Gathering Trainee Feedback to Improve Programs With Low Annual ACGME Survey Content Area Compliance: A Pilot Study.

PROBLEM: Systematically investigating annual Accreditation Council for Graduate Medical Education (ACGME) Resident/Fellow Survey results by directly gathering trainee feedback could uncover training program problems and clarify misunderstandings as they arise, leading to faster corrective actions and program improvement. APPROACH: The Focus Group Forum (FGF) was created based on the utilization-focused evaluation approach to systematically gather comprehensive, high-quality, actionable trainee feedback on specific annual ACGME survey results and involve trainees in program improvement (Henry Ford Hospital, 2021). Trainees from programs with survey results indicating <80% compliance within several content areas were invited to attend FGF sessions. During FGF sessions, neutral moderators experienced in conducting focus groups and creating psychologically safe spaces and neutral scribes gathered trainee feedback on survey results through structured, iterative discussions and an anonymous electronic polling system. Summaries of FGF findings were created, combined with actual annual ACGME survey data, and used to develop recommended corrective actions and monitoring plans. OUTCOMES: In 2021, 6 training programs had survey results below the institution's compliance threshold for 4-8 of the 9 content areas. Of the 180 trainees (from the 6 programs) invited to attend an FGF session, 79 (44%) participated. Five key issues were identified: misinterpretation of several survey questions, lack of knowledge of institutional policies and procedures, perceived inability to share feedback with faculty, feelings of being overwhelmed with administrative duties, and lack of sufficient protected time for educational activities and requirements. NEXT STEPS: The authors are developing an FGF process for faculty so that all stakeholders have a voice regarding annual ACGME survey results. They are also improving scheduling processes so that feedback from experienced trainees who are leaving the institution will not be missed and developing longer-term processes for tracking outcomes since time for implementing corrective actions before the next ACGME survey is limited.

Designing a Yoga Intervention Program to Improve Well-Being for Physician Trainees: Challenges and Lessons Learned.

Well-being activities may help to counteract physician burnout. Yoga is known to enhance well-being, but there are few studies of yoga as an intervention for physicians in training. This prospective methodology-development study aimed to explore how to establish a yoga-based well-being intervention for physician trainees in a large urban training hospital. We aimed to identify factors that contribute to trainee participation and explore an instrument to measure changes in self-reported well-being after yoga. Cohorts included a required-attendance group, a voluntary-attendance group, and an unassigned walk-in yoga group. Weekly 1-hour yoga sessions were led by a qualified yoga instructor for 4 weeks. The seven-question Resident Physician Well-Being Index (RPWBI) was used to measure resident well-being before yoga, after 4 weeks of yoga, and 6 months post-yoga. Trainees attending each session ranged from 17 for required yoga to 0-2 for voluntary yoga, 2-9 for lunchtime walk-in yoga, and 1-7 for evening walk-in yoga. In the required-yoga group (n = 17), overall RPWBI mean scores did not change significantly across the three query times, and participation in the survey declined over time. The mean baseline RPWBI score for the required group before yoga was in the non-distressed range and answers to the seven individual questions varied. Requiring a yoga activity for medical trainees may be a good strategy for promoting participation in yoga. The RPWBI may have limited utility for measuring changes in overall group well-being after a yoga intervention.

Does physician engagement affect satisfaction of patients or resident physicians?

Purpose: This study examined whether change in physician engagement affected outpatient or resident physician satisfaction using common US measures. Methods: Surveys were administered by Advisory Board Survey Solutions for staff physician engagement, Press Ganey for Clinician and Group Consumer Assessment of Healthcare Providers and Systems (CGCAHPS) for outpatient satisfaction, and Accreditation Council for Graduate Medical Education (ACGME) for the ACGME Resident/Fellow Survey. Survey sample sizes were 685, 697, and 763 for physician engagement and 621, 625, and 618 for resident satisfaction in 2014-2016, respectively; only respondents were available for CGCAHPS (24,302, 34,328, and 43,100 for 2014-2016, respectively). Two groups were analyzed across 3 years: (1) percentage of "engaged" staff physicians versus percentage of outpatient top box scores for physician communication, and (2) percentage of "engaged" staff physicians versus percentage of residents "positive" on program evaluation. For resident evaluation of faculty, the number of programs that met/exceeded ACGME national compliance scores were compared. Univariate chi-squared tests compared data between 2014, 2015, and 2016. Results: For 2014-2016, "engaged" physicians increased from 34% (169/497) to 44% (227/515) to 48% (260/542) (P<0.001) whereas CGCAHPS top box scores for physician communication remained unchanged at 90.9% (22,091/24,302), 90.8% (31,088/34,328), and 90.9% (39,178/43,100) (P=0.869). For the second group, "engaged" physicians increased from 33% (204/617) to 46% (318/692) to 50% (351/701) (P<0.001) and residents "positive" on program evaluation increased from 86% (534/618) in 2014 to 89% (556/624) in 2015 and 89% (550/615) in 2016 (P=0.174). The number of specialties that met/exceeded national compliance for all five faculty evaluation items grew from 44% (11/25) in 2014 to 68% (17/25) in 2015 and 64% (16/25) in 2016 (P=0.182). Conclusion: For our medical group, improvement in physician engagement across time did not coincide with meaningful change in the outpatient experience with physician communication or resident satisfaction with program and faculty.

