Abstract
Purpose
To measure the association between institutional investment in high-value care (HVC) performance improvement and resident HVC experiences.
Method
The authors analyzed data from two 2014 surveys assessing institutions’ investments in HVC performance improvement as reported by program directors (PDs) and residents’ perceptions of the frequency of HVC teaching, participation in HVC-focused quality improvement (QI), and views on HVC topics. The authors measured the association between institutional investment and resident-reported experiences using logistic regression, controlling for program and resident characteristics.
Results
The sample included 214 programs and 9,854 residents (59.3% of 361 programs, 55.2% of 17,851 residents surveyed). Most PDs (158/209; 75.6%) reported some support. Residents were more likely to report HVC discussions with faculty at least a few times weekly if they trained in programs that offered HVC-focused faculty development (odds ratio [OR] = 1.19; 95% confidence interval [CI] 1.04–1.37; P = .01), that supported such faculty development (OR = 1.21; 95% CI 1.04–1.41; P = .02), or that provided physician cost-of-care performance data (OR = 1.19; 95% CI 1.03–1.39; P = .02). Residents were more likely to report participation in HVC QI if they trained in programs with a formal HVC curriculum (OR = 1.83; 95% CI 1.48–2.27; P < .001) or with HVC-focused faculty development (OR = 1.46; 95% CI 1.15–1.85; P = .002).
Conclusions
Institutional investment in HVC-related faculty development and physician feedback on costs of care may increase the frequency of HVC teaching and resident participation in HVC-related QI.
High health care costs and suboptimal patient outcomes highlight concerns that the U.S. health care system does not provide sufficiently high value to patients. National efforts to address this range from the Choosing Wisely lists of low-value practices identified by medical professional societies and disseminated with the help of consumer advocacy groups1 to payment reforms shifting reimbursement from volume to value.2 To support these efforts, calls for teaching high-value care (HVC) in residency or graduate medical education (GME) have led to an increase in available resources to support curriculum and faculty development in this area.3,4 In 2012, when the Accreditation Council for Graduate Medical Education (ACGME) updated the core competencies to include provision of cost-effective care as a reporting milestone for internal medicine (IM) residents,5 only 15% of IM program directors (PDs) reported having an HVC curriculum, although over half reported a curriculum in development.6 In the same year, to facilitate national educational efforts, the American College of Physicians (ACP) and the Alliance for Academic Internal Medicine released a free off-the-shelf curriculum for teaching IM residents about HVC.7 Whether the increased national focus on HVC in GME has resulted either in an increased institutional investment in HVC-focused faculty or curriculum development or in an increased frequency of teaching residents about HVC is unknown.
Our goal was to measure, using two independently conducted surveys—of IM PDs and of IM residents—the association between investments by institutions to increase HVC teaching and corresponding residents’ perceptions of their experience with HVC teaching as well as their participation in HVC quality improvement (QI). We hypothesized that different types of institutional support would be differentially associated with teaching and QI activities in HVC. The objectives of this study were (1) to describe the types of institutional investment in HVC performance improvement reported by IM residency PDs, and (2) to measure the association between different types of institutional support and resident-perceived HVC teaching and QI.
Method
Surveys
We undertook a secondary analysis of two linked cross-sectional surveys independently conducted in 2014 (see Supplemental Digital Appendices 1 and 2 at http://links.lww.com/ACADMED/A551 and http://links.lww.com/ACADMED/A552). The Association of Program Directors in Internal Medicine Survey and Scholarship Committee e-mailed its annual survey to all member IM residency PDs, which represents 91% (361/396) of ACGME-approved IM PDs in the United States. The survey obtains data regarding program and PD characteristics and demographics, as well as PDs’ perceptions of current issues pertinent to GME. We used the results of the survey conducted from August through November of 2014 in this study. Questions relevant to HVC inquired about the presence of a formal curriculum, initiatives to promote HVC teaching, the presence of HVC-related faculty development efforts, and leadership support of HVC.
We used the results of the in-training examination (ITE) resident survey administered by the ACP to IM residents in October 2014. We excluded the data from residents who left the survey blank or selected the option to keep their responses unavailable for research purposes. The survey included questions about the IM ITE exam administration and preparation as well as questions about resident participants’ perceptions of HVC teaching at their institution: (1) the frequency of HVC teaching, (2) exposure to HVC-focused QI projects, and (3) views on topics related to HVC practice.
