Constantine A. Manthous, MD

Shock has been defined as failure to deliver and/or utilize adequate amounts of oxygen1 and is a common cause of critical illness. Few studies have examined the predictive utility of bedside clinical examination in predicting the category of shock. Scholars have suggested a bedside approach that uses simple examination techniques and applied physiology to rapidly identify a patients' circulation as high vs. low cardiac output. Those with a high‐output examination are designated as high‐output, most often septic shock. Low‐output patients are further categorized as heart full or heart empty to distinguish cardiogenic from hypovolemic categories of shock, respectively.2 The predictive characteristics of this simple algorithm have not been studied. In this study, we examine the operating characteristics of selected elements of this algorithm when administered at the bedside by trainees in Internal Medicine.

Methods

This study was performed after approval of the Institutional Review Board; informed consent was waived. Patients with nonsurgical problems who present to the hospital or who develop sustained hypotension are managed by medical house officers on the intensive care and/or rapid response team with the supervision of patients' attending physicians. All house officers were asked to document explicitly in their assessment notes the following examination findings: finger capillary refill (same/quicker vs. slower than examiner's), hand skin temperature (same/warmer vs. cooler than examiner's) and pulse pressure (ie, same/wider vs. thinner than examiner's), presence or absence of crackles >1/3 from base on bilateral lung examination and jugular venous pressure (JVP) vs. <8 cmH₂O. The documented examinations of either the rapid response team (PGY2; n = 14) or intensive care unit (ICU) resident (PGY3; n = 14) for patients evaluated between September 2008 and February 2009 were used for this study. Resuscitation was administered entirely by house officers, occasionally guided in person, but always supervised by attending physicians.

In May 2009, clinical data, including electrocardiograms/echocardiograms and laboratory (eg, cardiac enzymes, culture) results were abstracted from medical records of subjects. These were presented to a blinded senior clinician (DK) to review and apply evidence‐based or consensus criteria,36 whenever possible, to categorize the type of shock (septic vs. cardiogenic vs. hypovolemic) based on data acquired after the onset of shock. For example, patients with microbiologic and/or radiologic evidence of infection were classified as septic shock,1, 3, 4 those with acute left or right ventricular dysfunction on echocardiogram were classified as cardiogenic shock,1, 6 and those with clinical evidence of acute hemorrhage with hypovolemic shock.1, 5 While some of the patients were examined by DK as part of clinical care, he was blinded to the identity of patients and their algorithm‐related physical examination findings when he reviewed the abstracted data (>2 months after study closure) to adjudicate the final diagnosis of shock. These diagnoses were considered the reference standard for this study. The operating characteristics (sensitivity = true positive/true positive + false negative; specificity = true negative/true negative + false positive; negative predictive value (NPV) = true negative/all negatives; positive predictive value (PPV) = true positive/all positives; accuracy = true results/all results) were calculated for combinations of physical examination findings and correct final diagnosis (Figure 1).

Figure 1

Results

A total of 68 patients, averaging 71 16 years, were studied; 57% were male, and 66% were White, and 20% were Black. Table 1 lists characteristics of patients. A total of 37 patients were diagnosed as having septic shock, 11 had cardiogenic shock and 10 hypovolemic shock. Operating characteristics of the bedside examination techniques for predicting mechanism of shock are listed in Table 2. Capillary refill and skin temperature were 100% concordant yielding sensitivity of 89% (95% confidence interval [CI], 75‐97%), specificity of 68% (95% CI, 46‐83%), PPV of 77% (95% CI, 61‐88%), NPV of 84% (95% CI, 64‐96%) and overall accuracy of 79% (95% CI, 68‐88%) for diagnosis of high output (ie, septic shock). JVP 8 cmH₂O was more accurate than crackles for predicting cardiogenic shock in low‐output patients with sensitivity of 82% (95% CI, 48‐98%), specificity of 79% (95% CI, 41‐95%), PPV of 75% (95% CI, 43‐95%), NPV of 85% (95% CI, 55‐98%), and overall accuracy of 80% (95%CI, 59‐93%). Using just skin temperature and JVP, the bedside approach misdiagnosed 19 of 75 cases (overall accuracy, 75%; 95% CI, 16‐37%).

Discussion

This is the first study to examine the predictive characteristics of simple bedside physical examination techniques in correctly predicting the category/mechanism of shock. Overall, the algorithm performed well, and accurately predicted the category of shock in three‐quarters of patients. It also has the benefit of being very rapid, taking only seconds to complete, using bedside techniques that even inexperienced clinicians can apply.

Very few studies have examined the accuracy of examination techniques specifically for diagnosis of shock. In humans injected with endotoxin, body temperature and cardiac output increased, but skin temperature and capillary refill times are not well described.79 Schriger and Baraff10 reported that capillary refill >2 seconds was only 59% sensitive for diagnosing hypovolemia in patients with hypovolemic shock or orthostatic changes in blood pressure. Sensitivity was 77% in 13 patients with hypovolemic shock.10 However, some studies have demonstrated that age, sex, external temperature11 and fever12 can affect capillary refill times. Otieno et al.13 demonstrated a kappa statistic value of 0.49 for capillary refill 4 seconds, suggesting that reproducibility of this technique could be a major drawback. McGee et al.14 reviewed examination techniques for diagnosing hypovolemic states and concluded that postural changes in heart rate and blood pressure were the most accurate; capillary refill was not recommended. Stevenson and Perloff15 demonstrated that crackles and elevated JVP were absent in 18 of 43 patients with pulmonary capillary wedge pressures >22 mmHg. Butman et al.16 showed that elevated JVP was 82% accurate for predicting a wedge pressure >18 mmHg. Connors et al.17 demonstrated that clinicians' predictions of heart filling pressures and cardiac output were accurate (relative to pulmonary artery catheter measurements) in less than 50% of cases, though the examination techniques used were not qualified or quantified. No previous study has combined simple, semiobjective physical examination techniques for the purpose of distinguishing categories of shock.

Since identification of the pathogenesis of shock has important treatment/prognostic implications (eg, fluid and vasopressor therapies, early search for drainable focus of infection in sepsis, reestablishing vessel patency in myocardial infarction and pulmonary embolus), we believe that this simple, rapidly administered algorithm will prove useful in clinical medicine. In some clinical situations, the approach can lead to timely identification of the causative mechanism, allowing prompt definitive treatment. For example, a patient presenting with high‐output hypotension is so often sepsis/septic shock that treatment with antibiotics is justified (since success is time‐sensitive) even when the exact site/microbe has not yet been identified. Acute right heart overfilled low‐output hypotension should be considered pulmonary embolism (which also requires time‐sensitive therapies) until proven otherwise. Yet, a sizeable number of cases do not fit neatly into a single category. For example, 11% of patients with septic shock presented with cool extremities in the early phases of illness. In clinical decision‐making, 2 diagnostic‐therapeutic paradigms are common. In the first, the diagnosis is relatively certain and narrowly‐directed, mechanism‐specific treatment is appropriate. The second paradigm is 1 of significant uncertainty, when clinicians must treat empirically the most likely causes until more data become available to permit safe narrowing of therapies. For example, a patient presenting with hypotension, cool extremities, leukocytosis and apparent pneumonia should be treated empirically for septic shock while exploring explanations for the incongruous low‐output state (eg, profound hypovolemia, adrenal insufficiency, concurrent or precedent myocardial dysfunction). Patients often have several mechanisms contributing to hypotension. Since patients are not ideal forms, there can be no perfect decision‐tool; clinicians would be fool‐hardy to prematurely close decision‐making prior to definitive diagnosis. In the case of shock, such diagnostic arrogance would delay time‐sensitive therapies and thus contribute to morbidity and mortality. Nonetheless, this physical examination algorithmunderstanding its operating characteristics and limitationsmay add to the bedside clinician's diagnostic armamentarium.

Our study has several notable limitations. First, bedside examinations were performed by multiple observers who had limited (1 electronic mail) instruction on how to perform and document the data gathered for this study. So these results should be generalized cautiously until reproduced at other centers with greater numbers of observers (than the 28 of this study). The central supposition, that skin cooler, capillary refill longer, and pulse pressure more narrow than theirs, presupposes reasonable homogeneity of the normal state which is not necessarily true.11 Interobserver variability of physical examination further compromises the fidelity of findings recorded for this study.13 Since we conducted a retrospective review, and because of the emergency nature of the clinical problem, it would be difficult to conduct a study in which multiple examiners performed the same physical examinations to quantify interobserver variability. Irrespective, we would expect interobserver variability to systematically reduce accuracy; it is all‐the‐more impressive that trainees' examination results correctly diagnosed mechanism of shock in three‐quarters of cases. Also, examiners were not blinded to clinical history, so results of their examination could have been biased by their pre‐examination hypotheses of pathogenesis. Of course, they were not aware of the expert's final categorization of mechanism performed much later in time. Since there is no absolute reference standard for classification of the pathogenesis of shock, we depended upon careful review of selected data (same parameters for each patient) by a single senior investigatoralbeit armed with evidence‐based or consensus‐based standards of diagnosing shock. Finally, it can be argued that all forms of shock are mixed (with hypovolemia) early in the course; sepsis requires refilling of a leaky and dilated vasculature and the noncompliant ischemic ventricle often requires a higher filling pressure than normal to empty. To complicate even more, patients may have preexistent conditions (eg, chronic congestive heart failure, cirrhosis) that limit cardiovascular responses to acute shock. Our diagnostic approach was to identify the principal cause of the acute decompensation, assuming that many patients will have more than 1 single mechanism accounting for hypotension.

In conclusion, this is the first study to examine the utility of this simple physical examination algorithm to diagnose the mechanism of shock. Some have discounted or underemphasized examination techniques in favor of more time‐intensive and labor‐intensive diagnostic modalities, such as bedside echocardiography, which may waste precious time and resources. The simple physical examination algorithm assessed in this study has favorable operating characteristics and can be performed readily by even novice clinicians. If replicated at other centers and by greater numbers of observers, this approach could assist clinicians and teachers who train clinicians to rapidly diagnose and manage patients with shock.

Shock has been defined as failure to deliver and/or utilize adequate amounts of oxygen1 and is a common cause of critical illness. Few studies have examined the predictive utility of bedside clinical examination in predicting the category of shock. Scholars have suggested a bedside approach that uses simple examination techniques and applied physiology to rapidly identify a patients' circulation as high vs. low cardiac output. Those with a high‐output examination are designated as high‐output, most often septic shock. Low‐output patients are further categorized as heart full or heart empty to distinguish cardiogenic from hypovolemic categories of shock, respectively.2 The predictive characteristics of this simple algorithm have not been studied. In this study, we examine the operating characteristics of selected elements of this algorithm when administered at the bedside by trainees in Internal Medicine.

Methods

This study was performed after approval of the Institutional Review Board; informed consent was waived. Patients with nonsurgical problems who present to the hospital or who develop sustained hypotension are managed by medical house officers on the intensive care and/or rapid response team with the supervision of patients' attending physicians. All house officers were asked to document explicitly in their assessment notes the following examination findings: finger capillary refill (same/quicker vs. slower than examiner's), hand skin temperature (same/warmer vs. cooler than examiner's) and pulse pressure (ie, same/wider vs. thinner than examiner's), presence or absence of crackles >1/3 from base on bilateral lung examination and jugular venous pressure (JVP) vs. <8 cmH₂O. The documented examinations of either the rapid response team (PGY2; n = 14) or intensive care unit (ICU) resident (PGY3; n = 14) for patients evaluated between September 2008 and February 2009 were used for this study. Resuscitation was administered entirely by house officers, occasionally guided in person, but always supervised by attending physicians.

In May 2009, clinical data, including electrocardiograms/echocardiograms and laboratory (eg, cardiac enzymes, culture) results were abstracted from medical records of subjects. These were presented to a blinded senior clinician (DK) to review and apply evidence‐based or consensus criteria,36 whenever possible, to categorize the type of shock (septic vs. cardiogenic vs. hypovolemic) based on data acquired after the onset of shock. For example, patients with microbiologic and/or radiologic evidence of infection were classified as septic shock,1, 3, 4 those with acute left or right ventricular dysfunction on echocardiogram were classified as cardiogenic shock,1, 6 and those with clinical evidence of acute hemorrhage with hypovolemic shock.1, 5 While some of the patients were examined by DK as part of clinical care, he was blinded to the identity of patients and their algorithm‐related physical examination findings when he reviewed the abstracted data (>2 months after study closure) to adjudicate the final diagnosis of shock. These diagnoses were considered the reference standard for this study. The operating characteristics (sensitivity = true positive/true positive + false negative; specificity = true negative/true negative + false positive; negative predictive value (NPV) = true negative/all negatives; positive predictive value (PPV) = true positive/all positives; accuracy = true results/all results) were calculated for combinations of physical examination findings and correct final diagnosis (Figure 1).

