accordingly, the attending physician entered a do-not-resuscitate (DNR) order despite strong protest from the daughter. The patient died shortly thereafter without receiving CPR.

In this actual case where the daughter filed a negligence lawsuit against the hospital, which of the following statements is incorrect?

A. The defendant’s expert relied upon the position paper of the American Thoracic Society, which states that life support "can be limited without the consent of patient or surrogate when the intervention is judged to be futile."

B. Futile intervention may be defined as treatment that would be highly unlikely to result in a meaningful survival of the patient.

C. The jury found that if competent, the patient would have wanted CPR and would have wanted ventilation until death.

D. The jury found such treatment would be futile.

E. The jury entered a verdict of negligence.

Answer: E. The above narrative is based on Gilgunn v. Massachusetts General Hospital (verdict issued April 21, 1995) and adapted from an article by the prominent ethicist Alexander M. Capron.¹ All of the options listed were evident at trial, except that it was a defense verdict, i.e., no negligence. The case remains the best-known litigated example of medical futility. In earlier cases such as Wanglie² and Baby K,³ the courts had avoided addressing the issue directly.

Gilgunn supports the notion that futility of CPR can trump a patient’s family insistence on having such intervention. But being a trial court verdict, it lacks the precedential authority that an appellate decision would confer.

The verdict also was not without its critics. As Mr. Capron wrote:

"But to allow Mrs. Gilgunn’s physicians to impose this view, however widely held, on their patient is the equivalent of allowing them to abandon the patient. We still need means ... to reach a social consensus on whether health professionals should have authority to decide, among the interventions patients (or surrogates) will accept, which will actually be provided and which they may withhold based upon their evaluation of the worth of the outcome. When we come to adopt such policies, we would do well to ponder long and hard before adopting a utilitarian measure that affords the waning lives of the most vulnerable in our society less protection from unilateral decisions by powerful professionals."¹

In 1999, the District of Columbia Court of Appeals agreed with a trial court’s order to issue a DNR order for a neglected 2-year-old child who was born prematurely with serious medical problems and virtually nonexistent cognition, but who could still experience pain.⁴ The biological mother and putative father appealed, but the Court of Appeals affirmed. The court held, among other things, that the standard of proof required for issuance of a DNR order is clear and convincing evidence, and the applicable standard is the best interests of the child test rather than a substituted judgment standard. However, this case pitted the biological parents against a court-appointed guardian, rather than the medical providers.

The American Medical Association’s current Code of Ethics urges that when neither a patient nor surrogate is able or available to make a decision regarding CPR, an attending physician contemplating a DNR order should consult another physician or a hospital ethics committee if one is available.⁵ If the physician determines that a request for resuscitation would not be medically effective, "the physician should seek to resolve the conflict through a fair decision-making process, when time permits." In an earlier version,⁶ the AMA stated: "CPR may be withheld if, in the judgment of the treating physician, an attempt to resuscitate the patient would be futile."

DNR orders are at the heart of the futility conundrum, especially because CPR is a highly invasive, low-success procedure (notable exceptions exist, however). Medical futility denotes treatment that cannot confer an overall benefit on the whole person even if it can restore some physiologic variable.⁷ The Latin word "futilis" means leaky, and in Greek mythology, the daughters of Danaus were condemned in the underworld to draw water in leaky sieves, conveying the full meaning of futility.

The debate is over whether an intervention, however hopeless and ineffective, can ever be characterized as completely futile. Moreover, without absolute certainty, will this suffice to override a patient’s insistence on having that treatment?

Others have countered that absoluteness is an unrealistic standard, and that inhumane suffering and costs are relevant factors to ponder. The AMA has clearly stated that "physicians are not ethically obligated to deliver care that, in their best professional judgment, will not have a reasonable chance of benefiting their patients," and "patients should not be given treatments simply because they demand them."⁸

One legal consequence of discontinuing medical treatment that ends with a patient’s demise is the specter of criminal prosecution, although charges of homicide are unlikely to prevail.

In a landmark California case, Barber v. Superior Court of Los Angeles County,⁹ the court recognized that "a physician is authorized under the standards of medical practice to discontinue a form of therapy which in his medical judgment is useless. ... If the treating physicians have determined that continued use of a respirator is useless, then they may decide to discontinue it without fear of civil or criminal liability."

In Kansas v. Naramore,¹⁰ the state appeals court reversed and entered a verdict of acquittal despite a jury finding Dr. Naramore guilty over his provision of palliative treatment. The court noted that the burden of proof to establish the criminal guilt of a physician for acts arising out of providing medical treatment is higher than that necessary to find medical malpractice or to impose medical licensure discipline. It held that "with no direct evidence of criminal intent, it is highly disturbing that testimony by such an impressive array of apparently objective medical experts, who found the defendant’s actions to be not only noncriminal, but medically appropriate, can be dismissed as ‘unbelievable’ and not even capable of generating reasonable doubt."

Notwithstanding widely accepted ethical notions of medical futility, there are state and federal statutes touching on DNR orders that warrant careful attention.

For example, New York Public Health Law Section 2962, paragraph 1, states: "Every person admitted to a hospital shall be presumed to consent to the administration of cardiopulmonary resuscitation in the event of cardiac or respiratory arrest, unless there is consent to the issuance of an order not to resuscitate as provided in this article." This raises the question as to whether it is ever legally permissible to enter a unilateral DNR order against the wishes of the patient.

On the other hand, Hawaii Revised Statutes 327E-13(d) notes: "This chapter shall not authorize or require a health care provider or institution to provide health care contrary to generally accepted health care standards." This has been interpreted as allowing the prevailing standard of care to serve as the yardstick of propriety.

Finally, the federal "antidumping" law governing emergency treatment in hospitals, widely known as EMTALA (Emergency Medical Treatment & Labor Act), requires all emergency departments to provide treatment necessary to prevent the material deterioration of the individual’s condition. This would always include the use of CPR unless specifically rejected by the patient or surrogate, because the law does not contain a "standard of care" or futility exception.

References

1. Capron, A.M., Abandoning a Waning Life. Hastings Center Report 1995;25:24-6.

2. In re the conservatorship of Helga M. Wanglie, No. PX-91-283, District Probate Division, 4th Judicial District of the County of Hennepin, State of Minnesota.

3. In the Matter of Baby "K", 16 F.3d F. Supp. 590 (E.D. VA 1993). WL 38674 (4th Cir. 1994).

4. In re K.I., 735 A.2d 448 (D.C. Ct. App. 1999).

5. Code of Ethics of the AMA, section 2.22, 2012-2013 edition.

6. JAMA 1991;265:1868-71.

7. Ann. Int. Med. 1990;112:949-54.

8. Code of Ethics of the AMA, section 2.035, 2012-2013 edition.

9. Barber v. Superior Court of Los Angeles County, 147 Cal. App. 3d 1006 (1983).

10. Kansas v. Naramore, 965 P.2d 211 (Kan. 1998).

Dr. Tan is professor emeritus of medicine and former adjunct professor of law at the University of Hawaii. This article is meant to be educational and does not constitute medical, ethical, or legal advice. It is adapted from the author’s book, "Medical Malpractice: Understanding the Law, Managing the Risk" (2006). For additional information, readers may contact the author at [email protected].

Question: A 71-year-old woman with heart disease and breast cancer was hospitalized for uncontrolled diabetes and a hip fracture. There, she suffered two grand mal seizures that could not be controlled with anticonvulsants, and the patient lapsed into coma. Her daughter became the surrogate decision maker, and she made it clear that her mother always said she wanted everything done.

After several weeks, the physicians decided that further care would be futile. The chair of the ethics committee took the view that the family’s opinion was not relevant, because cardiopulmonary resuscitation (CPR) was not a genuine therapeutic option and would be "medically contraindicated, inhumane, and unethical." Accordingly, the attending physician entered a do-not-resuscitate (DNR) order despite strong protest from the daughter. The patient died shortly thereafter without receiving CPR.

In this actual case where the daughter filed a negligence lawsuit against the hospital, which of the following statements is incorrect?

A. The defendant’s expert relied upon the position paper of the American Thoracic Society, which states that life support "can be limited without the consent of patient or surrogate when the intervention is judged to be futile."

B. Futile intervention may be defined as treatment that would be highly unlikely to result in a meaningful survival of the patient.

C. The jury found that if competent, the patient would have wanted CPR and would have wanted ventilation until death.

D. The jury found such treatment would be futile.

E. The jury entered a verdict of negligence.

Answer: E. The above narrative is based on Gilgunn v. Massachusetts General Hospital (verdict issued April 21, 1995) and adapted from an article by the prominent ethicist Alexander M. Capron.¹ All of the options listed were evident at trial, except that it was a defense verdict, i.e., no negligence. The case remains the best-known litigated example of medical futility. In earlier cases such as Wanglie² and Baby K,³ the courts had avoided addressing the issue directly.

Gilgunn supports the notion that futility of CPR can trump a patient’s family insistence on having such intervention. But being a trial court verdict, it lacks the precedential authority that an appellate decision would confer.

The verdict also was not without its critics. As Mr. Capron wrote:

"But to allow Mrs. Gilgunn’s physicians to impose this view, however widely held, on their patient is the equivalent of allowing them to abandon the patient. We still need means ... to reach a social consensus on whether health professionals should have authority to decide, among the interventions patients (or surrogates) will accept, which will actually be provided and which they may withhold based upon their evaluation of the worth of the outcome. When we come to adopt such policies, we would do well to ponder long and hard before adopting a utilitarian measure that affords the waning lives of the most vulnerable in our society less protection from unilateral decisions by powerful professionals."¹

In 1999, the District of Columbia Court of Appeals agreed with a trial court’s order to issue a DNR order for a neglected 2-year-old child who was born prematurely with serious medical problems and virtually nonexistent cognition, but who could still experience pain.⁴ The biological mother and putative father appealed, but the Court of Appeals affirmed. The court held, among other things, that the standard of proof required for issuance of a DNR order is clear and convincing evidence, and the applicable standard is the best interests of the child test rather than a substituted judgment standard. However, this case pitted the biological parents against a court-appointed guardian, rather than the medical providers.

