Applicant Considerations Associated with Selection of an Emergency Medicine Residency Program

Abstract

Objectives:  The primary objective of this study was to assess variables that residency applicants ranked as influential in making residency choices. The secondary objective was to determine if residents were satisfied with their residency choices.

Methods:  A secondary analysis was performed on a cohort database from a stratified, random sampling of 322 emergency medicine (EM) residents collected in 1996–1998 and 2001–2004 from the American Board of Emergency Medicine Longitudinal Study on Emergency Medicine Residents (ABEM LSEMR). Residents rated the importance of 18 items in response to the question, “How much did each of the following factors influence your choice of residency program location?” The degree to which residents’ programs met prior expectations and the levels of satisfaction with residency programs were also assessed. All analyses were conducted using descriptive statistics.

Results:  Three-hundred twenty-two residents participated in the survey. Residents considered the following to be the most important variables: institutional reputation, hospital facilities, program director reputation, and spousal influence. Several geographic and gender differences were noted. Ninety percent (95% confidence interval [CI] = 86% to 93%) of residents surveyed in their final year answered that the residency program met or exceeded expectations. Seventy-nine percent (95% CI = 76% to 82%) of residents identified themselves as “highly satisfied” with their residency choice.

Conclusions:  The most influential factors in residency choice are institutional and residency director reputation and hospital facilities. Personal issues, such as recreational opportunities and spousal opinion, are also important, but are less influential. Significant geographic differences affecting residency choices exist, as do minor gender differences. A majority of residents were highly satisfied overall with their residency choices.

An abundance of manpower and financial resources are utilized each year recruiting applicants to emergency medicine (EM) residency programs. In 2008, 1,822 residency applicants applied for 1,399 available positions in 133 EM residency programs.¹ Residency applicants listed an average of eight programs with the National Residency Match Program.

Prior studies indicate that positive resident opinions of the program,² the content of a program’s website,³ and the availability of specialized study in ultrasound or international medicine⁴ can increase a program’s appeal to applicants. DeSantis and Marco⁵ reported that that the five most important considerations to applicants who selected their EM residency program were friendliness, environment, interview day, academics, and location. Elucidating important considerations for applicants to EM residency programs could influence program directors’ decisions in allocating resources and focusing efforts while recruiting residents.

The American Board of Emergency Medicine (ABEM) designed the Longitudinal Study of Emergency Medicine Residents (LSEMR) to describe the development of careers from EM-1 year of residency through retirement among representative emergency physicians (EPs). The long-term goals of the study were to examine how residents’ early goals, aspirations, and motivations are related to later choices made during their careers.

The primary objective of this study was to describe variables that residents participating in the ABEM LSEMR had ranked as influential when making National Residency Match Program rank decisions. The secondary objective was to determine if residents are satisfied with their choice of residency programs.

Methods

Study Design and Population

This is a secondary data analysis of data gathered from two rounds of the ABEM LSEMR. The LSEMR survey was administered to a random sample of EM residents during the last 3 years of their residencies (Years 1, 2, and 3 if a 3-year residency program and Years 2, 3, and 4 if a 4-year residency program). Data from all physicians who participated were included in the current analysis: 1996–1999 (Round 1 = 153 respondents, 96% response rate), 2001–2004 (Round 2 = 169 respondents, 91% response rate).

Cohorts were solicited in 1996 and 2001 for participation in the ABEM LSEMR. Details regarding tool development, including sampling stratification, sampling precision and sample sizes, survey design participation solicitation and return, informed consent, participant confidentiality, institutional approval, and scale development are similar to those previously described for the ABEM Longitudinal Study of Emergency Physicians (LSEP).⁶ Like the ABEM LSEP, initial, newly entering cohorts are solicited for participation every 5 years. Participants were selected using a random sampling from a listing of all EM residents stratified by program to ensure representation of all existing programs and regions in the initial sample. The ABEM Research Committee and Board of Directors approved this study.

Survey Content and Administration

The survey was created by an expert panel of EPs and included information in the following five domains: 1) professional interests, attitudes, and goals; 2) training, certification, and licensing; 3) professional experience; 4) well-being and leisure activities; and 5) demographic information. The initial survey was presented to another group of expert EPs who were asked to comment on the items. The survey was changed according to the panel’s recommendations and returned to them until no further changes were suggested. The final survey consisted of 22 questions on professional interests, attitudes, and goals; 8 questions on training, certification, and licensing; 21 questions on professional experience; 4 questions on well-being and leisure activities; and 8 questions about demographics.

A four-step procedure to ensure adequate return rates was used for the lengthy surveys: 1) a cover letter about the survey, a copy of the survey, and a postage-paid return envelope; 2) a postcard follow-up; 3) a second cover letter, a copy of the survey, and a postage-paid return envelope; and 4) a certified letter. All data were collected through self-reporting by physician participants.

Primary Data Analysis

All analyses were conducted using STATA/SE 9.2 (StataCorp, College Station, TX). We computed descriptive information on variables associated with residency selection using information gathered from the first year that residents participated in the lengthy surveys. The questions pertaining to variables associated with residency selection only appeared on the surveys sent to first-year participants and were not included in later years.

Information from the final year of residency was used to determine satisfaction with residency program choice. We chose to use only cross-sectional data from the residents’ final year for this analysis because we believed that replies at that time would best reflect the overall residency experience.

The 1996 and 2001 cohort databases were compared and found to be similar in terms of age, gender, and ethnicity and so were merged for purposes of this study. Estimates of fulfillment of residency expectations, fulfillment of EM as a specialty choice, and satisfaction with residency were calculated as percentages with 95% confidence intervals (CIs).

