Inequity in Assessment Among Pediatric Residents

TL;DR

This study explores how pediatric residents experience biased evaluations in their medical training.

Contribution

The study contributes insights into inequities in assessment practices within pediatric residency programs.

Findings

Residents reported experiencing biased verbal assessments.

Biased written assessments were also self-reported by participants.

Findings highlight the prevalence of inequity in medical education evaluations.

Abstract

This survey study evaluates self-reported experiences of biased verbal and written assessments in medical education.

Tables

Table 1.. Demographic Characteristics of Pediatric Residents in 15 Programs Responding to a Multisite Survey

Characteristic	Residents, No. (%)		Pearson χ2, unadjusted P valuea
	All (N = 479)	Reporting sometimes or usually experiencing inequity in assessment (n = 83)
Race and ethnicityb
Asian	108 (22.6)	20 (18.5)	.004
URIMc	82 (17.1)	24 (29.3)
White	289 (60.3)	39 (13.5)
Gender identity
Female	345 (72.0)	73 (21.2)	<.001
Male	134 (28.0)	10 (7.5)
Transgender, nonbinary, and otherd	NAe	NA
Sexual orientation
Lesbian, gay, bisexual, queer, and otherf	56 (11.7)	6 (10.7)	.16
Straight or heterosexual	423 (88.3)	77 (18.2)
Disability identity
With disability	39 (8.1)	13 (33.3)	.005
Without disability or did not report	440 (91.9)	70 (15.9)
Medical degree
IMG	25 (5.2)	8 (32.0)	.06
AMG	454 (94.8)	75 (16.5)
SES
Lowg	110 (23.0)	28 (25.5)	.04
Middleh	264 (55.1)	40 (15.2)
Highi	105 (21.9)	15 (14.3)

Table 2.. Experiences and Frequency of Inequity in Assessment Reported by Pediatric Residents

Group	Odds of ever experiencing inequity in assessmenta		Odds of experiencing inequity in verbal assessment		Odds of experiencing inequity in written assessment		Odds of frequently experiencing inequity in assessmentb
	OR (95% CI)	P valuec	OR (95% CI)	P valuec	OR (95% CI)	P value	OR (95% CI)	P valuec
Race and ethnicity
Asian	2.22 (1.27-3.87)	.005	3.76 (1.88-7.50)	<.001	1.00 (0.28-3.54)	.99	1.27 (0.98-1.63)	.07
URIMd	6.77 (4.42-10.36)	<.001	5.59 (3.08-10.13)	<.001	13.06 (6.07-28.10)	<.001	3.78 (3.07-4.67)	<.001
White	1 [Reference]	NA	1 [Reference]	NA	1 [Reference]	NA	1 [Reference]	NA
Gender identity
Male	1 [Reference]	NA	1 [Reference]	NA	1 [Reference]	NA	1 [Reference]	NA
Female	2.70 (1.95-3.73)	<.001	2.59 (1.67-4.01)	<.001	3.24 (1.70-6.16)	<.001	2.85 (2.51-3.24)	<.001
Intersectional group: URIM- female	0.44 (0.27-0.70)	.001	0.51 (0.26-1.01)	.05	0.19 (0.08-0.46)	<.001	0.75 (0.60-0.94)	.01
Intersectional group: Asian- female	0.50 (0.28-0.88)	.02	0.39 (0.19-0.80)	.01	0.81 (0.22-2.97)	.75	1.42 (1.08-1.85)	.01
Disability identity
Without disability	1 [Reference]	NA	1 [Reference]	NA	1 [Reference]	NA	1 [Reference]	NA
With disability	7.80 (5.88-10.36)	<.001	5.87 (3.92-8.78)	<.001	9.11 (5.57-14.88)	<.001	3.59 (3.22-4.00)	<.001
Sexual orientation
Straight or heterosexual	1 [Reference]	NA	1 [Reference]	NA	1 [Reference]	NA	1 [Reference]	NA
Lesbian, gay, bisexual, queer, and othere	0.61 (0.48-0.78)	<.001	0.82 (0.59-1.15)	.25	0.45 (0.28-0.73)	.001	0.72 (0.64-0.82)	<.001
Medical degree
AMG	1 [Reference]	NA	1 [Reference]	NA	1 [Reference]	NA	1 [Reference]	NA
IMG	4.63 (2.28-9.40)	<.001	1.73 (0.60-5.02)	.31	9.34 (3.54-21.60)	<.001	1.91 (1.62-2.26)	<.001
SES
Highf	1 [Reference]	NA	1 [Reference]	NA	1 [Reference]	NA	1 [Reference]	NA
Middleg	1.29 (1.05-1.59)	.02	1.10 (0.82-1.47)	.53	1.89 (0.68-5.28)	.23	0.90 (0.82-0.98)	.02
Lowh	1.24 (0.97-1.59)	.08	1.08 (0.77-1.52)	.66	1.03 (0.65-1.63)	.91	1.42 (1.28-1.57)	<.001