The American Medical Association’s current Code of Ethics urges that when neither a patient nor surrogate is able or available to make a decision regarding CPR, an attending physician contemplating a DNR order should consult another physician or a hospital ethics committee if one is available.⁵ If the physician determines that a request for resuscitation would not be medically effective, "the physician should seek to resolve the conflict through a fair decision-making process, when time permits." In an earlier version,⁶ the AMA stated: "CPR may be withheld if, in the judgment of the treating physician, an attempt to resuscitate the patient would be futile."

DNR orders are at the heart of the futility conundrum, especially because CPR is a highly invasive, low-success procedure (notable exceptions exist, however). Medical futility denotes treatment that cannot confer an overall benefit on the whole person even if it can restore some physiologic variable.⁷ The Latin word "futilis" means leaky, and in Greek mythology, the daughters of Danaus were condemned in the underworld to draw water in leaky sieves, conveying the full meaning of futility.

The debate is over whether an intervention, however hopeless and ineffective, can ever be characterized as completely futile. Moreover, without absolute certainty, will this suffice to override a patient’s insistence on having that treatment?

Others have countered that absoluteness is an unrealistic standard, and that inhumane suffering and costs are relevant factors to ponder. The AMA has clearly stated that "physicians are not ethically obligated to deliver care that, in their best professional judgment, will not have a reasonable chance of benefiting their patients," and "patients should not be given treatments simply because they demand them."⁸

One legal consequence of discontinuing medical treatment that ends with a patient’s demise is the specter of criminal prosecution, although charges of homicide are unlikely to prevail.

In a landmark California case, Barber v. Superior Court of Los Angeles County,⁹ the court recognized that "a physician is authorized under the standards of medical practice to discontinue a form of therapy which in his medical judgment is useless. ... If the treating physicians have determined that continued use of a respirator is useless, then they may decide to discontinue it without fear of civil or criminal liability."

In Kansas v. Naramore,¹⁰ the state appeals court reversed and entered a verdict of acquittal despite a jury finding Dr. Naramore guilty over his provision of palliative treatment. The court noted that the burden of proof to establish the criminal guilt of a physician for acts arising out of providing medical treatment is higher than that necessary to find medical malpractice or to impose medical licensure discipline. It held that "with no direct evidence of criminal intent, it is highly disturbing that testimony by such an impressive array of apparently objective medical experts, who found the defendant’s actions to be not only noncriminal, but medically appropriate, can be dismissed as ‘unbelievable’ and not even capable of generating reasonable doubt."

Notwithstanding widely accepted ethical notions of medical futility, there are state and federal statutes touching on DNR orders that warrant careful attention.

For example, New York Public Health Law Section 2962, paragraph 1, states: "Every person admitted to a hospital shall be presumed to consent to the administration of cardiopulmonary resuscitation in the event of cardiac or respiratory arrest, unless there is consent to the issuance of an order not to resuscitate as provided in this article." This raises the question as to whether it is ever legally permissible to enter a unilateral DNR order against the wishes of the patient.

On the other hand, Hawaii Revised Statutes 327E-13(d) notes: "This chapter shall not authorize or require a health care provider or institution to provide health care contrary to generally accepted health care standards." This has been interpreted as allowing the prevailing standard of care to serve as the yardstick of propriety.

Finally, the federal "antidumping" law governing emergency treatment in hospitals, widely known as EMTALA (Emergency Medical Treatment & Labor Act), requires all emergency departments to provide treatment necessary to prevent the material deterioration of the individual’s condition. This would always include the use of CPR unless specifically rejected by the patient or surrogate, because the law does not contain a "standard of care" or futility exception.

References

1. Capron, A.M., Abandoning a Waning Life. Hastings Center Report 1995;25:24-6.

2. In re the conservatorship of Helga M. Wanglie, No. PX-91-283, District Probate Division, 4th Judicial District of the County of Hennepin, State of Minnesota.

3. In the Matter of Baby "K", 16 F.3d F. Supp. 590 (E.D. VA 1993). WL 38674 (4th Cir. 1994).

4. In re K.I., 735 A.2d 448 (D.C. Ct. App. 1999).

5. Code of Ethics of the AMA, section 2.22, 2012-2013 edition.

6. JAMA 1991;265:1868-71.

7. Ann. Int. Med. 1990;112:949-54.

8. Code of Ethics of the AMA, section 2.035, 2012-2013 edition.

9. Barber v. Superior Court of Los Angeles County, 147 Cal. App. 3d 1006 (1983).

10. Kansas v. Naramore, 965 P.2d 211 (Kan. 1998).

Dr. Tan is professor emeritus of medicine and former adjunct professor of law at the University of Hawaii. This article is meant to be educational and does not constitute medical, ethical, or legal advice. It is adapted from the author’s book, "Medical Malpractice: Understanding the Law, Managing the Risk" (2006). For additional information, readers may contact the author at [email protected].

Question: Mrs. P sustained multiple fractures after a drunk driver struck her car. The orthopedic surgeon accidentally nicked her femoral artery during surgery, which resulted in profuse hemorrhage requiring six units of packed red blood cells. Although she survived, Mrs. P was left with irreversible renal failure, and she now requires lifelong dialysis. Which of the following is correct?

A. The drunk driver’s negligence may be both a factual and proximate cause of all of Mrs. P’s injuries.

B. The surgeon’s action may be deemed a factual, a concurring, or a superseding cause.

C. The surgeon will be successfully sued for malpractice.

D. Only A and B are correct.

E. All are correct.

Answer: D. The surgeon may be successfully sued for malpractice if the nicking of the artery is shown to be a negligent act. This is by no means a foregone conclusion, as a bad outcome is not necessarily indicative of negligence. The measure of legal negligence is what is to be ordinarily expected of a surgeon under similar situations.

For example, expert testimony may establish exculpatory circumstances such as an obscured surgical field, anomalous anatomy, emergency conditions, etc., that would free the surgeon from liability.

Causation issues have long plagued courts and scholars, earning epithets like "a thicket of complexities" and "a simplicity that is deceptive." Causation inquires into both factual and proximate cause. Factual cause deals with whether there is a physical and sequential cause-effect relationship between a defendant’s negligence and a plaintiff’s injuries. It uses the "but-for" test, which stipulates that the defendant’s conduct is a factual cause of a plaintiff’s injuries if the plaintiff’s harm would not have occurred but for defendant’s conduct, i.e., in the absence of the defendant’s tortious conduct.

A recent case is illustrative (BNM v. National University of Singapore, [2014] SGHC 05). An obese, middle-aged man drowned while swimming in the university pool. Lifeguards were on duty, but they were neither aware of where the emergency equipment was kept nor adequately trained in cardiopulmonary resuscitation. Efforts to revive the swimmer failed.

At autopsy, the victim was found to have cardiomegaly and advanced coronary artery disease, with old foci of myocardial scarring. The coroner testified that the victim probably suffered a major cardiac event, such as an arrhythmia, prior to drowning.

On the issue of causation, the court held that because of his severe underlying heart disease, the deceased was not likely to have survived, even if the lifeguards had acted more promptly, i.e., the negligent lifeguards did not factually cause the victim’s death.

A more important inquiry into causal connectivity is captured in the term "proximate cause," which is meant to prevent indeterminate liability. It is sometimes referred to as legal cause.

Unfortunately, there is no bright line to define what constitutes a sufficient causal nexus, and courts are therefore occasionally forced to base their decision on their sense of practical policy and justice. In some cases, an intervening event results in causing, or aggravating, harm suffered by the victim, but the original defendant may be freed of liability if the intervening event constitutes a superseding cause, i.e. unforeseeable event with unforeseeable results.

The opposite is a concurring cause. For example, rescuers may sometimes act negligently following a tortious event leading to aggravated injuries. If such conduct and damages were deemed foreseeable, that would constitute a concurring cause, and the initial wrongdoer then becomes liable for both the original and any aggravated injuries.

A variant of proximate cause is when the harm suffered is a natural expectation of the underlying condition, and the wrongdoer’s negligence simply deprived the victim of some chance of reducing that risk. This is known as the "loss of a chance" doctrine, which has been variously considered a part of causation analysis, a separate tort, or a means to apportion damages.

It asserts that the damage or loss that will materialize or had already materialized, could have been prevented or improved upon – had the victim not been deprived of an opportunity. Loss of a chance is simply another way of saying that the defendant’s conduct has increased the risk of harm to the plaintiff.

The doctrine is well established in contract law, especially where only economic losses are at issue, the key requirement there being that the lost opportunity be real or substantial, and not speculative. The seminal case, Chaplin v. Hicks ([1911] 2 KB 786), dealt with a late notification in a competition, which caused a plaintiff to lose her chance as a finalist to win a prize. The court ruled that the damage was not too remote, and the plaintiff did not have to prove that she would have won the competition.

In clinical negligence, allegations of loss of a chance often arise from an omission to treat or a failure to diagnose, which in turn may deprive the patient of an opportunity of a better outcome. In some U.S. and most Commonwealth jurisdictions, its successful application requires that the underlying condition prior to an individual’s negligence be associated with a better than even chance of cure to begin with.

The controversy arises over what level of risk reduction or lost opportunity is necessary to constitute proximate causation. How large the chance of an adverse outcome and how much of a reduction in that chance are required as a matter of law? Some courts assert that the lost opportunity is relevant only if the plaintiff’s prognosis for survival is better than 50% to begin with, whereas others reject this numerical imperative, especially in cases alleging a delayed diagnosis of cancer.

Proximate cause is the law’s intended mechanism to limit indeterminate liability. Thus, whether a defendant’s negligent conduct constitutes proximate cause is often dispositive in personal injury litigation. However, the term is frequently misunderstood, and has caused much confusion because it has sometimes been used interchangeably with legal cause, or used to include factual cause. And, of course, there can be more than one proximate cause for any given injury.