Approach
First, we linked the two surveys using the residency program identification numbers, and then we stripped the resident and program identifiers. We could not link 12 programs (of 361; 3.3%) because of a missing program number, and we excluded another 135 (37.4%) because the respondent did not answer the questions related to HVC teaching. To measure the correlation between leadership support of HVC teaching and resident perceptions, we modeled resident-reported HVC teaching as a function of PD-reported resources for HVC teaching. We controlled for overall IM-ITE scores to adjust for resident overall knowledge and attitudes, as well as for other program and resident characteristics including geographic region, size, and type of program, and postgraduate year (PGY) of resident. We modeled our other outcomes—that is, HVC QI involvement reported by the residents and their views on topics related to HVC—similarly.
Variables
Our exposures of interest were the types of institutional support offered by institutional leaders. PDs reported the types of institutional support for HVC teaching by answering the following question in a check-all-that-apply format: “How does your institution’s leadership support efforts to improve high-value, cost-conscious care?” Some of the choices were, for example, faculty development programs or financial support for curriculum implementation (see Supplemental Digital Appendix 1 at http://links.lww.com/ACADMED/A551 and Results). We measured the individual types of support reported, as well as a composite variable of “any support” reported. We conducted additional analyses using the following two questions: (1) “Does your program have a formal curriculum on high-value, cost-conscious care?” and (2) “Is faculty development on high-value, cost-conscious care offered at your institution?”
PDs also reported the number of QI projects developed in the past year that focused on HVC. The distribution of this variable was right-skewed, so we dichotomized the variable into “any” and “none.” We conducted additional analyses reporting the number of QI projects as a continuous variable, which did not alter our findings.
Our primary outcome was resident-perceived frequency of discussions with attending physicians about balancing the benefits and harms with costs of medical interventions, reported on a five-point Likert-type scale (1 = “Never”; 5 = “Every day”). We dichotomized the outcome into “at least a few times a week” (by combining the responses of “Few times a week” and “Every day”) and the other responses. Our secondary outcomes were whether or not residents participated in HVC-focused QI projects and residents’ views on topics related to HVC. The IM-ITE survey included 10 questions about their HVC-related views, all of which were scored on a five-point Likert scale (1 = “Completely disagree” to 5 = “Completely agree”). Please see Supplemental Digital Appendix 2 at http://links.lww.com/ACADMED/A552.
Analysis
Using logistic regression, we measured the correlation between institutional investment in HVC teaching and resident perceptions of teaching, reported participation in HVC QI projects, and views on HVC-related topics. We adjusted for resident and program characteristics such as PGY and program size. To account for overall program investment in education, we ran additional models controlling for overall IM-ITE scores. The results were stable, and we present adjusted results here. We adjusted all standard errors for clustering at the program level.
We conducted all analyses using STATA version 13.1 (StataCorp, College Station, Texas). The Institutional Review Board of the University of Pennsylvania deemed this study exempt research.
Results
The sample included 214 programs (59.3% of the 361 programs surveyed) and 9,854 residents (55.2% of the 17,851 residents surveyed). We present resident and program characteristics in Table 1. Nearly half of the 214 programs were community based and university affiliated (n = 104; 48.6%). Over one-third were university based (n = 85; 39.7%), and about a tenth were community-based programs (n = 23; 10.7%). The largest portion of the programs were located in the Northeast (n = 74; 34.6%), followed by the South (n = 54; 25.2%) and the Midwest (n = 51; 23.8%). Programs with fewer than 50 resident positions constituted the largest portion of the sample (n = 85; 39.7%), followed by larger-sized programs with 50 to 99 positions (n = 78; 36.5%) and programs with 100 to 149 positions (n = 37; 17.3%). Only 14 (6.5%) of the sampled programs had more than 150 positions. The 9,854 resident respondents represented the three PGY levels nearly about equally: 3,164 (32.1%) were PGY1 residents, 3,515 (35.7%) were PGY2, and 3,175 (32.2%) were PGY3. The distribution of resident respondents differed from program distribution: The majority of respondents were training in university-based programs (n = 5,929; 60.2%) and in larger programs with either 50 to 99 positions (n = 3,741; 38.0%) or programs with 100 to 149 positions (n = 2,988; 30.3%).