Figure 1

Results

A total of 68 patients, averaging 71 16 years, were studied; 57% were male, and 66% were White, and 20% were Black. Table 1 lists characteristics of patients. A total of 37 patients were diagnosed as having septic shock, 11 had cardiogenic shock and 10 hypovolemic shock. Operating characteristics of the bedside examination techniques for predicting mechanism of shock are listed in Table 2. Capillary refill and skin temperature were 100% concordant yielding sensitivity of 89% (95% confidence interval [CI], 75‐97%), specificity of 68% (95% CI, 46‐83%), PPV of 77% (95% CI, 61‐88%), NPV of 84% (95% CI, 64‐96%) and overall accuracy of 79% (95% CI, 68‐88%) for diagnosis of high output (ie, septic shock). JVP 8 cmH₂O was more accurate than crackles for predicting cardiogenic shock in low‐output patients with sensitivity of 82% (95% CI, 48‐98%), specificity of 79% (95% CI, 41‐95%), PPV of 75% (95% CI, 43‐95%), NPV of 85% (95% CI, 55‐98%), and overall accuracy of 80% (95%CI, 59‐93%). Using just skin temperature and JVP, the bedside approach misdiagnosed 19 of 75 cases (overall accuracy, 75%; 95% CI, 16‐37%).

Discussion

This is the first study to examine the predictive characteristics of simple bedside physical examination techniques in correctly predicting the category/mechanism of shock. Overall, the algorithm performed well, and accurately predicted the category of shock in three‐quarters of patients. It also has the benefit of being very rapid, taking only seconds to complete, using bedside techniques that even inexperienced clinicians can apply.

Very few studies have examined the accuracy of examination techniques specifically for diagnosis of shock. In humans injected with endotoxin, body temperature and cardiac output increased, but skin temperature and capillary refill times are not well described.79 Schriger and Baraff10 reported that capillary refill >2 seconds was only 59% sensitive for diagnosing hypovolemia in patients with hypovolemic shock or orthostatic changes in blood pressure. Sensitivity was 77% in 13 patients with hypovolemic shock.10 However, some studies have demonstrated that age, sex, external temperature11 and fever12 can affect capillary refill times. Otieno et al.13 demonstrated a kappa statistic value of 0.49 for capillary refill 4 seconds, suggesting that reproducibility of this technique could be a major drawback. McGee et al.14 reviewed examination techniques for diagnosing hypovolemic states and concluded that postural changes in heart rate and blood pressure were the most accurate; capillary refill was not recommended. Stevenson and Perloff15 demonstrated that crackles and elevated JVP were absent in 18 of 43 patients with pulmonary capillary wedge pressures >22 mmHg. Butman et al.16 showed that elevated JVP was 82% accurate for predicting a wedge pressure >18 mmHg. Connors et al.17 demonstrated that clinicians' predictions of heart filling pressures and cardiac output were accurate (relative to pulmonary artery catheter measurements) in less than 50% of cases, though the examination techniques used were not qualified or quantified. No previous study has combined simple, semiobjective physical examination techniques for the purpose of distinguishing categories of shock.

Since identification of the pathogenesis of shock has important treatment/prognostic implications (eg, fluid and vasopressor therapies, early search for drainable focus of infection in sepsis, reestablishing vessel patency in myocardial infarction and pulmonary embolus), we believe that this simple, rapidly administered algorithm will prove useful in clinical medicine. In some clinical situations, the approach can lead to timely identification of the causative mechanism, allowing prompt definitive treatment. For example, a patient presenting with high‐output hypotension is so often sepsis/septic shock that treatment with antibiotics is justified (since success is time‐sensitive) even when the exact site/microbe has not yet been identified. Acute right heart overfilled low‐output hypotension should be considered pulmonary embolism (which also requires time‐sensitive therapies) until proven otherwise. Yet, a sizeable number of cases do not fit neatly into a single category. For example, 11% of patients with septic shock presented with cool extremities in the early phases of illness. In clinical decision‐making, 2 diagnostic‐therapeutic paradigms are common. In the first, the diagnosis is relatively certain and narrowly‐directed, mechanism‐specific treatment is appropriate. The second paradigm is 1 of significant uncertainty, when clinicians must treat empirically the most likely causes until more data become available to permit safe narrowing of therapies. For example, a patient presenting with hypotension, cool extremities, leukocytosis and apparent pneumonia should be treated empirically for septic shock while exploring explanations for the incongruous low‐output state (eg, profound hypovolemia, adrenal insufficiency, concurrent or precedent myocardial dysfunction). Patients often have several mechanisms contributing to hypotension. Since patients are not ideal forms, there can be no perfect decision‐tool; clinicians would be fool‐hardy to prematurely close decision‐making prior to definitive diagnosis. In the case of shock, such diagnostic arrogance would delay time‐sensitive therapies and thus contribute to morbidity and mortality. Nonetheless, this physical examination algorithmunderstanding its operating characteristics and limitationsmay add to the bedside clinician's diagnostic armamentarium.

Our study has several notable limitations. First, bedside examinations were performed by multiple observers who had limited (1 electronic mail) instruction on how to perform and document the data gathered for this study. So these results should be generalized cautiously until reproduced at other centers with greater numbers of observers (than the 28 of this study). The central supposition, that skin cooler, capillary refill longer, and pulse pressure more narrow than theirs, presupposes reasonable homogeneity of the normal state which is not necessarily true.11 Interobserver variability of physical examination further compromises the fidelity of findings recorded for this study.13 Since we conducted a retrospective review, and because of the emergency nature of the clinical problem, it would be difficult to conduct a study in which multiple examiners performed the same physical examinations to quantify interobserver variability. Irrespective, we would expect interobserver variability to systematically reduce accuracy; it is all‐the‐more impressive that trainees' examination results correctly diagnosed mechanism of shock in three‐quarters of cases. Also, examiners were not blinded to clinical history, so results of their examination could have been biased by their pre‐examination hypotheses of pathogenesis. Of course, they were not aware of the expert's final categorization of mechanism performed much later in time. Since there is no absolute reference standard for classification of the pathogenesis of shock, we depended upon careful review of selected data (same parameters for each patient) by a single senior investigatoralbeit armed with evidence‐based or consensus‐based standards of diagnosing shock. Finally, it can be argued that all forms of shock are mixed (with hypovolemia) early in the course; sepsis requires refilling of a leaky and dilated vasculature and the noncompliant ischemic ventricle often requires a higher filling pressure than normal to empty. To complicate even more, patients may have preexistent conditions (eg, chronic congestive heart failure, cirrhosis) that limit cardiovascular responses to acute shock. Our diagnostic approach was to identify the principal cause of the acute decompensation, assuming that many patients will have more than 1 single mechanism accounting for hypotension.

In conclusion, this is the first study to examine the utility of this simple physical examination algorithm to diagnose the mechanism of shock. Some have discounted or underemphasized examination techniques in favor of more time‐intensive and labor‐intensive diagnostic modalities, such as bedside echocardiography, which may waste precious time and resources. The simple physical examination algorithm assessed in this study has favorable operating characteristics and can be performed readily by even novice clinicians. If replicated at other centers and by greater numbers of observers, this approach could assist clinicians and teachers who train clinicians to rapidly diagnose and manage patients with shock.

Shock has been defined as failure to deliver and/or utilize adequate amounts of oxygen1 and is a common cause of critical illness. Few studies have examined the predictive utility of bedside clinical examination in predicting the category of shock. Scholars have suggested a bedside approach that uses simple examination techniques and applied physiology to rapidly identify a patients' circulation as high vs. low cardiac output. Those with a high‐output examination are designated as high‐output, most often septic shock. Low‐output patients are further categorized as heart full or heart empty to distinguish cardiogenic from hypovolemic categories of shock, respectively.2 The predictive characteristics of this simple algorithm have not been studied. In this study, we examine the operating characteristics of selected elements of this algorithm when administered at the bedside by trainees in Internal Medicine.

Methods

This study was performed after approval of the Institutional Review Board; informed consent was waived. Patients with nonsurgical problems who present to the hospital or who develop sustained hypotension are managed by medical house officers on the intensive care and/or rapid response team with the supervision of patients' attending physicians. All house officers were asked to document explicitly in their assessment notes the following examination findings: finger capillary refill (same/quicker vs. slower than examiner's), hand skin temperature (same/warmer vs. cooler than examiner's) and pulse pressure (ie, same/wider vs. thinner than examiner's), presence or absence of crackles >1/3 from base on bilateral lung examination and jugular venous pressure (JVP) vs. <8 cmH₂O. The documented examinations of either the rapid response team (PGY2; n = 14) or intensive care unit (ICU) resident (PGY3; n = 14) for patients evaluated between September 2008 and February 2009 were used for this study. Resuscitation was administered entirely by house officers, occasionally guided in person, but always supervised by attending physicians.

In May 2009, clinical data, including electrocardiograms/echocardiograms and laboratory (eg, cardiac enzymes, culture) results were abstracted from medical records of subjects. These were presented to a blinded senior clinician (DK) to review and apply evidence‐based or consensus criteria,36 whenever possible, to categorize the type of shock (septic vs. cardiogenic vs. hypovolemic) based on data acquired after the onset of shock. For example, patients with microbiologic and/or radiologic evidence of infection were classified as septic shock,1, 3, 4 those with acute left or right ventricular dysfunction on echocardiogram were classified as cardiogenic shock,1, 6 and those with clinical evidence of acute hemorrhage with hypovolemic shock.1, 5 While some of the patients were examined by DK as part of clinical care, he was blinded to the identity of patients and their algorithm‐related physical examination findings when he reviewed the abstracted data (>2 months after study closure) to adjudicate the final diagnosis of shock. These diagnoses were considered the reference standard for this study. The operating characteristics (sensitivity = true positive/true positive + false negative; specificity = true negative/true negative + false positive; negative predictive value (NPV) = true negative/all negatives; positive predictive value (PPV) = true positive/all positives; accuracy = true results/all results) were calculated for combinations of physical examination findings and correct final diagnosis (Figure 1).

Figure 1

Results

A total of 68 patients, averaging 71 16 years, were studied; 57% were male, and 66% were White, and 20% were Black. Table 1 lists characteristics of patients. A total of 37 patients were diagnosed as having septic shock, 11 had cardiogenic shock and 10 hypovolemic shock. Operating characteristics of the bedside examination techniques for predicting mechanism of shock are listed in Table 2. Capillary refill and skin temperature were 100% concordant yielding sensitivity of 89% (95% confidence interval [CI], 75‐97%), specificity of 68% (95% CI, 46‐83%), PPV of 77% (95% CI, 61‐88%), NPV of 84% (95% CI, 64‐96%) and overall accuracy of 79% (95% CI, 68‐88%) for diagnosis of high output (ie, septic shock). JVP 8 cmH₂O was more accurate than crackles for predicting cardiogenic shock in low‐output patients with sensitivity of 82% (95% CI, 48‐98%), specificity of 79% (95% CI, 41‐95%), PPV of 75% (95% CI, 43‐95%), NPV of 85% (95% CI, 55‐98%), and overall accuracy of 80% (95%CI, 59‐93%). Using just skin temperature and JVP, the bedside approach misdiagnosed 19 of 75 cases (overall accuracy, 75%; 95% CI, 16‐37%).

Discussion

This is the first study to examine the predictive characteristics of simple bedside physical examination techniques in correctly predicting the category/mechanism of shock. Overall, the algorithm performed well, and accurately predicted the category of shock in three‐quarters of patients. It also has the benefit of being very rapid, taking only seconds to complete, using bedside techniques that even inexperienced clinicians can apply.

Very few studies have examined the accuracy of examination techniques specifically for diagnosis of shock. In humans injected with endotoxin, body temperature and cardiac output increased, but skin temperature and capillary refill times are not well described.79 Schriger and Baraff10 reported that capillary refill >2 seconds was only 59% sensitive for diagnosing hypovolemia in patients with hypovolemic shock or orthostatic changes in blood pressure. Sensitivity was 77% in 13 patients with hypovolemic shock.10 However, some studies have demonstrated that age, sex, external temperature11 and fever12 can affect capillary refill times. Otieno et al.13 demonstrated a kappa statistic value of 0.49 for capillary refill 4 seconds, suggesting that reproducibility of this technique could be a major drawback. McGee et al.14 reviewed examination techniques for diagnosing hypovolemic states and concluded that postural changes in heart rate and blood pressure were the most accurate; capillary refill was not recommended. Stevenson and Perloff15 demonstrated that crackles and elevated JVP were absent in 18 of 43 patients with pulmonary capillary wedge pressures >22 mmHg. Butman et al.16 showed that elevated JVP was 82% accurate for predicting a wedge pressure >18 mmHg. Connors et al.17 demonstrated that clinicians' predictions of heart filling pressures and cardiac output were accurate (relative to pulmonary artery catheter measurements) in less than 50% of cases, though the examination techniques used were not qualified or quantified. No previous study has combined simple, semiobjective physical examination techniques for the purpose of distinguishing categories of shock.

