Forward momentum: progressive change through diversity equity and inclusion initiatives in academic health care

Abstract

Background: This study evaluates the impact of a 2-year Diversity, Equity, and Inclusion (DEI) intervention program within a radiation oncology department. We analyzed employee perceptions of inclusivity, bias, training, and career development, recognizing the challenges of assessing DEI initiatives in an evolving sociopolitical context.

Methods: A voluntary survey was distributed in 2021 and 2023. The survey assessed 4 DEI domains: Inclusivity, Department Bias, Training/Education, and Career Development. Responses were analyzed using nonparametric tests. DEI initiatives included implicit bias training, allyship training, book clubs, anonymous feedback platforms, and a DEI website.

Results: Survey completion rates were 40% (2021) and 50% (2023). Significant improvements were observed in Inclusivity (3.72 vs 3.91, P = .042), Training/Education (3.57 vs 4.14, P < .001), and Career Development (3.39 vs 3.60, P = .019). Department Bias showed no significant change (P = .130). Anti-DEI sentiment increased in 2023, highlighting challenges in fostering inclusivity. Subgroup analyses revealed improvements for Black employees but persistent disparities for women.

Conclusions: This exploratory study suggests that targeted DEI initiatives can improve employee perceptions of workplace culture in academic health-care settings. Notably, the program improved perceptions of inclusivity, training opportunities, and career development. However, persistent gender inequities in training and career development highlight the need for monitoring and focused efforts. These findings can inform future DEI strategies and underscore the importance of continued vigilance in promoting an inclusive work environment. Future research should explore the program's downstream effects on patient care, clinical trial enrollment, and residency enrollment.

Figure 1.

DEI Index Scores (DISs) stratified by when the survey was done (2021 vs 2023). Box and whisker plots show grouped survey questions in classified DEI categories (A-D) and bar graphs highlight specific survey questions (a-d) from that category that illustrate differences between the 2 years. Each box and whisker plot shows the minimum and maximum using the furthest horizontal lines, the upper and lower quartile of data in the shaded box, and the median as the line within the shaded box. Outliers are represented as individual data points. Exact P values are displayed on the plots and insets, with statistically significant differences defined as P ≤ .05.

Figure 2.

DEI Index Scores (DISs) stratified by when the survey was done (2021 vs 2023) and various demographic categories (A-D). **P ≤ .05.

Figure 3.

Average coder scores of how the department performed (1 = very poor to 5 = very good) on persistent themes uncovered during the thematic analysis based on employee comments found on the survey in 2021 (black circles) and 2023 (gray circles). Error bars represent the SD of the coder scores.