Table 1.
Demographics of Internal Medicine Residents Responding to a Survey and Characteristics of Internal Medicine Programs According to Program Director Respondents, 2014
| Characteristic | Residents, no. (% of 9,854) | Programs, no. (% of 214) |
|---|---|---|
| Program typea | ||
| Community based | 571 (5.8) | 23 (10.7) |
| Community based, university affiliated | 3,304 (33.5) | 104 (48.6) |
| Military based | 50 (0.5) | 2 (0.9) |
| University based | 5,929 (60.2) | 85 (39.7) |
| Regionb | ||
| Midwest | 2,671 (27.1) | 51 (23.8) |
| Northeast | 3,483 (35.3) | 74 (34.6) |
| South | 2,322 (23.6) | 54 (25.2) |
| West | 1,311 (13.3) | 33 (15.4) |
| Other | 67 (0.7) | 2 (0.9) |
| Size | ||
| < 50 positions | 1,921 (19.5) | 85 (39.7) |
| 50–99 positions | 3,741 (38.0) | 78 (36.5) |
| 100–149 positions | 2,988 (30.3) | 37 (17.3) |
| 150+ positions | 1,204 (12.2) | 14 (6.5) |
| Postgraduate year level | ||
| 1 | 3,164 (32.1) | NA |
| 2 | 3,515 (35.7) | NA |
| 3 | 3,175 (32.2) | NA |
Abbreviation: NA indicates not applicable.
Source: FREIDA (Fellowship and Residency Electronic Interactive Database) Online.
Source: U.S. Census region.
Table 2 summarizes the types of support and institutional resources available to PDs. Overall, of 209 PDs, 158 (75.6%) reported some form of leadership support for HVC. The top categories of support were requiring prior approval for specific medications (n = 77; 36.8%), providing physicians with performance data on costs of care (n = 54; 25.8%), and supporting the construct of HVC without providing resources (n = 51; 24.4%). Over half of the PDs (n = 109; 52.2%) reported receiving two or more different types of support, but nearly a quarter (n = 51; 24.4%) reported receiving no leadership support to improve HVC. A third of programs (66/207; 32.9%) reported offering HVC-focused faculty development, and nearly two-thirds (129/210; 61.4%) reported having a formal curriculum on HVC. About three-quarters of the PDs (149/203; 73.4%) reported developing HVC-focused QI projects in the past year. Interestingly, only a few residents (13.2%; 1,303/9,854) reported participating in an HVC-focused QI project (data not shown).
Table 2.
Internal Medicine Program Director-Reported Types of Institutional Support and Infrastructure for HVC Teaching, 2014
| Description of support/infrastructure | Programs, no. (%) |
|---|---|
| How does your institution’s leadership support efforts to improve high-value, cost-conscious care?a | |
| Specific medications require attending approval prior to ordering | 77 (36.8) |
| Provides physician performance data with respect to costs of care | 54 (25.8) |
| Supports the construct without providing resourcesb | 51 (24.4) |
| Provides cost information (such as cost of specific tests/procedures) to our program | 44 (21.1) |
| Provides faculty development programs | 42 (20.1) |
| Provides decision support tools | 42 (20.1) |
| Provides oversight of tests ordered to reduce unnecessary testing | 27 (12.9) |
| Provides financial support for implementation of a high-value, cost-conscious care curriculum | 14 (6.7) |
| Provides cost information systematically at the point of care (e.g., when ordering tests in the EHR) | 11 (5.3) |
| Other support | 5 (2.4) |
| Multiple types of support (2 or more) | 109 (52.2) |
| No supportb | 51 (24.4) |
| Is faculty development on high-value, cost-conscious care offered at your institution?c | |
| Yes | 66 (32.9) |
| No | 126 (60.9) |
| I don’t know | 15 (7.2) |
| Does your program have a formal curriculum on high-value, cost-conscious care?d | |
| Yes | 129 (61.4) |
| No | 26 (12.4) |
| No, but we are working on it | 55 (26.2) |
| How many quality improvement projects developed in the past year have focused on high-value, cost-conscious care?e | |
| Any projects | 149 (73.4) |
| Resident-initiated and/or led | 117 (57.6) |
| Faculty-initiated but include residents | 83 (40.9) |
Abbreviations: HVC indicates high-value care; EHR, electronic health record.
