Mean improvement in weighted physician performance scores after feedback.
Vaidya VU, Greenberg LW, Patel KM, Strauss LH, Pollack MM. Teaching Physicians How to Break Bad NewsA 1-Day Workshop Using Standardized Parents. Arch Pediatr Adolesc Med. 1999;153(4):419-422. doi:10.1001/archpedi.153.4.419
To evaluate the effectiveness of a training program using standardized parents (SPs) to improve the performance of pediatric intensive care fellows in communicating bad news to parents.
Self-controlled crossover design.
Tertiary pediatric intensive care unit in a university-affiliated children's hospital.
Seven pediatric intensive care fellows and 4 trained volunteers (2 sets of SPs) participated in the study.
Two case scenarios of children admitted to the intensive care unit with a near-fatal diagnosis were used for the fellow's interactions with the SPs. The SPs had received 15 hours of training in role playing, performance evaluation, and giving feedback to the physicians. At the end of the first session, SPs provided feedback to the physicians under each of the 5 following categories: communication skills, content issues, support systems, interventions, and parent perceptions. During the second session, the parent meeting was repeated with a new but similar case scenario and a different set of SPs. Both sessions were videotaped, and a rater blinded to the order of the sessions used a weighted scale based on a checklist to score changes in physician performance.
The performance by the fellows showed a significant mean (±SEM) improvement in scores of 18.1 (±5.2) points (P=.007) between the first and the second sessions. Ranking of session scores revealed that physician performance improved significantly during the second session (Wilcoxon signed rank test, P=.002).
To our knowledge this is the first study that demonstrates short-term improvement in physician performance in conveying bad news in a pediatric intensive care setting using SPs in a 1-day workshop.
THE SUDDEN and unexpected admission of a previously healthy child with a potentially fatal illness to a pediatric intensive care unit is very emotionally stressful for parents.1 It is likewise stressful for physicians, who find it difficult to communicate this bad news to the parents.1- 4 Conveying bad news is a challenging yet important communication skill for physicians.5,6 Besides the moral and social obligation of conveying bad news in a sensitive and accurate manner, it is also a medicolegal responsibility of the physician to inform parents of a potentially fatal diagnosis. If handled improperly, communication of bad news can generate feelings of mistrust, anger, fear, and blame. It can have adverse long-term effects in the form of emotional consequences for the family.1,7 Despite the importance of this interaction, there are few reported formal training courses to teach residents or fellows how to convey this information.8- 10 Pediatric intensive care physicians frequently have to convey bad news to parents, yet there are no reports documenting that they receive any formal training in this area during fellowship training. We developed a 1-day workshop to train fellows in this area, with standardized parents (SPs) acting as educators and evaluating the short-term effectiveness of our intervention. To our knowledge, this is the first study to evaluate the usefulness of SPs trained in a 1-day workshop in improving the short-term performance of fellows in breaking bad news to parents.
All pediatric intensive care fellows (N=7) in a university-affiliated children's hospital participated in the study. There were 5 men and 2 women in the group; 2 were in the first year, 3 in the second year, and 2 in the third year of fellowship training. The study was conducted in a 1-day workshop, with 1 session in the morning and 1 in the afternoon. The 7 fellows were given a written summary of one case scenario pertinent to the assigned SPs (see below). Thirty minutes were allotted for each session. All sessions were videotaped, but unsupervised. At the end of each session, the SPs discussed the performance of the fellows and then reviewed the evaluation with the fellow to provide feedback and suggestions for improvement. Videotape segments were reviewed when indicated.