Since identification of the pathogenesis of shock has important treatment/prognostic implications (eg, fluid and vasopressor therapies, early search for drainable focus of infection in sepsis, reestablishing vessel patency in myocardial infarction and pulmonary embolus), we believe that this simple, rapidly administered algorithm will prove useful in clinical medicine. In some clinical situations, the approach can lead to timely identification of the causative mechanism, allowing prompt definitive treatment. For example, a patient presenting with high‐output hypotension is so often sepsis/septic shock that treatment with antibiotics is justified (since success is time‐sensitive) even when the exact site/microbe has not yet been identified. Acute right heart overfilled low‐output hypotension should be considered pulmonary embolism (which also requires time‐sensitive therapies) until proven otherwise. Yet, a sizeable number of cases do not fit neatly into a single category. For example, 11% of patients with septic shock presented with cool extremities in the early phases of illness. In clinical decision‐making, 2 diagnostic‐therapeutic paradigms are common. In the first, the diagnosis is relatively certain and narrowly‐directed, mechanism‐specific treatment is appropriate. The second paradigm is 1 of significant uncertainty, when clinicians must treat empirically the most likely causes until more data become available to permit safe narrowing of therapies. For example, a patient presenting with hypotension, cool extremities, leukocytosis and apparent pneumonia should be treated empirically for septic shock while exploring explanations for the incongruous low‐output state (eg, profound hypovolemia, adrenal insufficiency, concurrent or precedent myocardial dysfunction). Patients often have several mechanisms contributing to hypotension. Since patients are not ideal forms, there can be no perfect decision‐tool; clinicians would be fool‐hardy to prematurely close decision‐making prior to definitive diagnosis. In the case of shock, such diagnostic arrogance would delay time‐sensitive therapies and thus contribute to morbidity and mortality. Nonetheless, this physical examination algorithmunderstanding its operating characteristics and limitationsmay add to the bedside clinician's diagnostic armamentarium.

Our study has several notable limitations. First, bedside examinations were performed by multiple observers who had limited (1 electronic mail) instruction on how to perform and document the data gathered for this study. So these results should be generalized cautiously until reproduced at other centers with greater numbers of observers (than the 28 of this study). The central supposition, that skin cooler, capillary refill longer, and pulse pressure more narrow than theirs, presupposes reasonable homogeneity of the normal state which is not necessarily true.11 Interobserver variability of physical examination further compromises the fidelity of findings recorded for this study.13 Since we conducted a retrospective review, and because of the emergency nature of the clinical problem, it would be difficult to conduct a study in which multiple examiners performed the same physical examinations to quantify interobserver variability. Irrespective, we would expect interobserver variability to systematically reduce accuracy; it is all‐the‐more impressive that trainees' examination results correctly diagnosed mechanism of shock in three‐quarters of cases. Also, examiners were not blinded to clinical history, so results of their examination could have been biased by their pre‐examination hypotheses of pathogenesis. Of course, they were not aware of the expert's final categorization of mechanism performed much later in time. Since there is no absolute reference standard for classification of the pathogenesis of shock, we depended upon careful review of selected data (same parameters for each patient) by a single senior investigatoralbeit armed with evidence‐based or consensus‐based standards of diagnosing shock. Finally, it can be argued that all forms of shock are mixed (with hypovolemia) early in the course; sepsis requires refilling of a leaky and dilated vasculature and the noncompliant ischemic ventricle often requires a higher filling pressure than normal to empty. To complicate even more, patients may have preexistent conditions (eg, chronic congestive heart failure, cirrhosis) that limit cardiovascular responses to acute shock. Our diagnostic approach was to identify the principal cause of the acute decompensation, assuming that many patients will have more than 1 single mechanism accounting for hypotension.

In conclusion, this is the first study to examine the utility of this simple physical examination algorithm to diagnose the mechanism of shock. Some have discounted or underemphasized examination techniques in favor of more time‐intensive and labor‐intensive diagnostic modalities, such as bedside echocardiography, which may waste precious time and resources. The simple physical examination algorithm assessed in this study has favorable operating characteristics and can be performed readily by even novice clinicians. If replicated at other centers and by greater numbers of observers, this approach could assist clinicians and teachers who train clinicians to rapidly diagnose and manage patients with shock.

In 2007, the Joint Commission for Accreditation of Healthcare Organizations (JCAHO) recommended deployment of rapid response teams (RRTs) in U.S. hospitals to hasten identification and treatment of physiologically unstable hospitalized patients.1 Clinical studies that have focused on whether RRTs improve restorative care outcomes, frequency of cardiac arrest, and critical care utilization have yielded mixed results.2‐11 One study suggested that RRTs might provide an opportunity to enhance palliative care of hospitalized patients.11 In this study, RRT personnel felt that prior do‐not‐resuscitate orders would have been appropriate in nearly a quarter of cases. However, no previous study has examined whether the RRT might be deployed to identify acutely decompensating patients who either do not want or would not benefit from a trial of aggressive restorative treatments. We hypothesized that actuation of an RRT in our hospital would expedite identification of patients not likely to benefit from restorative care and would promote more timely commencement of end‐of‐life comfort care, thereby improving their quality of death (QOD).12‐16

Materials and Methods

Results

A total of 394 patients died on the hospital wards and were not admitted with palliative, end‐of‐life medical therapies. The combined (pre‐RRT and post‐RRT epochs) cohort had a mean age of 77.2 13.2 years. A total of 48% were male, 79% White, 12% Black, and 8% Hispanic. A total of 128 patients (33%) were admitted to the hospital from a skilled nursing facility and 135 (35%) had written advance directives.

A total of 197 patients met the inclusion criteria during the pre‐RRT (October 1, 2005 to May 31, 2006) and 197 during the post‐RRT epochs (October 1, 2006 to May 31, 2007). There were no differences in age, sex, advance directives, ethnicity, or religion between the groups (Table 1). Primary admission diagnoses were significantly different; pre‐RRT patients were 9% more likely to die with malignancy compared to post‐RRT patients and less likely to come from nursing homes (38% vs. 27%; P = 0.02).

Discussion

This pilot study hypothesizes that our RRT impacted patients' QOD. Deployment of the RRT in our hospital was associated with improvement in both symptom and psychospiritual domains of care. Theoretically, RRTs should improve quality‐of‐care via early identification/reversal of physiologic decompensation. By either reversing acute diatheses with an expeditious trial of therapy or failing to reverse with early actuation of palliative therapies, the duration and magnitude of human suffering should be reduced. Attenuation of both duration and magnitude of suffering is the ultimate goal of both restorative and palliative care and is as important an outcome as mortality or length of stay. Previous studies of RRTs have focused on efficacy in reversing the decompensation: preventing cardiopulmonary arrest, avoiding the need for invasive, expensive, labor‐intensive interventions. Our RRT, like others, had no demonstrable impact on restorative outcomes. However, deployment of the RRT was highly associated with improved QOD of our patients. The impact was significant across WHO‐specified domains: pain scores decreased by 19%; (documentation of) patients' distress decreased by 50%; and chaplains' visits were more often documented in the 24 hours prior to death. These relationships held across common disease diagnoses, so the association is unlikely to be spurious.

Outcomes were similarly improved in patients who did not receive RRT care in the post‐RRT epoch. Our hospital did not have a palliative care service in either time period. No new educational efforts among physicians or nurses accounted for this observation. While it is possible that temporal effects accounted for our observation, an equally plausible explanation is that staff observed RRT interventions and applied them to dying patients not seen by the RRT. Our hospital educated caregivers regarding the RRT triggers, and simply making hospital personnel more vigilant for signs of suffering and/or observing the RRT approach may have contributed to enhanced end‐of‐life care for non‐RRT patients.

There are a number of limitations in this study. First, the sample size was relatively small compared to other published studies,2‐11 promoting the possibility that either epoch was not representative of pre‐RRT and post‐RRT parent populations. Another weakness is that QOD was measured using surrogate endpoints. The dead cannot be interviewed to definitively examine QOD; indices of cardiopulmonary distress and psychosocial measures (eg, religious preparations, family involvement) are endpoints suggested by palliative care investigators12, 13 and the World Health Organization.14 While some validated tools17 and consensus measures18 exist for critically ill patients, they do not readily apply to RRT patients. Retrospective records reviews raise the possibility of bias in extracting objective and subjective data. While we attempted to control for this by creating uniform a priori rules for data acquisition (ie, at what intervals and in which parts of the record they could be extracted), we cannot discount the possibility that bias affected the observed results. Finally, improvements in end‐of‐life care could have resulted from temporal trends. This retrospective study cannot prove a causeeffect relationship; a prospective randomized trial would be required to answer the question definitively. Based on the available data suggesting some benefit in restorative outcomes2‐8 and pressure from federal regulators to deploy RRTs regardless,1 a retrospective cohort design may provide the only realistic means of addressing this question.

In conclusion, this is the first (pilot) study to examine end‐of‐life outcomes associated with deployment of an RRT. While the limitations of these observations preclude firm conclusions, the plausibility of the hypothesis, coupled with our observations, suggests that this is a fertile area for future research. While RRTs may enhance restorative outcomes, to the extent that they hasten identification of candidates for palliative end‐of‐life‐care, before administration of invasive modalities that some patients do not want, these teams may simultaneously serve patients and reduce hospital resource utilization.

Addendum

Prior to publication, a contemporaneous study was published that concluded: These findings suggest that rapid response teams may not be decreasing code rates as much as catalyzing a compassionate dialogue of end‐of‐life care among terminally ill patients. This ability to improve end‐of‐life care may be an important benefit of rapid response teams, particularly given the difficulties in prior trials to increase rates of DNR status among seriously ill inpatients and potential decreases in resource use. Chan PS, Khalid A, Longmore LS, Berg RA, Midhail Kosiborod M, Spertus JA. Hospital‐wide code rates and mortality before and after implementation of a rapid response team. JAMA 2008;300: 25062513.

In 2007, the Joint Commission for Accreditation of Healthcare Organizations (JCAHO) recommended deployment of rapid response teams (RRTs) in U.S. hospitals to hasten identification and treatment of physiologically unstable hospitalized patients.1 Clinical studies that have focused on whether RRTs improve restorative care outcomes, frequency of cardiac arrest, and critical care utilization have yielded mixed results.2‐11 One study suggested that RRTs might provide an opportunity to enhance palliative care of hospitalized patients.11 In this study, RRT personnel felt that prior do‐not‐resuscitate orders would have been appropriate in nearly a quarter of cases. However, no previous study has examined whether the RRT might be deployed to identify acutely decompensating patients who either do not want or would not benefit from a trial of aggressive restorative treatments. We hypothesized that actuation of an RRT in our hospital would expedite identification of patients not likely to benefit from restorative care and would promote more timely commencement of end‐of‐life comfort care, thereby improving their quality of death (QOD).12‐16

Materials and Methods

Results

A total of 394 patients died on the hospital wards and were not admitted with palliative, end‐of‐life medical therapies. The combined (pre‐RRT and post‐RRT epochs) cohort had a mean age of 77.2 13.2 years. A total of 48% were male, 79% White, 12% Black, and 8% Hispanic. A total of 128 patients (33%) were admitted to the hospital from a skilled nursing facility and 135 (35%) had written advance directives.

A total of 197 patients met the inclusion criteria during the pre‐RRT (October 1, 2005 to May 31, 2006) and 197 during the post‐RRT epochs (October 1, 2006 to May 31, 2007). There were no differences in age, sex, advance directives, ethnicity, or religion between the groups (Table 1). Primary admission diagnoses were significantly different; pre‐RRT patients were 9% more likely to die with malignancy compared to post‐RRT patients and less likely to come from nursing homes (38% vs. 27%; P = 0.02).

Discussion

This pilot study hypothesizes that our RRT impacted patients' QOD. Deployment of the RRT in our hospital was associated with improvement in both symptom and psychospiritual domains of care. Theoretically, RRTs should improve quality‐of‐care via early identification/reversal of physiologic decompensation. By either reversing acute diatheses with an expeditious trial of therapy or failing to reverse with early actuation of palliative therapies, the duration and magnitude of human suffering should be reduced. Attenuation of both duration and magnitude of suffering is the ultimate goal of both restorative and palliative care and is as important an outcome as mortality or length of stay. Previous studies of RRTs have focused on efficacy in reversing the decompensation: preventing cardiopulmonary arrest, avoiding the need for invasive, expensive, labor‐intensive interventions. Our RRT, like others, had no demonstrable impact on restorative outcomes. However, deployment of the RRT was highly associated with improved QOD of our patients. The impact was significant across WHO‐specified domains: pain scores decreased by 19%; (documentation of) patients' distress decreased by 50%; and chaplains' visits were more often documented in the 24 hours prior to death. These relationships held across common disease diagnoses, so the association is unlikely to be spurious.

