Printed from: http://www.stfm.org/NewsJournals/FamilyMedicine/PastIssues/ArticleView

Student Specialty Plans, Clinical Decision Making, and Health
Care Reform

Robert L. Williams, MD, MPH; Crystal Romney; Miria Kano, PhD; Randy Wright; Betty Skipper, PhD; Christina Getrich, PhD; Andrew L. Sussman, PhD, MRCP; Stephen J. Zyzanski, PhD

Background and Objectives: Health care reform aims to increase evidence-based, cost-conscious, and patient-centered care. Family medicine is seen as central to these aims in part due to evidence of lower cost and comparable quality care compared with other specialties. We sought evidence that senior medical students planning family medicine residency differ from peers entering other fields in decision-making patterns relevant to these health care reform aims.

Methods: We conducted a national, anonymous, internet-based survey of senior medical students. Students chose one of two equivalent management options for a set of patient vignettes based on preventive care, medication selection, or initial chronic disease management scenarios, representing in turn evidence-based care, cost-conscious care, and patient-centered care. We examined differences in student recommendations, comparing those planning to enter family medicine with all others using bivariate and weighted, multilevel, multivariable analyses.

Results: Among 4,656 surveys received from seniors at 84 participating medical schools, students entering family medicine were significantly more likely to recommend patient management options that were more cost conscious and more patient centered. We did not find a significant difference between the student groups in recommendations for evidence-based care vignettes.

Conclusions: This study provides preliminary evidence suggesting that students planning to enter family medicine may already have clinical decision-making patterns that support health care reform goals to a greater extent than their peers. If confirmed by additional studies, this could have implications for medical school admission and training processes.

(Fam Med 2014;46(5):340-7.)

Three of the pillars upon which US health care reform are built are (1) increasing adoption of evidence-based approaches to care, (2) increasing adoption of cost-conscious care, and (3) increasing patient-centered care. Primary care, in general, and family medicine, in particular, are seen as key to the success of expansion and reform of health care, in part due to evidence that family medicine can deliver equivalent quality care at lower cost than other specialties. While there has been some controversy regarding cost and outcome of care provided by primary care clinicians, the weight of evidence suggests equivalent or better outcomes and lower costs of care when provided by primary care specialists.1-7

Evidence-based, cost-conscious, and patient-centered approaches to care are not and should not be limited to family medicine, but differential levels of their adoption across specialties might suggest one mechanism for the observed equivalent outcome/lower cost care provided by family medicine. We sought to determine if such differential adoption would be apparent even at the medical student level. There is evidence that students who choose a career in family medicine differ in certain demographic and personality attributes from those who choose other specialties.8-11 However, we are unaware of any previous research exploring differences in patterns of clinical decision making among students grouped by future specialty choice.

Our hypothesis was that senior medical students planning a career in family medicine would be more likely to demonstrate evidence-based, cost-conscious, and patient-centered clinical decision making than their peers entering other fields.

 

Methods

 
 
Design and Overview

We recruited a national sample of senior medical students to complete a brief online survey as part of a broader study of student clinical decision making. For this study, we used the survey to explore student recommendations for a set of patient care vignettes constructed to determine student views in selected areas. We evaluated differences in student recommendations by their self-described career plans, comparing those planning to enter family medicine to those entering other fields.

Survey Instrument

Four academic practicing family physicians designed a set of 15 patient vignettes, presenting the student with clinical scenarios: (1) in areas of controversy regarding preventive care services (three vignettes; evidence-based care), (2) involving a choice between heavily promoted brand name and generic medications of equal effectiveness (eight vignettes; cost-conscious care), or (3) involving a choice between recommendations for lifestyle modification or medication for newly diagnosed chronic illness (four vignettes; patient-centered care). Preventive care vignettes contrasted recommendations at the time of the survey of the US Preventive Services Task Force with those of a well-known scientific and advocacy group for screening for breast, prostate, or cervical cancer. Vignettes presenting choices of brand versus generic medication involved scenarios for both acute and chronic care. Those involving a choice between lifestyle or medication focused on scenarios of mild depression, mild hypertension, mild hyperlipidemia, and prediabetes (Table 1).