The total number of respondents for the first question (“How does your institution’s leadership support efforts to improve high-value cost-conscious care?”) was 209.
“Supports the construct without providing resources” and “No support” were distinct answer choices. Observed answers did not overlap between these categories, suggesting respondents interpreted the two options as distinct.
The total number of respondents for the second question (“Is faculty development on high-value, cost-conscious care offered at your institution?”) was 207.
The total number of respondents for the third question (“Does your program have a formal curriculum on high-value, cost-conscious care?”) was 210.
The total number of respondents for the fourth question (“How many quality improvement projects developed in the past year have focused on high-value, cost-conscious care?”) was 203.
Table 3 summarizes the relationship of PD-reported institutional support for HVC performance improvement and residents’ experiences of HVC. Residents were more likely to report having discussions with faculty at least a few times a week if they trained in programs where institutional leadership provided faculty development program support (odds ratio [OR] = 1.21; 95% confidence interval [CI] 1.04–1.41; P = .02), in programs where physician performance data on costs of care were available (OR = 1.19; 95% CI 1.03–1.39; P = .02), or in programs where (regardless of leadership support) HVC-focused faculty development was offered (OR = 1.19; 95% CI 1.04–1.37; P = .01). The availability of faculty development on HVC was also associated with resident-reported participation in HVC QI projects (OR = 1.46; 95% CI 1.15–1.85; P = .002). Residents who trained in programs where there was a formal HVC curriculum were also more likely to report participation in HVC QI projects than residents in programs without a formal HVC curriculum (OR = 1.83; 95% CI 1.48–2.27; P < .001).
Table 3.
Association of IM Program Leader-Reported Institutional Support With IM Resident Perceptions of HVC Teaching and QI Projects, 2014
| Manifestations of institutional support of HVC | Faculty discuss balancing benefits and harms with cost when caring for patients at least a few times a week, ORa (95% CI) | Participation in a HVC QI project, ORa (95% CI) |
|---|---|---|
| Leadership supportb | ||
| Provides faculty development programs | 1.21c (1.04–1.41) | 1.14 (0.86–1.50) |
| Provides financial support for implementation of a high-value, cost-conscious care curriculum | 0.96 (0.75–1.24) | 1.30 (0.84–2.00) |
| Provides cost information (such as cost of specific tests/procedures) to our program | 1.17 (1.00–1.37) | 1.32 (1.00–1.74) |
| Provides physician performance data with respect to costs of care | 1.19c (1.03–1.39) | 1.07 (0.83–1.38) |
| Provides cost information systematically at the point of care (e.g., when ordering tests in the EHR) | 1.11 (0.83–1.49) | 0.98 (0.60–1.62) |
| Provides decision support tools | 0.99 (0.84–1.16) | 1.21 (0.92–1.59) |
| Provides oversight of tests ordered to reduce unnecessary testing | 1.04 (0.86–1.26) | 0.79 (0.56–1.11) |
| Specific medications require attending approval prior to ordering | 1.07 (0.94–1.22) | 0.95 (0.75–1.19) |
| Supports the construct without providing resources | 1.00 (0.87–1.16) | 1.08 (0.84–1.39) |
| Any leadership support | 1.12 (0.97–1.30) | 1.27 (0.99–1.65) |
| No leadership support | 0.89 (0.77–1.03) | 0.78 (0.61–1.02) |
| Faculty development on HVC offeredd | 1.19c (1.04–1.37) | 1.46c (1.15–1.85) |
| Presence of a formal HVC curriculume | 0.93 (0.82–1.06) | 1.83c (1.48–2.27) |
Abbreviations: IM indicates internal medicine; HVC, high-value care; QI, quality improvement; EHR, electronic health record; OR, odds ratio; CI, confidence interval; IM-ITE, internal medicine in-training examination; PGY, postgraduate year.
Adjusted for IM-ITE total score; size, region, and type of program; PGY year; and to account for clustering within programs. Unadjusted results consistent so not shown.
The total number of respondents for the questions on leadership support was 9,660.
Statistically significant at P < 0.05.
The total number of respondents for the question on HVC faculty development was 9,556.
The total number of respondents for the question on a formal HVC curriculum for residents was 9,587.