Outcomes were similarly improved in patients who did not receive RRT care in the post‐RRT epoch. Our hospital did not have a palliative care service in either time period. No new educational efforts among physicians or nurses accounted for this observation. While it is possible that temporal effects accounted for our observation, an equally plausible explanation is that staff observed RRT interventions and applied them to dying patients not seen by the RRT. Our hospital educated caregivers regarding the RRT triggers, and simply making hospital personnel more vigilant for signs of suffering and/or observing the RRT approach may have contributed to enhanced end‐of‐life care for non‐RRT patients.

There are a number of limitations in this study. First, the sample size was relatively small compared to other published studies,2‐11 promoting the possibility that either epoch was not representative of pre‐RRT and post‐RRT parent populations. Another weakness is that QOD was measured using surrogate endpoints. The dead cannot be interviewed to definitively examine QOD; indices of cardiopulmonary distress and psychosocial measures (eg, religious preparations, family involvement) are endpoints suggested by palliative care investigators12, 13 and the World Health Organization.14 While some validated tools17 and consensus measures18 exist for critically ill patients, they do not readily apply to RRT patients. Retrospective records reviews raise the possibility of bias in extracting objective and subjective data. While we attempted to control for this by creating uniform a priori rules for data acquisition (ie, at what intervals and in which parts of the record they could be extracted), we cannot discount the possibility that bias affected the observed results. Finally, improvements in end‐of‐life care could have resulted from temporal trends. This retrospective study cannot prove a causeeffect relationship; a prospective randomized trial would be required to answer the question definitively. Based on the available data suggesting some benefit in restorative outcomes2‐8 and pressure from federal regulators to deploy RRTs regardless,1 a retrospective cohort design may provide the only realistic means of addressing this question.

In conclusion, this is the first (pilot) study to examine end‐of‐life outcomes associated with deployment of an RRT. While the limitations of these observations preclude firm conclusions, the plausibility of the hypothesis, coupled with our observations, suggests that this is a fertile area for future research. While RRTs may enhance restorative outcomes, to the extent that they hasten identification of candidates for palliative end‐of‐life‐care, before administration of invasive modalities that some patients do not want, these teams may simultaneously serve patients and reduce hospital resource utilization.

Addendum

Prior to publication, a contemporaneous study was published that concluded: These findings suggest that rapid response teams may not be decreasing code rates as much as catalyzing a compassionate dialogue of end‐of‐life care among terminally ill patients. This ability to improve end‐of‐life care may be an important benefit of rapid response teams, particularly given the difficulties in prior trials to increase rates of DNR status among seriously ill inpatients and potential decreases in resource use. Chan PS, Khalid A, Longmore LS, Berg RA, Midhail Kosiborod M, Spertus JA. Hospital‐wide code rates and mortality before and after implementation of a rapid response team. JAMA 2008;300: 25062513.

In 2007, the Joint Commission for Accreditation of Healthcare Organizations (JCAHO) recommended deployment of rapid response teams (RRTs) in U.S. hospitals to hasten identification and treatment of physiologically unstable hospitalized patients.1 Clinical studies that have focused on whether RRTs improve restorative care outcomes, frequency of cardiac arrest, and critical care utilization have yielded mixed results.2‐11 One study suggested that RRTs might provide an opportunity to enhance palliative care of hospitalized patients.11 In this study, RRT personnel felt that prior do‐not‐resuscitate orders would have been appropriate in nearly a quarter of cases. However, no previous study has examined whether the RRT might be deployed to identify acutely decompensating patients who either do not want or would not benefit from a trial of aggressive restorative treatments. We hypothesized that actuation of an RRT in our hospital would expedite identification of patients not likely to benefit from restorative care and would promote more timely commencement of end‐of‐life comfort care, thereby improving their quality of death (QOD).12‐16

Materials and Methods

Results

A total of 394 patients died on the hospital wards and were not admitted with palliative, end‐of‐life medical therapies. The combined (pre‐RRT and post‐RRT epochs) cohort had a mean age of 77.2 13.2 years. A total of 48% were male, 79% White, 12% Black, and 8% Hispanic. A total of 128 patients (33%) were admitted to the hospital from a skilled nursing facility and 135 (35%) had written advance directives.

A total of 197 patients met the inclusion criteria during the pre‐RRT (October 1, 2005 to May 31, 2006) and 197 during the post‐RRT epochs (October 1, 2006 to May 31, 2007). There were no differences in age, sex, advance directives, ethnicity, or religion between the groups (Table 1). Primary admission diagnoses were significantly different; pre‐RRT patients were 9% more likely to die with malignancy compared to post‐RRT patients and less likely to come from nursing homes (38% vs. 27%; P = 0.02).

Discussion

This pilot study hypothesizes that our RRT impacted patients' QOD. Deployment of the RRT in our hospital was associated with improvement in both symptom and psychospiritual domains of care. Theoretically, RRTs should improve quality‐of‐care via early identification/reversal of physiologic decompensation. By either reversing acute diatheses with an expeditious trial of therapy or failing to reverse with early actuation of palliative therapies, the duration and magnitude of human suffering should be reduced. Attenuation of both duration and magnitude of suffering is the ultimate goal of both restorative and palliative care and is as important an outcome as mortality or length of stay. Previous studies of RRTs have focused on efficacy in reversing the decompensation: preventing cardiopulmonary arrest, avoiding the need for invasive, expensive, labor‐intensive interventions. Our RRT, like others, had no demonstrable impact on restorative outcomes. However, deployment of the RRT was highly associated with improved QOD of our patients. The impact was significant across WHO‐specified domains: pain scores decreased by 19%; (documentation of) patients' distress decreased by 50%; and chaplains' visits were more often documented in the 24 hours prior to death. These relationships held across common disease diagnoses, so the association is unlikely to be spurious.

Outcomes were similarly improved in patients who did not receive RRT care in the post‐RRT epoch. Our hospital did not have a palliative care service in either time period. No new educational efforts among physicians or nurses accounted for this observation. While it is possible that temporal effects accounted for our observation, an equally plausible explanation is that staff observed RRT interventions and applied them to dying patients not seen by the RRT. Our hospital educated caregivers regarding the RRT triggers, and simply making hospital personnel more vigilant for signs of suffering and/or observing the RRT approach may have contributed to enhanced end‐of‐life care for non‐RRT patients.

There are a number of limitations in this study. First, the sample size was relatively small compared to other published studies,2‐11 promoting the possibility that either epoch was not representative of pre‐RRT and post‐RRT parent populations. Another weakness is that QOD was measured using surrogate endpoints. The dead cannot be interviewed to definitively examine QOD; indices of cardiopulmonary distress and psychosocial measures (eg, religious preparations, family involvement) are endpoints suggested by palliative care investigators12, 13 and the World Health Organization.14 While some validated tools17 and consensus measures18 exist for critically ill patients, they do not readily apply to RRT patients. Retrospective records reviews raise the possibility of bias in extracting objective and subjective data. While we attempted to control for this by creating uniform a priori rules for data acquisition (ie, at what intervals and in which parts of the record they could be extracted), we cannot discount the possibility that bias affected the observed results. Finally, improvements in end‐of‐life care could have resulted from temporal trends. This retrospective study cannot prove a causeeffect relationship; a prospective randomized trial would be required to answer the question definitively. Based on the available data suggesting some benefit in restorative outcomes2‐8 and pressure from federal regulators to deploy RRTs regardless,1 a retrospective cohort design may provide the only realistic means of addressing this question.

In conclusion, this is the first (pilot) study to examine end‐of‐life outcomes associated with deployment of an RRT. While the limitations of these observations preclude firm conclusions, the plausibility of the hypothesis, coupled with our observations, suggests that this is a fertile area for future research. While RRTs may enhance restorative outcomes, to the extent that they hasten identification of candidates for palliative end‐of‐life‐care, before administration of invasive modalities that some patients do not want, these teams may simultaneously serve patients and reduce hospital resource utilization.

Addendum

Prior to publication, a contemporaneous study was published that concluded: These findings suggest that rapid response teams may not be decreasing code rates as much as catalyzing a compassionate dialogue of end‐of‐life care among terminally ill patients. This ability to improve end‐of‐life care may be an important benefit of rapid response teams, particularly given the difficulties in prior trials to increase rates of DNR status among seriously ill inpatients and potential decreases in resource use. Chan PS, Khalid A, Longmore LS, Berg RA, Midhail Kosiborod M, Spertus JA. Hospital‐wide code rates and mortality before and after implementation of a rapid response team. JAMA 2008;300: 25062513.

The cornerstones of American medical ethics include respect for patient autonomy and beneficence. Although informed consent is required for surgical procedures and transfusion of blood products, the overwhelming majority of medical treatments administered by physicians to hospitalized patients are given without discussing risks, benefits, and alternatives. Although patients may sign a general permission‐to‐treat form on admission to the hospital, informed consent for medical treatments is generally ad hoc, and there are no national standards or mandates. We hypothesized that given the choice, hospitalized patients would want to participate in informed decision making, especially for therapies associated with substantial risks and benefits.

METHODS

The Institutional Review Board of Bridgeport Hospital approved this study. Each day between June and August 2006, the hospital's admitting department provided investigators with a list that included names and locations of all patients admitted to the Department of Medicine inpatient service. All the patients were eligible for participation in the study. Patients were excluded if they were in a comatose state, were encephalopathic, or were judged to be severely demented. In addition, patients were assessed during the scripted intervention to ascertain whether they had the capacity to make informed decisions based on their ability: (a) to understand the presented information, (b) to consider the information in relation to their personal values, and (c) to communicate their wishes. If personnel doubted an individual's capacity in any of these 3 areas, they were not included in the study.

Study personnel read directly from the script (see Appendix) and recorded answers. Study personnel were permitted to reread questions but did not provide additional guidance beyond the questionnaire. Patients whose primary language was not English were interviewed through in‐house or 3‐way telephone (remote) translators.

Statistical analyses included the chi‐square test to examine responses across the 3 categories of answers (ie, always consent, qualified consent, waive consent) and simple comparisons of percentages. A P value < .05 was considered statistically significant.

RESULTS

A total of 634 patients were admitted to the medicine service during the study period June‐August 2006. Of these, 158 were judged to lack sufficient capacity by study personnel and were excluded from the study. Ninety‐five refused to participate, and 171 were discharged before the questionnaire could be administered. Two hundred and ten patients answered the questionnaire. They ranged in age from 18 to 96 years (mean age standard error, 63.3 1.1 years). One hundred and three (49%) were men, and 107 (51%) were women. A majority (67.5%) were white, 20% (42) were African American, and 11.9% (25) were Hispanic. Most (87.6%) had at least a high school education, and 35% had a college‐/graduate‐level education. Sixty‐seven percent had at least 2 comorbid conditions in addition to their principal reason for hospitalization. Their average acute physiology and chronic health care evaluation (APACHE II) score was 7.5 0.3 (median 7; range 0‐22).

Figure 1 shows the distribution of answers to each of the 4 questions.

Figure 1

DISCUSSION

The principal finding of this study is that most medical patients prefer to participate in making decisions about their medical care during acute hospitalization, even for relatively low‐risk treatments like potassium supplementation and administration of diuretics. Very few patients were prepared to waive consent and grant their physicians the absolute right to administer therapies such as thrombolysis, even if the risk of bleeding was estimated to be less than 5%. Whereas the elderly patients were less likely to prefer being asked to consent to treatments than were younger patients, most would want to be informed of even trivial therapies if time allowed.

In some situations older patients (65 years old) were more likely than younger patients (<65 years old) to allow their physicians to make unilateral decisions regarding their health care. This could be explained by those age 65 and older having grown up when physician paternalism was more prevalent in American medicine. In the 1970s physician paternalism waned, and respect for patient autonomy emerged as the dominant physicianpatient model. Patients who became adults after 1970 know only this relationship with their physician, and so it makes sense that they would be more inclined to prefer a participatory model.

These data complement and extend a series of studies we conducted with patients admitted to Bridgeport Hospital. Our data suggest that our patients wish to consent for end‐of‐life decisions,1, 2 invasive procedures,3 and, now, to be apprised of medical therapies administered during hospitalization. At the same time, we have found that consent practices at many centers are not consistent with these patient predilections.1, 2, 4 Our study suffered from having a small sample size obtained in one geographic location; so results should be generalized cautiously. Nonetheless, insofar as the expectations of patients for participation are not being met by the health care system in Connecticut (and we suspect elsewhere), clinicians, hospital administrators, and health care policy makers might consider whether more rigorous and explicit consent practices and policies are required. Another important limitation of the study was that patients included may not have entirely understood the implications of their answers (ie, how cumbersome to the system and bothersome to the patient seeking consent for every therapy could become). In fact, we cannot be certain that all patients truly understood the questions, some of which were complex. Nonetheless, these results support that considered in the abstract, most patients prefer to consent for medical therapies. Had the implications for safety and expediency been explained in detail, it is possible that patients would have waived the need to give consent for treatments with minimal risk. The questionnaire also presents an abbreviated list of risks and benefits for each intervention, and although it refers to the formal process of informed consent in its preamble, it uses terminology (ie, permission) that may not reflect the complexity of informed consent. Nonetheless, our goal was to examine the degree to which patients wished to participate in their medical decision making. Notwithstanding these weaknesses of the survey instrument, the data suggest patients want to be in the loop whenever possible.

