Newton, Dale A. MD; Grayson, Martha S. MD; Thompson, Lori Foster PhD
Many factors combine in complex and often poorly understood ways to influence medical students' career specialty choices.1,2 The potential lifestyle of a future career has been increasingly recognized as one important factor in specialty choice3 and may play a key role in the recent trend away from the primary care specialties and general surgery.4–7
Although prior studies have looked at lifestyle's influence on career choice, they have usually defined the controllable lifestyle careers on the basis of the investigators' a priori perceptions.3–5,7 For example, a recent study found that a controllable lifestyle was the most significant factor contributing to changing specialty choice trends.4 The authors examined this issue by tracking students' career preferences over time and then classifying choices in terms of lifestyle controllability, income, average work hours per week, and years of graduate medical education required. However, with respect to lifestyles, the authors' classifications may not be congruent with those of less experienced fourth-year medical students. Our study offers insights into medical students' perspectives of lifestyles associated with different career specialty choices.
In one of the earlier reports discussing the relationship between controllable lifestyle and career choice, the definition of “controllable lifestyle” was simply “control of work hours.”3, p. 606 A subsequent factor analysis by the same researchers reported that remuneration was also included within “lifestyle.”7 Indeed lifestyle and income are probably interrelated because medical students anticipating their future careers (compared to other physician specialists) may deem an “ample” income necessary to a satisfactory lifestyle. However, these two factors are not identical and do not always coincide. In accordance with recent work,4 we therefore examined lifestyle and income as two distinct factors in our study. We investigated the separate influence of each of these factors on career choice over time and unlike prior reports,3–5,7 by individual specialty.
We explored the variable influence of lifestyle and income on career choice via questionnaire data produced by students graduating from Brody School of Medicine at East Carolina University (BSOM) and New York Medical College (NYMC). For more than a decade, NYMC and BSOM have surveyed entering and graduating medical students about their career choices and the influences on these choices. The two schools are quite different: NYMC is a larger urban, private medical school and BSOM is a smaller rural, public school. Each has a traditional medical school curriculum and both emphasize primary care education which includes rotations in community-based practices. The demographic characteristics of the combined respondents from both schools were similar to national data in a prior study.8 The current version of NYMC's and BSOM's questionnaire has been used since 1998.
With the fourth-year medical students' responses to this instrument, we addressed the following research questions in our study:
▪ Have lifestyle and income become increasingly valued over time?
▪ What specialties do medical students view as lifestyle-friendly careers (that allow leisure time, opportunities to enjoy life outside of work, predictable work hours, time to pursue activities outside of work, and family time)?
▪ What is the relative influence of income and lifestyle on medical students' career specialty choices, and does this influence vary by specialty?
A total of 1,334 (73%) of the medical students graduating from BSOM (485 graduates) and NYMC (1,348 graduates) between 1998 and 2004 completed the IRB-approved questionnaires prior to residency “Match Day.” The 98-item questionnaire includes demographic and career-related variables. In our study, which was not externally funded, we focused specifically on the items addressing career specialty preferences, income concerns, and lifestyle values.
Early in the questionnaire, students were asked to specify which of 23 medical specialties they intended to pursue or to select an “other” option. Then students were asked to respond to multiple items describing various career attributes by using a four-point scale to rate the influence of each attribute on their selection of career specialties (1 = no influence; 2 = minor influence; 3 = moderate influence; 4 = major influence). We generated these attributes based on a synthesis of earlier literature.1,8
Prior to the present study, we randomly selected 900 first-year and 900 fourth-year students and their ratings were submitted to a principal axis factor analysis, followed by an oblique Promax rotation. Each questionnaire selected was from a different student (no duplications were allowed) and the sampling was from all respondents at both schools between 1998 and 2003. This analysis resulted in seven factors, two of which pertain to our study and supported our decision to examine lifestyle and income separately. Specifically, five of the items clustered to form a lifestyle-friendly factor, and three formed an income factor (see Table 1). Notably, none of these eight items loaded substantially (.3 or higher) on a second factor. Responses to the lifestyle items (alpha = .93) were therefore combined for analysis, as were responses to the income items (alpha = .85). In keeping with other research,9 parametric statistics (e.g., analysis of variance [ANOVA], t test) were used to examine the questionnaire data.