Reflecting this complexity, the California Supreme Court now disallows confusing jury instructions regarding proximate cause, suggesting instead that the jury be simply directed to determine whether the defendant’s conduct was a contributory factor in the plaintiff’s injury (Mitchell v. Gonzales, 819 P.2d 872 [Cal. 1991]).

Dr. Tan is professor emeritus of medicine and former adjunct professor of law at the University of Hawaii. This article is meant to be educational and does not constitute medical, ethical or legal advice. It is adapted from the author’s book, "Medical Malpractice: Understanding the Law, Managing the Risk" (2006) and his Halsbury treatise, "Medical Negligence and Professional Misconduct" (2012). For additional information, readers may contact the author at [email protected].

Question: Mrs. P sustained multiple fractures after a drunk driver struck her car. The orthopedic surgeon accidentally nicked her femoral artery during surgery, which resulted in profuse hemorrhage requiring six units of packed red blood cells. Although she survived, Mrs. P was left with irreversible renal failure, and she now requires lifelong dialysis. Which of the following is correct?

A. The drunk driver’s negligence may be both a factual and proximate cause of all of Mrs. P’s injuries.

B. The surgeon’s action may be deemed a factual, a concurring, or a superseding cause.

C. The surgeon will be successfully sued for malpractice.

D. Only A and B are correct.

E. All are correct.

Answer: D. The surgeon may be successfully sued for malpractice if the nicking of the artery is shown to be a negligent act. This is by no means a foregone conclusion, as a bad outcome is not necessarily indicative of negligence. The measure of legal negligence is what is to be ordinarily expected of a surgeon under similar situations.

For example, expert testimony may establish exculpatory circumstances such as an obscured surgical field, anomalous anatomy, emergency conditions, etc., that would free the surgeon from liability.

Causation issues have long plagued courts and scholars, earning epithets like "a thicket of complexities" and "a simplicity that is deceptive." Causation inquires into both factual and proximate cause. Factual cause deals with whether there is a physical and sequential cause-effect relationship between a defendant’s negligence and a plaintiff’s injuries. It uses the "but-for" test, which stipulates that the defendant’s conduct is a factual cause of a plaintiff’s injuries if the plaintiff’s harm would not have occurred but for defendant’s conduct, i.e., in the absence of the defendant’s tortious conduct.

A recent case is illustrative (BNM v. National University of Singapore, [2014] SGHC 05). An obese, middle-aged man drowned while swimming in the university pool. Lifeguards were on duty, but they were neither aware of where the emergency equipment was kept nor adequately trained in cardiopulmonary resuscitation. Efforts to revive the swimmer failed.

At autopsy, the victim was found to have cardiomegaly and advanced coronary artery disease, with old foci of myocardial scarring. The coroner testified that the victim probably suffered a major cardiac event, such as an arrhythmia, prior to drowning.

On the issue of causation, the court held that because of his severe underlying heart disease, the deceased was not likely to have survived, even if the lifeguards had acted more promptly, i.e., the negligent lifeguards did not factually cause the victim’s death.

A more important inquiry into causal connectivity is captured in the term "proximate cause," which is meant to prevent indeterminate liability. It is sometimes referred to as legal cause.

Unfortunately, there is no bright line to define what constitutes a sufficient causal nexus, and courts are therefore occasionally forced to base their decision on their sense of practical policy and justice. In some cases, an intervening event results in causing, or aggravating, harm suffered by the victim, but the original defendant may be freed of liability if the intervening event constitutes a superseding cause, i.e. unforeseeable event with unforeseeable results.

The opposite is a concurring cause. For example, rescuers may sometimes act negligently following a tortious event leading to aggravated injuries. If such conduct and damages were deemed foreseeable, that would constitute a concurring cause, and the initial wrongdoer then becomes liable for both the original and any aggravated injuries.

A variant of proximate cause is when the harm suffered is a natural expectation of the underlying condition, and the wrongdoer’s negligence simply deprived the victim of some chance of reducing that risk. This is known as the "loss of a chance" doctrine, which has been variously considered a part of causation analysis, a separate tort, or a means to apportion damages.

It asserts that the damage or loss that will materialize or had already materialized, could have been prevented or improved upon – had the victim not been deprived of an opportunity. Loss of a chance is simply another way of saying that the defendant’s conduct has increased the risk of harm to the plaintiff.

The doctrine is well established in contract law, especially where only economic losses are at issue, the key requirement there being that the lost opportunity be real or substantial, and not speculative. The seminal case, Chaplin v. Hicks ([1911] 2 KB 786), dealt with a late notification in a competition, which caused a plaintiff to lose her chance as a finalist to win a prize. The court ruled that the damage was not too remote, and the plaintiff did not have to prove that she would have won the competition.

In clinical negligence, allegations of loss of a chance often arise from an omission to treat or a failure to diagnose, which in turn may deprive the patient of an opportunity of a better outcome. In some U.S. and most Commonwealth jurisdictions, its successful application requires that the underlying condition prior to an individual’s negligence be associated with a better than even chance of cure to begin with.

The controversy arises over what level of risk reduction or lost opportunity is necessary to constitute proximate causation. How large the chance of an adverse outcome and how much of a reduction in that chance are required as a matter of law? Some courts assert that the lost opportunity is relevant only if the plaintiff’s prognosis for survival is better than 50% to begin with, whereas others reject this numerical imperative, especially in cases alleging a delayed diagnosis of cancer.

Proximate cause is the law’s intended mechanism to limit indeterminate liability. Thus, whether a defendant’s negligent conduct constitutes proximate cause is often dispositive in personal injury litigation. However, the term is frequently misunderstood, and has caused much confusion because it has sometimes been used interchangeably with legal cause, or used to include factual cause. And, of course, there can be more than one proximate cause for any given injury.

Reflecting this complexity, the California Supreme Court now disallows confusing jury instructions regarding proximate cause, suggesting instead that the jury be simply directed to determine whether the defendant’s conduct was a contributory factor in the plaintiff’s injury (Mitchell v. Gonzales, 819 P.2d 872 [Cal. 1991]).

Dr. Tan is professor emeritus of medicine and former adjunct professor of law at the University of Hawaii. This article is meant to be educational and does not constitute medical, ethical or legal advice. It is adapted from the author’s book, "Medical Malpractice: Understanding the Law, Managing the Risk" (2006) and his Halsbury treatise, "Medical Negligence and Professional Misconduct" (2012). For additional information, readers may contact the author at [email protected].

Question: Mrs. P sustained multiple fractures after a drunk driver struck her car. The orthopedic surgeon accidentally nicked her femoral artery during surgery, which resulted in profuse hemorrhage requiring six units of packed red blood cells. Although she survived, Mrs. P was left with irreversible renal failure, and she now requires lifelong dialysis. Which of the following is correct?

A. The drunk driver’s negligence may be both a factual and proximate cause of all of Mrs. P’s injuries.

B. The surgeon’s action may be deemed a factual, a concurring, or a superseding cause.

C. The surgeon will be successfully sued for malpractice.

D. Only A and B are correct.

E. All are correct.

Answer: D. The surgeon may be successfully sued for malpractice if the nicking of the artery is shown to be a negligent act. This is by no means a foregone conclusion, as a bad outcome is not necessarily indicative of negligence. The measure of legal negligence is what is to be ordinarily expected of a surgeon under similar situations.

For example, expert testimony may establish exculpatory circumstances such as an obscured surgical field, anomalous anatomy, emergency conditions, etc., that would free the surgeon from liability.

Causation issues have long plagued courts and scholars, earning epithets like "a thicket of complexities" and "a simplicity that is deceptive." Causation inquires into both factual and proximate cause. Factual cause deals with whether there is a physical and sequential cause-effect relationship between a defendant’s negligence and a plaintiff’s injuries. It uses the "but-for" test, which stipulates that the defendant’s conduct is a factual cause of a plaintiff’s injuries if the plaintiff’s harm would not have occurred but for defendant’s conduct, i.e., in the absence of the defendant’s tortious conduct.

A recent case is illustrative (BNM v. National University of Singapore, [2014] SGHC 05). An obese, middle-aged man drowned while swimming in the university pool. Lifeguards were on duty, but they were neither aware of where the emergency equipment was kept nor adequately trained in cardiopulmonary resuscitation. Efforts to revive the swimmer failed.

At autopsy, the victim was found to have cardiomegaly and advanced coronary artery disease, with old foci of myocardial scarring. The coroner testified that the victim probably suffered a major cardiac event, such as an arrhythmia, prior to drowning.

On the issue of causation, the court held that because of his severe underlying heart disease, the deceased was not likely to have survived, even if the lifeguards had acted more promptly, i.e., the negligent lifeguards did not factually cause the victim’s death.

A more important inquiry into causal connectivity is captured in the term "proximate cause," which is meant to prevent indeterminate liability. It is sometimes referred to as legal cause.

Unfortunately, there is no bright line to define what constitutes a sufficient causal nexus, and courts are therefore occasionally forced to base their decision on their sense of practical policy and justice. In some cases, an intervening event results in causing, or aggravating, harm suffered by the victim, but the original defendant may be freed of liability if the intervening event constitutes a superseding cause, i.e. unforeseeable event with unforeseeable results.

The opposite is a concurring cause. For example, rescuers may sometimes act negligently following a tortious event leading to aggravated injuries. If such conduct and damages were deemed foreseeable, that would constitute a concurring cause, and the initial wrongdoer then becomes liable for both the original and any aggravated injuries.

A variant of proximate cause is when the harm suffered is a natural expectation of the underlying condition, and the wrongdoer’s negligence simply deprived the victim of some chance of reducing that risk. This is known as the "loss of a chance" doctrine, which has been variously considered a part of causation analysis, a separate tort, or a means to apportion damages.

It asserts that the damage or loss that will materialize or had already materialized, could have been prevented or improved upon – had the victim not been deprived of an opportunity. Loss of a chance is simply another way of saying that the defendant’s conduct has increased the risk of harm to the plaintiff.