There are no national standards of consent for medical treatments. The Veterans' Administration, which has led the way in many areas of patients' rights, has a policy:

Compliance with this standard (ie, consent for every new medication) is not routine in most acute care hospitals. Although some clinicians obtain formal consent for high‐risk therapies (perhaps out of respect for autonomy, perhaps to reduce medical‐legal liability), there are no explicit decision rules to guide clinicians regarding for which treatments they should obtain formal consent. Accordingly, some might obtain formal consent for thrombolysis for massive pulmonary embolus, and others might not. It is not clear that the consent‐to‐treat form signed during hospital admission would legally cover all medical therapies during hospitalization. The legal standard for informed consent is what any reasonable patient would want to consent for. Our data suggest that most reasonable patients wish to at least assent and perhaps consent for much of what they receive during hospitalization. Although we have been unable to find case law predicated entirely on failure to obtain consent prior to administration of a therapy that caused a complication, it is plausible that the reasonable patient standard could be used in this manner. Regardless, it is impractical to require consent for the thousands of medical therapies administered each day in hospitals. Requiring consent for all therapies, if respected rigidly, would threaten the safety and efficiency of American hospitals. Naturally, a balance betweem respect for autonomy, that is, informed consent for the riskiest therapies, and efficiency is necessary. Explicit guidelines issued by accrediting agencies or the federal government would be helpful. The rules for consent (and/or assent) should be more explicit and less arbitrary, that is, determined independently by each clinician.

In conclusion, these data demonstrate that when considered in the abstract, that is, without explaining the logistical hurdles that it would create, inpatients wish to participate in decision making for both low‐ and high‐risk treatments. Clinicians are faced with demands and obligations that preclude full consent for the myriad low‐risk treatments administered daily to hospitalized patients. Some treatments are likely to be covered implicitly under the general consent‐to‐treat process and paperwork. Nonetheless, clinicians should consider explaining the principal risks and benefits of moderate‐risk treatments in order to secure informed assent. Full informed consent may be most appropriate for very high‐risk therapies. Patients expect and deserve frequent communication with caregivers that balances their safety with their right to self‐determination.

The cornerstones of American medical ethics include respect for patient autonomy and beneficence. Although informed consent is required for surgical procedures and transfusion of blood products, the overwhelming majority of medical treatments administered by physicians to hospitalized patients are given without discussing risks, benefits, and alternatives. Although patients may sign a general permission‐to‐treat form on admission to the hospital, informed consent for medical treatments is generally ad hoc, and there are no national standards or mandates. We hypothesized that given the choice, hospitalized patients would want to participate in informed decision making, especially for therapies associated with substantial risks and benefits.

METHODS

The Institutional Review Board of Bridgeport Hospital approved this study. Each day between June and August 2006, the hospital's admitting department provided investigators with a list that included names and locations of all patients admitted to the Department of Medicine inpatient service. All the patients were eligible for participation in the study. Patients were excluded if they were in a comatose state, were encephalopathic, or were judged to be severely demented. In addition, patients were assessed during the scripted intervention to ascertain whether they had the capacity to make informed decisions based on their ability: (a) to understand the presented information, (b) to consider the information in relation to their personal values, and (c) to communicate their wishes. If personnel doubted an individual's capacity in any of these 3 areas, they were not included in the study.

Study personnel read directly from the script (see Appendix) and recorded answers. Study personnel were permitted to reread questions but did not provide additional guidance beyond the questionnaire. Patients whose primary language was not English were interviewed through in‐house or 3‐way telephone (remote) translators.

Statistical analyses included the chi‐square test to examine responses across the 3 categories of answers (ie, always consent, qualified consent, waive consent) and simple comparisons of percentages. A P value < .05 was considered statistically significant.

RESULTS

A total of 634 patients were admitted to the medicine service during the study period June‐August 2006. Of these, 158 were judged to lack sufficient capacity by study personnel and were excluded from the study. Ninety‐five refused to participate, and 171 were discharged before the questionnaire could be administered. Two hundred and ten patients answered the questionnaire. They ranged in age from 18 to 96 years (mean age standard error, 63.3 1.1 years). One hundred and three (49%) were men, and 107 (51%) were women. A majority (67.5%) were white, 20% (42) were African American, and 11.9% (25) were Hispanic. Most (87.6%) had at least a high school education, and 35% had a college‐/graduate‐level education. Sixty‐seven percent had at least 2 comorbid conditions in addition to their principal reason for hospitalization. Their average acute physiology and chronic health care evaluation (APACHE II) score was 7.5 0.3 (median 7; range 0‐22).

Figure 1 shows the distribution of answers to each of the 4 questions.

Figure 1

DISCUSSION

The principal finding of this study is that most medical patients prefer to participate in making decisions about their medical care during acute hospitalization, even for relatively low‐risk treatments like potassium supplementation and administration of diuretics. Very few patients were prepared to waive consent and grant their physicians the absolute right to administer therapies such as thrombolysis, even if the risk of bleeding was estimated to be less than 5%. Whereas the elderly patients were less likely to prefer being asked to consent to treatments than were younger patients, most would want to be informed of even trivial therapies if time allowed.

In some situations older patients (65 years old) were more likely than younger patients (<65 years old) to allow their physicians to make unilateral decisions regarding their health care. This could be explained by those age 65 and older having grown up when physician paternalism was more prevalent in American medicine. In the 1970s physician paternalism waned, and respect for patient autonomy emerged as the dominant physicianpatient model. Patients who became adults after 1970 know only this relationship with their physician, and so it makes sense that they would be more inclined to prefer a participatory model.

These data complement and extend a series of studies we conducted with patients admitted to Bridgeport Hospital. Our data suggest that our patients wish to consent for end‐of‐life decisions,1, 2 invasive procedures,3 and, now, to be apprised of medical therapies administered during hospitalization. At the same time, we have found that consent practices at many centers are not consistent with these patient predilections.1, 2, 4 Our study suffered from having a small sample size obtained in one geographic location; so results should be generalized cautiously. Nonetheless, insofar as the expectations of patients for participation are not being met by the health care system in Connecticut (and we suspect elsewhere), clinicians, hospital administrators, and health care policy makers might consider whether more rigorous and explicit consent practices and policies are required. Another important limitation of the study was that patients included may not have entirely understood the implications of their answers (ie, how cumbersome to the system and bothersome to the patient seeking consent for every therapy could become). In fact, we cannot be certain that all patients truly understood the questions, some of which were complex. Nonetheless, these results support that considered in the abstract, most patients prefer to consent for medical therapies. Had the implications for safety and expediency been explained in detail, it is possible that patients would have waived the need to give consent for treatments with minimal risk. The questionnaire also presents an abbreviated list of risks and benefits for each intervention, and although it refers to the formal process of informed consent in its preamble, it uses terminology (ie, permission) that may not reflect the complexity of informed consent. Nonetheless, our goal was to examine the degree to which patients wished to participate in their medical decision making. Notwithstanding these weaknesses of the survey instrument, the data suggest patients want to be in the loop whenever possible.

There are no national standards of consent for medical treatments. The Veterans' Administration, which has led the way in many areas of patients' rights, has a policy:

Compliance with this standard (ie, consent for every new medication) is not routine in most acute care hospitals. Although some clinicians obtain formal consent for high‐risk therapies (perhaps out of respect for autonomy, perhaps to reduce medical‐legal liability), there are no explicit decision rules to guide clinicians regarding for which treatments they should obtain formal consent. Accordingly, some might obtain formal consent for thrombolysis for massive pulmonary embolus, and others might not. It is not clear that the consent‐to‐treat form signed during hospital admission would legally cover all medical therapies during hospitalization. The legal standard for informed consent is what any reasonable patient would want to consent for. Our data suggest that most reasonable patients wish to at least assent and perhaps consent for much of what they receive during hospitalization. Although we have been unable to find case law predicated entirely on failure to obtain consent prior to administration of a therapy that caused a complication, it is plausible that the reasonable patient standard could be used in this manner. Regardless, it is impractical to require consent for the thousands of medical therapies administered each day in hospitals. Requiring consent for all therapies, if respected rigidly, would threaten the safety and efficiency of American hospitals. Naturally, a balance betweem respect for autonomy, that is, informed consent for the riskiest therapies, and efficiency is necessary. Explicit guidelines issued by accrediting agencies or the federal government would be helpful. The rules for consent (and/or assent) should be more explicit and less arbitrary, that is, determined independently by each clinician.

In conclusion, these data demonstrate that when considered in the abstract, that is, without explaining the logistical hurdles that it would create, inpatients wish to participate in decision making for both low‐ and high‐risk treatments. Clinicians are faced with demands and obligations that preclude full consent for the myriad low‐risk treatments administered daily to hospitalized patients. Some treatments are likely to be covered implicitly under the general consent‐to‐treat process and paperwork. Nonetheless, clinicians should consider explaining the principal risks and benefits of moderate‐risk treatments in order to secure informed assent. Full informed consent may be most appropriate for very high‐risk therapies. Patients expect and deserve frequent communication with caregivers that balances their safety with their right to self‐determination.

The cornerstones of American medical ethics include respect for patient autonomy and beneficence. Although informed consent is required for surgical procedures and transfusion of blood products, the overwhelming majority of medical treatments administered by physicians to hospitalized patients are given without discussing risks, benefits, and alternatives. Although patients may sign a general permission‐to‐treat form on admission to the hospital, informed consent for medical treatments is generally ad hoc, and there are no national standards or mandates. We hypothesized that given the choice, hospitalized patients would want to participate in informed decision making, especially for therapies associated with substantial risks and benefits.

METHODS

The Institutional Review Board of Bridgeport Hospital approved this study. Each day between June and August 2006, the hospital's admitting department provided investigators with a list that included names and locations of all patients admitted to the Department of Medicine inpatient service. All the patients were eligible for participation in the study. Patients were excluded if they were in a comatose state, were encephalopathic, or were judged to be severely demented. In addition, patients were assessed during the scripted intervention to ascertain whether they had the capacity to make informed decisions based on their ability: (a) to understand the presented information, (b) to consider the information in relation to their personal values, and (c) to communicate their wishes. If personnel doubted an individual's capacity in any of these 3 areas, they were not included in the study.

Study personnel read directly from the script (see Appendix) and recorded answers. Study personnel were permitted to reread questions but did not provide additional guidance beyond the questionnaire. Patients whose primary language was not English were interviewed through in‐house or 3‐way telephone (remote) translators.

Statistical analyses included the chi‐square test to examine responses across the 3 categories of answers (ie, always consent, qualified consent, waive consent) and simple comparisons of percentages. A P value < .05 was considered statistically significant.

RESULTS

A total of 634 patients were admitted to the medicine service during the study period June‐August 2006. Of these, 158 were judged to lack sufficient capacity by study personnel and were excluded from the study. Ninety‐five refused to participate, and 171 were discharged before the questionnaire could be administered. Two hundred and ten patients answered the questionnaire. They ranged in age from 18 to 96 years (mean age standard error, 63.3 1.1 years). One hundred and three (49%) were men, and 107 (51%) were women. A majority (67.5%) were white, 20% (42) were African American, and 11.9% (25) were Hispanic. Most (87.6%) had at least a high school education, and 35% had a college‐/graduate‐level education. Sixty‐seven percent had at least 2 comorbid conditions in addition to their principal reason for hospitalization. Their average acute physiology and chronic health care evaluation (APACHE II) score was 7.5 0.3 (median 7; range 0‐22).

Figure 1 shows the distribution of answers to each of the 4 questions.

Figure 1

DISCUSSION

The principal finding of this study is that most medical patients prefer to participate in making decisions about their medical care during acute hospitalization, even for relatively low‐risk treatments like potassium supplementation and administration of diuretics. Very few patients were prepared to waive consent and grant their physicians the absolute right to administer therapies such as thrombolysis, even if the risk of bleeding was estimated to be less than 5%. Whereas the elderly patients were less likely to prefer being asked to consent to treatments than were younger patients, most would want to be informed of even trivial therapies if time allowed.

In some situations older patients (65 years old) were more likely than younger patients (<65 years old) to allow their physicians to make unilateral decisions regarding their health care. This could be explained by those age 65 and older having grown up when physician paternalism was more prevalent in American medicine. In the 1970s physician paternalism waned, and respect for patient autonomy emerged as the dominant physicianpatient model. Patients who became adults after 1970 know only this relationship with their physician, and so it makes sense that they would be more inclined to prefer a participatory model.

