The concept of “hire your own” has attractive and intuitive benefits. Cultural fit is already understood by both parties, and there is little or no need to orient new hires. Additionally, there is better alignment of expectations between the health care system and physician. In a study by Rivera, in which 120 interviews of “elite professional service firms,” cultural fit is an important factor in hiring and can outweigh concerns about absolute productivity [12]. Yet, to our knowledge, the concept of “fit” and how it impacts other measurable areas of performance has not been studied previously. By conducting a retrospective analysis of KPSC GME hires over the past 20 years, we have demonstrated that physicians who had previously trained either in residency and/or fellowship at KPSC outperformed non-KPSC GME-trained physicians in five key metrics: Retention, Leadership (Current & Historical), Physician Relations, MAPPS, and Board Certification.
We believe KPSC is uniquely suited to address this question for several reasons. As an integrated healthcare system serving over 4.5 million members across 15 medical centers, it has a large physician labor pool requirement for constant, steady physician hires across its comprehensive medical and surgical specialties. This contrasts with most academic GME programs, where full time physician positions are much less likely to be available in proportion to the number of their GME graduates. Many traditional academic program GME graduates are likely to be hired in the community or other healthcare systems. Many, if not most, KPSC graduates will consider working within the KPSC after graduation.
As such, publications like ours are sparse. One study by Kohler et al., investigating physician retention in Michigan, found that 45% of in-state GME graduates practiced within Michigan at some point following graduation [12]. However, this data focused on state-wide retention rather than healthcare system specific retention. Bazemore, in response, argues there may be “imprinting” effects by GME programs on behaviors of trainees that contribute to how and where they practice [13]. To our knowledge, ours is the largest study examining the benefits of a large GME program within a fully integrated health care system that has the ability to hire its own GME graduates into staff physician positions within the medical group (or faculty practice).
There is not currently a standardized GME graduate recruitment process at KPSC. Physician recruitment is heterogenous and can depend on department culture and staffing needs. Over the past 20 years, KPSC GME hires represent just 12.3% of the total physician hires. However, they outperform non-KP GME hires in all five metrics that were analyzed. One of the most significant metrics in terms of overall impact to a healthcare system is retention. Hiring is a costly, time consuming process, and 82% of physicians hired into KPSC from an internal GME program between 2000 and 2020 are still active, while 76% of non-KPSC GME trained physician hires are active (202 inactive KPSC GME hires vs 4198 inactive non-KPSC GME hires). This difference in retention could be explained by increased familiarity with institutional culture and understanding of workflows.
Leadership helps inform and establish culture within a health care system. Greater proportions of internal hires who progress to positions of leadership can be beneficial for maintaining institutional knowledge and culture. GME graduate hires will have pre-existing relationships and thus have an assumed advantage in obtaining leadership positions, particularly in positions selected by peers. Proportionally, current and historical leaders have been over-represented by KPSC GME trained physicians. This could be, at least partly, explained by greater investment in the institution, as well as greater knowledge and understanding of the underlying systems. Leadership has also been shown as critical to a culture of safety within a health care system; safety performance is directly impacted by leadership [14].
Lower rates of Physician Relations cases were seen among KPSC GME trained physicians. These cases were related to behavioral and/or performance deficiencies that were managed by Physician Risk Officers. Disruptive behaviors or other adverse behavioral issues among physicians can have significant impacts on patient care [15]. While many do not recognize their behaviors, or the stress they are under, organizations still have a responsibility to address any underlying issues. Health care systems that have lower rates of PR cases can expect to expend fewer resources dealing with those deficiencies and to provide better overall patient care.
Patient satisfaction, as recorded by MAPPS surveys, was significantly higher among KPSC GME trained physicians. Higher patient satisfaction scores are not only important to subjective patient experiences, but are also associated with improved outcomes, including lower 30-day hospital readmission rates [16]. KPSC GME graduates are familiar with the mission and culture of the organization as well as the expectations of health plan members. Higher rates of patient satisfaction could be, in part, due to physicians having prior experience with the patient population and better understanding their needs.
Board certification criteria are determined by each respective specialty and imply that the physician has achieved a minimum level of qualifications and skills. It allows an organization to guarantee a baseline practice standard. Current board certification was found to be significantly higher (81% vs 74%) among KPSC GME trained physicians compared with non-KPSC trained peers. Higher rates of board certification are commonly associated with higher quality of care [17, 18].
There are some limitations of our study. First, Kaiser Permanente is unique; it is the largest integrated health care system in the United States, and KPSC is one of the largest health care organizations in the state of California. Therefore, our findings may not be fully generalizable to other health care systems or community health care groups. Second, the KP culture is also differentiated by many factors, including a focus on preventive care, and physicians who are salaried rather than paid on a fee-for-service basis. It can be argued that these and other cultural factors attract residents and fellows who preselect for affinity to KPSC, as the organization is well known across the regions it serves. Furthermore, demographic information, as well as exact time of hiring start and stop dates were not examined, and may have provided a better baseline comparison of the study groups. Finally, statistical samples are non-random, group sizes are asymmetric, and study variables are non-standard, as previously defined.
However, we strongly believe that this analysis represents an accurate picture of the value that GME trained hires can bring to their own health care systems, and that the setting in which this study is conducted is unique to address the question asked. We believe this results demonstrate the additional value that internal physician hires can bring to medical groups, integrated systems of care, and their patients.