Resident-reported agreement with statements related to HVC varied (Table 4). Of the 9,854 residents responding, 21.9% (n = 2,155) agreed that “it is unreasonable to expect physicians to understand their patients’ out-of-pocket costs”; 27.1% (n = 2,670) agreed that “physicians are too busy to worry about the costs of tests and procedures”; and 29.4% (n = 2,896) agreed that “physicians should order tests that might prevent a malpractice suit, even if those tests are unlikely to alter patient management.” We detected no consistent associations between any type of leadership support or institutional infrastructure for HVC teaching and resident attitudes on topics related to HVC; however, compared with residents in programs without formal HVC curricula, residents in programs that provided a formal curriculum were less likely to agree that “physicians should order tests that might provide additional information, even if those tests are unlikely to alter patient management” (OR = 0.86; 95% CI 0.75–1.00; P = .04) and that “physicians should order tests that might prevent a malpractice suit, even if those tests are unlikely to alter patient management” (OR = 0.88; 95% CI 0.77–1.00; P = .04). In addition, residents in programs that provided faculty development were less likely than residents in programs that do not offer HVC faculty development to agree that “physicians should order tests that might provide additional information, even if those tests are unlikely to alter patient management” (OR = 0.84; 95% CI 0.72–0.98; P = .02), and they were more likely to agree that “physicians need to take a prominent role in limiting use of unnecessary tests” (OR = 1.16; 95% CI 1.00–1.35; P = .04).
Table 4.
Internal Medicine Residents’ Views on Topics Related to High-Value Care and Their Association With the Presence of Faculty Development, Formal Curriculum, and Leadership Support as Reported by Internal Medicine Program Directors, 2014
| Viewpoint on high-value care | Disagree, no. (% of 9,854)a | Agree, no. (% of 9,854)a | Faculty development, OR of agreement (P value) | Formal curriculum, OR of agreement (P value) | Any support, OR of agreement (P value) |
|---|---|---|---|---|---|
| 1. Physicians should order tests that might provide additional clinical information, even if those tests are unlikely to alter patient management. | 5,748 (58.3) | 2,494 (25.3) | 0.84 (0.02)b | 0.86 (0.04)b | 0.87 (0.08) |
| 2. Physicians should take an active role in identifying waste within their own hospital system(s). | 490 (5.0) | 8,568 (87.0) | 1.12 (0.22) | 1.07 (0.40) | 1.10 (0.33) |
| 3. There is currently too much emphasis on costs of tests and procedures. | 4,560 (46.3) | 2,836 (28.8) | 0.92 (0.47) | 0.88 (0.21) | 0.88 (0.30) |
| 4. I should sometimes deny beneficial but costly services to certain patients because resources should go to other patients who need them more. | 5,300 (53.8) | 2,059 (20.9) | 0.91 (0.27) | 0.89 (0.15) | 0.99 (0.99) |
| 5. Physicians need to take a prominent role in limiting use of unnecessary tests. | 517 (5.3) | 8,355 (84.8) | 1.16 (0.04)b | 1.07 (0.33) | 0.98 (0.80) |
| 6. It is unreasonable to expect physicians to understand their patients’ out-of-pocket costs. | 5,819 (59.1) | 2,155 (21.9) | 0.97 (0.55) | 0.95 (0.34) | 0.94 (0.30) |
| 7. Physicians are too busy to worry about the costs of tests and procedures. | 5,182 (52.6) | 2,670 (27.1) | 0.91 (0.12) | 0.96 (0.46) | 0.97 (0.59) |
| 8. Physicians should order tests that might prevent a malpractice suit, even if those tests are unlikely to alter patient management. | 4,332 (44.0) | 2,896 (29.4) | 0.97 (0.65) | 0.88 (0.04)b | 0.99 (0.94) |
| 9. The cost of a test or medication is only important if the patient has to pay for it out-of-pocket. | 7,059 (71.6) | 1246 (12.6) | 0.90 (0.34) | 0.91 (0.34) | 1.02 (0.87) |
| 10. Physicians should discourage the use of interventions that have a small advantage over standard interventions but carry a greater potential for harm. | 891 (9.0) | 6,626 (67.2) | 1.02 (0.72) | 1.01 (0.89) | 1.11 (0.08) |
Abbreviation: OR indicates odds ratio.
Percentages do not add up to 100% because the neutral answer category “Neither agree nor disagree” was omitted from the table.
