Physician stress in the era of COVID-19 vaccine disparity: a multi-institutional survey

Methods

On January 25, 2022, this study was approved by the Lincoln Memorial University Institutional Review Board (IRB, IRB# 1063 V.1), which expedited it due to the survey research format employed (45 CFR 46.110). Also, on February 8, 2022, this study was approved by the Rowan University IRB (IRB# PRO-2-22-7), which deemed it exempt. Grant funding was not pursued for this study, and this was not a clinical trial. Compensation was not offered or provided to incentivize participation in this study.

This study utilized a mixed-methods survey research format. Physician preceptor email lists were obtained from the Clinical Education Departments at each of the four osteopathic medical schools represented by the researchers. Lists were not specific to any one geographical region. A total of 4,842 physicians were sent the study invitation, but only 218 physician preceptors responded to the survey. Twenty-two incomplete responses were removed from the data summary, resulting in 196 complete responses for analysis.

An anonymous Qualtrics^® survey utilizing a modified version of the questionnaire from the expanded Physician Well-being Index (ePWBI) designed by MedEd Web Solutions (MEWS) was utilized in this study (permission obtained). The ePWBI was created by the Mayo Clinic to promote organizational and population health utilizing ongoing data gathering practices for continuous quality improvement. Because strategies are needed to reduce burnout and mitigate it when it occurs in physicians, the ePWBI was developed to enable patients to receive excellent care from compassionate, collaborative, and competent physicians. The ePWBI is a nine-item questionnaire developed to further ascertain the levels of distress and to compare relationships between “career satisfaction, meaning in work, high and low quality of life, fatigue, the likelihood of reporting a recent major medical error, recent suicidal ideation, and intent to reduce work hours or leave current medical practice. This ePWBI was validated by 6,880 physicians.”

The Qualtrics^® survey (link) was distributed via email, and periodic reminders were sent to nonresponders (Supplementary Material). The email described the study, including research questions and the IRB number with CITI-approved informed consent language attached. In the survey design options, the box that reads, “Don’t record respondents’ IP Address, location data, and contact Info in the survey. was chosen to ensure anonymity. Because this option was selected, the researchers did not have access to any identifiable information. In the first survey question, participants were asked to acknowledge or decline consent before the survey research would ensue. This question also indicated that completion of the survey would serve as consent. If respondents answered “No” to the consent question, the survey would automatically close. Study participants were also informed and were able to opt out at any time during the survey. In addition to the ePWBI question set, the following demographic information was collected: age, gender, COVID-19 vaccination status, United States region of medical practice, and medical credentials.

Respondents were also asked to indicate their agreement with COVID-19 vaccination status of healthcare workers and patients, the associated stress, and burnout levels. Respondents were also asked about the treatment and prescribing behaviors associated with working with COVID-19 vaccinated vs unvaccinated patients. Data analysis and sharing occurred with the research team. Only aggregated, de-identified data were available.

Descriptive statistics were utilized to analyze and compare responses from the survey. Survey data were collected and analyzed in Qualtrics. Cross-tabulation in Qualtrics XM was utilized to determine the association between sociodemographic variables (DO, MD, age, gender) and perceptions utilizing chi-square analysis. The level of significance for tests of association was set at a p value of less than 0.05. To analyze potential themes within responses to the open-ended questions, the researchers mapped the data to Excel and utilized inductive qualitative coding and analysis [14]. Utilizing this method, a coding schematic was developed based on the emerging data and the underlying themes were documented. Two researchers analyzed and reviewed the qualitative data and compared themes to promote the reliability of the findings. Utilizing a threshold of p≤0.05 in the Qualtrics XM platform, a chi-square analysis revealed many statistically significant relationships between the quantitative variables.

Results

A total of 4,842 physicians received the study invitation, and 218 physicians completed the survey, yielding a response rate of 4.5%. However, 22 incomplete responses were discarded, bringing the responses down to 196 total responses for analysis. Among those 196 responses, 187 identified as either an MD (100, 51.0%) or DO (87, 44.4%). Utilizing a threshold of p≤0.05, data analysis revealed many statistically significant relationships between the variables (Table 1). First, respondents’ vaccination status (question 6) impacted their views on the vaccination status of others as well as perceptions of burnout and stress. Physician respondents who had received the COVID-19 vaccine (96) were more likely (at a statistically significant level) to indicate that all healthcare workers should be vaccinated (p<0.001) and that all patients should be vaccinated (p<0.001). Interestingly, vaccination status also showed a statistically significant relationship with respondents’ perceptions of stress when working with unvaccinated healthcare team members (p<0.001) and/or patients who were unvaccinated (p<0.001). Vaccinated respondents (96) were also more likely to feel burned out due to the pandemic during the last 2 years (p=0.024) and indicated that they have observed burnout characteristics in healthcare team members when caring for unvaccinated patients that is different than when they care for vaccinated patients (p=0.026).