Table1

We designed all vignettes to present two equally clinically justifiable options for recommendations for care. Each participating student received a total of five of the described vignettes—one of which was a preventive care vignette—randomly mixed across the sample to reduce the chance of neighboring students’ receiving the same vignettes. Instructions for the survey stressed the clinical equivalence of the options presented. After the vignettes were completed, we asked students to provide selected demographic information and their postgraduate career plans.

The survey was conducted on-line and consisted of informational and consent pages followed by a series of screens for each of the five vignettes and the demographic information. Whenever a student skipped a vignette, the survey program returned the student to that vignette up to two times with requests to consider responding to it; however, students were not required to answer any vignette. We recorded time spent on each screen to permit us to note length of time students took to answer each vignette. After completing the survey, students were provided an opportunity to be entered into drawings for one of two $50 gift certificates to be drawn from participants at each medical school in addition to a second drawing for one of 20 iPad2s to be awarded from among participating students nationally. We maintained student anonymity by separating survey data and contact email addresses in unlinked databases.

Survey Piloting

The survey was piloted sequentially first with a group of family medicine residents and then with earlier graduating seniors at three medical schools. Readers may view the vignettes and a copy of the final version of the survey at http://med.unm.edu/mdm.

Distribution Process

Students were recruited to participate through their medical schools. We sent invitations with links to the survey directly to the students (with up to five solicitations to non-respondents), to a student listserv, or to a school contact person who in turn forwarded the invitation to the senior students, depending on each school’s preferred method. Re-invitation messages were sent every 7 days after the initial invitation. As a student clicked on the survey link, a unique version of the survey in the mix of vignettes was generated, using stochastic distribution methods. To guard against first position bias, we randomly varied the order in which the options for each vignette were presented to the student. We ran Monte Carlo simulations to test the equality of randomization.

Review and Approval Process

The study protocol was determined to be exempt by the University of New Mexico Institutional Review Board. We sought approval to survey their seniors from administrators at 130 of the 131 US allopathic medical school campuses that graduated seniors in 2012. At those schools where the administrators agreed to the survey, we also sought approval from the local Institutional Review Boards.

Analysis

We eliminated from analysis returned surveys in which the mean recorded time spent viewing each vignette was less than 10 seconds. We assessed this as the minimum time required to validly read and respond to a vignette and dropped those below this standard to reduce contamination by students participating solely to be entered in the incentive drawings. We then compared our remaining sample using standard descriptive statistics to senior students responding to the Association of American Medical Colleges Graduate Questionnaire in 2012.12 We initially summarized student recommendations across the sample for each vignette category (eg, preventive care) and then tested for variation in recommendations for each vignette category by student career plans, using chi-square statistics. Finally, we used weighted, multilevel, multivariate models to estimate the relationship of student career plans on student recommendations while controlling for student demographic descriptors, as well as clustering by medical school and for variation in medical school senior class sizes and response rates. We looked for response patterns indicating a first position response bias (P<.05), and found it in two of the brand versus generic medication vignettes, so we included a variable for response
order.

 

Results

 
 
Sample Characteristics

We successfully contacted administrators at 109 of the 130 campuses during the survey enrollment period from May 2011 to March 2012; 25 declined participation. The 84 participating schools represented 77% of successfully contacted schools and 65% of eligible schools. Sixty- eight schools permitted the full five rounds of survey invitations to students (Figure 1).

Figure1

We received 4,815 returned surveys (Figure 1), of which we eliminated 159 with mean response times of less than 10 seconds per vignette. The remaining 4,656 of the 11,438 seniors from the 84 participating schools represented an overall response rate of 40.7%. Comparison of our survey participants to the seniors participating in the 2012 Association of American Medical Colleges Graduate Questionnaire showed our participants were slightly more likely to be white, to be non-Hispanic, and to be entering family medicine, internal medicine, or pediatrics than the Graduate Questionnaire respondents (Table 2). Ninety-six percent of the students answered all five vignettes.