Full text

Introduction

In undergraduate and graduate medical education (GME), verbal and written assessments provide feedback to learners and inform decisions about advancement.^1^ These assessments are prone to inequity, reflecting biases leading to disparities in performance assessment.^2^ Disparities in scores and differential, biased language in written and verbal assessments of learners underrepresented in medicine (URIM) and female learners have been described in undergraduate medical education and several GME specialties, but not pediatrics.^2,3^ Understanding whether these inequities exist in pediatrics is important because inequities accumulate over time, harming trainees’ career trajectories and society’s interest in a diverse physician workforce.^4,5^ In this study, we sought to determine whether pediatric residents had experienced inequity in assessments.

Methods

From May 2023 to June 2024 as part of a multicenter study, we surveyed residents in 15 US pediatric residency programs. Cincinnati Children’s Institutional Review Board deemed this survey study exempt from review due to minimal risk survey procedures. Participants provided written consent. We followed the AAPOR reporting guideline.

We asked residents for their demographic characteristics, whether they had ever experienced inequity in written or verbal assessment, and if so, how frequently (eAppendix in Supplement 1). Demographic groups included race and ethnicity (Asian, URIM [African American or Black, American Indian or Alaskan Native, Hispanic, Native Hawaiian or Other Pacific Islander], White), gender identity, sexual orientation, disability identity, medical degree (international or American medical graduate [IMG/AMG]), and socioeconomic status. We defined ever experiencing inequity in assessment as reporting inequity in either written or verbal assessments and frequently experiencing inequity in assessment as responding sometimes, usually, or almost always.

We compared reported frequency of inequities across demographic groups using the Pearson χ^2^ test. We then modeled the association between demographic groups (plus 2 intersectional groups: URIM-female, Asian-female) and 2 dependent variables (ever and frequently experiencing inequity in assessment) using multivariable logistic regression. To determine intersectional relationships, we combined interactions between the 2 dependent variables with main outcomes. We included a random program intercept to adjust for residents clustered in programs. Data analysis was performed with R 4.0.2 (R Core Team).

Results

Overall, 479 of 803 pediatric residents (59.7%) completed the survey (345 females (72.0%); 108 Asian [22.6%], 82 URIM [17.1%], 289 White [60.3%] individuals) (Table 1). In adjusted models, odds of ever experiencing inequity in either written or verbal assessment were significantly higher among residents with a disability (odds ratio [OR], 7.80; 95% CI, 5.88-10.36) and those identifying as Asian (OR, 2.22; 95% CI, 1.27-3.87), URIM (OR, 6.77; 95% CI, 4.42-10.36), IMG (OR, 4.63; 95% CI, 2.28-9.40), or female (OR, 2.70; 95% CI, 1.95-3.73) than residents without a disability, White residents, or male residents (Table 2). Residents identifying as lesbian, gay, bisexual, queer, or other sexual orientation (OR, 0.61; 95% CI, 0.48-0.78) reported significantly lower odds of ever experiencing inequities than other identities. The intersectional interactions of URIM-female (OR, 0.44; 95% CI, 0.27-0.70) and Asian-female (OR, 0.50; 95% CI, 0.28-0.88) resulted in lower odds of ever experiencing inequities than odds for individual female, URIM, or Asian identities. We found similar results in residents reporting frequently experiencing inequity in assessment (Table 2).

Discussion

Pediatric residents with a disability and those identifying as URIM reported 7- to 8-fold greater odds of experiencing inequities in assessment than residents without a disability or White residents. Residents identifying as IMG, Asian, or female also had greater odds of experiencing inequities, but to a lesser extent. Findings among URIM-female residents parallel findings from other specialties, indicating that race and ethnicity– and gender-based inequities are also a problem in pediatrics. The finding that URIM-female and Asian-female groups reported more experiences of inequity but not additive of levels reported by individual identities is similar to results in other specialties.^6^

A study limitation was our reliance on retrospective self-report. Future research is needed to confirm the findings and develop interventions to reduce inequities.