We also found a statistically significant relationship (p=0.026) between gender and respondents’ perception that their work schedule leaves them enough time for personal/family life. Respondents who identified as male (127; 64.8%) reported higher means (3.47) on this question than respondents who identified as female (2.98; 66; 33.7%). This could be due to a wide array of influences in the workplace and family dynamics, but it is an interesting finding nevertheless.

Lastly, respondents’ perceptions related to never experiencing burnout appeared to increase with age, with 60% (6/10) of the respondents aged 68+ years indicating that they had not experienced burnout, compared to only 20.9% (9/43) of respondents aged 24–34 years indicating they had not experienced burnout (see Figure 1). This may be due to generational differences in acknowledging burnout or speaking openly about it with others.

Figure 1:

Age-associated differences in respondent perceptions of experiencing burnout.

Open-ended questions provided feedback on the many challenges that physicians are encountering, including short-tempered patients, increased administrative responsibilities due to COVID-19 regulations, and patient bias/distrust of healthcare workers. Sample comments related to challenges included “people needing to push their viewpoints and opinions onto me” or “dealing with administration and other physicians who consistently treat unvaccinated patients, hospital staff, and colleagues with a contempt unseen in my lifetime” and “stress about vaccination requirements.” The most common comments (36%) centered around patient attitudes; however, the comments varied. The second most common response (14%) involved volume of work and changing healthcare guidelines, followed by concern over the attitude of healthcare toward caring for the unvaccinated (10%). Interestingly, personal/public safety was not among the top comments (8%). This information indicates that the stress that physicians are feeling extends beyond whether specific individuals are vaccinated and likely includes systematic and cultural challenges that weigh on mental well-being as well.

Discussion

Osteopathic physician training includes the foundational principle of a dynamic interaction between the mind, body, and spirit. In keeping with this philosophy, it is appropriate to examine if social determinants, such as vaccination status, impact physician well-being. In an effort to investigate if the vaccination status of physicians, their patients, or their colleagues impacts mindset, descriptive statistics of anonymous survey results from physicians in clinical education were analyzed and revealed that physicians vaccinated against COVID-19 still experienced significant stress.

The perception that stress for physicians was increased when working with unvaccinated patients and/or unvaccinated healthcare team members despite being vaccinated against COVID-19 themselves was evident in the survey results. This finding was surprising in light of the nature of COVID-19 vaccine in protecting from severe illness in most cases. The qualitative comments from physicians supported this conclusion and provided evidence that the stress came from a number of different areas, including patient attitudes, changing guidelines, eroding trust in healthcare providers, and attitudes of healthcare providers toward unvaccinated patients. This indicates that individual health alone was not the source of stress, but the issues are complex and relate to community health as well as the relationships that physicians build to care for their patients. Workplaces may consider if opportunities to compassionately provide medically relevant information to both patients and healthcare workers would help to relieve the stress that comes with difficult patient care topics.

Two other interesting findings were that more female respondents perceived that there was not enough time for personal/family life due to work schedules, and that younger individuals expressed burnout more often. These findings could reflect the stage of life, family dynamics, and the role of caretakers for family members. Women and younger physicians are often balancing family obligations, social calendars, and dealing with work tasks that could be less satisfying due to not being a senior. This finding, combined with the previous findings of work stress due to caring for unvaccinated patients or working with unvaccinated team members, indicates that job strain is being felt not only at work but also leaking into an employee’s personal time. This aligns with other studies showing that the strain of multiple roles, work tasks, and personal responsibilities contributes to mental stress [15, 16]. Instituting resources and programs to aid physician balance can help well-being and improve health system performance.

Work-related stress impacts staff and physician burnout by increasing costs to the healthcare system by increasing patient load beyond what is safe, causing expensive complications due to inappropriate medication decisions, leading to reduced adherence to treatment plans, and resulting in workers leaving the field early [13]. Survey results along with the stark comments made by physicians in the open comment section show extensive frustration, burnout, and strong feelings against both peers and patients. This should be addressed and repaired in order to build back trust and health for the system overall.

This survey of clinical preceptors from around the continental United States indicated that the pandemic has brought new challenges to the healthcare system. Burnout and stress are felt among vaccinated healthcare workers due to a wide array of factors and is more acute in the younger and female population surveyed. Being mindful of the well-being of our healthcare teams makes it incumbent upon those in administration to seek strategies that improve work life so that the workforce and patients can contribute to a healthier community.

Conclusions

Physician perspectives on working through COVID-19 show that there are many causes of stress. Importantly, vaccination status of those encountered during patient care contributed to perceived burnout. Aiding the physician workforce with resources to reduce burnout and improve patient outcomes should be part of the practice of all institutions that support and provide care to patients.