Four volunteers were selected on the basis of their personal experiences as parents or professionals in pediatrics and were trained for 15 hours to act as 2 sets of SPs. The SPs and the fellows did not know each other prior to the study. The 2 sets of SPs were each assigned to study one case scenario. First, they were trained to respond to physician questions about "their" child and to ask relevant medical questions. The training focused on teaching the SPs to react in a manner similar to actual parents in the pediatric intensive care unit. The SPs were able to effectively simulate feelings of anger, frustration, denial, self-blame, and grief. As a part of the training sessions, we held simulated parent-physician meetings with the SPs, which were videotaped and discussed to further enhance their acting abilities as SPs. Second, they were trained to evaluate the performance of physicians conveying bad news and to provide feedback. They were taught to observe verbal and nonverbal physician actions during the meeting, to summarize the performance, and to give suggestions for improvement at the conclusion of the meeting. The SP training was supervised by 1 of us (L.W.G.) who has extensive experience with SPs. Training was considered adequate once both sets of SPs consistently demonstrated effective role-playing and were able to evaluate physician performance and provide feedback in a consistent and reproducible manner. Both sets of SP responses were comparable in these practice sessions.
The case scenarios selected for role playing in the study represented realistic and common scenarios that have potentially fatal outcomes. The first case scenario involved a previously well 6-month-old infant with severe pneumococcal meningitis, intractable seizures, and evidence on a computed tomographic scan of diffuse global infarction and herniation. The second case scenario was of a 13-year-old girl involved in a severe motor vehicle accident. She suffered cardiac arrest at the scene, intractable increased intracranial pressure, multiple skull fractures, and herniation of brain tissue through the ear.
In a randomized fashion, half of the fellows presented case 1 to the first set of SPs during the morning session, while the other half presented case 2 to the second set of SPs. The groups were switched for the afternoon session. Physician performance evaluation was based on 26 items (Table 1) assessed during parent-physician interaction. We decided to use a single rater (L.H.S.) who was blinded to the order of the sessions and viewed the videotaped recordings to rate physician performance. The rater knew the fellows, but was also a key player in informing parents about their child's illness in the pediatric intensive care unit, ie, she was very familiar with the standard criterion for conveying bad news. The 26 items were grouped into 5 categories: communication skills, content issues, support systems, interventions, and the Patient Perception Questionnaire. Three independent observers were asked to rank each of the 5 categories in order of their importance and to rank each item within a category according to its importance. The 3 rankings were averaged and normalized over and within categories. These normalized weights were used to obtain a weighted rater evaluation score. All scores are expressed as y=(x × 100), where x indicates weighted score. Each physician acted as his or her own control. The performance of all pediatric intensive care unit fellows together before and after receiving feedback was used to calculate the mean change in score after feedback. Changes in scores were analyzed for each of the 5 categories and also for a combined overall performance of all 5 categories. Statistical analysis was done using the paired t and Wilcoxon signed rank tests.
Although all of the fellows had communicated bad news to parents in varying degrees during the course of their training, none had received prior formal education in this skill. Table 2 andFigure 1 show the mean improvement in weighted physician performance scores from the first to the second session with the SPs. The 5 categories are shown separately, and then combined for an overall score in performance. When each category was analyzed separately, there was a statistically significant improvement of scores during the feedback session in the Patient Perception Questionnaire category (all values are given as mean [±SEM]): change in score, 35.2 [±14.4] points, paired t test, P=.02; Wilcoxon signed rank test, P=.005). Evaluation of the overall performance in all 5 categories combined revealed a highly significant improvement in mean scores of 18.1 (±5.2) points (paired t test, P=.007; Wilcoxon signed rank test, P = .002).
Interpersonal communication skills are essential for all physicians to facilitate the communication of bad news to patients and parents.1 Parental reactions on hearing the news that their child is likely to die can be devastating. It is important that this information be conveyed in an accurate, yet compassionate and caring manner. If handled improperly, it can result in short-term miscommunication and long-term emotional consequences for the family.11,12 Good interpersonal skills help improve the patient-physician relationship and have been shown to influence patient satisfaction, compliance, and even health status.13- 15
Studies that address the issue of communicating bad news need to answer 3 fundamental questions: how should bad news be conveyed, what content issues should be included, and how should physicians be trained to acquire the necessary skills. Many published articles address the first 2 questions and provide guidelines on content issues and the appropriate manner of conveying bad news. Most of these, however, are based on opinions of physicians, nurses, social workers, or parents; some are consensus guidelines, but very few are based on randomized controlled studies.5,16,17 Each parent-physician interaction is unique and the guidelines should be adjusted to meet individual circumstances. The third question, regarding teaching communication skills to physicians, has not been adequately addressed in the medical literature. Physicians need to receive formal training in this area to develop good interpersonal skills. Most medical training programs do not have a structured format for teaching communication skills.8- 10 Some programs have made an effort in this direction by adopting varied teaching strategies to educate physicians in communicating bad news.18- 20 These methods include didactic sessions, audiotape and videotape recording of parent-physician interaction, and exposure to real-life experiences.21 Whereas lectures and select readings provide theoretical knowledge in this area, providing simulated opportunities that are the same for each trainee allows transformation of theory into practice.