The doctrine is well established in contract law, especially where only economic losses are at issue, the key requirement there being that the lost opportunity be real or substantial, and not speculative. The seminal case, Chaplin v. Hicks ([1911] 2 KB 786), dealt with a late notification in a competition, which caused a plaintiff to lose her chance as a finalist to win a prize. The court ruled that the damage was not too remote, and the plaintiff did not have to prove that she would have won the competition.

In clinical negligence, allegations of loss of a chance often arise from an omission to treat or a failure to diagnose, which in turn may deprive the patient of an opportunity of a better outcome. In some U.S. and most Commonwealth jurisdictions, its successful application requires that the underlying condition prior to an individual’s negligence be associated with a better than even chance of cure to begin with.

The controversy arises over what level of risk reduction or lost opportunity is necessary to constitute proximate causation. How large the chance of an adverse outcome and how much of a reduction in that chance are required as a matter of law? Some courts assert that the lost opportunity is relevant only if the plaintiff’s prognosis for survival is better than 50% to begin with, whereas others reject this numerical imperative, especially in cases alleging a delayed diagnosis of cancer.

Proximate cause is the law’s intended mechanism to limit indeterminate liability. Thus, whether a defendant’s negligent conduct constitutes proximate cause is often dispositive in personal injury litigation. However, the term is frequently misunderstood, and has caused much confusion because it has sometimes been used interchangeably with legal cause, or used to include factual cause. And, of course, there can be more than one proximate cause for any given injury.

Reflecting this complexity, the California Supreme Court now disallows confusing jury instructions regarding proximate cause, suggesting instead that the jury be simply directed to determine whether the defendant’s conduct was a contributory factor in the plaintiff’s injury (Mitchell v. Gonzales, 819 P.2d 872 [Cal. 1991]).

Dr. Tan is professor emeritus of medicine and former adjunct professor of law at the University of Hawaii. This article is meant to be educational and does not constitute medical, ethical or legal advice. It is adapted from the author’s book, "Medical Malpractice: Understanding the Law, Managing the Risk" (2006) and his Halsbury treatise, "Medical Negligence and Professional Misconduct" (2012). For additional information, readers may contact the author at [email protected].

Question: A doctor writes a refill for 0.075-mg Synthroid, which the pharmacist substituted with a generic l-thyroxine preparation. The doctor’s prescription did not specify "do not substitute," and the pharmacist did not call for approval before switching. One year later, the patient was noted to be in atrial fibrillation with a suppressed serum TSH (thyroid-stimulating hormone) level, but there was no earlier EKG or serum TSH measurement for comparison. Which of the following is best?

A. A generic drug is identical to the branded version in chemical composition and bioavailability.

B. Thyroxine has a narrow therapeutic index, so substitution with a generic version is forbidden.

C. The doctor should have checked serum TSH every 3 months.

D. The pharmacist should have asked for physician approval before switching.

E. In a malpractice lawsuit against the doctor, the patient’s biggest legal hurdle is to prove breach and causation.

Answer: E. Although a generic drug contains the same active chemical as its brand-name analogue and is considered "bioequivalent," it is not identical as to bioavailability (see below).

For some drugs, the health care provider should exercise greater caution when using generics, as the therapeutic window is narrow and toxicity may result. Thyroxine is such an example, but use of the generic version is still entirely within the standard of care. Checking serum TSH levels in this case is a good idea, but this does not have to be done every 3 months unless there is another change in preparation or clinical signs and/or symptoms so dictate.

Under many state statutes, there is no requirement for a pharmacist to seek physician approval prior to generic substitution, unless it is plainly written into the prescription. Choice E is best in this hypothetical situation, because the tort of negligence requires a plaintiff to prove the defendant’s breach of duty as well as causation, and both elements will face defense rebuttal under the given facts.

The U.S. Food and Drug Administration is the governmental body that regulates prescription drugs. In 1984, Congress passed the Drug Price Competition and Patent Term Restoration Act, popularly known as the Hatch-Waxman Act. This allowed generic versions of drugs to proliferate, but it also extended patent life for proprietary drug manufacturers whose profits were threatened by the onslaught of the cheaper generic copycats.

In order for the FDA to ensure that generics were equivalently safe and effective without requiring the same stringent clinical trials, it relied on the concept of bioequivalence, which is a statistical interpretation of a drug’s bioavailability. The latter measures the rate and extent of a drug’s absorption, generally defined by its maximum plasma concentration (C-max) and the area under the curve (AUC).

The FDA defines bioequivalence as "the absence of a significant difference in the rate and extent to which the active ingredient or active moiety in pharmaceutical equivalents or pharmaceutical alternatives becomes available at the site of drug action when administered at the same molar dose under similar conditions in an appropriately designed study." In statistical terms, this translates into allowing a generic to vary from 80% to 125% of the original drug.

Bioequivalent drugs are tabulated in the FDA’s Orange Book, which pharmacists rely upon when switching preparations.

In the vast majority of cases, a difference in bioavailability of a generic drug ranging from 20% less to 25% more than a brand-name drug exerts no effect on therapeutic outcome. However, in drugs with a narrow therapeutic index (the FDA’s preferred term is narrow therapeutic ratio), there is only a small difference between therapeutic and toxic plasma levels, usually less than a twofold difference between the median lethal dose and the median effective dose. Generic substitution should therefore proceed, if at all, with careful dose titration and monitoring. Some examples of drugs with a narrow therapeutic index are warfarin, digoxin, phenytoin, cyclosporin, levothyroxine, lithium, carbamazepine, clonidine, minoxidil, and theophylline.

The case of Winn Dixie of Montgomery, Inc. v. Colburn (709 So.2d 1222 [Ala. 1998]) illustrates some of the issues faced by drug substitution, although this case did not involve an actual generic switch.

In Colburn, the doctor wrote a prescription for Sedapap, a non–codeine-containing compound, to treat his patient’s migraine, but the pharmacist negligently substituted it with Fiorinal #3, which contains codeine. The doctor had checked off on the line "product selection permitted," which meant a generic substitution was acceptable. The pharmacy’s computer drug profile erroneously listed Sedapap and Fiorinal #3 as being equivalent, whereas they are in fact different compounds, therapeutically equivalent but not having generic bioequivalence.

It turned out the patient had a known serious allergy to codeine and developed an anaphylactic reaction after taking the substituted drug. The court found for the plaintiff, and held both the pharmacist and the drugstore liable for damages of $130,000. There was evidence at trial that the pharmacist had called the doctor’s office, which did not give approval for the substitution!

It is estimated that three-quarters of prescriptions are filled with generic substitutes, the law and the substantial cost savings prompting formularies to offer them to an eager public. Numerous state statutes have been enacted to regulate generic substitutions, although these laws vary in important aspects, such as permissible versus mandatory (automatic) substitutions, the roles of pharmacist and prescribing doctor, and exceptions to generic drug switch.

Prescriptions drugs cost $269.2 billion in 2011, a significant portion of total national health spending of $2.7 trillion, and are expected to cost even more, according to estimates by the Centers for Medicare and Medicaid Services. Increasing use of generics should attenuate this escalating price tag. For example, a recent study of drugs in disease prevention reported that whereas blood pressure reduction with a brand-name drug would cost an estimated $53,000 per quality-adjusted life-year (QALY), this figure would drop dramatically to less than $8,000 with the use of a generic substitute (Health Aff. (Millwood) 2011;30:1351-7).

Two 2013 U.S. Supreme Court cases address additional issues surrounding generic drugs.

In Mutual Pharmaceutical Co. v. Bartlett (133 S. Ct. 2466 [2013]), the court held that a generic manufacturer cannot be held liable for inadequate warnings if its labeling faithfully tracks that of the proprietary drug – in this case, the nonsteroidal anti-inflammatory drug sulindac. Federal law forbids any deviation from the parent label, and trumps any other legal premises otherwise afforded by state tort law. It was a case that dealt with failure to adequately warn of the rare but serious complication of toxic epidermal necrolysis, which the patient developed after using a generic version of the drug.

In Federal Trade Commission v. Actavis (133 S. Ct. 2223 [2013]), the court held that “reverse payment” agreements are subject to antitrust scrutiny to ensure they are not anticompetitive. The general issue concerns vulnerable brand-name drugs about to lose their patent protection. A generic firm would place its version into the marketplace before the parent drug’s expiration date. To avoid a costly legal battle and loss of profits in the interim, the proprietary drug manufacturer would be incentivized to negotiate for a rollback of the generic drug’s release in exchange for a payout (“pay-for-delay”).

Dr. Tan is emeritus professor of medicine and former adjunct professor of law at the University of Hawaii. This article is meant to be educational and does not constitute medical, ethical, or legal advice. It is adapted from the author’s book, "Medical Malpractice: Understanding the Law, Managing the Risk" (2006). For additional information, readers may contact the author at [email protected].

Question: A doctor writes a refill for 0.075-mg Synthroid, which the pharmacist substituted with a generic l-thyroxine preparation. The doctor’s prescription did not specify "do not substitute," and the pharmacist did not call for approval before switching. One year later, the patient was noted to be in atrial fibrillation with a suppressed serum TSH (thyroid-stimulating hormone) level, but there was no earlier EKG or serum TSH measurement for comparison. Which of the following is best?

A. A generic drug is identical to the branded version in chemical composition and bioavailability.

B. Thyroxine has a narrow therapeutic index, so substitution with a generic version is forbidden.

C. The doctor should have checked serum TSH every 3 months.

D. The pharmacist should have asked for physician approval before switching.

E. In a malpractice lawsuit against the doctor, the patient’s biggest legal hurdle is to prove breach and causation.

Answer: E. Although a generic drug contains the same active chemical as its brand-name analogue and is considered "bioequivalent," it is not identical as to bioavailability (see below).

For some drugs, the health care provider should exercise greater caution when using generics, as the therapeutic window is narrow and toxicity may result. Thyroxine is such an example, but use of the generic version is still entirely within the standard of care. Checking serum TSH levels in this case is a good idea, but this does not have to be done every 3 months unless there is another change in preparation or clinical signs and/or symptoms so dictate.

