Pursuing Osteopathic Recognition: A National Survey on US Program Director Perspectives

In 2014, the Accreditation Council for Graduate Medical Education (ACGME) and the American Osteopathic Association (AOA) announced the formation of a single accreditation system. Consequently, the AOA stopped accrediting residency or fellowship programs in any specialty after June 30, 2020. The former AOA programs or the ACGME-accredited residency programs that were pre-accredited after June 2015 are eligible to apply for Osteopathic Recognition (OR) by submitting an online application in the ACGME Accreditation Data System.¹  The ACGME Osteopathic Recognition Committee confers OR to the residency programs that demonstrate a commitment to education in osteopathic principles and practice (OPP).

For the majority of osteopathic medical students, OR is perceived as a mechanism to continue to train and practice as osteopathic physicians.²  The number of osteopathic students graduating from ACGME-accredited programs increased from 10 999 in 2014-2015 to 22 069 in 2018-2019.³  OR has been attractive not only to DO graduates but also to MD graduates.⁴  Prior studies suggest that programs apply for OR to increase osteopathic education across regions and increase their applicant pool.⁵  Since its inception in 2015, OR expanded almost 7-fold in its first 6 years, from 36 programs in 2015-2016 to 259 in 2020-2021.⁶  However, from 2021 to 2024, this total number has not changed, plateauing at 259. Why have many residency training programs that prioritized osteopathic graduate medical education prior to the single accreditation system not applied for OR? Scholars have suggested a lack of a sense of belonging as one reason.⁷  National data, covering a wide range of US regions and specialties, is lacking.

We sought to answer 3 research questions: For the formerly AOA-accredited residency programs that have not applied for OR, what are the perceived barriers and values of obtaining OR? Does the level of interest in pursuing OR differ in terms of geographic location, specialty, or other program demographics? What resources would be needed to motivate a program to apply for OR?

The survey (online supplementary data Appendix 1) first asked program directors whether the program had interest in applying for OR. The remainder of the survey had 4 sections: (1) barriers for pursuing OR; (2) value of pursuing OR; (3) residency program demographics; and (4) resources that would be helpful to motivate their program to apply for OR. Questions on the perceived barriers of pursuing OR were modified from a study by Hortos et al⁸  in which the authors gathered responses from medical students, residents, and faculty. We included some of their items with modification of wording and added more barrier items based on other studies and commentaries.^4,9,10  For perceived value of pursuing OR items, we used the items in the survey developed by Maier et al¹¹  with minor wording changes. We collected program demographic information such as the percentage of faculty with a DO degree, program location, and program specialty. The final section included an open-ended question about the resources that would motivate the program to apply for OR.

The survey was piloted with a convenience sample of 4 program directors from previously AOA-accredited programs who were asked to undergo cognitive interviews for the survey.¹²  Based on their suggestions, several questions were modified, and new items were added to the survey: physical space to practice OPP, presence of qualified faculty, and faculty education history.

The anonymous survey was distributed from May 23, 2023 to October 31, 2023, via the survey software Qualtrics. Program director contacts were provided by the American Association of Colleges of Osteopathic Medicine. The 458 programs surveyed in this study were among the 640 formerly AOA-accredited programs that transitioned to the ACGME, and for which we had contact email addresses, minus those that had already achieved OR. A $25 ClinCard was offered for participation in the research study.

If a survey was opened but the program director did not answer any questions, it was removed from data analysis. For the perceived barriers and value items, we performed contingency table analysis/chi-square test of independence. We combined the scale of “not at all challenging/important” with “slightly challenging/important” and combined “very challenging/important” with “extremely challenging/important” to help with the interpretation of the results and to avoid low cell count (<5) as much as possible. A statistically significant relationship between the group (Yes, Maybe, No) and a particular survey item was determined by the P value of the chi-square test being smaller than .05. The effect size of the chi-square test was measured by Cramer’s V. In addition, since there were 3 groups, we further examined the standardized residuals of each cell to pinpoint where exactly the difference occurred (ie, which 2 groups had statistically significant difference). If the absolute value of a standardized residual was larger than 1.96, it indicated a statistically significant difference. We also reported the expected count, which refers to the count if assuming there is no relationship between a particular item and level of interest. We conducted descriptive crosstab analysis to examine the distribution of the level of interest in pursuing OR by program demographics.

For the open-ended survey item, we conducted Braun and Clarke’s thematic coding analysis¹³  and reported the emergent themes. This approach consists of the following 6 phases: (1) familiarizing with data, (2) generating initial codes, (3) searching for themes, (4) reviewing themes, (5) defining and naming themes, and (6) producing the report. This approach was applied in analyzing open-ended survey responses in previous research.^14-16  Each open-ended survey response was coded by 2 researchers (M.S., T.D.) independently. Cohen’s kappa for the 2 raters was 0.88.

The study was approved by Uniformed Services University of the Health Sciences Institutional Review Board with protocol number DBS.2022.458.

