This study was funded by Dayton Area Graduate Medical Education Consortium Resident Research Grant, Spring 2007, total award $1000.
Medical Malpractice: Utilization of Layered Simulation for Resident Education
Version of Record online: 8 JUL 2008
© 2008 by the Society for Academic Emergency Medicine
Academic Emergency Medicine
Special Issue: Proceedings of The 2008 AEM Consensus Conference: The Science of Simulation in Healthcare: Defining and Developing Clinical Expertise Guest Editors:Amy Kaji, MD, PhD David C. Cone, MD
Volume 15, Issue 11, pages 1175–1180, November 2008
How to Cite
Schlicher, N. R. and Ten Eyck, R. P. (2008), Medical Malpractice: Utilization of Layered Simulation for Resident Education. Academic Emergency Medicine, 15: 1175–1180. doi: 10.1111/j.1553-2712.2008.00165.x
- Issue online: 3 NOV 2008
- Version of Record online: 8 JUL 2008
- Received February 28, 2008; revision received April 16, 2008; accepted April 17, 2008.
- patient simulation;
- graduate education;
- legal liability;
- legislation and jurisprudence;
- medical errors
Objectives: The authors present a novel approach to the use of simulation in medical education with a two-event layered simulation. A patient care simulation with an adverse outcome was followed by a simulated deposition.
Methods: Senior residents in an academic emergency medicine (EM) program were solicited as simulation research volunteers. Other than stating that the research involved adverse outcomes, no identifying information was given. Seven volunteers participated in a simulation involving a forced error (nurse confederate gave an incorrect medication dose). Six weeks later based on the initial simulation, one physician completed a simulated deposition in a teaching conference conducted by a licensed attorney with malpractice experience. The audience, consisting of residents, attendings, and students, watched a recording of the patient care, witnessed the deposition, and evaluated the experience using a 17-question survey with 5-point Likert scales.
Results: Participants felt that overall the training program was a useful educational tool (mean ± standard deviation [SD] Likert score = 4.63 ± 0.49) that would change aspects of their practice (3.31 ± 0.85). Participants stated that they would be more careful in their documentation (3.88 ± 0.60), review high-risk situations with staff (4.00 ± 0.71), and monitor more carefully for errors (3.95 ± 0.74). There was increased fear of the litigation process (3.95 ± 1.18), but participants felt the experience would help improve the risk profile of their practices (3.71 ± 0.68).
Conclusions: A novel approach to medical education was successful in changing attitudes and provided an expanded educational experience for participants. Layered simulation can be successfully incorporated into educational programs for numerous issues including medical malpractice.
Emergency medicine (EM) remains a highly litigated specialty. Physicians practicing in emergency departments (EDs) across the country identify fear of litigation as an influential force in both the areas they practice and the decisions they make regarding patient care.1“Defensive medicine,” the practice of ordering additional tests for litigation prevention, has only grown with this increasing fear of the medical liability system.2
Physician education regarding medical malpractice has generally been confined to lecture-based discussions and case reenactments. In two studies, case reenactments were utilized involving medical students3 and health care management students.4 These studies recreated previous malpractice cases that were not based directly on the participants’ medical care. Simulation has gained widespread utilization in medicine, but we were unable to find any reports of high-fidelity human patient simulators being used in medical malpractice education. Historically, depositions and other case reenactments were based on role-playing from written case material. This study evaluated a new educational model. We hypothesized that the performance of a deposition based on the care provided by the physician in a simulated forced-error patient encounter would be an effective method of resident education, as judged by the participants.
This was a simulated physician–patient encounter on a high-fidelity simulator (SimMan, Laerdal Medical, Wappinger Falls, NY) with a planned forced-error performed by a standardized nurse participant. The protocols for the simulated patient encounter, three separate informed consent forms for all stages of the simulation, and the research questionnaires were approved by the Wright State University Institutional Review Board. Participants at each stage signed an informed consent document before participation in any part of the research.