These data complement and extend a series of studies we conducted with patients admitted to Bridgeport Hospital. Our data suggest that our patients wish to consent for end‐of‐life decisions,1, 2 invasive procedures,3 and, now, to be apprised of medical therapies administered during hospitalization. At the same time, we have found that consent practices at many centers are not consistent with these patient predilections.1, 2, 4 Our study suffered from having a small sample size obtained in one geographic location; so results should be generalized cautiously. Nonetheless, insofar as the expectations of patients for participation are not being met by the health care system in Connecticut (and we suspect elsewhere), clinicians, hospital administrators, and health care policy makers might consider whether more rigorous and explicit consent practices and policies are required. Another important limitation of the study was that patients included may not have entirely understood the implications of their answers (ie, how cumbersome to the system and bothersome to the patient seeking consent for every therapy could become). In fact, we cannot be certain that all patients truly understood the questions, some of which were complex. Nonetheless, these results support that considered in the abstract, most patients prefer to consent for medical therapies. Had the implications for safety and expediency been explained in detail, it is possible that patients would have waived the need to give consent for treatments with minimal risk. The questionnaire also presents an abbreviated list of risks and benefits for each intervention, and although it refers to the formal process of informed consent in its preamble, it uses terminology (ie, permission) that may not reflect the complexity of informed consent. Nonetheless, our goal was to examine the degree to which patients wished to participate in their medical decision making. Notwithstanding these weaknesses of the survey instrument, the data suggest patients want to be in the loop whenever possible.

There are no national standards of consent for medical treatments. The Veterans' Administration, which has led the way in many areas of patients' rights, has a policy:

Compliance with this standard (ie, consent for every new medication) is not routine in most acute care hospitals. Although some clinicians obtain formal consent for high‐risk therapies (perhaps out of respect for autonomy, perhaps to reduce medical‐legal liability), there are no explicit decision rules to guide clinicians regarding for which treatments they should obtain formal consent. Accordingly, some might obtain formal consent for thrombolysis for massive pulmonary embolus, and others might not. It is not clear that the consent‐to‐treat form signed during hospital admission would legally cover all medical therapies during hospitalization. The legal standard for informed consent is what any reasonable patient would want to consent for. Our data suggest that most reasonable patients wish to at least assent and perhaps consent for much of what they receive during hospitalization. Although we have been unable to find case law predicated entirely on failure to obtain consent prior to administration of a therapy that caused a complication, it is plausible that the reasonable patient standard could be used in this manner. Regardless, it is impractical to require consent for the thousands of medical therapies administered each day in hospitals. Requiring consent for all therapies, if respected rigidly, would threaten the safety and efficiency of American hospitals. Naturally, a balance betweem respect for autonomy, that is, informed consent for the riskiest therapies, and efficiency is necessary. Explicit guidelines issued by accrediting agencies or the federal government would be helpful. The rules for consent (and/or assent) should be more explicit and less arbitrary, that is, determined independently by each clinician.

In conclusion, these data demonstrate that when considered in the abstract, that is, without explaining the logistical hurdles that it would create, inpatients wish to participate in decision making for both low‐ and high‐risk treatments. Clinicians are faced with demands and obligations that preclude full consent for the myriad low‐risk treatments administered daily to hospitalized patients. Some treatments are likely to be covered implicitly under the general consent‐to‐treat process and paperwork. Nonetheless, clinicians should consider explaining the principal risks and benefits of moderate‐risk treatments in order to secure informed assent. Full informed consent may be most appropriate for very high‐risk therapies. Patients expect and deserve frequent communication with caregivers that balances their safety with their right to self‐determination.

Respect for patient autonomy is a primary ethical principle guiding the practice of medicine in the United States.1. The Patient Self‐Determination Act (PSDA), enacted to enhance autonomy at the end of life, has not fulfilled its promise for a number of reasons.24 No state mandates that on admission, hospitalized patients be asked to provide informed consent for end‐of‐life procedures. Despite informed consent being a requirement for all other invasive procedures when there is sufficient opportunity to obtain it (eg, in nonemergent situations with a capable patient),5 cardiopulmonary resuscitation (CPR) and mechanical ventilation are assumed, until otherwise stipulated, to be procedures that all patients want. It also has been assumed that patients would believe that a request for informed consent for such procedures on hospital admission implied they had significant risk of cardiopulmonary failure and that this would discourage or disturb acutely ill patients.6 Another impediment to obtaining informed consent is that many physicians may not have sufficient time or level of comfort to be able to routinely approach end‐of‐life discussions. In this prospective study, we hypothesized that acutely ill medical patients would be willing to provide informed consent for CPR and mechanical ventilation and to create written advance directives.

METHODS

This study was approved by the hospital's institutional review board. Patients admitted to the Department of Medicine from December 2003 through February 2004 were candidates for this study. Patients admitted for cardiac catheterization (and similar same‐day medical procedures) or critical illness (admitted to intensive care units) were excluded from the study. In our hospital, all patients are asked by admitting personnel (clerk and nurse) whether they already have advance directives. Some patients are also queried by their physicians about whether they wish to have CPR in the event of cardiopulmonary arrest during hospitalization. Patients who are not asked are assumed to be full codes, that is, they are to receive CPR and mechanical ventilation in the event of cardiac and respiratory failure. For those who are asked, there are generally 3 possible outcomes: (1) the patient chooses to accept CPR and mechanical ventilation, and nothing further is documented; (2) the patient chooses a code status, and it is documented in the admission orders and/or a formal code designation form with a progress note describing the discussion; or (3) the patient defers the decision.

Our data processing department generated a daily list of the patients admitted to the hospital on the previous day. Patients satisfying inclusion criteria were randomized (by a random number generator) to the intervention or the control group. Medical records of all patients were examined to ascertain demographic information, admission Acute Physiology and Chronic Health Evaluation (APACHE) II score, primary diagnosis, number of comorbid illnesses, and documentation of whether the patient had a preexisting advance directive or wishes regarding CPR and mechanical ventilation for that admission.

Patients in the control group were not approached by study personnel, but medical records were surveyed for their in‐hospital outcomes and changes in code or advance directive status. Patients randomized to the intervention arm were approached by 1 of 4 study physicians, who read from a script detailed information about life‐sustaining therapies and advance directives (see Appendix). This script was developed with hospital clinician‐experts and approved by members of the Department of Medicine.

Patients whose primary language was not English were interviewed through in‐house or 3‐way telephone (remote) translators. All patients in the treatment group were assessed during the scripted intervention to ascertain whether they had the capacity to make informed decisions, which was determined based on their ability: (a) to understand the information presented, (b) to consider the information in relation to their personal values, and (c) to communicate their wishes. If personnel doubted an individual's capacity in any of these 3 areas, then he or she was not included in the study (ie, excluded after randomization). In the control group, patients with documented dementia or delirium were also excluded.

As specified in the script, patients in the intervention group were asked at the end of the interview whether they wished to choose their in‐hospital CPR status for that admission. If a patient definitely wanted to change the status indicated in the hospital record, study personnel would communicate the patient's wishes to the admitting physician. Attending physicians were given the opportunity to speak with their patients before changing a code status, but if the physicians agreed with the change, study personnel would document it in the formal orders. Patients were also asked whether they wished to create advance directives; if so, staff from the hospital's patient relations department would meet with them to draft the documents.

The following outcomes were measured: 1) willingness of patients assigned to the intervention group to listen to the script about end‐of‐life/life‐sustaining therapies; 2) opinions of patients about whether the information in the intervention was useful versus whether it was disturbing; 3) the frequency with which patients who had proactively received the information chose or changed their code status; and 4) the frequency with which patients without a preexisting advance directive created one while hospitalized. Simple proportions of each of these variables (ie, observed number divided by total number) in the intervention and control groups were compared using software that calculates the significance of the difference between two percentages (Statistica). The demographics of the patients were compared using the unpaired Student's t test. A P value of < .05 was considered statistically significant.

RESULTS

A total of 585 patients admitted to the Department of Medicine between December 2003 and February 2004 were randomized for the study. Patients were excluded if they had insufficient capacity (133) or if they were rapidly discharged from the hospital (155). Patients who were excluded tended to be more ill (APACHE 8.1 vs. 7.3, P = .06) and were more likely to die while hospitalized (8% vs. 4%, P = .04). A total of 297 patients were included in the study, 136 in the intervention group and 161 in the control group. Baseline characteristics were similar between the 2 groups (see Table 1).

Did Patients Who Received the Intervention Clarify Their CPR Preference?

Of the 136 patients in the intervention arm, 49 (36%) had explicit documentation of their code status on admission, compared to 55 of the 161 patients in the control group (34%; P = .7). Documentation included listing the CPR status in the admission orders or in a completed code designation form. After receiving the intervention, 125 of the 136 patients in the intervention arm (92%) clarified their preferences about CPR and mechanical ventilation.

Of the 49 patients in the intervention group who had documented CPR status on admission, 48 were listed as full code (both CPR and mechanical ventilation), and 1 was documented as refusing both CPR and mechanical ventilation. Of the 48 patients who were full codes, 3 stated they did not want CPR and mechanical ventilation under any circumstances after the intervention. Their preferences were subsequently documented as formal orders. The remaining 45 (94%) stayed full codes (see Figure 1).

Figure 1

Of the 87 patients in the intervention group who had no explicit documentation of CPR status on hospital admission, 76 clarified their preference and 11 did not. Of the 76 patients, 71 wished to receive both CPR and mechanical ventilation, and 5 wanted neither. The status of the latter as no code, no ventilator was subsequently documented in the medical record with the consent of their attending physicians. One of these 5 patients became increasingly ill during hospitalization, with reduced capacity, and family members later asked that he receive only comfort care.

Of the 161 patients in the control group, 55 (34%) had documentation of their code status (ie, to receive CPR if needed) in the admission hospital record. By the end of hospitalization, 1 patient requested no CPR and no mechanical ventilation, and 2 received comfort care with cessation of other active life‐prolonging interventions. Of the 106 without initial code documentation, 4 were later documented as being no code, no ventilator and 2 as being comfort care (see Figure 1).

DISCUSSION

This study demonstrates that most (95%) hospitalized medical patients welcomed the opportunity to provide prospective informed consent for CPR and mechanical ventilation. Although only a small minority (4%) opted out of CPR/mechanical ventilation, a majority (92%) of those who received the educational intervention chose to accept those therapies if required. This study also demonstrates that hospitalization can be one point‐of‐care where patients can consider and create advance directives. The results of this study are consistent with those of the SUPPORT group4 and other7 studies about patient interest in making choices on CPR. Our study suggests that physicians can elicit patients' wishes about and record formal orders on CPR around the time of hospital admission.

The default action has been to administer CPR and mechanical ventilation after cardiopulmonary failure or arrest, that is, patients receive these procedures unless they state explicitly that they do not want them. Unlike with all other invasive procedures, no national regulation mandates obtaining informed consent prospectively, when possible, for these treatments, because it is assumed that patients would want these therapies rather than the alternative (ie, death). Indeed, it is appropriate to perform lifesaving procedures in emergencies without consent if the patient lacks capacity and a surrogate decision maker cannot be contacted quickly. This clinical approach is consistent with medical ethics: to err on the side of life when a patient's wishes are unknown or unclear. Nonetheless, having a full code as the default action denies patients the opportunity to provide informed consent for these highly invasive procedures because there often is ample opportunity to ask their permission. If patient self‐determination is the categorical imperative of American medicine, then current practice violates that principle at the moment when it may be most important, that is, when a patient's decision about whether to risk life‐sustaining therapies could promote survival or prolong dying. Our study demonstrates that a simple interventionsimply askingpromotes a decision and therefore patient autonomy in most cases.

