Hospitalists must create rapport and communicate large amounts of information in a short amount of time without having a prior relationship with the patient.1 High-quality communication can improve satisfaction and compliance, while poor communication leaves patients ill prepared to transition back to the community.2–10
Many medical schools use standardized patients (SPs) to both train and evaluate their students’ communication skills. To our knowledge, no published studies describe using SPs to assess or teach communication skills for hospitalists.
Our objective in this study was to use SPs to assess for deficits in our hospitalists’ communication skills and to determine whether feedback provided by SPs could improve hospitalist confidence in and performance of optimal communication behaviors.
Setting and Participants
Scenario and Checklist Development
We developed 3 SP encounters around common hospitalist-patient interactions: daily rounding, discharge, and interacting with a difficult patient. In order to assess communication skills, we developed a checklist with 3 core domains: Courtesy and Respect, Listen, and Explain. Each domain corresponded to 1 of 3 questions on the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey that pertained to doctor’s communications skills: (1) How often did doctors treat you with courtesy and respect? (2) How often did doctors listen carefully to you? (3) How often did doctors explain things in a way you could understand? We then developed checklist items that corresponded to essential communication skills within each of the 3 domains. These communication skills were based on best practices and published literature.
Discharge Encounter (Table 2): Patient admitted the night before with abdominal pain by another hospitalist. The checklist was based on AIDET®, an effective communication skills training protocol that our hospitalist group had been trained on.11
Daily Rounding Encounter (Table 3): Patient being discharged after an admission for congestive heart failure. The checklist was developed from the Society of Hospital Medicine discharge toolkit.12
Difficult Patient Encounter (Table 4): A patient and his daughter who were unhappy because of a previously missed lung mass that was now found to be cancer. Our checklist was based on characteristics of therapeutic bedside manner.13
The checklist items were each scored using a 3-point scale of adequate, partial, or inadequate performance. A description of checklist items within each of the 3 domains is listed in Table 1. A postintervention survey was filled out by all hospitalists after the 3 encounters.
All 3 encounters occurred on the same day and each one lasted 1 hour (20 minutes for the encounter, 10 minutes for a posttest survey, and 30 minutes of feedback from the SP). For each case, a task list was given to the hospitalist before walking into the room (Appendix 1). During the feedback session, the SP gave the hospitalist feedback using the case checklist items. They then watched a video of the encounter and the SP further emphasized areas for improvement.
SP training consisted of three 3-hour training sessions, which included review of the case, script, guidance on scoring the checklist items, role plays with attending hospitalists, and feedback training. Each SP was assigned to only 1 case.
Seven of the 24 encounters for each SP were reviewed independently by 2 investigators who created a final score for each checklist item which was compared to the SP’s checklist item score. The kappa (k) statistic was used to evaluate inter-observer reliability using the SAS system software (SAS Institute Inc.).
The percent of hospitalists who performed each checklist item adequately within in each of the 3 domains (Courtesy and Respect, Listen, and Explain) was calculated. To compare the 3 domains, t tests were used.
We calculated the percent that our hospitalist group received on the 3 HCAHPS doctor’s questions 1 year prior to our SP exercise and 1 year after the SP exercise.
Twenty-three hospitalists completed all 3 encounters. For the 3 domains (Courtesy and Respect, Listen, and Explain), hospitalists performed significantly better in the Listen domain compared to the other 2 domains, with a mean percent adequate score of 90.2 % (95% confidence interval [CI], 72.2%-100%; P < 0.05), and significantly worse in the Explain domain compared to the other 2 domains, with a mean percent adequate score of 65.0% (95% CI, 49.2%-83.6%; P < 0.05). The mean percent adequate score for the Courtesy and Respect domain was 81.6% (95% CI, 56%-100%). This was significantly higher than the Explain domain and significantly lower than the Listen domain.
Posttest survey results showed that hospitalists had an increased level of confidence in their bedside manner, patient satisfaction skills, and high-quality discharge discussion skills.
Inter-rater reliability for the discharge encounter, the daily rounding encounter, and the difficult patient encounter were 0.74 (95% CI, 0.64-0.84), 0.73 (95% CI, 0.63-0.82), and 0.73 (95% CI, 0.63-0.83), respectively.
Four hundred sixteen HCAHPS surveys were returned in the year prior to our SP exercise, and the percent of patients who answered always to the questions on Courtesy and Respect, Listen, and Explain were 80.4%, 74.2 %, and 69.4 %, respectively. In the year after our SP exercise, 492 surveys were returned, and there was no significant change in HCAHP scores for the group (80.9% for Courtesy and Respect, 70.2% for the Listen question, and 70.5% for Explain).
We have shown that SPs can be used to assess deficits in hospitalist communication skills and provide feedback that can improve hospitalist confidence in performing optimal communication behaviors. We have also shown that hospitalists perceive the exercise as beneficial in improving their communication skills and perceive them as similar to their real patient encounters.
The Explain domain was significantly worse than the Courtesy and Respect and Listen domains for our hospitalists. Analysis of the checklist items within the Explain domain found that the items within this domain that were most problematic for hospitalists were summarizing information at the end of the encounter, using teach-back (a communication confirmation method where a healthcare provider asks a patient to repeat what was said to confirm understanding), encouraging additional questions by using open-ended statements (What questions do you have?) instead of close ended statements (Do you have any questions?), managing team and self-up, setting expectations on length of stay, and timing of tests. This correlated with our patient satisfaction HCAHPS data, which showed that patients consistently rated our hospitalists’ ability to explain things in a way they could understand lowest among the 3 questions. HCAHPS scores did not change after our SP exercise, and this lack of improvement may indicate that meaningful improvement in communication skills requires longitudinal interventions and real-time feedback rather than a single exercise, as was shown in a recent study looking at daily patient satisfaction score feedback given to internal medicine residents.14
Our study had several limitations. First, hospitalists knew they were being videotaped and observed, which may have altered their behaviors and may not reflect our hospitalists’ actual behaviors with patients. Furthermore, we did not examine whether the feedback given was incorporated into our hospitalists’ daily patient communications and whether this impacted our patients care other than examining HCAHPS scores.
SPs can be used to identify deficiencies in communication skills and provide specific guidance that improves hospitalist confidence in their communication skills.
This trial was funded by a grant from The Doctor’s Company Foundation.
None of the authors report any conflicts of interest.