A total of 178 out of 458 program directors (38.9%) responded to the survey. We compared the geographic region and specialty of these 178 programs with those of the 458 programs to determine if the collected samples were representative. Not all of the respondents provided information on geographic region and specialty; we used valid percentages for calculation. The US geographic region distribution of the collected samples (vs total population) was: Midwest 42.4% (36.8%), Northeast 34.5% (27.0%), Southeast 10.0% (18.5%), Southwest 8.6% (7.2%), and West 4.3% (10.5%). The specialty distribution of the collected samples (vs total population) was primary care/front line care 62.1% (51.2%), nonsurgical subspecialties 15.1% (30.1%), and surgical subspecialties 22.8% (18.7%).

Of the 178 programs, 67 (37.6%) indicated interest in applying for OR, 36 (20.2%) responded maybe, and 75 (42.1%) were not interested. The quantitative analysis is reported from 4 perspectives—perceived barriers, perceived values, program location and specialty, and program demographics.

The top 5 barriers, rated very or extremely challenging by the largest number of program directors, were: osteopathic recognition continued recognition requirement (68 of 178, 38.2%), lack of funding (62 of 178, 34.8%), osteopathic evaluation requirement (54 of 178, 30.3%), lack of support from colleagues (49 of 178, 27.5%), and insufficient time for OPP training in addition to normal residency schedule (48 of 178, 27.0%).

Out of the 16 survey items on barriers, 13 demonstrated overall/omnibus statistically significant difference between the groups (ie, Yes, Maybe, No in terms of interest in applying for OR). The 13 barriers were lack of interest from potential applicants to the program (χ²(4)=24.88, P<.001, Cramer’s V=0.29), lack of support from colleagues (χ²(4)=34.78, P<.001, Cramer’s V=0.35), insufficient time for OPP training in addition to normal residency schedule (χ²(4)=28.07, P<.001, Cramer’s V=0.31), osteopathic recognition evaluation requirement (χ²(4)=13.36, P=.01, Cramer’s V=0.22), osteopathic recognition continued recognition requirement (χ²(4)=21.13, P<.001, Cramer’s V=0.27), lack of osteopathic medicine in practice at site (χ²(4)=36.18, P<.001, Cramer’s V=0.35), lack of physical space to conduct OPP (χ²(4)=21.46, P<.001, Cramer’s V=0.27), uncertainty in integration of OPP in patient care (χ²(4)=16.44, P=.002, Cramer’s V=0.24), lack of relevance to specialty (χ²(4)=22.98, P<.001, Cramer’s V=0.28), lack of funding (χ²(4)=20.94, P<.001, Cramer’s V=0.27), uncertainty over documentation (χ²(4)=16.06, P=.003, Cramer’s V=0.24), and uncertainty over billing (χ²(4)=19.89, P<.001, Cramer’s V=0.27). As the degree of freedom was 4, the threshold of a large effect size was 0.25. Thus, most of these significant overall/omnibus differences had large or nearly large effect sizes (Table 1).

As indicated by bolded cells in Table 1, the Yes vs No group had significant differences in rating the following items as very or extremely challenging: lack of support from colleagues, insufficient time for OPP training in addition to normal residency schedule, OR continued recognition requirement, lack of osteopathic medicine in practice at site, lack of psychical space to conduct OPP, uncertainty in integration of OPP in patient care, lack of relevance of specialty, uncertainty over documentation, and uncertainty over billing.

Further, the direction of the differences was consistent—the Yes group had a lower proportion of rating these items as very or extremely challenging, while the No group had a higher proportion. For example, assuming there is no difference between groups, we would expect about 18 program directors in the Yes group to rate lack of support from colleagues as a very big challenge in applying for OR, but the actual count was only 7. In contrast, we would expect about 20 program directors in the No group to rate lack of support from colleagues as a very big challenge in applying for OR, but the actual count was as high as 33. All the expected counts were reported in parentheses in Table 1.

Overall, recruitment of DO students to the program was rated very or extremely important by the largest number of program directors (73 of 178, 41.0%). The groups of different interest level in pursuing OR were different on the value of recruitment of DO students (χ²(4)=23.62, P<.001), ability to provide instruction that includes OPP (χ²(4)=11.71, P=.02), and inclusion of osteopathic principles in enhancing patient satisfaction and fostering health promotion in your residency practice (χ²(4)=13.97, P=.007). The effect sizes were medium to large with Cramer’s V being 0.29, 0.21, and 0.22, respectively. Again, the direction of differences was consistent—the Yes group had a higher proportion of rating these items as very or extremely important, while the No group had a lower proportion. For example, assuming there is no difference between groups, we would expect about 27 program directors in the Yes group to rate recruitment of DO students to the program as a very or extremely important value of achieving OR, but the actual count was as high as 39. In contrast, we would expect about 30 program directors in the No group to rate recruitment of DO students to the program as a very or extremely important value of achieving OR, but the actual count was only 21 (Table 2).