Table2
 

Relationship of Student Career Plans to Student Recommendations

Table 3 presents the results of bivariate analyses of student recommendations for each of the three categories of vignette patients, based on student postgraduate career plans. Overall, students were slightly more likely to agree with the US Preventive Services Task Force recommendations for the selected cancer screening services than with those of the scientific/advocacy group. About 75% of students recommended equivalent generic rather than brand name medications, and almost 60% recommended initial lifestyle change in preference to medication for vignette patients with newly diagnosed, mild, chronic illness. When examined by participant future career plans, there was a small, non-significant tendency for those entering family medicine to more commonly agree with the Preventive Services Task Force than their peers entering other fields. (A post-hoc power analysis showed the possibility of a type 2 error in this comparison with power estimated at 22%.) However, with vignettes comparing generic versus brand name recommendations, future family medicine trainees were significantly more likely to recommend the less costly, equivalent generic medication (P=.03) than those planning other training. Likewise, those planning family medicine training were significantly more likely to recommend a lifestyle change than medication for new onset mild chronic illness (P=.005) than their peers entering other fields.

Table3
 

Multilevel Analysis

Results of the multilevel modeling (Table 4) showed findings consistent with the bivariate analyses. There was no significant relationship of future career plans and the proportion of students agreeing with Preventive Services Task Force recommendations, while students planning to enter family medicine were significantly more likely to recommend generic medications (P=.01) and lifestyle change (P<.001) than their peers entering other fields.

Table4
 
 

Discussion

 
 

This study provides preliminary evidence that students planning to enter residency training in family medicine may already have decision-making patterns consistent with health care reform goals to a greater extent than peers entering other fields. Our findings suggest future family medicine residents are more likely to select lower-cost medications and to recommend lifestyle changes over medication. We did not find evidence that these future family physicians were more likely to recommend evidence-based preventive care in selected circumstances. However, post-hoc power analysis suggested that within our sample the number of students planning to pursue family medicine may have been too small to demonstrate a difference of the magnitude observed in this study area. While it is important to emphasize the preliminary nature of these findings, they do appear to be consistent with other research suggesting that primary care clinicians provide lower cost care, with equal health outcomes.1-7 Since all fourth-year medical students regardless of future plans would have received generally comparable training at the time in which they participated in this survey, it might be difficult to attribute the observed differences in survey response patterns to differences in training of students headed toward different fields. Prior research has provided some evidence that students selecting a career in family medicine differ in certain ways from students selecting other post-graduate training.8-11 Future family medicine trainees are described as “people oriented, driven by diversity in diagnosis and treatment,”13 “as sympathetic, trusting, cooperative, and altruistic,”14 and as having higher humanism scores.15 It is possible that our survey is uncovering other differences between future family physicians and their peers that are more directly related to clinical decision-making and approaches to care.

Limitations

In considering these findings, we recognize several possible limitations to this study. First, the response rate could suggest the possibility that those responding to the survey may not be fully representative of all senior medical students. The close similarity of demographics between our survey respondents and those of the AAMC’s Graduate Questionnaire, however, gives reassurance of representativeness. Second, our survey items may not have validly assessed the intended decision-making patterns. To our knowledge, this area of research is previously unexplored, and no validated survey exists. While this raises the possibility that the results are not reflective of the concepts we aimed to test, an alternative explanation for the differences, apart from that presented, is not apparent. A third possible limitation, regarding the use of vignettes to assess actual clinical practice, is less of a concern. Several previous studies have shown that clinical vignettes are valid estimators of actual clinical practice.16-19

 

Conclusions

 
 

These survey results are intriguing, as they suggest that students who plan a future career in family medicine may have differences from their peers with regard to health care decision making even before they have completed their training. These differences could be quite important as we move toward a health care system that values greater cost-consciousness and patient-centered care. If further study were to replicate these findings, it would be important to explore the origins of those differences, and to consider their implications for both medical school admissions and training processes.

Acknowledgments: Research reported in this publication was supported by the National Institute on Minority Health and Health Disparities of the National Institutes of Health under Award Number R01MD006073. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

This study was presented at the 2012 North American Primary Care Research Group Annual Meeting, New Orleans.

The authors greatly appreciate the important contributions of a number of persons without whose support this work would not have been possible: Denise Ruybal (administrative support); Charles North, MD, MSPH, Brian Solan, MD, MPH, and Daniel Stulberg, MD (vignette design); Jacque Garcia, MPH (data collection); Catherine Pino, BA and Joseph Colbert, BA (student research assistants). In addition, many faculty and staff persons at participating medical schools provided invaluable assistance to the study through their assistance with approval processes and survey distribution.

