Background

African Americans are disproportionately affected by hepatitis C (HCV) and are less likely to undergo HCV treatment. Underserved populations are especially at risk for experiencing health disparity. Aim. To identify reasons for HCV non-treatment among underserved African Americans in a large safetynet system.

Material and methods

Medical records of HCV-infected African Americans evaluated at San Francisco General Hospital liver specialty clinic from 2006-2011 who did not receive HCV treatment were reviewed. Treatment eligibility and reasons for non-treatment were assessed. Factors associated with treatment ineligibility were assessed using logistic regression modeling.

Results

Among 118 patients, 42% were treatment ineligible, 18% treatment eligible, and 40% were undergoing work-up to determine eligibility. Reasons for treatment ineligibility were medical (54%), non-medical (14%), psychiatric (4%), or combined (28%). When controlling for age and sex, active/recent substance abuse (OR 6.65, p = 0.001) and having two or more medical comorbidities (OR 3.39, p = 0.005) predicted treatment ineligibility. Excluding those ineligible for treatment, 72% of all other patients were lost to follow-up; they were older (55 vs. 48 years, p = 0.01) and more likely to be undergoing work up to determine treatment eligibility (86 vs. 21%, p < 0.0001) than those not lost to follow-up.

Conclusions

Medical comorbidities and substance abuse predicted HCV treatment ineligibility in underserved African Americans. Importantly, the majority of those undergoing work-up to determine HCV treatment eligibility were lost to follow-up. While newer anti-HCV agents may increase treatment eligibility, culturally appropriate interventions to increase compliance with evaluation and care remain critical to HCV management in underserved African Americans.

Background

Competency-based medical education relies on meaningful resident assessment. Implicit gender bias represents a potential threat to the integrity of resident assessment. We sought to examine the available evidence of the potential for and impact of gender bias in resident assessment in graduate medical education.

Methods

A systematic literature review was performed to evaluate the presence and influence of gender bias on resident assessment. We searched Medline and Embase databases to capture relevant articles using a tiered strategy. Review was conducted by two independent, blinded reviewers. We included studies with primary objective of examining the impact of gender on resident assessment in graduate medical education in the USA or Canada published from 1998 to 2018.

Results

Nine studies examined the existence and influence of gender bias in resident assessment and data included rating scores and qualitative comments. Heterogeneity in tools, outcome measures, and methodologic approach precluded meta-analysis. Five of the nine studies reported a difference in outcomes attributed to gender including gender-based differences in traits ascribed to residents, consistency of feedback, and performance measures.

Conclusion

Our review suggests that gender bias poses a potential threat to the integrity of resident assessment in graduate medical education. Future study is warranted to understand how gender bias manifests in resident assessment, impact on learners and approaches to mitigate this bias.

Background

Disparities in objective assessments in graduate medical education such as the In-Training Examination (ITE) that disadvantage women and those self-identifying with race/ethnicities underrepresented in medicine (URiM) are of concern.

Objective

Examine ITE trends longitudinally across post-graduate year (PGY) with gender and race/ethnicity.

Design

Longitudinal analysis of resident ITE metrics at 7 internal medicine residency programs, 2014-2019. ITE trends across PGY of women and URiM residents compared to non-URiM men assessed via ANOVA. Those with ITE scores associated with less than 90% probability of passing the American Board of Internal Medicine certification exam (ABIM-CE) were identified and odds of being identified as at-risk between groups were assessed with chi square.

Participants

A total of 689 IM residents, including 330 women and URiM residents (48%).

Main measures

ITE score KEY RESULTS: There was a significant difference in ITE score across PGY for women and URiM residents compared to non-URiM men (F_{(2, 1321)} 4.46, p=0.011). Adjusting for program, calendar year, and baseline ITE, women and URiM residents had smaller ITE score gains (adjusted mean change in score between PGY1 and PGY3 (se), non-URiM men 13.1 (0.25) vs women and URiM residents 11.4 (0.28), p<0.001). Women and URiM residents had greater odds of being at potential risk for not passing the ABIM-CE (OR 1.75, 95% CI 1.10 to 2.78) with greatest odds in PGY3 (OR 3.13, 95% CI 1.54 to 6.37).

Conclusion

Differences in ITE over training were associated with resident gender and race/ethnicity. Women and URiM residents had smaller ITE score gains across PGY translating into greater odds of potentially being seen as at-risk for not passing the ABIM-CE. Differences in ITE over training may reflect differences in experiences of women and URiM residents during training and may lead to further disparities.