When patients have opted for life‐sustaining therapies that subsequently have been administered or when patients have received such therapies by default, physicians and patients can be left in 2 situations. In one outcome the patient retains capacity, and the dialogue about life‐sustaining therapies can continue between patient and physician. In the second, frequent scenario, the patient is incapacitated. Until patient capacity can be restored, the physician must work with surrogate decision makers and preexisting advance directives to infer a patient's wishes about continuation of life‐sustaining care. Our data demonstrate that hospital admission is one point‐of‐care at which patients can be offered and can complete, albeit in small numbers, advance directives.8 Previous work with our patients demonstrated that many patients misunderstood advance directives and the degree of effort required to create them.9 We reasoned that more patients might create advance directives if we offered the service for free during hospitalization. We were very surprised at how infrequently patients created advance directives in this study, although this finding is consistent with others in the published literature.8 It is speculated that hospitalized patients may feel too ill to exert themselves and/or are not psychologically prepared to consider end‐of‐life directives (ie, I came to the hospital to get better, not to consider what should be done when I'm terminal ). Some patients may not trust physicians to use advance directives reliably.4, 10

Our study had several important limitations. First, and most important, not all patients who were randomized were enrolled in the study. The most common reasons for exclusion were rapid discharge from the hospital and mental status change calling into question a patient's capacity to make end‐of‐life decisions. Nonetheless, it is only competent patients who can be engaged to decide these questions for themselves. Surrogates (ie, loved ones), guided by advance directives, are left to address resuscitation decisions for those lacking capacity. In addition, patients' predilections may change with time,11 especially as death becomes more imminent. However, insofar as many patients have several hospital admissions as they approach the end of life and are more likely to possess capacity to consider CPR decisions during early admissions, their choices can be recorded repeatedly over time (with each admission or even as status changes during an admission) to inform decisions if they develop incapacity. Little more can be done to enhance autonomy regarding CPR beyond repeatedly educating and asking, as disease and specific illnesses progress. It can be argued that this intervention had little real overall effectmost patients who would have received CPR by default did in fact want it when informed and asked. This is an ethically problematic position for two reasons: it neglects the right of patients to decide for themselves, and it potentially subjects the small group of patients who would reject CPR if asked to an unwanted risky procedure (ie, one that may prolong dying). Another limitation of the present study is that patients were approached by doctors‐in‐training with whom they had had no prior therapeutic relationship. Although it would have been optimal for patients to be approached by their primary care physicians, this was not feasible. Even if we could have convinced all of our medical staff members to implement the intervention, it is unlikely that all would have adhered to a study script, which is what enabled standardization of the information shared with patients. Some physicians may disagree with the script's content. But the goal of this study was not to determine if specific information would affect outcomes; rather, it was to determine if patients were receptive to discussing these issues and making proactive choices regarding life‐sustaining therapies during hospitalization for acute illness. It is possible that using different scripts delivered by different personnel, ideally the patients' own doctors, might have elicited even greater rates of consent and proactive decision making. Finally, the degree to which these results can be generalized may vary based on the population sampled. White and well‐educated patients are more likely to engage in end‐of‐life decision making than non‐White and poorly educated patients.9, 12

In conclusion, this study suggests that capable patients hospitalized for medical problems are willing to give informed consent for (or reject) CPR and mechanical ventilation in the event of cardiopulmonary failure. The approach of the study was very simple. It took roughly 510 minutes to inform patients and elicit their choices. Allowing patients to choose, rather than assuming that CPR is the choice of patients by default, strenuously honors patient autonomy. If these findings are replicated in larger cohorts and at different centers, there would be little justification for not informing patients about and asking them to choose their CPR preferences for each hospitalization. In the meantime, caregivers might consider the appropriateness of addressing these issues when they admit acutely ill patients to the hospital.

Respect for patient autonomy is a primary ethical principle guiding the practice of medicine in the United States.1. The Patient Self‐Determination Act (PSDA), enacted to enhance autonomy at the end of life, has not fulfilled its promise for a number of reasons.24 No state mandates that on admission, hospitalized patients be asked to provide informed consent for end‐of‐life procedures. Despite informed consent being a requirement for all other invasive procedures when there is sufficient opportunity to obtain it (eg, in nonemergent situations with a capable patient),5 cardiopulmonary resuscitation (CPR) and mechanical ventilation are assumed, until otherwise stipulated, to be procedures that all patients want. It also has been assumed that patients would believe that a request for informed consent for such procedures on hospital admission implied they had significant risk of cardiopulmonary failure and that this would discourage or disturb acutely ill patients.6 Another impediment to obtaining informed consent is that many physicians may not have sufficient time or level of comfort to be able to routinely approach end‐of‐life discussions. In this prospective study, we hypothesized that acutely ill medical patients would be willing to provide informed consent for CPR and mechanical ventilation and to create written advance directives.

METHODS

This study was approved by the hospital's institutional review board. Patients admitted to the Department of Medicine from December 2003 through February 2004 were candidates for this study. Patients admitted for cardiac catheterization (and similar same‐day medical procedures) or critical illness (admitted to intensive care units) were excluded from the study. In our hospital, all patients are asked by admitting personnel (clerk and nurse) whether they already have advance directives. Some patients are also queried by their physicians about whether they wish to have CPR in the event of cardiopulmonary arrest during hospitalization. Patients who are not asked are assumed to be full codes, that is, they are to receive CPR and mechanical ventilation in the event of cardiac and respiratory failure. For those who are asked, there are generally 3 possible outcomes: (1) the patient chooses to accept CPR and mechanical ventilation, and nothing further is documented; (2) the patient chooses a code status, and it is documented in the admission orders and/or a formal code designation form with a progress note describing the discussion; or (3) the patient defers the decision.

Our data processing department generated a daily list of the patients admitted to the hospital on the previous day. Patients satisfying inclusion criteria were randomized (by a random number generator) to the intervention or the control group. Medical records of all patients were examined to ascertain demographic information, admission Acute Physiology and Chronic Health Evaluation (APACHE) II score, primary diagnosis, number of comorbid illnesses, and documentation of whether the patient had a preexisting advance directive or wishes regarding CPR and mechanical ventilation for that admission.

Patients in the control group were not approached by study personnel, but medical records were surveyed for their in‐hospital outcomes and changes in code or advance directive status. Patients randomized to the intervention arm were approached by 1 of 4 study physicians, who read from a script detailed information about life‐sustaining therapies and advance directives (see Appendix). This script was developed with hospital clinician‐experts and approved by members of the Department of Medicine.

Patients whose primary language was not English were interviewed through in‐house or 3‐way telephone (remote) translators. All patients in the treatment group were assessed during the scripted intervention to ascertain whether they had the capacity to make informed decisions, which was determined based on their ability: (a) to understand the information presented, (b) to consider the information in relation to their personal values, and (c) to communicate their wishes. If personnel doubted an individual's capacity in any of these 3 areas, then he or she was not included in the study (ie, excluded after randomization). In the control group, patients with documented dementia or delirium were also excluded.

As specified in the script, patients in the intervention group were asked at the end of the interview whether they wished to choose their in‐hospital CPR status for that admission. If a patient definitely wanted to change the status indicated in the hospital record, study personnel would communicate the patient's wishes to the admitting physician. Attending physicians were given the opportunity to speak with their patients before changing a code status, but if the physicians agreed with the change, study personnel would document it in the formal orders. Patients were also asked whether they wished to create advance directives; if so, staff from the hospital's patient relations department would meet with them to draft the documents.

The following outcomes were measured: 1) willingness of patients assigned to the intervention group to listen to the script about end‐of‐life/life‐sustaining therapies; 2) opinions of patients about whether the information in the intervention was useful versus whether it was disturbing; 3) the frequency with which patients who had proactively received the information chose or changed their code status; and 4) the frequency with which patients without a preexisting advance directive created one while hospitalized. Simple proportions of each of these variables (ie, observed number divided by total number) in the intervention and control groups were compared using software that calculates the significance of the difference between two percentages (Statistica). The demographics of the patients were compared using the unpaired Student's t test. A P value of < .05 was considered statistically significant.

RESULTS

A total of 585 patients admitted to the Department of Medicine between December 2003 and February 2004 were randomized for the study. Patients were excluded if they had insufficient capacity (133) or if they were rapidly discharged from the hospital (155). Patients who were excluded tended to be more ill (APACHE 8.1 vs. 7.3, P = .06) and were more likely to die while hospitalized (8% vs. 4%, P = .04). A total of 297 patients were included in the study, 136 in the intervention group and 161 in the control group. Baseline characteristics were similar between the 2 groups (see Table 1).

Did Patients Who Received the Intervention Clarify Their CPR Preference?

Of the 136 patients in the intervention arm, 49 (36%) had explicit documentation of their code status on admission, compared to 55 of the 161 patients in the control group (34%; P = .7). Documentation included listing the CPR status in the admission orders or in a completed code designation form. After receiving the intervention, 125 of the 136 patients in the intervention arm (92%) clarified their preferences about CPR and mechanical ventilation.

Of the 49 patients in the intervention group who had documented CPR status on admission, 48 were listed as full code (both CPR and mechanical ventilation), and 1 was documented as refusing both CPR and mechanical ventilation. Of the 48 patients who were full codes, 3 stated they did not want CPR and mechanical ventilation under any circumstances after the intervention. Their preferences were subsequently documented as formal orders. The remaining 45 (94%) stayed full codes (see Figure 1).

Figure 1

Of the 87 patients in the intervention group who had no explicit documentation of CPR status on hospital admission, 76 clarified their preference and 11 did not. Of the 76 patients, 71 wished to receive both CPR and mechanical ventilation, and 5 wanted neither. The status of the latter as no code, no ventilator was subsequently documented in the medical record with the consent of their attending physicians. One of these 5 patients became increasingly ill during hospitalization, with reduced capacity, and family members later asked that he receive only comfort care.

Of the 161 patients in the control group, 55 (34%) had documentation of their code status (ie, to receive CPR if needed) in the admission hospital record. By the end of hospitalization, 1 patient requested no CPR and no mechanical ventilation, and 2 received comfort care with cessation of other active life‐prolonging interventions. Of the 106 without initial code documentation, 4 were later documented as being no code, no ventilator and 2 as being comfort care (see Figure 1).

DISCUSSION

This study demonstrates that most (95%) hospitalized medical patients welcomed the opportunity to provide prospective informed consent for CPR and mechanical ventilation. Although only a small minority (4%) opted out of CPR/mechanical ventilation, a majority (92%) of those who received the educational intervention chose to accept those therapies if required. This study also demonstrates that hospitalization can be one point‐of‐care where patients can consider and create advance directives. The results of this study are consistent with those of the SUPPORT group4 and other7 studies about patient interest in making choices on CPR. Our study suggests that physicians can elicit patients' wishes about and record formal orders on CPR around the time of hospital admission.

The default action has been to administer CPR and mechanical ventilation after cardiopulmonary failure or arrest, that is, patients receive these procedures unless they state explicitly that they do not want them. Unlike with all other invasive procedures, no national regulation mandates obtaining informed consent prospectively, when possible, for these treatments, because it is assumed that patients would want these therapies rather than the alternative (ie, death). Indeed, it is appropriate to perform lifesaving procedures in emergencies without consent if the patient lacks capacity and a surrogate decision maker cannot be contacted quickly. This clinical approach is consistent with medical ethics: to err on the side of life when a patient's wishes are unknown or unclear. Nonetheless, having a full code as the default action denies patients the opportunity to provide informed consent for these highly invasive procedures because there often is ample opportunity to ask their permission. If patient self‐determination is the categorical imperative of American medicine, then current practice violates that principle at the moment when it may be most important, that is, when a patient's decision about whether to risk life‐sustaining therapies could promote survival or prolong dying. Our study demonstrates that a simple interventionsimply askingpromotes a decision and therefore patient autonomy in most cases.

When patients have opted for life‐sustaining therapies that subsequently have been administered or when patients have received such therapies by default, physicians and patients can be left in 2 situations. In one outcome the patient retains capacity, and the dialogue about life‐sustaining therapies can continue between patient and physician. In the second, frequent scenario, the patient is incapacitated. Until patient capacity can be restored, the physician must work with surrogate decision makers and preexisting advance directives to infer a patient's wishes about continuation of life‐sustaining care. Our data demonstrate that hospital admission is one point‐of‐care at which patients can be offered and can complete, albeit in small numbers, advance directives.8 Previous work with our patients demonstrated that many patients misunderstood advance directives and the degree of effort required to create them.9 We reasoned that more patients might create advance directives if we offered the service for free during hospitalization. We were very surprised at how infrequently patients created advance directives in this study, although this finding is consistent with others in the published literature.8 It is speculated that hospitalized patients may feel too ill to exert themselves and/or are not psychologically prepared to consider end‐of‐life directives (ie, I came to the hospital to get better, not to consider what should be done when I'm terminal ). Some patients may not trust physicians to use advance directives reliably.4, 10

Our study had several important limitations. First, and most important, not all patients who were randomized were enrolled in the study. The most common reasons for exclusion were rapid discharge from the hospital and mental status change calling into question a patient's capacity to make end‐of‐life decisions. Nonetheless, it is only competent patients who can be engaged to decide these questions for themselves. Surrogates (ie, loved ones), guided by advance directives, are left to address resuscitation decisions for those lacking capacity. In addition, patients' predilections may change with time,11 especially as death becomes more imminent. However, insofar as many patients have several hospital admissions as they approach the end of life and are more likely to possess capacity to consider CPR decisions during early admissions, their choices can be recorded repeatedly over time (with each admission or even as status changes during an admission) to inform decisions if they develop incapacity. Little more can be done to enhance autonomy regarding CPR beyond repeatedly educating and asking, as disease and specific illnesses progress. It can be argued that this intervention had little real overall effectmost patients who would have received CPR by default did in fact want it when informed and asked. This is an ethically problematic position for two reasons: it neglects the right of patients to decide for themselves, and it potentially subjects the small group of patients who would reject CPR if asked to an unwanted risky procedure (ie, one that may prolong dying). Another limitation of the present study is that patients were approached by doctors‐in‐training with whom they had had no prior therapeutic relationship. Although it would have been optimal for patients to be approached by their primary care physicians, this was not feasible. Even if we could have convinced all of our medical staff members to implement the intervention, it is unlikely that all would have adhered to a study script, which is what enabled standardization of the information shared with patients. Some physicians may disagree with the script's content. But the goal of this study was not to determine if specific information would affect outcomes; rather, it was to determine if patients were receptive to discussing these issues and making proactive choices regarding life‐sustaining therapies during hospitalization for acute illness. It is possible that using different scripts delivered by different personnel, ideally the patients' own doctors, might have elicited even greater rates of consent and proactive decision making. Finally, the degree to which these results can be generalized may vary based on the population sampled. White and well‐educated patients are more likely to engage in end‐of‐life decision making than non‐White and poorly educated patients.9, 12

In conclusion, this study suggests that capable patients hospitalized for medical problems are willing to give informed consent for (or reject) CPR and mechanical ventilation in the event of cardiopulmonary failure. The approach of the study was very simple. It took roughly 510 minutes to inform patients and elicit their choices. Allowing patients to choose, rather than assuming that CPR is the choice of patients by default, strenuously honors patient autonomy. If these findings are replicated in larger cohorts and at different centers, there would be little justification for not informing patients about and asking them to choose their CPR preferences for each hospitalization. In the meantime, caregivers might consider the appropriateness of addressing these issues when they admit acutely ill patients to the hospital.

