Confessions of Physicians: What Systemic Reporting Does Not Uncover

It's July, and the newest group of anesthesiology residents is sitting around the conference table. With clear memories of previous failed efforts to get them to “open up,” the program director is trying something new. Sixteen pieces of folded paper are arranged in a pile, one for each resident in the room. Each paper contains a typed statement, and there are no names attached to these “confessions.”

The papers are shuffled and dealt, poker-style, to the group. Participants are reminded, “Even if you get your own statement, do not reveal the authorship, as anonymity is critical to the session.”

The first statement is read:

The facilitator decides to remain silent for a few moments. After an uncomfortable pause, a resident wonders aloud, “Who cleans the equipment in between cases? How do we know when we have to do it, or when the technician is responsible?” Another resident responds confidently, “I just do it. I mean, my Mom could be the next patient . . . why take any chances?” The conversation lulls and the next resident reads:

A conversation regarding situational awareness and vigilance ensues. With each confessional statement, the tension that marked the outset of the session dissipates further. The next confession is read:

What this new approach to resident conversations is trying to achieve is to overcome the stigma of reporting and discussing incidents and situations that negatively affect the quality and safety of care. The most important barrier to improving safety is the lack of a standardized metric of adverse events and incident reports. A few programs have been successful at increasing reporting in certain disciplines¹  and globally,²  although more research is clearly needed.³  Since the overall incidence of poor-quality outcomes is low, the use of such metrics as report cards and insurance claims to benchmark specific concerns tends to be underpowered. Consequently, there is a general dearth of data that could bring about systematic improvements in patient safety.

The Institute of Medicine drastically “changed the conversation”³  regarding error reporting by shifting the focus to systems rather than on individuals. In doing so, the goal is to more readily engage all stakeholders, including health care workers, to become part of the solution. Hospital staff is constantly encouraged to report errors and near-misses in event-reporting systems. Although US health care institutions have begun promoting transparency to improve the safety of care, there is no widespread achievement of this goal. Current hospital-based reporting systems are profoundly underused by physicians compared with nurses and other staff⁴  (1.1% versus 45.3% and 53.6%, respectively). A perceived threat to autonomy and fear of reprisal reflect some of the impediments to establishing accountability and accomplishing effective incident reporting. The 2012 Agency for Healthcare Research and Quality Hospital Survey on Patient Safety Culture revealed that nearly 50% of staff felt that there is a nonpunitive response to error: only 44% reported that their mistakes and event reports were not held against them, and that information on these events was not kept in their personnel file.⁵

With a lack of measures reflecting the incidence and quality of patient safety events, and in the face of poor reporting mechanisms, the time is ripe for adopting innovative approaches to this problem. Perhaps we need to move away from the systems approach and get back to the individual, after all.

Back at the conference table, these confessions are shared:

The more statements recited, the clearer it became that the newly minted Confessions program had far-reaching implications for the clinical learning environment. Quality improvement, on-boarding management, patient safety concerns, and a host of other issues arose from the first session. For example, the wrong drug (ephedrine) injected statement uncovers a near-miss that would otherwise not have been reported.

We started the Confessions program in 2010. In prior years, we tried other ways to understand resident stressors, including group discussions, to ease anxiety associated with the transition from medical school to residency. Michael Balint had the same thought in the 1950s when he proposed facilitated, small group discussions of the physician-patient relationship. There were ground rules: no notes were taken and all discussions were confidential. Balint's wife, Enid, a social worker and cofacilitator of these groups, commented, “… the essence … has always been to share experiences and enable people to observe and rethink aspects of their relationships with patients and their work as doctors.”

Early references to confessions exist in the Bible (Numbers 5:7). To evoke the term “confessions” seems to imply a heightened honesty and transparency to the shared information. The anonymity of confession statements adds to the residents' willingness to contribute candid vignettes. Over the years, Confessions sessions have succeeded when (1) sessions were led by experienced facilitators; (2) a safe environment was ensured for the sessions; (3) ground rules were clearly defined and not violated; and (4) statements that warranted intervention resulted in clinical learning environment improvements, and quality and patient safety improvements, with measurable outcomes that were reported back to the participants. See the box for our rules for the sessions.

Our experience has provided us with some specific “dos and don'ts.” We have found that the ideal group consists of 8 to 12 participants, and the optimal setting allows for everyone to sit in a circle. Allotting about 1 hour for each session seems to be enough time, and we try our best to get through everyone's confessions. The facilitator should try not to hijack the conversation and is advised to act more as a guide and to not be overly authoritarian. The facilitator also should take notes to ensure that necessary and timely follow-up occurs. For example, a confession about a drug error stimulated a conversation among the residents about how drugs are located in the storage cart. Rather than blame the confessor about the drug error, a systemic problem was sought and a recommendation was made to reorganize the drug drawers. The facilitator followed up with the pharmacist and then reported back to the group that the desired change had occurred.

We also have met and overcome some challenges in holding these sessions. For example, in one program, a designated facilitator jeopardized the group trust when he stated, “You guys are in residency; it's supposed to be challenging. Stop whining about it.” He then proceeded to exit the room. The residents felt betrayed, stating they were encouraged to confess their fears, and were assured that there would be no negative responses. This experience threatened to derail the safety and trust in the environment. In another program, when difficulty initiating conversations was noted, a facilitator slipped a “fake confession” into the pool. The planted confession was regarding the false documentation of heart and lung auscultation when it had not actually been performed. This particular issue had arisen with a previous group of residents, so the program director believed it would help stimulate frank dialogue—which it did.

Our use of Confessions sessions has provided an opportunity to reflect, discuss, and admit without fear of punitive actions. It allows for early intervention on the issues that are relevant to physician trainees. If used effectively by program directors, and by department chairs and division leaders with faculty, the Confessions program could guide modification of patient care and further quality care efforts in focused and specific ways that are currently not being addressed in any other manner.