Health Care Disparities: a Practical Approach to Teach Residents about Self-Bias and Patient Communication.

Studies have shown that the education of resident physicians on health care disparities (HCDs) needs improvement. We implemented a system-wide program on HCD for residents and evaluated outcomes across 1 year. Designed in 2015 by a multidisciplinary team, the HCD program incorporated information about our health system's patient population and the tenets of unconscious bias. We used the ask-tell-ask model of communication to teach trainees how to identify patients' barriers to health care. In 2016, resident participants in the HCD program were asked to complete a modified version of the Bonham and Sellers RACE survey, which measures consideration of race in clinical care, at four time-points (pre-, post-, 3-months post-, and 1-year post-intervention). Of 186 PGY2 residents who completed the HCD program, 108 (58%) completed all 4 surveys across 1 year. The modified Bonham and Sellers RACE survey yielded a Cronbach's alpha of 0.885 and communality for the six questions ranging from 0.543 to 0.727. Using the modified RACE survey, resident respondents showed overall significantly increased consideration of race in clinical care from pre- to post-intervention time-points (p < 0.001). This study of our program on health care disparities showed that resident survey respondents increased self-reported consideration of race in clinical care after the intervention across 1 year.

Continuity clinic satisfaction and valuation in residency training.

BACKGROUND: Internal Medicine residency training in ambulatory care has been judged inadequate, yet how trainees value continuity clinic and which aspects of clinic affect attitudes are unknown. OBJECTIVES: To determine the value that Internal Medicine residents place on continuity clinic and how clinic precepting, operations, and patient panels affect its valuation. DESIGN AND MEASUREMENTS: A survey on ambulatory care was developed, including questions on career choice and the value of clinical training experiences. Independent variables were Likert-scale ratings (1 = disagree strongly/no value; 3 = neutral; 5 = agree strongly/high value) on preceptors, patients, operations, and resident characteristics. Odds ratios and stepwise multivariate logistic regression with clustering were used to evaluate associations between clinic valuation and independent variables. SUBJECTS: Internal medicine residents at 3 residency programs. RESULTS: 218 of 260 residents (83.8%) completed the survey. Resident ratings were highest on diversity of illness seen (4.1), medical record systems used (4.1), and contact with preceptors who were receptive to questions (4.8). Resident ratings were lowest on economic diversity of patients (2.7), interruptions from inpatient wards (3.1), and contact with preceptors who taught history and physical exam skills (3.5). High ratings on all precepting issues and nearly all operational issues were associated with valuing clinic. With multivariate analysis, high ratings of preceptors as role models were most strongly associated with valuing clinic (corrected relative risk 3.44). A planned career in general Internal Medicine was not associated with valuing clinic. CONCLUSIONS: Satisfaction with preceptors, particularly as role models, and clinic operations correlate with the value residents place on continuity clinic.

Socioeconomic and Racial Disparities: a Case-Control Study of Patients Receiving Transcatheter Aortic Valve Replacement for Severe Aortic Stenosis.

BACKGROUND: We sought to quantify socioeconomic disparities in patients with severe aortic stenosis (AS) undergoing transcatheter aortic valve replacement (TAVR) at an urban, tertiary referral center. METHODS: This retrospective case-control study identified 67 patients with severe AS (aortic valve [AV] area ≤1 cm2 or AV area index ≤0.60 cm2/m2 or AV velocity ≥40 mmHg) who underwent TAVR from November 5, 2013 to June 10, 2014. Study subjects were matched to controls with severe AS without TAVR in a 4:1 age-frequency match. Demographic data were collected using electronic medical records. Area-based median household income was obtained by geocoding patients' addresses and linking with census data. Charlson comorbidity index for all subjects was calculated. RESULTS: Income disparity was significant in that with every $10,000 increase in income, the odds of receiving TAVR increased by 10% (p = 0.05). Non-blacks were significantly more likely to receive TAVR than blacks (odds ratio [OR] 2.812, confidence interval [CI] 1.007-7.853; p = 0.048). No differences in comorbidities were found between the two groups. Post hoc analysis to identify etiologies of the found disparities examined differences of AV area and AV area index, indication for two-dimensional echocardiography (echo), symptoms prior to echo, and action after echo within the control group. Black race significantly impacted the TAVR status despite the same AV area (OR 0.33, CI 0.09-0.97, p = 0.043). After echo, blacks were more likely to decline AVR, be lost to follow-up, and not be referred to cardiology (OR 4.41, CI 1.43-13.64; p = 0.010). CONCLUSION: Socioeconomic and racial disparities were associated with patients with severe AS receiving TAVR at a major referral center. This study emphasizes the importance of improving access to standard of care for these subgroups of cardiac patients.