Statistically significant at P < 0.05.
Discussion and Conclusions
Our findings, based on our analysis of two linked surveys (of IM PDs and of IM residents), highlight the importance of faculty development to improve HVC teaching in residency. As many educators have probably suspected, studies have shown that residents model their practice after their attending physicians.8,9 In the clinic setting, for example, a multisite study of generic statin prescribing found that PGY1 residents were twice as likely to prescribe a brand name drug when precepted by attending physicians who were in the highest versus the lowest quartile of brand name prescribing.8 Research indicates that cost-effective practice patterns developed during residency training persist later in independent practice. A study of Medicare billing data for general internists and family physicians who completed residency between 1992 and 2010 found that those physicians’ spending patterns in independent practice closely reflected the patterns of the region where they trained.9 Our findings, considered with this previous research,8,9 suggest that investments in faculty development may have long-lasting effects on spending patterns across different settings.
Residents training in programs that reported receiving support specifically for faculty development in HVC were more likely to report both experiencing HVC teaching by faculty at least a few times a week and participating in HVC QI efforts. Given the relatively low levels of reported institutional support for HVC teaching—specifically, that 25% of IM PDs report no support from leadership and only 20% report receiving support for faculty development in HVC teaching—these findings are concerning. Providing physician performance reports about costs of care was also associated with resident-perceived higher frequency of HVC teaching, and having a formal HVC curriculum was associated with resident participation in HVC-related QI projects. Our findings suggest that medical center leaders looking to make an investment in HVC teaching should focus their efforts on faculty development, as faculty knowledge of HVC translates into increased use of these concepts when interacting with residents.
We observed a dramatic rise in HVC curriculum adoption between 2012—when only 15% of PDs reported on a survey that their institution had a formal HVC curriculum6—and 2014 when, according to our study, over 60% reported a formal HVC curriculum. These findings, consistent with the 50% of PDs who reported in 2012 that they were planning to implement an HVC curriculum,6 provide empiric evidence of the commitment of the academic medicine community to HVC education.10 This trend may also reflect the uptake of the Alliance for Academic Internal Medicine–ACP HVC curriculum launched in 2012,7 which was designed to require minimal institutional investment to implement. Furthermore, the last module in the curriculum focuses exclusively on HVC QI, which may explain the association we observed between having a formal HVC curriculum and resident-reported participation in HVC QI. Notably, having a formal HVC curriculum was not significantly associated with resident-perceived HVC discussions with attendings, again highlighting the need for faculty development and performance feedback reports (in addition to didactic teaching) to improve HVC teaching on rounds. While some programs have started reporting cost-of-care data to residents,11 the role of direct feedback on resident practice patterns to complement training in HVC requires further study.
While three-quarters of the PDs reported having HVC-related QI projects, only 13% of residents reported participation in HVC QI. This difference between PD and resident responses suggests that HVC QI projects were concentrated within residents or possibly involved only faculty and may reflect insufficient time allotted for residents to participate in QI. Many residents were not aware of HVC QI in their programs, suggesting that broader dissemination and participation in HVC-focused QI projects may increase reported resident awareness of ongoing HVC efforts. Patel and colleagues12 observed similar discordance between resident and PD perceptions as reported in surveys in 2012; according to their findings, only about 20% of paired PDs and residents agreed that faculty consistently modeled cost-conscious care. Furthermore, a relatively high proportion of residents reported attitudes poorly aligned with HVC, such as malpractice concerns and the belief that physicians are too busy to worry about costs. Although some of the observed attitudes may be a consequence of vague question wording resulting in alternative interpretations by the respondents (e.g., tests that would not alter management but provide additional information may be appropriate to inform patient decisions at the end of life), other reported attitudes (such as malpractice concerns) highlight the need for focused and persistent intervention. Our findings suggest that specific investments by institutions may lead to improvements. For example, residents in programs with a formal HVC curriculum were less likely than their peers in programs without such a curriculum to agree that malpractice concerns should motivate test ordering.