Study Setting and Population
The study was conducted in conjunction with a 3-year accredited EM residency training program involving residents and attending faculty physicians from participating teaching hospitals. All participants were EM-affiliated. The simulated patient encounters were conducted in the department’s Center for Immersive Education and Research. The deposition was performed as part of the regularly scheduled protected resident curriculum.
Resident physician participants in the simulated patient care portion of the study were solicited from the 14-member senior class of the residency program. An e-mail was sent to all potential participants requesting study volunteers for a simulation research project. All residents attending the regularly scheduled conference were asked to participate in the deposition portion of the research.
The study protocol involved multiple layers of simulation, participation, and informed consent. Seven of the 14 members of the senior resident class responded affirmatively to the request for study participants. No information was disclosed at that time regarding the intended forced error, the research objectives of medical malpractice education, or the later deposition. All 7 resident physicians participated in the simulated patient encounter of an anaphylactic shock patient in extremis. The nurse participant was instructed before the simulation to administer 1 mg of 1:1,000 epinephrine intravenously if any form of epinephrine was ordered, regardless of the requested dose, concentration, or route. The simulation was programmed to end in the demise of the patient regardless of care provided. All 7 participants were then asked to dictate a medical record that was transcribed, reviewed by them, and signed. An observer to the simulation recorded a code sheet, which was verified with the simulation computer.
After the simulated patient encounter, a single physician participant was selected to be deposed in the second stage of the study. After a planned latency period, a live deposition of a participating resident physician was conducted based on the care provided. Audience participants were shown a videotape of the simulated care before the deposition. All participants were then surveyed anonymously regarding the benefits of the simulation, how it impacted their perceived risk, and their ability to decrease their future risk.
The participant was selected based on the lack of clarity in his order for epinephrine. This ensured shared responsibility between the nurse with the planted forced error and the physician who would be deposed. The physician was informed of the intent of the study, and the plan for a deposition to be performed as part of the residency conference. Informed consent was again obtained. A copy of the medical record and code sheet was provided to the physician for review before the deposition.
One month after the simulated clinical encounter, the deposition was performed during a scheduled resident conference. A resident physician and licensed attorney with experience in medical malpractice (NRS) performed the role of plaintiff’s attorney. A faculty member selected by the simulated defendant physician served the role of defense attorney. The deposition lasted 1 hour, after a viewing of the 15-minute patient encounter.
After the deposition, the audience members each anonymously completed a 17-question survey. The first 12 questions used a 5-point Likert scale with a range from strongly agree (5) to neutral (3) to strongly disagree (1). These questions were utilized to evaluate the effectiveness of the training as an educational experience, impact on behavior change, and opinions about the malpractice system. The remaining 5 questions evaluated the audience perception of liability of the parties and accuracy of the physician’s recollection.
The statistical consulting service of the sponsoring university collated and analyzed the data. Statistical averaging of the scores on the Likert scale and percentage questions was performed.
Characteristics of Study Subjects
There were 48 participants in the study. The distribution by discipline, gender, and level of training is shown in Table 1. Of 40 residents eligible to participate, 34 participated. As a mixed military and civilian residency, the mean number of years in medical practice for all participants was higher than what would otherwise be expected for a residency program, with a mean (± standard deviation [SD]) of 6.6 (±0.43) years.
The mean overall satisfaction based on the Likert scale was 4.63 (±0.49), with no rating below “agree (4)” on the satisfaction questions, as shown in Table 2. Respondents felt that their understanding of malpractice was enhanced by the experience, with a mean score of 4.54 (±0.60). The educational experience demonstrated the importance of accurate documentation (4.54 ± 0.67). There was an increased fear of the litigation process (3.95 ± 1.18); however, respondents noted that the training would help them change their practice patterns (3.31 ± 0.85). Overall, respondents felt that the experience would help them improve the risk profile of their practice.