Under many state statutes, there is no requirement for a pharmacist to seek physician approval prior to generic substitution, unless it is plainly written into the prescription. Choice E is best in this hypothetical situation, because the tort of negligence requires a plaintiff to prove the defendant’s breach of duty as well as causation, and both elements will face defense rebuttal under the given facts.

The U.S. Food and Drug Administration is the governmental body that regulates prescription drugs. In 1984, Congress passed the Drug Price Competition and Patent Term Restoration Act, popularly known as the Hatch-Waxman Act. This allowed generic versions of drugs to proliferate, but it also extended patent life for proprietary drug manufacturers whose profits were threatened by the onslaught of the cheaper generic copycats.

In order for the FDA to ensure that generics were equivalently safe and effective without requiring the same stringent clinical trials, it relied on the concept of bioequivalence, which is a statistical interpretation of a drug’s bioavailability. The latter measures the rate and extent of a drug’s absorption, generally defined by its maximum plasma concentration (C-max) and the area under the curve (AUC).

The FDA defines bioequivalence as "the absence of a significant difference in the rate and extent to which the active ingredient or active moiety in pharmaceutical equivalents or pharmaceutical alternatives becomes available at the site of drug action when administered at the same molar dose under similar conditions in an appropriately designed study." In statistical terms, this translates into allowing a generic to vary from 80% to 125% of the original drug.

Bioequivalent drugs are tabulated in the FDA’s Orange Book, which pharmacists rely upon when switching preparations.

In the vast majority of cases, a difference in bioavailability of a generic drug ranging from 20% less to 25% more than a brand-name drug exerts no effect on therapeutic outcome. However, in drugs with a narrow therapeutic index (the FDA’s preferred term is narrow therapeutic ratio), there is only a small difference between therapeutic and toxic plasma levels, usually less than a twofold difference between the median lethal dose and the median effective dose. Generic substitution should therefore proceed, if at all, with careful dose titration and monitoring. Some examples of drugs with a narrow therapeutic index are warfarin, digoxin, phenytoin, cyclosporin, levothyroxine, lithium, carbamazepine, clonidine, minoxidil, and theophylline.

The case of Winn Dixie of Montgomery, Inc. v. Colburn (709 So.2d 1222 [Ala. 1998]) illustrates some of the issues faced by drug substitution, although this case did not involve an actual generic switch.

In Colburn, the doctor wrote a prescription for Sedapap, a non–codeine-containing compound, to treat his patient’s migraine, but the pharmacist negligently substituted it with Fiorinal #3, which contains codeine. The doctor had checked off on the line "product selection permitted," which meant a generic substitution was acceptable. The pharmacy’s computer drug profile erroneously listed Sedapap and Fiorinal #3 as being equivalent, whereas they are in fact different compounds, therapeutically equivalent but not having generic bioequivalence.

It turned out the patient had a known serious allergy to codeine and developed an anaphylactic reaction after taking the substituted drug. The court found for the plaintiff, and held both the pharmacist and the drugstore liable for damages of $130,000. There was evidence at trial that the pharmacist had called the doctor’s office, which did not give approval for the substitution!

It is estimated that three-quarters of prescriptions are filled with generic substitutes, the law and the substantial cost savings prompting formularies to offer them to an eager public. Numerous state statutes have been enacted to regulate generic substitutions, although these laws vary in important aspects, such as permissible versus mandatory (automatic) substitutions, the roles of pharmacist and prescribing doctor, and exceptions to generic drug switch.

Prescriptions drugs cost $269.2 billion in 2011, a significant portion of total national health spending of $2.7 trillion, and are expected to cost even more, according to estimates by the Centers for Medicare and Medicaid Services. Increasing use of generics should attenuate this escalating price tag. For example, a recent study of drugs in disease prevention reported that whereas blood pressure reduction with a brand-name drug would cost an estimated $53,000 per quality-adjusted life-year (QALY), this figure would drop dramatically to less than $8,000 with the use of a generic substitute (Health Aff. (Millwood) 2011;30:1351-7).

Two 2013 U.S. Supreme Court cases address additional issues surrounding generic drugs.

In Mutual Pharmaceutical Co. v. Bartlett (133 S. Ct. 2466 [2013]), the court held that a generic manufacturer cannot be held liable for inadequate warnings if its labeling faithfully tracks that of the proprietary drug – in this case, the nonsteroidal anti-inflammatory drug sulindac. Federal law forbids any deviation from the parent label, and trumps any other legal premises otherwise afforded by state tort law. It was a case that dealt with failure to adequately warn of the rare but serious complication of toxic epidermal necrolysis, which the patient developed after using a generic version of the drug.

In Federal Trade Commission v. Actavis (133 S. Ct. 2223 [2013]), the court held that “reverse payment” agreements are subject to antitrust scrutiny to ensure they are not anticompetitive. The general issue concerns vulnerable brand-name drugs about to lose their patent protection. A generic firm would place its version into the marketplace before the parent drug’s expiration date. To avoid a costly legal battle and loss of profits in the interim, the proprietary drug manufacturer would be incentivized to negotiate for a rollback of the generic drug’s release in exchange for a payout (“pay-for-delay”).

Dr. Tan is emeritus professor of medicine and former adjunct professor of law at the University of Hawaii. This article is meant to be educational and does not constitute medical, ethical, or legal advice. It is adapted from the author’s book, "Medical Malpractice: Understanding the Law, Managing the Risk" (2006). For additional information, readers may contact the author at [email protected].

Question: A doctor writes a refill for 0.075-mg Synthroid, which the pharmacist substituted with a generic l-thyroxine preparation. The doctor’s prescription did not specify "do not substitute," and the pharmacist did not call for approval before switching. One year later, the patient was noted to be in atrial fibrillation with a suppressed serum TSH (thyroid-stimulating hormone) level, but there was no earlier EKG or serum TSH measurement for comparison. Which of the following is best?

A. A generic drug is identical to the branded version in chemical composition and bioavailability.

B. Thyroxine has a narrow therapeutic index, so substitution with a generic version is forbidden.

C. The doctor should have checked serum TSH every 3 months.

D. The pharmacist should have asked for physician approval before switching.

E. In a malpractice lawsuit against the doctor, the patient’s biggest legal hurdle is to prove breach and causation.

Answer: E. Although a generic drug contains the same active chemical as its brand-name analogue and is considered "bioequivalent," it is not identical as to bioavailability (see below).

For some drugs, the health care provider should exercise greater caution when using generics, as the therapeutic window is narrow and toxicity may result. Thyroxine is such an example, but use of the generic version is still entirely within the standard of care. Checking serum TSH levels in this case is a good idea, but this does not have to be done every 3 months unless there is another change in preparation or clinical signs and/or symptoms so dictate.

Under many state statutes, there is no requirement for a pharmacist to seek physician approval prior to generic substitution, unless it is plainly written into the prescription. Choice E is best in this hypothetical situation, because the tort of negligence requires a plaintiff to prove the defendant’s breach of duty as well as causation, and both elements will face defense rebuttal under the given facts.

The U.S. Food and Drug Administration is the governmental body that regulates prescription drugs. In 1984, Congress passed the Drug Price Competition and Patent Term Restoration Act, popularly known as the Hatch-Waxman Act. This allowed generic versions of drugs to proliferate, but it also extended patent life for proprietary drug manufacturers whose profits were threatened by the onslaught of the cheaper generic copycats.

In order for the FDA to ensure that generics were equivalently safe and effective without requiring the same stringent clinical trials, it relied on the concept of bioequivalence, which is a statistical interpretation of a drug’s bioavailability. The latter measures the rate and extent of a drug’s absorption, generally defined by its maximum plasma concentration (C-max) and the area under the curve (AUC).

The FDA defines bioequivalence as "the absence of a significant difference in the rate and extent to which the active ingredient or active moiety in pharmaceutical equivalents or pharmaceutical alternatives becomes available at the site of drug action when administered at the same molar dose under similar conditions in an appropriately designed study." In statistical terms, this translates into allowing a generic to vary from 80% to 125% of the original drug.

Bioequivalent drugs are tabulated in the FDA’s Orange Book, which pharmacists rely upon when switching preparations.

In the vast majority of cases, a difference in bioavailability of a generic drug ranging from 20% less to 25% more than a brand-name drug exerts no effect on therapeutic outcome. However, in drugs with a narrow therapeutic index (the FDA’s preferred term is narrow therapeutic ratio), there is only a small difference between therapeutic and toxic plasma levels, usually less than a twofold difference between the median lethal dose and the median effective dose. Generic substitution should therefore proceed, if at all, with careful dose titration and monitoring. Some examples of drugs with a narrow therapeutic index are warfarin, digoxin, phenytoin, cyclosporin, levothyroxine, lithium, carbamazepine, clonidine, minoxidil, and theophylline.

The case of Winn Dixie of Montgomery, Inc. v. Colburn (709 So.2d 1222 [Ala. 1998]) illustrates some of the issues faced by drug substitution, although this case did not involve an actual generic switch.

In Colburn, the doctor wrote a prescription for Sedapap, a non–codeine-containing compound, to treat his patient’s migraine, but the pharmacist negligently substituted it with Fiorinal #3, which contains codeine. The doctor had checked off on the line "product selection permitted," which meant a generic substitution was acceptable. The pharmacy’s computer drug profile erroneously listed Sedapap and Fiorinal #3 as being equivalent, whereas they are in fact different compounds, therapeutically equivalent but not having generic bioequivalence.

It turned out the patient had a known serious allergy to codeine and developed an anaphylactic reaction after taking the substituted drug. The court found for the plaintiff, and held both the pharmacist and the drugstore liable for damages of $130,000. There was evidence at trial that the pharmacist had called the doctor’s office, which did not give approval for the substitution!

It is estimated that three-quarters of prescriptions are filled with generic substitutes, the law and the substantial cost savings prompting formularies to offer them to an eager public. Numerous state statutes have been enacted to regulate generic substitutions, although these laws vary in important aspects, such as permissible versus mandatory (automatic) substitutions, the roles of pharmacist and prescribing doctor, and exceptions to generic drug switch.