Prior studies of trainee attitudes toward HVC found that informal role modeling by attending physicians was more influential than formal didactics.13,14 Similar to evidence from fields such as patient safety, data show that organizational culture affects HVC learning and can be objectively characterized and distinguished across organizations.15 Consistent with these studies, a recent systematic review of interventions to teach HVC found that a supportive environment, along with knowledge transfer and reflective practice, was a key common component of successful interventions to improve HVC teaching.16 Our findings regarding the importance of investment in faculty development for HVC teaching highlight a potential goal for medical center leaders to work toward: to instill a culture of HVC among frontline providers. Other ideas, specifically for faculty development, have been proposed and disseminated by the Consensus of the Millennium Conference on Teaching High Value Care.17 On the basis of recommendations from seven medical schools, the Consensus provided evidence-based advice for academic leaders looking to implement faculty development in HVC principles as a strategy to effect culture change.17
This study has several limitations. First, it represents a secondary analysis of survey data. Our interpretation of the results is limited by the respondents’ recall, priming, and social desirability biases common to survey studies; nevertheless, our two independent sources of information—the resident and PD surveys—appear to be generally consistent, which supports the accuracy of the responses. Furthermore, although the responses of residents and PDs alike were subject to recall and social desirability biases, we somewhat alleviated priming by triangulating the two different surveys. Second, our ability to make causal inferences based on the associations we observed was limited by the cross-sectional observational study design. Future longitudinal studies should investigate how institutional prioritization of HVC drives individual resident decision making.18 Third, although we obtained a relatively high response rate for surveys of physicians, the overall sample represented only half of the targeted population, heightening concerns about selection bias. Regrettably, we were unable to compare the respondents’ demographics with those of the populations surveyed because those data were not collected for the nonrespondent residents and PDs. Fourth, only a single specialty is represented, although findings are likely generalizable to other specialties. Lastly, neither survey was formally validated; however, both surveys have been used in the field, and other published studies have used data/ results from prior versions.13,19
In sum, we observed a strong association between institutional support for faculty development in HVC teaching and resident-reported teaching on HVC as well as resident participation in HVC-focused QI projects. The availability of a formal curriculum on HVC and access to physician performance data on costs of care were also correlated with resident-reported perceptions of HVC QI and teaching. Nevertheless, a significant percentage of PDs reported little or no allocation of institutional resources to support of HVC initiatives. More consistent institutional support for HVC education, particularly in faculty development, may improve individual residents’ ability to “choose wisely” and train a workforce better prepared to practice within value-based payment models.
Supplementary Material
Acknowledgments:
The authors gratefully acknowledge Mike Adams, Darcy Reed, and the Association of Program Directors in Academic Medicine (APDIM) Survey and Scholarship Committee for their contributions to developing the questions on the APDIM survey.
Funding/Support: Dr. Ryskina is supported by the National Institute on Aging Career Development Award (K08AG052572). This study received no external funding.
Footnotes
Other disclosures: None reported.
Ethical approval: This study was deemed exempt from review by The Institutional Review Board of the University of Pennsylvania.
Supplemental digital content for this article is available at http://links.lww.com/ACADMED/A551 and http://links.lww.com/ACADMED/A552.
Contributor Information
Kira L. Ryskina, Division of General Internal Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania.
Cynthia D. Smith, Clinical Programs, American College of Physicians, and adjunct associate professor of medicine, Division of General Internal Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania.
Vineet M. Arora, Graduate Medical Education Clinical Learning Environment Innovation, Section of General Internal Medicine, Department of Medicine, University of Chicago, Chicago, Illinois.
Aimee K. Zaas, Division of Infectious Diseases and International Health, and program director, Duke Internal Medicine Residency, Duke University School of Medicine, Duke University, Durham, North Carolina..
Andrew J. Halvorsen, Office of Educational Innovations, Internal Medicine Residency Program, Mayo Clinic, Rochester, Minnesota.
Arlene Weissman, Research Center, American College of Physicians, Philadelphia, Pennsylvania..
Sandhya Wahi-Gururaj, Section of General Internal Medicine, and program director, Internal Medicine Residency, Department of Internal Medicine, University of Nevada, Las Vegas School of Medicine, Las Vegas, Nevada..