|Question Asked||Mean ± SD||95% CI|
|Useful training experience||4.63 ± 0.49||4.48, 4.78|
|Understanding of medical malpractice improved||4.54 ± 0.60||4.35, 4.72|
|Demonstrated accurate documentation need||4.54 ± 0.67||4.33, 4.74|
|Will not discuss cases with colleagues||1.98 ± 0.82||1.72, 2.23|
|Result in change in practice||3.31 ± 0.85||3.06, 3.58|
|Increased fear of litigation||3.95 ± 1.18||3.59, 4.31|
|Improve the risk profile of practice||3.71 ± 0.68||3.50, 3.92|
Impact on physician risk management behavior was evaluated in the Likert questions shown in Table 3. Participants stated that it would help them change their practice, including more careful documentation (3.88 ± 0.60), review of high-risk situations with staff (4.00 ± 0.71), and monitoring more carefully for errors (3.95 ± 0.74). When asked if they would no longer discuss difficult cases with colleagues, the mean response of 1.98 (±0.82) indicated that the experience had not dissuaded them from sharing their difficulties.
|Question Asked||Mean ± sd||95% CI|
|Review high risk incidents with staff||4.00 ± 0.71||3.78, 4.22|
|Improve documentation||3.88 ± 0.60||3.69, 4.06|
|More cautious in medical practice||3.51 ± 0.95||3.22, 3.80|
|Limit discussion to protected locations||3.27 ± 1.12||2.93, 3.61|
|Monitor for errors more vigilantly||3.95 ± 0.74||3.72, 4.18|
The study also examined the effectiveness of the simulated case in creating a realistic medical malpractice claim, legally referred to as an “issue of fact.” Given the variable nature and inability to predict the outcome, there was no guarantee that the physician would share in the alleged negligence as it was designed with a forced error by the nurse. Without informing the physician of the intent of the study and decreasing the realism of the scenario, there was no assurance that the physician would take an action that would be construed as potential negligence. Overall, 78% of the audience felt that malpractice had occurred and would have found for the plaintiff if the case had gone to trial. The audience split liability between the care providers, with a mean liability of 62% (95% confidence interval [CI] = 55.5% to 68.44%) for the nurse and 38% (95% [CI] = 31.5% to 44.1%) for the physician.
To further evaluate the efficacy of the simulation in reproducing a deposition consistent with a real case, the issue of accuracy of recollection was evaluated. Generally, the latency period from incident to deposition is measured in years. In the present case, we separated the simulated care and deposition by 1 month. Most witnesses have limited recollection independent of the medical records, and there are often sections of the case that are recalled differently among testifying participants. The audience, after reviewing the care provided on videotape and witnessing the deposition, felt that the deponent’s recollection was on average only 48% (95% CI = 40.41% to 55.93%) accurate. This reflects the discordant testimony of witnesses in a real case.
Medical malpractice has been an issue in the practice and education of physicians over the past 30 years since the first “malpractice crisis” in the 1970s. The practice of medicine has changed drastically over that time, as has the litigation of its practitioners. There have been arguably three major malpractice crises. Many argue that the most recent crisis is still in need of legislative remedy.5–7 As these malpractice crises have blossomed, the associated fear and resulting changes in practice have been well documented.8 Studies have reported that fear of litigation among physicians has achieved rates as high as 66%.9 Our internal survey results were consistent with these rates. Respondents expressed a fear of litigation and estimated a mean 60% likelihood of being sued in their career.
The impact of this crisis has been felt throughout medicine. Effects have included increased litigation, defensive medicine practices, and other changes in physician practice patterns. As documented in the Missouri malpractice crisis, shortages have begun to develop in life-saving specialties.10 This loss of specialists is not limited to attending physicians. There has been documented concern that new resident graduates would be in part determining their practice location based on malpractice risk.11
Tort reform has been proposed as the solution to the problem from a number of sources.12,13 From President George W. Bush’s State of the Union Speech in 2003, to a multitude of House and Senate Bills, the issue of tort reform has been a political issue for the past 5 years.14 This issue has been addressed in a number of legislatures, with a peak of 36 separate states considering various forms of tort reform in 2006.15
Despite the importance of these issues in the practice of medicine, resident decision-making regarding future employment, and the need for physician participation in legislation reform, little has been done to enhance the education of resident physicians in the area of malpractice. The mainstay of malpractice and legal training in residency remains based in morbidity and mortality conferences and lectures on legal constructs.16 The call for improved education in medical malpractice has been longstanding, but without specific revisions suggested for the educational system.17 A separate study examined a new, well-received approach to education through a one week rotation in the office of a medical liability insurance company reviewing claims and cases involving emergency physicians.18 Other educational groups have used mock trials for graduate training.