Prescriptions drugs cost $269.2 billion in 2011, a significant portion of total national health spending of $2.7 trillion, and are expected to cost even more, according to estimates by the Centers for Medicare and Medicaid Services. Increasing use of generics should attenuate this escalating price tag. For example, a recent study of drugs in disease prevention reported that whereas blood pressure reduction with a brand-name drug would cost an estimated $53,000 per quality-adjusted life-year (QALY), this figure would drop dramatically to less than $8,000 with the use of a generic substitute (Health Aff. (Millwood) 2011;30:1351-7).

Two 2013 U.S. Supreme Court cases address additional issues surrounding generic drugs.

In Mutual Pharmaceutical Co. v. Bartlett (133 S. Ct. 2466 [2013]), the court held that a generic manufacturer cannot be held liable for inadequate warnings if its labeling faithfully tracks that of the proprietary drug – in this case, the nonsteroidal anti-inflammatory drug sulindac. Federal law forbids any deviation from the parent label, and trumps any other legal premises otherwise afforded by state tort law. It was a case that dealt with failure to adequately warn of the rare but serious complication of toxic epidermal necrolysis, which the patient developed after using a generic version of the drug.

In Federal Trade Commission v. Actavis (133 S. Ct. 2223 [2013]), the court held that “reverse payment” agreements are subject to antitrust scrutiny to ensure they are not anticompetitive. The general issue concerns vulnerable brand-name drugs about to lose their patent protection. A generic firm would place its version into the marketplace before the parent drug’s expiration date. To avoid a costly legal battle and loss of profits in the interim, the proprietary drug manufacturer would be incentivized to negotiate for a rollback of the generic drug’s release in exchange for a payout (“pay-for-delay”).

Dr. Tan is emeritus professor of medicine and former adjunct professor of law at the University of Hawaii. This article is meant to be educational and does not constitute medical, ethical, or legal advice. It is adapted from the author’s book, "Medical Malpractice: Understanding the Law, Managing the Risk" (2006). For additional information, readers may contact the author at [email protected].

Case 1: Follow-up in Community-Acquired Methicillin- Resistant Staphylococcus Aureus Pharyngitis

A 73-year-old woman presented to the ED complaining of injuries she sustained following a minor fall at home. The patient’s past medical history was remarkable for diabetes mellitus (DM), hypertension, cerebrovascular accident, and a history of chronic pain. During the course of evaluation, the patient mentioned that she had experienced a sore throat. In addition to X-rays, the emergency physician (EP) ordered a rapid strep test (RST). All studies were normal and the patient was discharged home.

Over the next 2 months, the patient received medical treatment from other medical professionals, but not at the previous hospital. Reportedly, she did not complain of a sore throat during this time period. The patient did return to the same ED approximately 2 months later, complaining of cough and difficulty breathing; she died 5 days thereafter. No autopsy was performed.

A lawsuit was filed on behalf of the patient, stating the hospital breached the standard of care by not reporting the finding of MRSA directly to the treating physician, and that this led directly to the patient’s death by MRSA pneumonia. The jury returned a verdict in favor of the plaintiff for $32 million.

Discussion

This case illustrates two simple but important points: managing community-acquired MRSA (CA-MRSA) infections is a growing challenge for physicians; and hospital follow-up systems for positive findings (ie, fractures, blood cultures, etc) need to be consistently reliable.

It is estimated that 30% of healthy people carry S aureus in their anterior nares; colonization rates for the throat are much less studied. In one recent study, 265 throat swabs were collected from patients aged 14 to 65 years old, who complained of pharyngitis in an outpatient setting.1 A total of 165 S aureus isolates (62.3%) were recovered from the 265 swabs. For the S aureus isolates, 38.2% (63) grew CA-MRSA; the remaining 68.1% (102) were methicillin-sensitive S aureus (MSSA). Interestingly, of the 63 MRSA-positive swabs, over half also grew Group A Streptococcus. The natural disease progression of CA-MRSA pharyngitis is still unknown, as is what to do with a positive throat swab for CA-MRSA. While there are a few case reports of bacteremia and Lemmiere syndrome possibly related to CA-MRSA pharyngitis,2,3 more information is clearly needed. For this case, it is not possible to definitively determine the role of the positive throat swab for CAMRSA in the patient’s subsequent death.

The other teaching point in this case is much simpler and well defined. Simply put, patients expect to be informed of positive findings, whether the result is known at the time of their ED visit or sometime afterward. There needs to be a system in place that consistently and reliably provides important information to either the patient’s treating physician or to the patient. The manners in which this information is communicated are myriad and should take into account hospital resources, the role of the EP and nurse, and what works best for your locality. The “who” and “how” of the contact is not important: reliability, timeliness, and consistency need to be the key drivers of the system.

Case 2: Arterial Occlusion

A 61-year-old woman called emergency medical services (EMS) after noticing her feet felt cold to the touch and having difficulty ambulating. The paramedics noted the patient had normal vital signs and normal circulation in her legs, and she was transported to the ED without incident. Upon arrival to the ED, she was triaged as nonurgent and placed in the minor-care area. On nursing assessment, the patient’s feet were found to be cold, but with palpable pulses bilaterally. Her past medical history was significant for hypertension and DM.

The patient was seen by a physician assistant (PA), who found both feet cool to the touch, but with bilateral pulses present. She was administered intravenous morphine for pain and laboratory studies were ordered. At the time, the PA was concerned about arterial occlusion versus deep vein thrombosis (DVT) versus cellulitis. A venous ultrasound examination was ordered and shown to be negative for DVT. A complete blood count was remarkable only for mild leukocytosis. The PA discussed the case with the supervising EP; they agreed on a diagnosis of cellulitis and discharged the patient home with antibiotics and analgesics.

Approximately 12 hours after discharge, the patient presented back to the ED via ambulance. At that time, she was hypotensive and tachypneic, with a thready palpable pulse. On repeat examination, she no longer had pulses present in her feet. An arteriogram found complete occlusion of her arterial circulation at the level of the knees bilaterally, requiring bilateral belowthe- knee amputation.

The patient sued both the emergency medicine physician and the PA for failure to provide her with the necessary care during her initial ED visit, resulting in loss of limbs. The defendants claimed the patient could not prove gross negligence by clear and convincing evidence, as required by state law. Following the ensuing trial, the jury returned a $5 million verdict in favor of the plaintiff.

Discussion

First, it is important to remember that just because a patient has been triaged to a low-acuity area does not mean she or he must have a minor problem. The provider still must maintain a high level of vigilance— regardless of the location of the patient in the ED.

Second, was this patient in atrial fibrillation, which is responsible in approximately 65% of all peripheral emboli? The abrupt onset of this patient’s symptoms is much more compatible with an embolic origin of her symptoms rather than a thrombus (ie, symptoms of claudication).

Lastly, a diagnosis of cellulitis is inconsistent with the physical findings of the PA, as well as those of the triage nurse and paramedics. This patient’s feet were cool to the touch whereas cellulitis presents with erythema and increased warmth. While the presence of pulses was somewhat reassuring, the cool temperature of the feet and complaint of pain were indicators of the need to evaluate for a possible arterial origin of these findings. However, if this were an embolic phenomenon, peripheral arterial ultrasound would have probably been normal, and the outcome unchanged.

Case 3: Failure to Communicate

A 59-year-old man presented to his primary care physician (PCP) for intermittent right-hand weakness and numbness and tingling in his right arm during the previous 24 hours. As the PCP was concerned that the patient might be experiencing a transient ischemic attack (TIA), the patient and his wife were instructed to go directly to the ED of the local hospital. The PCP wrote a note stating that the patient needed “a stroke work up,” gave the note to the patient, and told him to give it to the ED staff.

The patient went directly to the ED and was immediately seen by a “rapid triage nurse.” He gave the note to the nurse and told her of the PCP’s concerns. The nurse documented on the hospital assessment form that the patient was high priority and needed to be seen immediately. She attached the PCP’s note to the front of the form.

The patient was then seen by the traditional triage nurse. After evaluation, she changed the priority from high to low acuity. The triage nurse later stated later that she never saw the PCP’s note, nor did she obtain any history regarding the concerns of the PCP.

The patient was then evaluated by an EP in the low-acuity (or minor-care) area of the ED. The EP later stated that he never saw the note from the PCP, and had not received any information regarding a suspected TIA or stroke. The EP ordered a right wrist X-ray, diagnosed carpel tunnel syndrome, and prescribed an anti-inflammatory medication as well as follow-up with a hand surgeon.

The initial (rapid triage) nurse saw the patient leaving the ED at the time of discharge and thought he had not been in the ED long enough to have undergone a stroke work up. She reviewed his paperwork and saw the patient had not received the indicated work up. The nurse called the patient’s house and left a message on the answering machine notifying him of the need to return to the ED. The patient arrived back to the ED approximately 2 hours later.

On the second ED presentation that day, the patient was evaluated by a different EP. The patient had blood drawn, an electrocardiogram, and a noncontrast computed tomography scan of the brain, the results of which were all normal. The EP concluded the patient required admission to the hospital for additional work up (eg, carotid Doppler ultrasound). The hospitalist was paged, and told the EP he would be there “as soon as possible.” However, after several hours delay, and no hospitalist, the patient became impatient and expressed the desire to go home. The EP urged the patient to follow up with his PCP in the morning to complete the evaluation.

The next day, before seeing his PCP, the patient suffered an ischemic stroke with right-sided hemiparesis. The case went to trial, and the jury found in favor of the plaintiff.

Discussion

Unfortunately, multiple opportunities were lost in obtaining the correct care at the right time for this patient. Lack of communication and poor communication are frequently cited as causes in medical malpractice cases, and this case perfectly illustrates this problem.

First, the PCP should have called the ED and spoken to the EP directly. This would have provided the PCP the opportunity to express his concerns directly to the treating physician. This kind of one-on-one communication between physicians will always be superior to a hand-written note.