References
- 1.Grady D On board with the Choosing Wisely campaign. JAMA Intern Med. 2014;174:1396. [DOI] [PubMed] [Google Scholar]
- 2.Miller P, Mosley K. Physician reimbursement: From fee-for-service to MACRA, MIPS and APMs. J Med Pract Manage. 2016;31:266–269. [PubMed] [Google Scholar]
- 3.Weinberger SE. Providing high-value, cost-conscious care: A critical seventh general competency for physicians. Ann Intern Med. 2011;155:386–388. [DOI] [PubMed] [Google Scholar]
- 4.Eden J, Berwick DM, Wilensky G, eds. Institute of Medicine; Board on Health Care Services; Committee on the Governance and Financing of Graduate Medical Education Graduate Medical Education That Meets the Nation’s Health Needs. Washington, DC: National Academies Press; 2014. [PubMed] [Google Scholar]
- 5.Accreditation Council for Graduate Medical Education and American Board of Internal Medicine. The Internal Medicine Milestone Project. http://www.acgme.org/portals/0/pdfs/milestones/internalmedicinemilestones. pdf. Published July 2015. Accessed March 20, 2018.
- 6.Patel MS, Reed DA, Loertscher L, McDonald FS, Arora VM. Teaching residents to provide cost-conscious care: A national survey of residency program directors. JAMA Intern Med. 2014;174:470–472. [DOI] [PubMed] [Google Scholar]
- 7.Smith CD; Alliance for Academic Internal Medicine–American College of Physicians High Value, Cost-Conscious Care Curriculum Development Committee. Teaching high-value, cost-conscious care to residents: The Alliance for Academic Internal Medicine–American College of Physicians curriculum. Ann Intern Med. 2012;157:284–286. [DOI] [PubMed] [Google Scholar]
- 8.Ryskina KL, Dine CJ, Kim EJ, Bishop TF, Epstein AJ. Effect of attending practice style on generic medication prescribing by residents in the clinic setting: An observational study. J Gen Intern Med. 2015;30:1286–1293. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Chen C, Petterson S, Phillips R, Bazemore A, Mullan F. Spending patterns in region of residency training and subsequent expenditures for care provided by practicing physicians for Medicare beneficiaries. JAMA. 2014;312:2385–2393. [DOI] [PubMed] [Google Scholar]
- 10.Smith CD, Levinson WS; Internal Medicine HVC Advisory Board. A commitment to high-value care education from the internal medicine community. Ann Intern Med. 2015;162:639–640. [DOI] [PubMed] [Google Scholar]
- 11.Kurtzman G, Dine J, Epstein A, et al. Internal medicine resident engagement with a laboratory utilization dashboard: Mixed methods study. J Hosp Med. 2017;12:743–746. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Patel MS, Reed DA, Smith C, Arora VM. Role-modeling cost-conscious care—A national evaluation of perceptions of faculty at teaching hospitals in the United States. J Gen Intern Med. 2015;30:1294–1298. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ryskina KL, Korenstein D, Weissman A, Masters P, Alguire P, Smith CD. Development of a high-value care subscore on the internal medicine in-training examination. Ann Intern Med. 2014;161:733–739. [DOI] [PubMed] [Google Scholar]
- 14.Leep Hunderfund AN, Dyrbye LN, Starr SR, et al. Role modeling and regional health care intensity: U.S. medical student attitudes toward and experiences with cost-conscious care. Acad Med. 2017;92:694–702. [DOI] [PubMed] [Google Scholar]
- 15.Gupta R, Moriates C, Harrison JD, et al. Development of a high-value care culture survey: A modified Delphi process and psychometric evaluation. BMJ Qual Saf. 2017;26:475–483. [DOI] [PubMed] [Google Scholar]
- 16.Stammen LA, Stalmeijer RE, Paternotte E, et al. Training physicians to provide high-value, cost-conscious care: A systematic review. JAMA. 2015;314:2384–2400. [DOI] [PubMed] [Google Scholar]
- 17.Huang GC, Tibbles CD, Newman LR, Schwartzstein RM. Consensus of the Millennium Conference on Teaching High Value Care. Teach Learn Med. 2016;28:97–104. [DOI] [PubMed] [Google Scholar]
- 18.Sedrak MS, Patel MS, Ziemba JB, et al. Residents’ self-report on why they order perceived unnecessary inpatient laboratory tests. J Hosp Med. 2016;11:869–872. [DOI] [PubMed] [Google Scholar]
- 19.Ryskina KL, Halpern SD, Minyanou NS, Goold SD, Tilburt JC. The role of training environment care intensity in US physician cost consciousness. Mayo Clin Proc. 2015;90:313–320. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.