The Southwest region showed the highest proportion of interest (9 of 12, 75%), and Midwest (n=19) and Northeast (n=16) had the highest number of interested residency programs (Table 3). In terms of program specialty, 40 of 90 (44.4%) primary care/front line care survey respondents were certain about the willingness to pursue OR, while 33 (37%) were not interested. For nonsurgical subspecialities, only 3 of 22 (14%) respondents indicated interest while 16 (73%) were not interested. In surgical specialties, the distribution of Yes, Maybe, and No was nearly even with 11 of 33 (33%) expressing clear interest, 10 (30%) not sure, and 12 (36%) indicating no interest (Table 4).

The program demographics included: percentage of faculty having a DO degree, percentage of faculty who can teach the OPP curriculum, percentage of residents who came from osteopathic medical schools, percentage of DO residents who complete the residency programs, and whether the residency program was community-based or university-based. Only one of these items showed a statistically significant relationship with level of interest in pursuing OR. There was a larger-than-expected number of programs with 26% to 50% of faculty who could teach OPP curriculum that expressed a clear interest in pursuing OR (χ²(6)=19.99, P=.003). The Yes group had a higher proportion of university-based residency programs compared to the Maybe or No group (online supplementary data Appendix 2).

A total of 128 program directors responded to the open-ended question, “What resources do you need to be able to help your program to apply for OR?” Initially, there were 32 code applications, which were refined into 24 primary codes. Ten themes were identified, and they were largely aligned with the Likert scale items results. The most frequently mentioned theme was “lack of resources regarding faculty” (n=44), followed by “guidance with the application process,” such as having an application template provided and assistance with navigating the Accreditation Data System (n=29). Other common resource issues included “administrative burden” (n=28), “fear of receiving citations and their program maintaining compliance and accreditation” (n=26), and “institutional and program support” (n=24). Several program directors stated that their programs were “already osteopathic at the core and hence already practicing OPP and OMT (osteopathic manipulative treatment)” (n=21), which was as prevalent as the theme “lack of resources regarding funding.” The least mentioned themes were “time” (n=17), “faculty development” (n=13), and “support from MD colleagues” (n=11). Online supplementary data Appendix 3 displays the count and ranking of the themes and online supplementary data Appendix 4 shows the exemplary quotes.

In this national survey of formerly AOA-accredited and now single-accredited residency programs that did not apply for OR since the transition, the majority of program directors expressed some or strong interest in pursuing OR. However, multiple barriers to obtaining OR were also reported, including the OR continued recognition requirement, lack of funding, the osteopathic evaluation requirement, lack of support from colleagues, and insufficient time for OPP training in schedules.

The primary care specialty residency programs expressed the highest interest in applying for OR. This is not surprising as the originally AOA-accredited residency programs included a large proportion of primary care specialty programs. The majority of perceived barriers to OR were viewed differently across the 3 levels of OR interest (Yes, Maybe, No), with statistical significance and medium to large effect sizes. That is, for the Yes group, the number of program directors who considered the barriers not challenging or just slightly challenging was smaller than expected, while for the No group, the number of program directors who regarded the barriers very or extremely challenging was larger than expected. From the survey, we cannot determine if the perceived barriers are contributing to a program director’s lack of interest, or whether lack of interest enhances the program director’s perception of OR barriers. The barriers of faculty, time, and funding needs were again mentioned in the open-ended survey question and may present targets for interventions. These findings are consistent with prior research showing that program directors need strong internal and external support systems to enable program development,¹⁷  as well as findings reported by 3 other studies that used different target populations.^8,11,18 

Underlying motivation to acquire OR appears to play a role as well. Program directors who indicated “Yes” interest in OR also perceived OR as valuable in the recruitment of DO students to their program. Surprisingly, program director survey responses showed that having faculty with a DO degree did not necessarily align with having faculty who can teach the OPP curriculum. This may indicate a need for faculty development, especially for programs that expressed interest in pursuing OR.

This study was limited by the low response rate, which may bias the findings in unknown directions. The survey was piloted with a small number of program directors for clarity and content but was not tested for further validity evidence. Thus, respondents may have interpreted questions differently than intended, and the closed-ended questions may have omitted key variables, which were not added in the single open-ended question. Using an open-ended survey question severely limits understanding, as additional clarification is not possible and requires further study.

Future steps to determine interventions to facilitate programs seeking OR would benefit from qualitative research methods. Of note, a study using in-depth interviews and focus groups with program directors as well as ACGME and AOA leadership, to explore the value of OR and how program leadership navigate its adoption and maintenance, is planned.

In this national survey of formerly AOA-accredited residency programs that have not applied for OR since the transition to a single accreditation system in 2015, the majority of program directors expressed some or strong interest in pursuing OR. There were many reported barriers to obtaining OR. These were particularly focused on resources and may affect the level of interest reported and the number of programs with OR.

We would like to thank the AACOM UME-GME Task Force for their support of the study.

The online supplementary data contains the survey used in the study and further data from the study.

Funding Sources: This study was funded by the Osteopathic Heritage Foundations and the American Association of Colleges of Osteopathic Medicine (AACOM).

Conflict of interest: The authors declare they have no competing interests.

Disclaimer: The opinions and assertions contained herein are the sole ones of the authors and are not to be construed as reflecting the views of the Uniformed Services University or the Department of Defense.