INTRODUCTION

Physicians are less likely to vote than the general population,1 but there is no description of specific barriers to voting that physicians encounter.

METHODS

We conducted a single-center, cross-sectional survey of resident physicians across 8 of the largest specialties at University of Texas Southwestern Medical Center from May 1 to June 30, 2020, to better understand voting practices. Investigators reached out to Chief Residents of various specialties (see Table 1) to assess interest in participating. The survey was paired with verifying voter registration status and assisting physicians in registering to vote. We organized virtual outreach events during times when the majority of resident physicians were available (i.e., during mandatory teaching conferences).

Table 1 Demographics of Resident Physicians Eligible to Vote
Full size table

Our primary outcome was voting barriers in the 2020 election. We used a five-point Likert scale to rate four common barriers: voter registration, logistical hurdles (when or where to vote), time, and psychological (perception that one’s vote does not count).2 We calculated the proportions of physicians who cited each barrier using Microsoft Excel and used Cochran’s Q test to detect the difference between barriers, including pairwise comparisons. Secondary outcomes were self-reported voter registration and voter participation, the percent eligible to vote who voted in the last 4 elections (2016 presidential, 2018 midterm, 2019 local, and 2020 primary). UT Southwestern IRB deemed the study exempt. Informed consent was obtained prior to beginning the survey.

RESULTS

We reached out to 464 resident physicians, and 187 (40%) completed the survey. Of the respondents, 12 (6.4%) were ineligible to vote due to non-US citizenship. Of the 175 eligible voters, mean age was 29 years with 52% males, 43% were married. A majority (51%) identified as non-Hispanic Caucasian. Internal medicine (37%) was the most common specialty (Table 1).

Time was the most commonly cited barrier (61%) to voting in the 2020 election and was significantly higher than each other barrier (P < 0.001 for all pairwise comparisons, Fig. 1). Twenty-three percent reported psychological barriers, 14% logistical hurdles, and 4% voter registration. Forty-six individuals (26%) cited multiple barriers.

Figure 1
figure 1

Self-reported barriers to voting in the 2020 election by resident physicians. Survey participants were given a Likert scale (strongly agree, agree, neither agree nor disagree, disagree, strongly disagree) to rate and select all barriers to voting in the 2020 election. Those who selected strongly agree or agree are displayed above. Denominator is all resident physicians (n = 175) who were eligible to register to vote. Registration denotes “I’m not registered to vote.” Logistical hurdles refer to not knowing “when and where to vote.” Time refers to having “long work hours.” Psychological denotes “I feel like my vote does not count.” Cochran’s Q test, P < 0.001 for Time vs. Psychological, Time vs. Logistical Hurdles, and Time vs. Registration.

Full size image

Nearly all respondents (96%) reported active voter registration with 89% voter participation in 2016. Voter participation in 2018 was lower (69%), declined further in 2019 (36%), and was 45% for the 2020 primary election (Table 1).

DISCUSSION

This is the first report of barriers to voting for physicians. In this diverse sample of resident physicians, the majority cited long work hours as a significant barrier to voting in the 2020 election. Nearly one-fourth reported feeling their individual vote lacked impact. While our results are from a single center, our sample demographics were roughly reflective of the Accreditation Council for Graduate Medical Education.3

These findings, given long resident work hours4 and increasing physician burnout, illustrate the need for residency programs to address underlying barriers to voting. Training programs are beginning to encourage engagement in social advocacy for patients,5 but they may be unknowingly discouraging a bedrock form of civic engagement. More than 1400 employers already offer paid time off to vote as part of the Time To Vote campaign, and training programs could offer this as well. Training programs already go to great organizational lengths to accommodate comparatively longer absences for fellowship and post-residency jobs. Additional solutions include providing logistical information about when and where to vote and voting by mail.

As a voluntary survey, our findings may reflect response bias and social desirability bias. However, our results are generally consistent with population surveys.6

Early career physicians may be accultured into prioritizing working over voting. It is clear that physicians vote less than the general population, and developing institutional norms to encourage physician voter engagement is much needed.