METHODS
This cross-sectional study was conducted from February to April 2019 at a large academic institution with 6 regional ambulatory care family medicine clinics, each one with 11,000 to 18,000 patient visits annually. Faculty work at all 6 clinics and residents at 2 of them. All MAs are hired, paid, and managed by a central administrative department rather than by the family medicine department. The family medicine clinics are currently PCMH certified, with a mix of fee-for-service and capitated reimbursement.
We developed and piloted a voluntary, anonymous 39-question (29 closed-ended and 10 brief open-ended) online Qualtrics survey, which we distributed via an email link to all the MAs in the department. The survey included clinic site, years as an MA, perceptions of the clinic environment, perception of teamwork at their site, identification of efficient practices, and feedback for physicians to improve efficiency and flow. Most questions were Likert-style with 5 choices ranging from “strongly agree” to “strongly disagree” or short answer. Age and gender were omitted to protect confidentiality, as most MAs in the department are female. Participants could opt to enter in a drawing for three $25 gift cards. The survey was reviewed by the University of Michigan Institutional Review Board and deemed exempt.
We asked MAs to nominate peers in their clinic who were “especially efficient and do their jobs well—people that others can learn from.” The staff members who were nominated most frequently by their peers were invited to share additional perspectives via a 10- to 30-minute semi-structured interview with the first author. Interviews covered highly efficient practices, barriers and facilitators to efficient care, and physician behaviors that impaired efficiency. We interviewed a minimum of 2 MAs per clinic and increased the number of interviews through snowball sampling, as needed, to reach data saturation (eg, the point at which we were no longer hearing new content). MAs were assured that all comments would be anonymized. There was no monetary incentive for the interviews. The interviewer had previously met only 3 of the 18 MAs interviewed.
Analysis. Summary statistics were calculated for quantitative data. To compare subgroups (such as individual clinics), a chi-square test was used. In cases when there were small cell sizes (< 5 subjects), we used the Fisher’s Exact test. Qualitative data was collected with real-time typewritten notes during the interviews to capture ideas and verbatim quotes when possible. We also included open-ended comments shared on the Qualtrics survey. Data were organized by theme using a deductive coding approach. Both authors reviewed and discussed observations, and coding was conducted by the first author. Reporting followed the STROBE Statement checklist for cross-sectional studies.10 Results were shared with MAs, supervisory staff, and physicians, which allowed for feedback and comments and served as “member-checking.” MAs reported that the data reflected their lived experiences.
Continue to: RESULTS...