Importance Gender bias may affect assessment in competency-based medical education. Objective To evaluate the association of gender with assessment of internal medicine residents. Design, Setting, and Participants This multisite, retrospective, cross-sectional study included 6 internal medicine residency programs in the United States. Data were collected from July 1, 2016, to June 30, 2017, and analyzed from June 7 to November 6, 2019. Exposures Faculty assessments of resident performance during general medicine inpatient rotations. Main Outcomes and Measures Standardized scores were calculated based on rating distributions for the Accreditation Council for Graduate Medical Education’s core competencies and internal medicine Milestones at each site. Standardized scores are expressed as SDs from the mean. The interaction of gender and postgraduate year (PGY) with standardized scores was assessed, adjusting for site, time of year, resident In-Training Examination percentile rank, and faculty rank and specialty. Results Data included 3600 evaluations for 703 residents (387 male [55.0%]) by 605 faculty (318 male [52.6%]). Interaction between resident gender and PGY was significant in 6 core competencies. In PGY2, female residents scored significantly higher than male residents in 4 of 6 competencies, including patient care (mean standardized score [SE], 0.10 [0.04] vs 0.22 [0.05]; P = .04), systems-based practice (mean standardized score [SE], −0.06 [0.05] vs 0.13 [0.05]; P = .003), professionalism (mean standardized score [SE], −0.04 [0.06] vs 0.21 [0.06]; P = .001), and interpersonal and communication skills (mean standardized score [SE], 0.06 [0.05] vs 0.32 [0.06]; P < .001). In PGY3, male residents scored significantly higher than female patients in 5 of 6 competencies, including patient care (mean standardized score [SE], 0.47 [0.05] vs 0.32 [0.05]; P = .03), medical knowledge (mean standardized score [SE], 0.47 [0.05] vs 0.24 [0.06]; P = .003), systems-based practice (mean standardized score [SE], 0.30 [0.05] vs 0.12 [0.06]; P = .02), practice-based learning (mean standardized score [SE], 0.39 [0.05] vs 0.16 [0.06]; P = .004), and professionalism (mean standardized score [SE], 0.35 [0.05] vs 0.18 [0.06]; P = .03). There was a significant increase in male residents’ competency scores between PGY2 and PGY3 (range of difference in mean adjusted standardized scores between PGY2 and PGY3, 0.208-0.391; P ≤ .002) that was not seen in female residents’ scores (range of difference in mean adjusted standardized scores between PGY2 and PGY3, −0.117 to 0.101; P ≥ .14). There was a significant increase in male residents’ scores between PGY2 and PGY3 cohorts in 6 competencies with female faculty and in 4 competencies with male faculty. There was no significant change in female residents’ competency scores between PGY2 to PGY3 cohorts with male or female faculty. Interaction between faculty-resident gender dyad and PGY was significant in the patient care competency (β estimate [SE] for female vs male dyad in PGY1 vs PGY3, 0.184 [0.158]; β estimate [SE] for female vs male dyad in PGY2 vs PGY3, 0.457 [0.181]; P = .04). Conclusions and Relevance In this study, resident gender was associated with differences in faculty assessments of resident performance, and differences were linked to PGY. In contrast to male residents’ scores, female residents’ scores displayed a peak-and-plateau pattern whereby assessment scores peaked in PGY2. Notably, the peak-and-plateau pattern was seen in assessments by male and female faculty. Further study of factors that influence gender-based differences in assessment is needed.