In our study, we explored the usefulness of SPs to train physicians in communication skills in the setting of a 1-day workshop. Categories of evaluation (Table 1) used in this study to rate physician performance were based on those used in previous studies.19,20,22,23 The Patient Perception Questionnaire (Table 1, category 5) has been developed and used by the National Board of Medical Examiners in interactions between SPs and physician candidates for licensure and has been shown to be a valid and reliable instrument.24 The results of our study showed improvement in physician performance in all 5 categories in the feedback session, although it was statistically significant only in the Patient Perception Questionnaire (improvement, 35.2 [±14.4], P=.018) (Table 2, Figure 1). The combined score for all 5 categories showed a statistically significant improvement in the postfeedback session (improvement, 18.1 [±5.2] points, P=.007). These results demonstrate that SPs can simulate real-life experiences and can be used successfully to teach physicians essential interpersonal communication skills. A major limitation of the study is the small number of trainees who participated. We recognized this barrier before the study, but believed that the potential value of the process overrided this concern, especially if we were able to demonstrate changes in the trainees after training. The cost of the study was kept to a minimum by using volunteers as SPs.
To our knowledge, using a 1-day workshop to detect short-term improvement in performance after receiving feedback from SPs has not been reported previously in the literature. The 14 physician-SP sessions were believed to be the maximum number possible for a 1-day workshop. Previous studies by Greenberg et al19,23 have shown that SPs can be used successfully to educate trainees in improving skills in conveying bad news. The trainees rated the use of SPs as a highly effective tool in learning these skills. Kent et al25 showed that students tend to retain impressions and information obtained from patients more vividly than from theoretical classroom learning. Ideally, for any educational strategy to be considered effective, it must produce both short-term and long-term measurable changes in knowledge and performance. Studies have shown that the benefits of feedback training in patient-physician communication skills can be retained and have positive long-term effects.26 We were able to show beneficial short-term changes in performance following feedback from SPs. Repeated evaluation of communication skills over a longer period would be necessary to study the long-term benefits of this program. Our study did not include plans to determine if fellows maintained or improved their skills over the long-term.
Using SPs to teach physicians how to communicate bad news resulted in a short-term improvement in performance. Standardized parents can be used successfully to train physicians in this area.
Accepted for publication October 22, 1998.
Dr Vaidya was a fellow when this study was conducted.
This study was supported in part by an educational grant to the Children's National Medical Center from the Joshua Stouck Memorial Fund, Potomac, Md. The workshop was carried out as a part of the annual Joshua Stouck Memorial Educational series.
Presented at the 10th Annual Pediatric Critical Care Colloquium, Little Rock, Ark, September 19, 1997; the 36th Annual Research in Medical Education Conference, Washington, DC, November 4, 1997; and the 1998 Annual Meeting of the Pediatric Academic Societies, New Orleans, La, May 3, 1998.
We thank the Stouck family and the volunteers who acted as standardized parents.
This study is dedicated to the memory of Leslie H. Strauss, MSW, LICSW, who died shortly after the completion of this study.
Reprints: Larrie W. Greenberg, MD, Director, Office of Medical Education, Children's National Medical Center, 111 Michigan Ave NW, Washington, DC, 20010 (e-mail: firstname.lastname@example.org).
Editor's Note: The sample size is small and the study shows only short-term success, at least that's all that's been measured thus far. However, the concept is sound, and we need all the help we can get in teaching physicians how to communiate bad news.—Catherine D. DeAngelis, MD