Of the 1,334 fourth-year medical student respondents, 1,327 indicated a career preference. Seven students chose careers selected by fewer than nine other students. These students (and their career choices) were dropped from the analyses because their career groupings lacked the sample size needed to achieve adequate statistical power during between-specialty comparisons. In addition, we eliminated the 32 respondents who selected the “other” category because we presumed their career interests were heterogeneous. The final sample therefore consisted of 1,288 respondents spanning 20 specialty preferences. Of the 1,258 people who reported their gender, 616 (49%) were female. The average age was 28.1 ± 3.2 years. Of the 1,280 respondents who reported their race or ethnicity, 751 (59%) were white, 324 (25%) were Asian or Pacific Islander, 89 (7%) were African American, 34 (3%) were Hispanic, 16 (1%) were American Indian or Alaskan natives, and the remaining 66 (5%) self-identified as “other.” Regarding affiliation, the sample was proportional to the relative size of the two medical school classes; 69% of the respondents were from NYMC.
With our first research question, we investigated whether students were becoming increasingly concerned with lifestyle and income when making career choices. We computed a Pearson correlation coefficient to examine the association between year of graduation and ratings on the lifestyle and income items. Results demonstrated that lifestyle became slightly more important from 1998 to 2004 (r = .07, n = 1,286, p = .018, two-tailed) with a similar rise in the importance of income (r = .07, n = 1,286, p = .011, two-tailed) (see Figure 1).
With our second research question, we sought to empirically distinguish lifestyle-friendly and lifestyle-unfriendly careers. We compared respondents intending to pursue the 20 specialties in terms of the degree to which they valued lifestyle. As Figure 2 shows, lifestyle played a notable role in medical students' decisions to specialize in areas such as radiology and physical medicine/rehabilitation, whereas students who valued lifestyle highly seemed relatively unlikely to pursue specialties such as general surgery and obstetrics–gynecology. A one-way ANOVA confirmed that lifestyle concerns differentially influenced the decision to pursue the various specialties (p < .001). Post hoc analyses compared every specialty pair in terms of lifestyle concerns after reducing the .05 significance level via a Bonferroni correction to control for family-wise error (i.e., the inflated type I error risk that occurs when a set of comparisons are made).10 Table 2 shows the degree to which graduating medical students considered each specialty lifestyle friendly. Students who chose lifestyle-friendly careers (radiology, physical medicine/rehabilitation, emergency medicine, ophthalmology, anesthesiology, urology, dermatology, and otolaryngology) valued lifestyle significantly more than did students pursuing many of the other specialties. Conversely, lifestyle matters influenced students pursuing the two lifestyle-unfriendly careers (general surgery and obstetrics–gynecology) significantly less than lifestyle matters influenced students who were pursuing the majority of the alternative specialties.
With our final research question, we looked at the importance of lifestyle compared to income by examining a mixed model ANOVA, which included value (lifestyle versus income) as the within-subjects factor and career specialty as the between-subjects factor. Results revealed a nonsignificant within-subjects effect (p = .949), indicating that the average value placed on lifestyle (2.70 ± .85) did not reliably differ from the average value placed on income (2.72 ± .76). A significant interaction term (p < .001) suggested that this balanced concern with lifestyle and income did not characterize all career specialties. After using a Bonferroni correction to adjust the .05 level of significance to control for family-wise error, we conducted a series of paired samples t tests to compare the value placed on lifestyle versus income within each specialty. As Figure 2 shows, medical students who chose seven specialties were significantly more likely to value lifestyle over income: radiology, physical medicine/rehabilitation, emergency medicine, ophthalmology, psychiatry, family practice, and general pediatrics. Meanwhile, students who chose four specialties were significantly more likely to value income over lifestyle: internal medicine subspecialties, orthopedics, general surgery, and obstetrics–gynecology.
Our data representing lifestyle and income's influences on medical students' career specialty choices show that between 1998 and 2004, both lifestyle and income became increasingly important to the graduating medical students in our study. However, the correlations between year of graduation and lifestyle/income were small, suggesting that recent rises in the importance of lifestyle and income may not be as dramatic as indicated by past work looking at career choice trends in the absence of data assessing students' perceptions of the factors they considered important when selecting a specialty.4
While lifestyle and income became only slightly more important over time to the students in our study, the absolute importance of these two factors (irrespective of time) should not be overlooked. When students were asked to indicate the influence of lifestyle and income, the average rating assigned to both of these factors was higher than 2 on the four-point scale (1 = no influence, 4 = major influence). It is interesting to note that the influence of lifestyle, as compared to income, did not differ significantly for the study group as a whole. However, when we examined specialty careers individually, we found a significant difference in the influence of income versus lifestyle for 11 of the 20 specialties studied.