Second, it is unclear why the triage nurse changed the initial nurse’s correct assessment. It is also unclear what happened to the PCP’s note—it was never seen again. Clearly there was miscommunication at this point between the triage nurse and the patient. This case further illustrates the importance of good triage. Once a patient is directed down the wrong pathway (ie, to minor care rather than the main treatment area), the situation becomes much more difficult to correct.

Next, the EP in the low-acuity area was probably falsely assured this patient had only a “minor” problem problem, and not something serious. Emergency physicians must be vigilant to the possibility that the patient can have something seriously wrong even if he or she has been triaged to a low-acuity area. A minor sore throat can turn out to be epiglottitis and a viral stomachache can turn out to be appendicitis. These patients deserve the same quality of history, physical examination, and differential diagnosis as any other patient in the ED.

Finally, while we are not responsible for hospitalists or consultants, we do have a responsibility to our patients. We need to ensure that the care they receive is the appropriate care. Possible alternatives to discharging this patient would have been to call another hospitalist for admission or to seek the input of the chief of the medical staff or the on-call hospital administrator. As EPs, we are frequently required to serve as the primary advocate for our patients.

There is the possibility that even if the patient had been admitted to the hospital, the outcome would have been the same. However, since he might have been a candidate for tissue plasminogen activator or interventional radiology if he had suffered the cerebrovascular accident as an inpatient, he lost his best chance for a good outcome.

Case 1: Follow-up in Community-Acquired Methicillin- Resistant Staphylococcus Aureus Pharyngitis

A 73-year-old woman presented to the ED complaining of injuries she sustained following a minor fall at home. The patient’s past medical history was remarkable for diabetes mellitus (DM), hypertension, cerebrovascular accident, and a history of chronic pain. During the course of evaluation, the patient mentioned that she had experienced a sore throat. In addition to X-rays, the emergency physician (EP) ordered a rapid strep test (RST). All studies were normal and the patient was discharged home.

Over the next 2 months, the patient received medical treatment from other medical professionals, but not at the previous hospital. Reportedly, she did not complain of a sore throat during this time period. The patient did return to the same ED approximately 2 months later, complaining of cough and difficulty breathing; she died 5 days thereafter. No autopsy was performed.

A lawsuit was filed on behalf of the patient, stating the hospital breached the standard of care by not reporting the finding of MRSA directly to the treating physician, and that this led directly to the patient’s death by MRSA pneumonia. The jury returned a verdict in favor of the plaintiff for $32 million.

Discussion

This case illustrates two simple but important points: managing community-acquired MRSA (CA-MRSA) infections is a growing challenge for physicians; and hospital follow-up systems for positive findings (ie, fractures, blood cultures, etc) need to be consistently reliable.

It is estimated that 30% of healthy people carry S aureus in their anterior nares; colonization rates for the throat are much less studied. In one recent study, 265 throat swabs were collected from patients aged 14 to 65 years old, who complained of pharyngitis in an outpatient setting.1 A total of 165 S aureus isolates (62.3%) were recovered from the 265 swabs. For the S aureus isolates, 38.2% (63) grew CA-MRSA; the remaining 68.1% (102) were methicillin-sensitive S aureus (MSSA). Interestingly, of the 63 MRSA-positive swabs, over half also grew Group A Streptococcus. The natural disease progression of CA-MRSA pharyngitis is still unknown, as is what to do with a positive throat swab for CA-MRSA. While there are a few case reports of bacteremia and Lemmiere syndrome possibly related to CA-MRSA pharyngitis,2,3 more information is clearly needed. For this case, it is not possible to definitively determine the role of the positive throat swab for CAMRSA in the patient’s subsequent death.

The other teaching point in this case is much simpler and well defined. Simply put, patients expect to be informed of positive findings, whether the result is known at the time of their ED visit or sometime afterward. There needs to be a system in place that consistently and reliably provides important information to either the patient’s treating physician or to the patient. The manners in which this information is communicated are myriad and should take into account hospital resources, the role of the EP and nurse, and what works best for your locality. The “who” and “how” of the contact is not important: reliability, timeliness, and consistency need to be the key drivers of the system.

Case 2: Arterial Occlusion

A 61-year-old woman called emergency medical services (EMS) after noticing her feet felt cold to the touch and having difficulty ambulating. The paramedics noted the patient had normal vital signs and normal circulation in her legs, and she was transported to the ED without incident. Upon arrival to the ED, she was triaged as nonurgent and placed in the minor-care area. On nursing assessment, the patient’s feet were found to be cold, but with palpable pulses bilaterally. Her past medical history was significant for hypertension and DM.

The patient was seen by a physician assistant (PA), who found both feet cool to the touch, but with bilateral pulses present. She was administered intravenous morphine for pain and laboratory studies were ordered. At the time, the PA was concerned about arterial occlusion versus deep vein thrombosis (DVT) versus cellulitis. A venous ultrasound examination was ordered and shown to be negative for DVT. A complete blood count was remarkable only for mild leukocytosis. The PA discussed the case with the supervising EP; they agreed on a diagnosis of cellulitis and discharged the patient home with antibiotics and analgesics.

Approximately 12 hours after discharge, the patient presented back to the ED via ambulance. At that time, she was hypotensive and tachypneic, with a thready palpable pulse. On repeat examination, she no longer had pulses present in her feet. An arteriogram found complete occlusion of her arterial circulation at the level of the knees bilaterally, requiring bilateral belowthe- knee amputation.

The patient sued both the emergency medicine physician and the PA for failure to provide her with the necessary care during her initial ED visit, resulting in loss of limbs. The defendants claimed the patient could not prove gross negligence by clear and convincing evidence, as required by state law. Following the ensuing trial, the jury returned a $5 million verdict in favor of the plaintiff.

Discussion

First, it is important to remember that just because a patient has been triaged to a low-acuity area does not mean she or he must have a minor problem. The provider still must maintain a high level of vigilance— regardless of the location of the patient in the ED.

Second, was this patient in atrial fibrillation, which is responsible in approximately 65% of all peripheral emboli? The abrupt onset of this patient’s symptoms is much more compatible with an embolic origin of her symptoms rather than a thrombus (ie, symptoms of claudication).

Lastly, a diagnosis of cellulitis is inconsistent with the physical findings of the PA, as well as those of the triage nurse and paramedics. This patient’s feet were cool to the touch whereas cellulitis presents with erythema and increased warmth. While the presence of pulses was somewhat reassuring, the cool temperature of the feet and complaint of pain were indicators of the need to evaluate for a possible arterial origin of these findings. However, if this were an embolic phenomenon, peripheral arterial ultrasound would have probably been normal, and the outcome unchanged.

Case 3: Failure to Communicate

A 59-year-old man presented to his primary care physician (PCP) for intermittent right-hand weakness and numbness and tingling in his right arm during the previous 24 hours. As the PCP was concerned that the patient might be experiencing a transient ischemic attack (TIA), the patient and his wife were instructed to go directly to the ED of the local hospital. The PCP wrote a note stating that the patient needed “a stroke work up,” gave the note to the patient, and told him to give it to the ED staff.

The patient went directly to the ED and was immediately seen by a “rapid triage nurse.” He gave the note to the nurse and told her of the PCP’s concerns. The nurse documented on the hospital assessment form that the patient was high priority and needed to be seen immediately. She attached the PCP’s note to the front of the form.

The patient was then seen by the traditional triage nurse. After evaluation, she changed the priority from high to low acuity. The triage nurse later stated later that she never saw the PCP’s note, nor did she obtain any history regarding the concerns of the PCP.

The patient was then evaluated by an EP in the low-acuity (or minor-care) area of the ED. The EP later stated that he never saw the note from the PCP, and had not received any information regarding a suspected TIA or stroke. The EP ordered a right wrist X-ray, diagnosed carpel tunnel syndrome, and prescribed an anti-inflammatory medication as well as follow-up with a hand surgeon.

The initial (rapid triage) nurse saw the patient leaving the ED at the time of discharge and thought he had not been in the ED long enough to have undergone a stroke work up. She reviewed his paperwork and saw the patient had not received the indicated work up. The nurse called the patient’s house and left a message on the answering machine notifying him of the need to return to the ED. The patient arrived back to the ED approximately 2 hours later.

On the second ED presentation that day, the patient was evaluated by a different EP. The patient had blood drawn, an electrocardiogram, and a noncontrast computed tomography scan of the brain, the results of which were all normal. The EP concluded the patient required admission to the hospital for additional work up (eg, carotid Doppler ultrasound). The hospitalist was paged, and told the EP he would be there “as soon as possible.” However, after several hours delay, and no hospitalist, the patient became impatient and expressed the desire to go home. The EP urged the patient to follow up with his PCP in the morning to complete the evaluation.

The next day, before seeing his PCP, the patient suffered an ischemic stroke with right-sided hemiparesis. The case went to trial, and the jury found in favor of the plaintiff.

Discussion

Unfortunately, multiple opportunities were lost in obtaining the correct care at the right time for this patient. Lack of communication and poor communication are frequently cited as causes in medical malpractice cases, and this case perfectly illustrates this problem.

First, the PCP should have called the ED and spoken to the EP directly. This would have provided the PCP the opportunity to express his concerns directly to the treating physician. This kind of one-on-one communication between physicians will always be superior to a hand-written note.

Second, it is unclear why the triage nurse changed the initial nurse’s correct assessment. It is also unclear what happened to the PCP’s note—it was never seen again. Clearly there was miscommunication at this point between the triage nurse and the patient. This case further illustrates the importance of good triage. Once a patient is directed down the wrong pathway (ie, to minor care rather than the main treatment area), the situation becomes much more difficult to correct.

Next, the EP in the low-acuity area was probably falsely assured this patient had only a “minor” problem problem, and not something serious. Emergency physicians must be vigilant to the possibility that the patient can have something seriously wrong even if he or she has been triaged to a low-acuity area. A minor sore throat can turn out to be epiglottitis and a viral stomachache can turn out to be appendicitis. These patients deserve the same quality of history, physical examination, and differential diagnosis as any other patient in the ED.