Respect for patient autonomy is a primary ethical principle guiding the practice of medicine in the United States.1. The Patient Self‐Determination Act (PSDA), enacted to enhance autonomy at the end of life, has not fulfilled its promise for a number of reasons.24 No state mandates that on admission, hospitalized patients be asked to provide informed consent for end‐of‐life procedures. Despite informed consent being a requirement for all other invasive procedures when there is sufficient opportunity to obtain it (eg, in nonemergent situations with a capable patient),5 cardiopulmonary resuscitation (CPR) and mechanical ventilation are assumed, until otherwise stipulated, to be procedures that all patients want. It also has been assumed that patients would believe that a request for informed consent for such procedures on hospital admission implied they had significant risk of cardiopulmonary failure and that this would discourage or disturb acutely ill patients.6 Another impediment to obtaining informed consent is that many physicians may not have sufficient time or level of comfort to be able to routinely approach end‐of‐life discussions. In this prospective study, we hypothesized that acutely ill medical patients would be willing to provide informed consent for CPR and mechanical ventilation and to create written advance directives.

METHODS

This study was approved by the hospital's institutional review board. Patients admitted to the Department of Medicine from December 2003 through February 2004 were candidates for this study. Patients admitted for cardiac catheterization (and similar same‐day medical procedures) or critical illness (admitted to intensive care units) were excluded from the study. In our hospital, all patients are asked by admitting personnel (clerk and nurse) whether they already have advance directives. Some patients are also queried by their physicians about whether they wish to have CPR in the event of cardiopulmonary arrest during hospitalization. Patients who are not asked are assumed to be full codes, that is, they are to receive CPR and mechanical ventilation in the event of cardiac and respiratory failure. For those who are asked, there are generally 3 possible outcomes: (1) the patient chooses to accept CPR and mechanical ventilation, and nothing further is documented; (2) the patient chooses a code status, and it is documented in the admission orders and/or a formal code designation form with a progress note describing the discussion; or (3) the patient defers the decision.

Our data processing department generated a daily list of the patients admitted to the hospital on the previous day. Patients satisfying inclusion criteria were randomized (by a random number generator) to the intervention or the control group. Medical records of all patients were examined to ascertain demographic information, admission Acute Physiology and Chronic Health Evaluation (APACHE) II score, primary diagnosis, number of comorbid illnesses, and documentation of whether the patient had a preexisting advance directive or wishes regarding CPR and mechanical ventilation for that admission.

Patients in the control group were not approached by study personnel, but medical records were surveyed for their in‐hospital outcomes and changes in code or advance directive status. Patients randomized to the intervention arm were approached by 1 of 4 study physicians, who read from a script detailed information about life‐sustaining therapies and advance directives (see Appendix). This script was developed with hospital clinician‐experts and approved by members of the Department of Medicine.

Patients whose primary language was not English were interviewed through in‐house or 3‐way telephone (remote) translators. All patients in the treatment group were assessed during the scripted intervention to ascertain whether they had the capacity to make informed decisions, which was determined based on their ability: (a) to understand the information presented, (b) to consider the information in relation to their personal values, and (c) to communicate their wishes. If personnel doubted an individual's capacity in any of these 3 areas, then he or she was not included in the study (ie, excluded after randomization). In the control group, patients with documented dementia or delirium were also excluded.

As specified in the script, patients in the intervention group were asked at the end of the interview whether they wished to choose their in‐hospital CPR status for that admission. If a patient definitely wanted to change the status indicated in the hospital record, study personnel would communicate the patient's wishes to the admitting physician. Attending physicians were given the opportunity to speak with their patients before changing a code status, but if the physicians agreed with the change, study personnel would document it in the formal orders. Patients were also asked whether they wished to create advance directives; if so, staff from the hospital's patient relations department would meet with them to draft the documents.

The following outcomes were measured: 1) willingness of patients assigned to the intervention group to listen to the script about end‐of‐life/life‐sustaining therapies; 2) opinions of patients about whether the information in the intervention was useful versus whether it was disturbing; 3) the frequency with which patients who had proactively received the information chose or changed their code status; and 4) the frequency with which patients without a preexisting advance directive created one while hospitalized. Simple proportions of each of these variables (ie, observed number divided by total number) in the intervention and control groups were compared using software that calculates the significance of the difference between two percentages (Statistica). The demographics of the patients were compared using the unpaired Student's t test. A P value of < .05 was considered statistically significant.

RESULTS

A total of 585 patients admitted to the Department of Medicine between December 2003 and February 2004 were randomized for the study. Patients were excluded if they had insufficient capacity (133) or if they were rapidly discharged from the hospital (155). Patients who were excluded tended to be more ill (APACHE 8.1 vs. 7.3, P = .06) and were more likely to die while hospitalized (8% vs. 4%, P = .04). A total of 297 patients were included in the study, 136 in the intervention group and 161 in the control group. Baseline characteristics were similar between the 2 groups (see Table 1).

Did Patients Who Received the Intervention Clarify Their CPR Preference?

Of the 136 patients in the intervention arm, 49 (36%) had explicit documentation of their code status on admission, compared to 55 of the 161 patients in the control group (34%; P = .7). Documentation included listing the CPR status in the admission orders or in a completed code designation form. After receiving the intervention, 125 of the 136 patients in the intervention arm (92%) clarified their preferences about CPR and mechanical ventilation.

Of the 49 patients in the intervention group who had documented CPR status on admission, 48 were listed as full code (both CPR and mechanical ventilation), and 1 was documented as refusing both CPR and mechanical ventilation. Of the 48 patients who were full codes, 3 stated they did not want CPR and mechanical ventilation under any circumstances after the intervention. Their preferences were subsequently documented as formal orders. The remaining 45 (94%) stayed full codes (see Figure 1).

Figure 1

Of the 87 patients in the intervention group who had no explicit documentation of CPR status on hospital admission, 76 clarified their preference and 11 did not. Of the 76 patients, 71 wished to receive both CPR and mechanical ventilation, and 5 wanted neither. The status of the latter as no code, no ventilator was subsequently documented in the medical record with the consent of their attending physicians. One of these 5 patients became increasingly ill during hospitalization, with reduced capacity, and family members later asked that he receive only comfort care.

Of the 161 patients in the control group, 55 (34%) had documentation of their code status (ie, to receive CPR if needed) in the admission hospital record. By the end of hospitalization, 1 patient requested no CPR and no mechanical ventilation, and 2 received comfort care with cessation of other active life‐prolonging interventions. Of the 106 without initial code documentation, 4 were later documented as being no code, no ventilator and 2 as being comfort care (see Figure 1).

DISCUSSION

This study demonstrates that most (95%) hospitalized medical patients welcomed the opportunity to provide prospective informed consent for CPR and mechanical ventilation. Although only a small minority (4%) opted out of CPR/mechanical ventilation, a majority (92%) of those who received the educational intervention chose to accept those therapies if required. This study also demonstrates that hospitalization can be one point‐of‐care where patients can consider and create advance directives. The results of this study are consistent with those of the SUPPORT group4 and other7 studies about patient interest in making choices on CPR. Our study suggests that physicians can elicit patients' wishes about and record formal orders on CPR around the time of hospital admission.

The default action has been to administer CPR and mechanical ventilation after cardiopulmonary failure or arrest, that is, patients receive these procedures unless they state explicitly that they do not want them. Unlike with all other invasive procedures, no national regulation mandates obtaining informed consent prospectively, when possible, for these treatments, because it is assumed that patients would want these therapies rather than the alternative (ie, death). Indeed, it is appropriate to perform lifesaving procedures in emergencies without consent if the patient lacks capacity and a surrogate decision maker cannot be contacted quickly. This clinical approach is consistent with medical ethics: to err on the side of life when a patient's wishes are unknown or unclear. Nonetheless, having a full code as the default action denies patients the opportunity to provide informed consent for these highly invasive procedures because there often is ample opportunity to ask their permission. If patient self‐determination is the categorical imperative of American medicine, then current practice violates that principle at the moment when it may be most important, that is, when a patient's decision about whether to risk life‐sustaining therapies could promote survival or prolong dying. Our study demonstrates that a simple interventionsimply askingpromotes a decision and therefore patient autonomy in most cases.

When patients have opted for life‐sustaining therapies that subsequently have been administered or when patients have received such therapies by default, physicians and patients can be left in 2 situations. In one outcome the patient retains capacity, and the dialogue about life‐sustaining therapies can continue between patient and physician. In the second, frequent scenario, the patient is incapacitated. Until patient capacity can be restored, the physician must work with surrogate decision makers and preexisting advance directives to infer a patient's wishes about continuation of life‐sustaining care. Our data demonstrate that hospital admission is one point‐of‐care at which patients can be offered and can complete, albeit in small numbers, advance directives.8 Previous work with our patients demonstrated that many patients misunderstood advance directives and the degree of effort required to create them.9 We reasoned that more patients might create advance directives if we offered the service for free during hospitalization. We were very surprised at how infrequently patients created advance directives in this study, although this finding is consistent with others in the published literature.8 It is speculated that hospitalized patients may feel too ill to exert themselves and/or are not psychologically prepared to consider end‐of‐life directives (ie, I came to the hospital to get better, not to consider what should be done when I'm terminal ). Some patients may not trust physicians to use advance directives reliably.4, 10

Our study had several important limitations. First, and most important, not all patients who were randomized were enrolled in the study. The most common reasons for exclusion were rapid discharge from the hospital and mental status change calling into question a patient's capacity to make end‐of‐life decisions. Nonetheless, it is only competent patients who can be engaged to decide these questions for themselves. Surrogates (ie, loved ones), guided by advance directives, are left to address resuscitation decisions for those lacking capacity. In addition, patients' predilections may change with time,11 especially as death becomes more imminent. However, insofar as many patients have several hospital admissions as they approach the end of life and are more likely to possess capacity to consider CPR decisions during early admissions, their choices can be recorded repeatedly over time (with each admission or even as status changes during an admission) to inform decisions if they develop incapacity. Little more can be done to enhance autonomy regarding CPR beyond repeatedly educating and asking, as disease and specific illnesses progress. It can be argued that this intervention had little real overall effectmost patients who would have received CPR by default did in fact want it when informed and asked. This is an ethically problematic position for two reasons: it neglects the right of patients to decide for themselves, and it potentially subjects the small group of patients who would reject CPR if asked to an unwanted risky procedure (ie, one that may prolong dying). Another limitation of the present study is that patients were approached by doctors‐in‐training with whom they had had no prior therapeutic relationship. Although it would have been optimal for patients to be approached by their primary care physicians, this was not feasible. Even if we could have convinced all of our medical staff members to implement the intervention, it is unlikely that all would have adhered to a study script, which is what enabled standardization of the information shared with patients. Some physicians may disagree with the script's content. But the goal of this study was not to determine if specific information would affect outcomes; rather, it was to determine if patients were receptive to discussing these issues and making proactive choices regarding life‐sustaining therapies during hospitalization for acute illness. It is possible that using different scripts delivered by different personnel, ideally the patients' own doctors, might have elicited even greater rates of consent and proactive decision making. Finally, the degree to which these results can be generalized may vary based on the population sampled. White and well‐educated patients are more likely to engage in end‐of‐life decision making than non‐White and poorly educated patients.9, 12

In conclusion, this study suggests that capable patients hospitalized for medical problems are willing to give informed consent for (or reject) CPR and mechanical ventilation in the event of cardiopulmonary failure. The approach of the study was very simple. It took roughly 510 minutes to inform patients and elicit their choices. Allowing patients to choose, rather than assuming that CPR is the choice of patients by default, strenuously honors patient autonomy. If these findings are replicated in larger cohorts and at different centers, there would be little justification for not informing patients about and asking them to choose their CPR preferences for each hospitalization. In the meantime, caregivers might consider the appropriateness of addressing these issues when they admit acutely ill patients to the hospital.