Our respondents viewed eight specialties as lifestyle friendly: radiology, physical medicine/rehabilitation, emergency medicine, ophthalmology, anesthesiology, urology, dermatology, and otolaryngology. Students who chose each of these eight specialties indicated that lifestyle was significantly more influential in their career choices than did students who chose a number of other specialties. In addition, no other specialty received significantly higher lifestyle ratings than did these eight specialties. To the contrary, students with relatively minor lifestyle concerns tended to choose general surgery and obstetrics–gynecology, specialties considered more lifestyle unfriendly than any of the other 18 specialty choices.
Ten specialties (psychiatry, family practice, neurology, general pediatrics, general internal medicine, plastic surgery, the pediatric and internal medicine subspecialties, combined internal medicine–pediatrics, and orthopedics) were between the extremes, or lifestyle intermediate, as the students viewed them as more lifestyle friendly than some specialties and less so than others. It is of note that we found significant variation within the lifestyle-intermediate group. Both neurology and psychiatry were highly rated and only considered to be less lifestyle friendly than one specialty, radiology. In contrast, students rated three specialties at the lower end of the lifestyle-intermediate group: internal medicine subspecialties, combined internal medicine–pediatrics, and orthopedics. Although students gave these three specialties a low lifestyle rating, they rated them significantly higher than general surgery and obstetrics–gynecology.
While prior studies about the influence of lifestyle have defined controllable lifestyle careers on the basis of the investigators' a priori perceptions,3–5,7 we looked at the perceptions of fourth-year medical students in our study. The data revealed some important differences between the views of these students and the views of investigators published in the literature. First, in our study urology emerged as a lifestyle-friendly career specialty even though past studies have placed it in the uncontrollable-lifestyle category.4 Second, it was of great interest to note that in our study students rated physical medicine/rehabilitation, a specialty not included in prior studies, as the second most lifestyle-friendly specialty. Additionally, our data suggest that in the past investigators' tendency to dichotomize careers into lifestyle-controllable versus lifestyle-uncontrollable categories may have masked important complexities. The data in Table 2 reveal finer gradations and lead to the conclusion that in the eyes of fourth-year medical students, some of the careers typically lumped into the uncontrollable-lifestyle category (e.g., family practice, general pediatrics, general internal medicine) may actually reside between the lifestyle-friendly and lifestyle-unfriendly extremes.
In our study, we defined a lifestyle-friendly career as one that allows leisure time, opportunities to enjoy life outside of work, predictable work hours, time to pursue activities outside of work, and family time. Our definition of “lifestyle friendly” is therefore somewhat different than the definition of “controllable lifestyle” used in past research.7 Recent researchers who have based their work on defining controllable lifestyle as simply control of work hours acknowledged that controllable lifestyle may “capture other less tangible and even less easily quantified influences that ultimately affect lifestyle.”4,p.1,177 Thus, our expansion of the lifestyle construct may provide a useful addition to the literature.
From a practical standpoint, our study can be used to examine a number of the career trends in the medical community. For example, there has been an overall decrease in applications to primary care residencies, but an increase in the applications to medicine and pediatrics subspecialties.11 Because general internal medicine, general pediatrics, combined internal medicine–pediatrics, family practice, and medicine and pediatrics subspecialties were included in our list of potential career choices, our data provide insights into perceptions about ultimate career paths for students entering the primary care residencies. In general, students' perceptions of the primary care fields as a whole were consistent, all clustering in the lifestyle intermediate range. However, careers in internal medicine subspecialties or in combined internal medicine-pediatrics were rated as more lifestyle unfriendly than any of the other generalist specialties. We are unclear why a career in combined internal medicine–pediatrics was viewed less favorably than was a career in either of the two separate specialties, general internal medicine and general pediatrics. Although students rated a career in a subspecialty of medicine or pediatrics as lifestyle intermediate, we realize that this rating may have masked differences in students' perceptions of each of the various subspecialties. It is possible that students viewed one subspecialty, such as endocrinology, as more lifestyle friendly than another one, such as cardiology, thus leading to a mean score within the lifestyle-intermediate range.
