Code Status Assessment and Documentation in Inpatient Psychiatry: A Call for Increased Training

All psychiatry residents (N=49) at a single institution were invited via email to complete a 21-item, multiple-choice, anonymous, voluntary, Web-based survey in January 2019. A raffle entry for a $10 Amazon gift card was offered as a small incentive to complete the survey. Per the institution’s Human Research Protection Program, the project did not meet the definition of human subjects research, and thus no institutional review board approval was required. The survey queried residents about their demographics, training information, perceptions about the importance of CSAs in inpatient psychiatry, and barriers to completing CSAs. A 5-point Likert scale was used to assess the frequency with which they completed CSAs (always, most of the time, about half the time, sometimes, and never). A 5-point Likert scale was also used by residents to rate whether their training had prepared them to conduct CSAs in inpatient psychiatry and whether they thought this was their responsibility (definitely yes, probably yes, might or might not, probably not, and definitely not). Descriptive statistics were used to compare the responses.

Characteristics of the 23 residents (47%) who completed the survey are presented in Table 1. Ninety percent (N=20) felt it was their responsibility to assess psychiatric inpatients’ code status. Over 50% (N=13) noted that they never assess a patient’s code status within 24 hours of admission. Additionally, 61% (N=14) reported that they do not consistently document code status. Fifty-seven percent (N=13) felt they were probably or definitely not prepared to discuss code status with psychiatric inpatients.

Respondents noted barriers encountered when assessing and documenting code status (Table 2 and Box 1). When respondents were asked whether they knew the differences between do not resuscitate, do not intubate, and do not escalate (DNR/DNI/DNE), 17% (N=4) stated that they did not. Only one respondent (4%) felt that they definitely would know what to do if their patient lacked capacity to make code status decisions. Eighty-three percent of respondents (N=19) thought that it was extremely or very important for psychiatry residents to be competent in having discussions about code status. Seventy percent (N=16) reported receiving no formal training in how to address code status of psychiatric inpatients, and all 23 respondents indicated that they would benefit from more formal training.

Despite legal and ethical obligations to conduct CSAs for all hospitalized patients, the findings of this study support the existing literature suggesting that CSAs are underutilized among psychiatric inpatients (1–3). Nearly all psychiatry residents surveyed in this study (N=20) felt that it was definitely or probably their responsibility as psychiatry residents to assess psychiatric inpatients’ code status, yet most (N=14) did not consistently assess and document their patients’ code status.

Our survey suggests that this discrepancy can be partly explained by lack of training. The Accreditation Council for Graduate Medical Education currently requires that all residents—including psychiatry residents—learn to "communicate with patients and families … to assess their care goals, including, when appropriate, end-of-life goals" (8). Yet only a small minority of respondents in this study felt that their training definitely or probably prepared them to complete CSAs for psychiatric inpatients, and most reported receiving no formal training. Perhaps the most informative piece of data was that all 23 residents surveyed believed that they would benefit from more formal education on how to conduct CSAs for psychiatric inpatients. These data underscore the deficits in end-of-life training that have been consistently reported across other specialties (9), although such training has been understudied in psychiatry despite its relevance. In the only other known published study surveying psychiatrists on their CSA practices, 53% (N=8/15) of trainees did not feel well trained in this aspect of clinical care (7).

There are unique challenges in conducting CSAs for psychiatric inpatients. For one, concerns about decisional incapacity and loss of autonomy often arise and must be balanced with the ethical principles of beneficence, nonmaleficence, and justice. In particular, psychiatric conditions, such as severe depression and suicidality, mania, psychosis, catatonia, and delirium, may affect patients’ capacity to make clinical decisions in general and end-of-life care decisions in particular (10, 11). Clarifying these various factors can be a daunting task even in the best of circumstances and may be particularly trying in the acute setting of inpatient psychiatry. In our study, comfort level in assessing psychiatric inpatients’ capacity to decide code status was cited by 70% of residents as a barrier to conducting CSAs. There have been numerous published studies testing the efficacy of CSA skills training in various medical populations (12, 13), yet there are no such studies among psychiatric inpatients.

Barriers to CSAs related to knowledge deficits have been reported in other specialties (14), but no such studies have been conducted in psychiatry. The results of this study suggest that, in addition to training deficits, gaps in knowledge related to code status preference and cardiopulmonary resuscitation also interfere with respondents’ ability to conduct CSAs. For example, 17% of surveyed residents in this study were unaware of the difference between DNR/DNI/DNE, which is basic terminology used in CSAs. Additionally, a majority of respondents were unaware of next steps to take if a patient lacked capacity to make code status decisions. Finally, some respondents noted that CSAs for psychiatric inpatients felt irrelevant unless their patients were medically ill. A similar observation was made in the only other published study on this topic (7).

There are limitations to this study. First, generalizability was limited, given that the survey was part of a larger quality improvement project and was not designed to evaluate a research hypothesis. Additionally, this study was conducted in a small single-site sample. Our response rate of 47% is slightly higher than what is expected from the literature (15). This might suggest that our respondents had a special interest in this topic, which could have introduced selection bias. Finally, the open-ended responses (Box 1) suggest that some residents viewed conducting CSAs as irrelevant, and the absence of this as a response option in our survey was a limitation.

As psychiatrists, we have a legal, ethical, and clinical responsibility to conduct CSAs for psychiatric inpatients. This work adds to the small number of studies that show a deficit in CSAs in inpatient psychiatry, as well as a lack of training on this topic (3, 4, 7). We hope this study serves as a call to action for other institutions to develop quality improvement initiatives to examine and improve the frequency of CSA and documentation practices in their units and track quality of resident training on this important topic. In addition, greater attention is needed from the research community to study the evaluation of end-of-life wishes among psychiatric patients, as well as to research and develop specialized training for psychiatrists to conduct effective CSAs in this vulnerable population. The psychiatric and medical communities need to engage in conversation about this clinical issue and develop consensus guidelines on how to assess and document code status for psychiatric patients.

The authors thank Dr. Daniel Kim for his guidance and mentorship.