Health Disparities in Hepatitis C Screening and Linkage to Care at an Integrated Health System in Southeast Michigan.

With recommended screening for hepatitis C among the 1945-1965 birth cohort and advent of novel highly effective therapies, little is known about health disparities in the Hepatitis C care cascade. Our objective was to evaluate hepatitis C screening rates and linkage to care, among patients who test positive, at our large integrated health system. We used electronic medical records to retrospectively identify patients, in the birth cohort, who were seen in 21 Internal Medicine clinics from July 2014 to June 2015. Patients previously screened for hepatitis C and those with established disease were excluded. We studied patients' sociodemographic and medical conditions along with provider-specific factors associated with likelihood of screening. Patients who tested positive for HCV antibody were reviewed to assess appropriate linkage to care and treatment. Of 40,561 patients who met inclusion criteria, 21.3% (8657) were screened, 1.3% (109) tested positive, and 30% (30/100) completed treatment. Multivariate logistic regression showed that African American race, male gender, electronic health engagement, residency teaching clinic visit, and having more than one clinic visit were associated with higher odds of screening. Patients had a significant decrease in the likelihood of screening with sequential interval increase in their Charlson comorbidity index. When evaluating hepatitis C treatment in patients who screened positive, electronic health engagement was associated with higher odds of treatment whereas Medicaid insurance was associated with significantly lower odds. This study shows that hepatitis C screening rates and linkage to care continue to be suboptimal with a significant impact of multiple sociodemographic and insurance factors. Electronic health engagement emerges as a tool in linking patients to the hepatitis C care cascade.

Mapping Direct Observations From Objective Structured Clinical Examinations to the Milestones Across Specialties.

BACKGROUND: Little is known about residents' performance on the milestones at the institutional level. Our institution formed a work group to explore this using an institutional-level curriculum and residents' evaluation of the milestones. OBJECTIVE: We assessed whether beginner-level milestones for interpersonal and communication skills (ICS) related to observable behaviors in ICS-focused objective structured clinical examinations (OSCEs) for postgraduate year (PGY) 1 residents across specialties. METHODS: The work group compared ICS subcompetencies across 12 programs to identify common beginner-level physician-patient communication milestones. The selected ICS milestone sets were compared for common language with the ICS-OSCE assessment tool-the Kalamazoo Essential Elements of Communication Checklist-Adapted (KEECC-A). To assess whether OSCE scores related to ICS milestone scores, all PGY-1 residents from programs that were part of Next Accreditation System Phase 1 were identified; their OSCE scores from July 2013 to June 2014 and ICS subcompetency scores from December 2014 were compared. RESULTS: The milestones for 10 specialties and the transitional year had at least 1 ICS subcompetency that related to physician-patient communication. The language of the ICS beginner-level milestones appears similar to behaviors outlined in the KEECC-A. All 60 residents with complete data received at least a beginner-level ICS subcompetency score and at least a satisfactory score on all 3 OSCEs. CONCLUSIONS: The ICS-OSCE scores for PGY-1 residents appear to relate to beginner-level milestones for physician-patient communication across multiple specialties.

A time-efficient web-based teaching tool to improve medical knowledge and decrease ABIM failure rate in select residents.

AIM: The American Board of Internal Medicine (ABIM) exam's pass rate is considered a quality measure of a residency program, yet few interventions have shown benefit in reducing the failure rate. We developed a web-based Directed Reading (DR) program with an aim to increase medical knowledge and reduce ABIM exam failure rate. METHODS: Internal medicine residents at our academic medical center with In-Training Examination (ITE) scores ≤ 35 th percentile from 2007 to 2013 were enrolled in DR. The program matches residents to reading assignments based on their own ITE-failed educational objectives and provides direct electronic feedback from their teaching physicians. ABIM exam pass rates were analyzed across various groups between 2002 and 2013 to examine the effect of the DR program on residents with ITE scores ≤ 35 percentile pre- (2002-2006) and post-intervention (2007-2013). A time commitment survey was also given to physicians and DR residents at the end of the study. RESULTS: Residents who never scored ≤ 35 percentile on ITE were the most likely to pass the ABIM exam on first attempt regardless of time period. For those who ever scored ≤ 35 percentile on ITE, 91.9% of residents who participated in DR passed the ABIM exam on first attempt vs 85.2% of their counterparts pre-intervention (p < 0.001). This showed an improvement in ABIM exam pass rate for this subset of residents after introduction of the DR program. The time survey showed that faculty used an average of 40±18 min per week to participate in DR and residents required an average of 25 min to search/read about the objective and 20 min to write a response. CONCLUSIONS: Although residents who ever scored ≤ 35 percentile on ITE were more likely to fail ABIM exam on first attempt, those who participated in the DR program were less likely to fail than the historical control counterparts. The web-based teaching method required little time commitment by faculty.