In this study, we introduced a previously untested approach to medical malpractice education utilizing simulated patient care followed by a simulated deposition. The data suggests that this educational model achieved the goal of fostering the antecedents that are needed to change physician behavior. The positive effect on physician beliefs and attitudes regarding communication, documentation, and error monitoring are evident in our study results. Audience participation and attendance throughout the process was high, with 85% of residents participating in a voluntary educational experience.
Behind the issue of medical malpractice litigation and defensive medicine are the behaviors of the physician defendants that increase the risk of litigation. With estimates from the Institute of Medicine’s study showing errors resulting in as many as 98,000 patient deaths per year, there is reason for concern.19 It is likely impractical and impossible to completely eliminate errors from the practice of medicine. However, through education and case experience, we were able to show a change in attitude to improve vigilance for monitoring to avoid errors. There was also a noted improvement in attitudes toward debriefing with staff after high-risk situations. That can improve both the feelings of participants and the documentation of the encounter. By addressing these issues, malpractice litigation is less likely than in cases where poor documentation and angry feelings are present.
With any study that attempts to change physician behavior and implement an effective educational tool, there will be a limit on what can be studied. In the present case, the study population indicated a desire and plan to change behavior. However, there was no effective means to evaluate whether a true impact was accomplished due to a combination of factors including cost, resident variables, and multiple facilities in which residents practice. Faculty and residents have reflected over the past year that they have improved their documentation and hope they have improved the risk profile of their practice. As some of these residents began moonlighting shifts as attending physicians, some have recalled the malpractice education experience and stated it has had a positive impact on their behavior. Since the training, a vibrant medical malpractice and quality improvement forum has continued in the residency curriculum, with multiple residents leading discussions on cases in which care could be better optimized. Further study could help to validate this simulation model as an effective tool in changing behavior as well as attitudes.
There was concern when the study was developed about the possibility of an undesired impact on the medical practices of the participants. The issue of medical malpractice liability has been well documented to increase fear of litigation in physicians as discussed herein. The results did demonstrate increased fear and a heightened perception of risk of being sued and the malpractice issue. However, this fear was balanced with an improvement in attitudes that can lower the risk profile of the participants’ medical practice and improve vigilance toward monitoring for errors. Although ideally any educational model would not increase physician fear, the reality is that the topic is laced with highly charged emotions. Given the overwhelming response that the experience was highly educational and useful, the authors believe that this increased fear is balanced out by the positive impact.
One of the difficulties with simulation is making the experience a realistic reproduction of a real-life scenario. In our study this difficulty was compounded by creating a simulated deposition layered on a simulated patient care encounter. To make the deposition realistic, the underlying patient care had to include an issue of possible malpractice. The nurse confederate was predestined to make an error. However, to make the deposition experience more realistic, some aspect of the physician’s performance needed to be questioned to involve him or her in the potential liability. Four of the physicians participating in the patient care simulation did not specify a complete order (drug, dose, concentration, route, frequency) or made other decisions in management that could potentially be construed as erroneous. This created an accurate basis for the simulated deposition as an effective teaching tool. This was demonstrated in the study by the audience response that a majority would have found an issue of malpractice and split liability between the nurse and physician.
What could not be quantified in the study was the ability to vest the participants in the experience. Although only subjective from the researchers’ perspective, the response of the defendant physician was realistic and consistent with that of a physician testifying in her or his own defense case. After the deposition, there was an excited exchange between the nurse and physician that was not a planned part of the simulation. The audience perception that they thought the physician was less than 50% accurate in his recollection reflects the generalized decline in knowledge retention experienced in most malpractice cases. This degree of inaccuracy reflected the personal investment of the parties and the realities of delayed recollection.