Finally, while the analysis of recent changes in career trends has been largely attributed to lifestyle issues,4–6,12 we found that income concerns were also on the rise. As Figure 2 shows, lifestyle and income played distinct roles that varied by chosen specialty. Students entering fields such as orthopedics, general surgery, obstetrics–gynecology, and internal medicine subspecialties were more influenced by income issues than by lifestyle.
Our study had several limitations. First, the sample sizes characterizing certain career specialties were rather small, thereby limiting our power to detect statistically significant mean differences when examining those particular specialties. We took some measures to diminish this concern (i.e., careers chosen by fewer than ten students were dropped from the data set). Nevertheless, we looked at several careers with relatively modest sample sizes (e.g., plastic surgery, otolaryngology). Despite potential power limitations, we chose to include these careers to afford readers the opportunity to examine the ratings associated with these important specialties. Second, we conducted our study at two schools that both emphasize primary care education. Although the emphasis on primary care education may have influenced the absolute numbers of students who chose primary care careers, we do not believe this emphasis affected the relationship between student-rated influences and career choice. Also, the characteristics of the two schools are quite different, and the combined study population had demographic characteristics very similar to national data.8 Third, our results are students' reports of anticipated long-term specialty choice, rather than their actual entry to a preferred residency or fellowship. It is possible that a student stating plans to enter a career, such as a pediatrics subspecialty, would change his or her mind over the three years of a pediatrics residency. Fourth, we used a four-point scale that allowed students to express equal weights for lifestyle and income, if they were so inclined. We are unable to determine which factor would have emerged as the primary factor if respondents were forced to choose between the two. Fifth, we did not determine the accuracy of students' insights into the factors shaping their career preferences. However, choosing among specialties tends to be a very deliberative process; accordingly, we believe students' ratings provide meaningful information about the factors driving their career decisions. Finally, by design in this study we focused on only two factors (lifestyle and income) that influence career choice. We did not investigate the influence of other factors (e.g., altruism, debt, demographic characteristics).
Although recent literature has focused on the influence of lifestyle on medical students' career choices, we found that both lifestyle and income play a notable role. In order to reverse the current downward trend in applications to fields considered lifestyle unfriendly or economically unrewarding, the current models of practice and reimbursement may need to be restructured. Certain specialties have the potential to become more lifestyle friendly, with a reorganization of expected work hours within shared practices or with the increased use of physician extenders. It is possible that the wider use of technology, such as the electronic medical record, may allow some physicians the flexibility to do part of their work at home. Finally, the disparity in reimbursement by specialty for physician services continues to have a major influence on specialty choice. Policymakers need to take these considerations into account as they plan for future workforce needs.
1.Bland CJ, Meurer LN, Maldonado G. Determinants of primary care specialty choice: a non-statistical meta-analysis of the literature. Acad Med. 1995;70:620–41.
2.Newton DA, Grayson MS, Whitley TW. What predicts medical student career choice? J Gen Intern Med. 1998;13:200–3.
3.Schwartz RW, Jarecky RK, Strodel WE, Haley JV, Young B, Griffen WO. Controllable lifestyle: a new factor in career choice by medical students. Acad Med. 1989;64:606–9.
4.Dorsey ER, Jarjoura D, Rutecki GW. Influence of controllable lifestyle on recent trends in specialty choice by US medical students. JAMA. 2003;290:1173–78.
5.Lind DS, Cendan JC. Two decades of student career choice at the University of Florida: increasingly a lifestyle decision. Am Surg. 2003;69:53–55.
6.Thornton J, Esposto F. How important are economic factors in choice of medical specialty? Health Econ. 2003;12:67–73.
7.Schwartz RW, Haley JV, Williams C, et al. The controllable lifestyle factor and students' attitudes about specialty selection. Acad Med. 1990;65:207–10.
8.Grayson MS, Newton DA, Whitley T. First-year medical students' knowledge of and attitudes toward primary care careers. Fam Med. 1996;28:337–42.
10.Keppel G. Design and Analysis: A Researcher's Handbook. 3rd ed. Englewood Cliffs, NJ: Prentice Hall, 1991.
11.Newton DA, Grayson MS. Trends in career choice by us medical school graduates. JAMA. 2003;290:1179–82.
12.Gelfand D, Podnos Y, Wilson S, Cooke J, Williams R. Choosing general surgery: insights into career choices of current medical students. Arch Surg. 2002;137:941–47.