System-based approach to educating internal medicine residents on preventable hospital readmissions.

BACKGROUND: A literature gap exists in educating internal medicine residents about hospital readmissions and how to prevent them. INTERVENTION: The study aimed to implement a readmissions education initiative for general internal medicine inpatient resident teams in 3 general practice units at an urban, tertiary hospital. METHODS: Senior residents were given access to a daily list of readmissions, used a readmission assessment tool to investigate causes and to assess whether each readmission was preventable, led a monthly general practice unit team meeting to discuss each case, and presented their findings at the monthly multidisciplinary readmissions meeting for additional feedback. For program evaluation, we hypothesized that the "preventable" readmissions count tracked via the readmissions assessment tool would increase as residents became better educated on the root causes of readmissions. We also conducted a survey to assess perception of the readmissions education initiative. RESULTS: "Preventable" readmissions increased from 21% for the first 3 months of the intervention (September-November 2010) to 46% for the most recent 3 months (January-March 2011). The survey showed that 98% (41 of 42) of respondents who had attended a multidisciplinary readmissions meeting felt involved in an effort to review or improve the rate of hospital readmissions, whereas only 40% (21 of 53) of the group that never attended a session shared the same answer. CONCLUSIONS: This initiative required few resources, and it appeared to help residents identify "preventable" reasons for readmissions, as well as increased their perceptions of being actively involved in reducing hospital readmissions. The intervention was not associated with a statistically significant reduction in readmissions, which may be influenced primarily by multiple factors outside residents' control.

Internal medicine residents' time study: paperwork versus patient care.

BACKGROUND: Multiple factors affect residency education, including duty-hour restrictions and documentation requirements for regulatory compliance. We designed a work sampling study to determine the proportion of time residents spend in structured education, direct patient care, indirect patient care that must be completed by a physician, indirect patient care that 5 be delegated to other health care workers, and personal activities while on an inpatient general practice unit. METHODS: The 3-month study in 2009 involved 14 categorical internal medicine residents who volunteered to use personal digital assistants to self-report their location and primary tasks while on an inpatient general practice unit. RESULTS: Residents reported spending most of their time at workstations (43%) and less time in patient rooms (20%). By task, residents spent 39% of time on indirect patient care that must be completed by a physician, 31% on structured education, 17% on direct patient care, 9% on indirect patient care that 5 be delegated to other health care workers, and 4% on personal activities. From these data we estimated that residents spend 34 minutes per patient per day completing indirect patient care tasks compared with 15 minutes per patient per day in direct patient care. CONCLUSIONS: This single-institution time study objectively quantified a current state of how and where internal medicine residents spend their time while on a general practice unit, showing that residents overall spend less time on direct patient care compared with other activities.

Patient-centered plan-of-care tool for improving clinical outcomes.

PURPOSE: In developing our Patient-Centered Team Care (PCTC) program, we designed a Plan-of-Care (POC) tool to facilitate the physician-patient discussion for setting health goals. This study aimed to determine the effectiveness of the POC tool in improving clinical outcomes. METHODS: We compared baseline and 6-month or greater follow-up values for hemoglobin A1c (HbA1c), low-density lipoprotein (LDL), systolic blood pressure (SP), diastolic blood pressure (DP), and weight for PCTC patients (intervention group) and non-PCTC patients (control group). For the intervention group, we also compared POC tool completeness scores (1, low and 8, high). RESULTS: Of 1366 patients, 1110 (593 intervention, 517 control) had baseline and follow-up clinical measures for analysis. After adjusting for demographics, significant effects occurred in the intervention group for HbA1c (P = .0067), LDL (P = .012), and DP (P = .091). For completeness of POC, a significant association occurred between more fully completed forms (scores, 5-8) and change in HbA1c (P < .001) and SP (P = .011). CONCLUSION: Patients receiving a POC showed significant improvement in 3 of 5 clinical outcomes compared with those without the tool, and those with more fully completed forms had significant improvement in 2 of 5 clinical outcomes compared with those with partially completed forms.