Corresponding Author: Address correspondence to Dr Williams, Department of Family and Community Medicine, MSC09 5040, 1 University of New Mexico, Albuquerque, NM 87131. 505-272-2165. Fax: 505-272-8045.
rlwilliams@salud.unm.edu.

 

References

 
 
  1. Starfield B, Shi L, Macinko J. Contribution of primary care to health systems and health. Milbank Q 2005;83(3):457-502.
  2. Franks P, Fiscella K. Primary care physicians and specialists as personal physicians. Health care expenditures and mortality experience. J Fam Pract 1998;47(2):105-9.
  3.  
  4. Shi L, Macinko J, Starfield B, Wulu J, Regan J, Politzer R. The relationship between primary care, income inequality, and mortality in US states, 1980–1995. J Am Board Fam Pract 2003;16(5):412-22.
  5. Baicker K, Chandra A. Medicare spending, the physician workforce, and beneficiaries’ quality of care. Health Aff (Millwood) 2004;Suppl Web Exclusives:W4-184-97.
  6. Parchman ML, Culler S. Primary care physicians and avoidable hospitalizations. J Fam Pract 1994;39(2):123-8.
  7. Kravet SJ, Shore AD, Miller R, Green GB, Kolodner K, Wright SM. Health care utilization and the proportion of primary care physicians. Am J Med 2008;121(2):142-8.
  8. Whittle J, Lin CJ, Lave JR, et al. Relationship of provider characteristics to outcomes, process, and costs of care for community-acquired pneumonia. Med Care 1998;36(7):977-87.
  9. Stilwell NA, Wallick MM, Thal SE, Burleson JA. Myers-Briggs type and medical specialty choice: a new look at an old question. Teach Learn Med 2000;12:14-20.
  10. Merrill JM, Laux LF, Lorimor R, Thornby JI, Vallbona C. Using attribution theory models to predict senior medical students’ perceptions of patients and career choice. Acad Med 1996;71(1 Suppl):S67–9.
  11. Buddeberg-Fischer B, Klaghofer R, Abel T, Buddeberg C. Swiss residents’ speciality choices—impact of gender, personality traits, career motivation and life goals. BMC Health Serv Res 2006;6:137.
  12. Rogers LQ, Fincher RM, Lewis LA. Factors influencing medical students to choose primary care or non-primary care specialties. Acad Med 1990;65(9 Suppl):S47-S48.
  13. Association of American Medical Colleges. Medical School Graduation Questionnaire 2012 All Schools Summary Report. Available at https://www.aamc.org/download/300448/data/2012gqallschoolssummaryreport.pdf. Accessed June 4, 2013.
  14. Kusurkar RA, Ten Cate TJ, van Asperen M, Croiset G. Motivation as an independent and a dependent variable in medical education: a review of the literature. Med Teach 2011;33(5):e242-62.
  15. Borges NJ, Savickas ML. Personality and medical specialty choice: a literature review and integration. Journal of Career Assessment 2002;10:362-80.
  16. Coutts-van Dijk LC, Bray JH, Moore S, Rogers J. Prospective study of how students’ humanism and psychosocial beliefs relate to specialty matching. Acad Med 1997;72(12):1106-8.
  17. Peabody JW, Luck J, Glassman P, Dresselhaus TR, Lee M. Comparison of vignettes, standardized patients, and chart abstraction. JAMA 2000;283:1715-22.
  18. Kirwan JR, Chaput de Saintonge DM, Joyce CRB, Currey HLF. Clinical judgment in rheumatoid arthritis. I. Rheumatologists’ opinions and the development of “paper patients.” Ann Rheum Dis 1983;42:644-7.
  19. Rice N, Robone S, Smith P. Analysis of the validity of the vignette approach to correct for heterogeneity in reporting health system responsiveness. Eur J Health Econ 2011;12:141-62.
  20. Peabody JW, Luck J, Glassman P, et al. Measuring the quality of physician practice by using clinical vignettes: a prospective validation study. Ann Intern Med 2004;141:771-80.

From the Department of Family and Community Medicine (Dr Williams, Ms Romney, Ms Kano, Dr Skipper, Dr Getrich, Dr Sussman) and Health Sciences Library and Informatics Center (Mr Wright), University of New Mexico; and Department of Family Medicine and Community Health, Case Western Reserve University (Dr Zyzanski).


Copyright 2017 by Society of Teachers of Family Medicine