Finally, while we are not responsible for hospitalists or consultants, we do have a responsibility to our patients. We need to ensure that the care they receive is the appropriate care. Possible alternatives to discharging this patient would have been to call another hospitalist for admission or to seek the input of the chief of the medical staff or the on-call hospital administrator. As EPs, we are frequently required to serve as the primary advocate for our patients.

There is the possibility that even if the patient had been admitted to the hospital, the outcome would have been the same. However, since he might have been a candidate for tissue plasminogen activator or interventional radiology if he had suffered the cerebrovascular accident as an inpatient, he lost his best chance for a good outcome.

Case 1: Follow-up in Community-Acquired Methicillin- Resistant Staphylococcus Aureus Pharyngitis

A 73-year-old woman presented to the ED complaining of injuries she sustained following a minor fall at home. The patient’s past medical history was remarkable for diabetes mellitus (DM), hypertension, cerebrovascular accident, and a history of chronic pain. During the course of evaluation, the patient mentioned that she had experienced a sore throat. In addition to X-rays, the emergency physician (EP) ordered a rapid strep test (RST). All studies were normal and the patient was discharged home.

Over the next 2 months, the patient received medical treatment from other medical professionals, but not at the previous hospital. Reportedly, she did not complain of a sore throat during this time period. The patient did return to the same ED approximately 2 months later, complaining of cough and difficulty breathing; she died 5 days thereafter. No autopsy was performed.

A lawsuit was filed on behalf of the patient, stating the hospital breached the standard of care by not reporting the finding of MRSA directly to the treating physician, and that this led directly to the patient’s death by MRSA pneumonia. The jury returned a verdict in favor of the plaintiff for $32 million.

Discussion

This case illustrates two simple but important points: managing community-acquired MRSA (CA-MRSA) infections is a growing challenge for physicians; and hospital follow-up systems for positive findings (ie, fractures, blood cultures, etc) need to be consistently reliable.

It is estimated that 30% of healthy people carry S aureus in their anterior nares; colonization rates for the throat are much less studied. In one recent study, 265 throat swabs were collected from patients aged 14 to 65 years old, who complained of pharyngitis in an outpatient setting.1 A total of 165 S aureus isolates (62.3%) were recovered from the 265 swabs. For the S aureus isolates, 38.2% (63) grew CA-MRSA; the remaining 68.1% (102) were methicillin-sensitive S aureus (MSSA). Interestingly, of the 63 MRSA-positive swabs, over half also grew Group A Streptococcus. The natural disease progression of CA-MRSA pharyngitis is still unknown, as is what to do with a positive throat swab for CA-MRSA. While there are a few case reports of bacteremia and Lemmiere syndrome possibly related to CA-MRSA pharyngitis,2,3 more information is clearly needed. For this case, it is not possible to definitively determine the role of the positive throat swab for CAMRSA in the patient’s subsequent death.

The other teaching point in this case is much simpler and well defined. Simply put, patients expect to be informed of positive findings, whether the result is known at the time of their ED visit or sometime afterward. There needs to be a system in place that consistently and reliably provides important information to either the patient’s treating physician or to the patient. The manners in which this information is communicated are myriad and should take into account hospital resources, the role of the EP and nurse, and what works best for your locality. The “who” and “how” of the contact is not important: reliability, timeliness, and consistency need to be the key drivers of the system.

Case 2: Arterial Occlusion

A 61-year-old woman called emergency medical services (EMS) after noticing her feet felt cold to the touch and having difficulty ambulating. The paramedics noted the patient had normal vital signs and normal circulation in her legs, and she was transported to the ED without incident. Upon arrival to the ED, she was triaged as nonurgent and placed in the minor-care area. On nursing assessment, the patient’s feet were found to be cold, but with palpable pulses bilaterally. Her past medical history was significant for hypertension and DM.

The patient was seen by a physician assistant (PA), who found both feet cool to the touch, but with bilateral pulses present. She was administered intravenous morphine for pain and laboratory studies were ordered. At the time, the PA was concerned about arterial occlusion versus deep vein thrombosis (DVT) versus cellulitis. A venous ultrasound examination was ordered and shown to be negative for DVT. A complete blood count was remarkable only for mild leukocytosis. The PA discussed the case with the supervising EP; they agreed on a diagnosis of cellulitis and discharged the patient home with antibiotics and analgesics.

Approximately 12 hours after discharge, the patient presented back to the ED via ambulance. At that time, she was hypotensive and tachypneic, with a thready palpable pulse. On repeat examination, she no longer had pulses present in her feet. An arteriogram found complete occlusion of her arterial circulation at the level of the knees bilaterally, requiring bilateral belowthe- knee amputation.

The patient sued both the emergency medicine physician and the PA for failure to provide her with the necessary care during her initial ED visit, resulting in loss of limbs. The defendants claimed the patient could not prove gross negligence by clear and convincing evidence, as required by state law. Following the ensuing trial, the jury returned a $5 million verdict in favor of the plaintiff.

Discussion

First, it is important to remember that just because a patient has been triaged to a low-acuity area does not mean she or he must have a minor problem. The provider still must maintain a high level of vigilance— regardless of the location of the patient in the ED.

Second, was this patient in atrial fibrillation, which is responsible in approximately 65% of all peripheral emboli? The abrupt onset of this patient’s symptoms is much more compatible with an embolic origin of her symptoms rather than a thrombus (ie, symptoms of claudication).

Lastly, a diagnosis of cellulitis is inconsistent with the physical findings of the PA, as well as those of the triage nurse and paramedics. This patient’s feet were cool to the touch whereas cellulitis presents with erythema and increased warmth. While the presence of pulses was somewhat reassuring, the cool temperature of the feet and complaint of pain were indicators of the need to evaluate for a possible arterial origin of these findings. However, if this were an embolic phenomenon, peripheral arterial ultrasound would have probably been normal, and the outcome unchanged.

Case 3: Failure to Communicate

A 59-year-old man presented to his primary care physician (PCP) for intermittent right-hand weakness and numbness and tingling in his right arm during the previous 24 hours. As the PCP was concerned that the patient might be experiencing a transient ischemic attack (TIA), the patient and his wife were instructed to go directly to the ED of the local hospital. The PCP wrote a note stating that the patient needed “a stroke work up,” gave the note to the patient, and told him to give it to the ED staff.

The patient went directly to the ED and was immediately seen by a “rapid triage nurse.” He gave the note to the nurse and told her of the PCP’s concerns. The nurse documented on the hospital assessment form that the patient was high priority and needed to be seen immediately. She attached the PCP’s note to the front of the form.

The patient was then seen by the traditional triage nurse. After evaluation, she changed the priority from high to low acuity. The triage nurse later stated later that she never saw the PCP’s note, nor did she obtain any history regarding the concerns of the PCP.

The patient was then evaluated by an EP in the low-acuity (or minor-care) area of the ED. The EP later stated that he never saw the note from the PCP, and had not received any information regarding a suspected TIA or stroke. The EP ordered a right wrist X-ray, diagnosed carpel tunnel syndrome, and prescribed an anti-inflammatory medication as well as follow-up with a hand surgeon.

The initial (rapid triage) nurse saw the patient leaving the ED at the time of discharge and thought he had not been in the ED long enough to have undergone a stroke work up. She reviewed his paperwork and saw the patient had not received the indicated work up. The nurse called the patient’s house and left a message on the answering machine notifying him of the need to return to the ED. The patient arrived back to the ED approximately 2 hours later.

On the second ED presentation that day, the patient was evaluated by a different EP. The patient had blood drawn, an electrocardiogram, and a noncontrast computed tomography scan of the brain, the results of which were all normal. The EP concluded the patient required admission to the hospital for additional work up (eg, carotid Doppler ultrasound). The hospitalist was paged, and told the EP he would be there “as soon as possible.” However, after several hours delay, and no hospitalist, the patient became impatient and expressed the desire to go home. The EP urged the patient to follow up with his PCP in the morning to complete the evaluation.

The next day, before seeing his PCP, the patient suffered an ischemic stroke with right-sided hemiparesis. The case went to trial, and the jury found in favor of the plaintiff.

Discussion

Unfortunately, multiple opportunities were lost in obtaining the correct care at the right time for this patient. Lack of communication and poor communication are frequently cited as causes in medical malpractice cases, and this case perfectly illustrates this problem.

First, the PCP should have called the ED and spoken to the EP directly. This would have provided the PCP the opportunity to express his concerns directly to the treating physician. This kind of one-on-one communication between physicians will always be superior to a hand-written note.

Second, it is unclear why the triage nurse changed the initial nurse’s correct assessment. It is also unclear what happened to the PCP’s note—it was never seen again. Clearly there was miscommunication at this point between the triage nurse and the patient. This case further illustrates the importance of good triage. Once a patient is directed down the wrong pathway (ie, to minor care rather than the main treatment area), the situation becomes much more difficult to correct.

Next, the EP in the low-acuity area was probably falsely assured this patient had only a “minor” problem problem, and not something serious. Emergency physicians must be vigilant to the possibility that the patient can have something seriously wrong even if he or she has been triaged to a low-acuity area. A minor sore throat can turn out to be epiglottitis and a viral stomachache can turn out to be appendicitis. These patients deserve the same quality of history, physical examination, and differential diagnosis as any other patient in the ED.

Finally, while we are not responsible for hospitalists or consultants, we do have a responsibility to our patients. We need to ensure that the care they receive is the appropriate care. Possible alternatives to discharging this patient would have been to call another hospitalist for admission or to seek the input of the chief of the medical staff or the on-call hospital administrator. As EPs, we are frequently required to serve as the primary advocate for our patients.

There is the possibility that even if the patient had been admitted to the hospital, the outcome would have been the same. However, since he might have been a candidate for tissue plasminogen activator or interventional radiology if he had suffered the cerebrovascular accident as an inpatient, he lost his best chance for a good outcome.