Background Inpatient hyperglycemia in adult patients with and without a history of diabetes is a predictor of poor clinical outcome. No previous studies; however, have examined the association of hyperglycemia and clinical outcome in children admitted to a community pediatric hospital. Methods Retrospective observational cohort of pediatric patients admitted to a community children's hospital, from January 2004 to August 2004. Medical records of 903 consecutive children admitted to critical and non-critical care areas were reviewed. Of them, 342 patients (38%) had no blood glucose measurements during the hospital stay. In the remaining patients, we determined the prevalence of hyperglycemia and examined the association of hyperglycemia and clinical outcome. Results A total of 406 patients (75%) had an admission blood glucose ≤ 120 mg/dl [(mean 98 ± 1 mg/dl (± SEM)], 103 children (19%) had an admission blood glucose level between 121–179 mg/dl (mean 143 ± 2 mg/dl), and 32 patients (5.9%) had a blood glucose level ≥ 180 mg/dl (mean 260 ± 18 mg/dl). Seventeen patients (13%) had a known history of diabetes prior to admission. Children with hyperglycemia were more likely to be admitted to the ICU (p<0.001) and had a longer length of ICU stay (p<0.001), but admission hyperglycemia was not associated with longer hospital stay or higher hospital mortality. Summary Hyperglycemia is present in one-fourth of children admitted to the hospital, most of them without a history of diabetes prior to admission. Hyperglycemia was associated with a greater need for ICU care and longer ICU stay but not with increased in-hospital mortality.

Despite evidence of stabilization in some areas of the USA, HIV infection in black women is not declining in the Deep South. Using a phenomenological approach to qualitative inquiry, we investigated women's experiences influencing their adherence to highly active antiretroviral therapy (HAART) in an urban setting. Inclusion criteria specified black women who had been aware of their HIV status for at least two years and were engaged in HIV outpatient care. Twelve single face-to-face confidential in-depth semi-structured interviews were conducted from a sample of predominantly middle-aged women retained in care at an HIV clinic in Atlanta, Georgia. Data were analyzed by two independent reviewers and three themes emerged from the group of women's accounts of their experiences. First, sentinel events led to changes in perspective and motivated women to adhere to HAART. Second, recognition that one had the personal strength necessary to cope with HIV fostered adherence. Finally, relationships with healthcare providers especially trust issues surrounding this relationship, impacted adherence both positively and negatively. These findings suggest that HAART adherence is a complex issue among middle-aged urban black women with HIV in the Deep South. Providers caring for this patient population should recognize that sentinel events, personal strength, and positive healthcare relationships are opportunities to improve adherence.

Context: Bite-sized learning is an instructional method that utilizes brief, focused learning units. This approach may be beneficial in medical education given demands on learner time and cognitive load. This study aims to assess the impact of this approach on knowledge acquisition and learner attitudes in postgraduate medical education. Methods: An instructional method, termed Bite-Sized Teaching (BST), was implemented within the curriculum at a US Internal Medicine postgraduate training program. In BST, content is distilled into manageable units focused on relevant schemas and delivered via brief peer teaching. A two-fold assessment of BST was performed that included cross sectional survey to assess learner attitudes and experiences and a controlled study to assess knowledge acquisition with BST and case-based teaching control. Results: One hundred and six of 171 residents (62% response rate) completed the survey. Most residents (79.8%) reported BST was among the best conference types in the curriculum. Important components of BST cited by residents include the distilled content, multiple short talk format and peer teaching. Residents report incorporating what they learned via BST into their teaching (76.1%) and clinical practice (74.1%). Resident who had participated as speaker were significantly more likely to report incorporating learning from BST into their teaching (87.2% vs 63.0%, p < 0.01, Cramer’s V effect size = 0.37) and clinical practice (89.7% vs 65.3%, p = 0.02, Cramer’s V effect size 0.28). Fifty-one residents participated in the knowledge assessment. Residents taught via BST scored significantly higher on immediate post-test compared to case-based teaching (score [SE] 62.5% [1.9] vs 55.2% [2.4], p = 0.03, Hedges g effect size 0.66). While both groups improved over pretest, there was no significant difference in scores between BST and case-based teaching at two (score [SE] 57.1 [2.1] vs 54.8 [2.5], p = 0.54) and six weeks (score [SD] 55.9 [2.1] vs 53.0 [2.9], p = 0.43). Conclusions: Teaching via brief, focused learning units delivered by peers is well received by learners and appears to have a significantly greater impact on immediate knowledge recall than case-based teaching. Further study on long term knowledge retention and behaviors is needed. Bite-Sized Teaching may be a promising instructional approach in medical education.