The reaction of the other providers to having a medical error in a simulation was likewise profound and emotional. Most simulations in our educational venues end with the patient surviving or there being an unavoidable outcome. In this case, for the first time in their careers for most participants, the physician care providers were confronted with a real-time medical error and forced to cope with it while continuing to attempt to resuscitate the patient. Reactions varied from anger toward the nurse to shock and surprise. In debriefing the participants afterward, remarks included a statement by one member who explained her concern that even though it was a simulation and the error was a forced part of the event, it revealed a gap in her knowledge that she did not know was there.
This degree of realism and investment in the process reveals the value of simulation as an educational tool that cannot be quantitated well in a study. While all physicians hope never to be faced with this situation, the reality is that errors will be made while we care for patients, and we will have to deal with cases that are lost when we do not believe they should be. This experience, employing simulated patient care, while arguably traumatic, provided a practice opportunity for one of the most difficult and life-changing experiences of a physician’s career. Through the participation in this study, the physicians had an opportunity to evaluate and improve their own skills at dealing with adversity, errors, and malpractice in a safe and controlled environment that would otherwise not be available to them.
The major limitation of the study was an issue of sample size. A single residency program was utilized with a total of 48 participants. This limitation reduced the ability to draw inferences from subgroup analysis, although trends did appear in the data. The layered nature of the process, temporal separation of the study events, and the time requirement limited our ability to conduct the study at multiple sites. Further expansion of the study population in the future would help to address these issues.
The size of the survey, limited to 17 questions for time-related issues, compromised the ability of the survey to internally validate results. Although questions could be grouped to evaluate general topics, the size of the survey limited the ability to repeat questions utilizing different terms to evaluate for consistency of the results. Future studies, separate from tightly scheduled residency education, would resolve study questionnaire size and time limitation issues.
Finally, with any simulation, there is the question of reproducibility. Given the planned nature of the error and frequency of shared responsibility based on our seven participants, generating the basis for the case is possible, although not assured in every simulation. The other reproducibility limitation is having an attorney willing to participate in the simulation. In our case, the first author and resident researcher is an actively practicing lawyer, which made recruitment easier. However, the role of the lawyer could realistically be performed by a community lawyer without difficulty.
With the utilization of simulation, we were able to develop a new approach to medical malpractice education that appears to have been effective as an educational tool and achieved in the participants an intention to change their own behavior. The approach was well received and the participants felt it was beneficial. The experience brought a degree of participant investment and realism to the experience that was previously unobtainable. Further integration of simulation into traditionally lecture-based educational components of medical education using novel approaches, including multilayered simulation, should be explored.
This study was supported by the Dayton Area Greater Medical Education Consortium through grant funding and the sponsoring emergency medicine residency. Without their assistance, this study would not have been possible or practical. Their generosity is greatly appreciated. In addition, a special thanks to Dr. Kirk Hinkley and Dr. Michael Ballester for their participation in the simulated deposition process.
- 2Malpractice: how fear changes practice. Med Econ. 2005; 82:80–4..
- 6Weiss Ratings, Inc. The Impact of Non-Economic Damage Caps on Physician Premiums, Claims Payout Levels, and Availability of Coverage. Available at: http://www.weissratings.com/malpractice.asp. Accessed Feb 21, 2008.
- 7Our nation’s liability system threatens patients’ access to health care. Excerpted from Dr. Nelson’s testimony before the House Government Reform Subcommittee on Wellness and Human Rights. October 1, 2003..
- 8Living in fear: physicians face mounting liability insurance crisis. Tex Med. 2002; 98:35–40..
- 14The imposition of federal caps in medical malpractice liability actions: will they cure the current crisis in health care? Akron L Rev. 2004; 37:417–68..
- 15National Congress of State Legislatures. Medical Malpractice Tort Reform. http://www.ncsl.org/standcomm/sclaw/medmaloverview.htm. Accessed Feb 12, 2008.
- 19Institute of Medicine. To Err is Human: Building a Safer Health System. Washington, DC: National Academies Press, 1999.