Assessing Variables Associated with Selection of a Residency Program.  Residents were asked to rate a list of 18 items in response to the question, “How much did each of the following variables influence your choice of residency program location?” The rated items were financial advantages, grew up in the area, hospital facilities, influence of spouse or significant other, mentor recommended site, opportunities for moonlighting, opportunities for recreation, opportunity to serve a particular group of people, personal ties or contacts in the area, professional sports in the area, professional ties or contacts in the area, recruited by colleagues, reputation of the department chair, reputation of the institution, reputation of the program director, salary, size of community, and weather. The answer choices were presented in a Likertlike scale and ranged from 1 (not at all) to 5 (very much). For purposes of determining variables of greatest importance and least importance to these cohorts, an answer choice of 4 or 5 was recoded as 1 “considered to be very important” and an answer choice of 1 or 2 was recoded as 0 “not important.” Answer choices of “3” were not included in the analysis.

Responses were also evaluated by gender and by geographic distribution. The latter analysis was divided into the four U.S. census regions (Midwest, Northeast, South, and West) and based on the location of the program that the residents matched, rather than their location as applicants. Residents were asked to rate the following questions on a 5-point Likertlike scale: 1) How well did your residency program meet your expectations? 2) How well did the specialty of EM meet your expectations? 3) How satisfied are you overall with your residency program?

Results

A total of 322 residents responded to the survey for a response rate of 96% (n = 153) in 1996 and 91% (n = 169) in 2001. Women comprised 31% of respondents. The median age at entry to the study was 30 years. Additional demographic data are available in Table 1. We are unable to determine whether residents from each residency program elected to participate because we did not include information identifying programs in the questionnaire.

Table 1.    Demographics of Sample

Characteristic	1996 (n = 153)	2001 (n = 169)
Values are n (%) unless otherwise specified. SD = standard deviation.
Women	47 (31)	53 (31)
Mean age, years (±SD)	30.5 (3.8)	30 (3.0)
Ethnic group
White	129 (84)	135 (80)
African American	6 (4)	5 (3)
Hispanic	8 (5)	8 (5)
Asian	9 (6)	19 (11)
Married	78 (51)	83 (49)
Children	34 (22)	37 (22)

Table 2 summarizes the overall importance of the various variables and lists the four variables that over half of the sampled residency applicants identified as “very important” (with “reputation of the institution” ranked the most important), as well as the eight variables that at least half considered as “not important” in choosing a residency program (with “professional sports in the area” ranked the least important).

Table 2.    Self-reported Importance of Variables on Choosing a Residency Program

Variable	Very Important	Not Important
All values are n (%). N = 322
The reputation of the institution	251 (78.1)	32 (10.1)
Hospital facilities	186 (57.8)	50 (15.6)
Reputation of the program director	163 (50.7)	125 (38.9)
Influence of spouse, significant other	163 (50.7)	11 (35.9)
Opportunities for recreation	156 (48.4)	80 (25)
Size of community	146 (45.3)	78 (24.2)
Personal ties or contacts in the area	143 (44.5)	150 (46.8)
Reputation of the department chair	100 (31.2)	153 (47.6)
Mentor recommended site	95 (29.6)	161 (50)
Opportunity to serve a particular population	93 (28.9)	161 (50)
Weather	92 (28.6)	143 (44.5)
Grew up in the area	73 (22.6)	228 (71)
Moonlighting opportunities	70 (21.8)	191 (59.3)
Financial advantages	63 (19.5)	191 (59.3)
Salary	43 (13.2)	216 (67.1)
Recruited by colleagues	43 (13.2)	222 (69)
Professional ties or contacts in the area	45 (14)	246 (76.5)
Professional sports in the area	9 (3)	276 (85.9)

Gender Differences

Several differences were noted between men and women. Recreational opportunities are more important to men than women when choosing a residency location (56% males [95% CI = 48% to 66%] vs. 31% females [95% CI = 20% to 46%]). The reputation of the program director was ranked very important by more men than women (58% males [95% CI = 49% to 68%] vs. 33% females [95% CI = 23% to 50%]). The proportion of men who rated opportunities to serve certain populations as very important was roughly one-half that of women (23% men [95% CI = 18% to 29%] vs. 41% women [95% CI = 28% to 48%]). The reputation of the department chair was not important to women, with 63% (95% CI = 48% to 74%) citing this variable as not important and only 11% (95% CI = 5% to 25%) identifying it as an important variable in their decision-making. Thirty-nine percent of men (95% CI = 32% to 50%) thought that reputation of the department chair was an important factor.

Analysis by Regions

Residents who matched in programs in the Western region ranked recreational opportunities, weather, and the opportunity to serve a particular group of people as very important. Midwestern residents identified community size as an important variable in determining location for postgraduate education, and those in the Northeast indicated personal ties or contacts in the area affected their decision to select a program in that region of the country (Table 3).

Table 3.    Regional Differences in Importance of Variables Related to Residency Selection

Variable	Very Important, % (95% CI)
Puerto Rico is not included in the regional analysis. N = 322. Midwest, n = 90; Northeast, n = 94; South, n = 81; West, n = 55.
Recreation opportunities
Midwest	39 (22, 56)
Northeast	37 (22, 52)
South	46 (27, 65)
West	83 (68, 99)
Opportunity to serve a particular group of people
Midwest	24 (9, 39)
Northeast	26 (12, 39)
South	11 (0, 22)
West	63 (42, 82)
Personal ties or contacts in the area
Midwest	36 (20, 53)
Northeast	63 (48, 77)
South	36 (17, 54)
West	33 (14, 53)
Size of community
Midwest	61 (44, 77)
Northeast	44 (29, 59)
South	39 (21, 58)
West	33 (14, 53)
Weather
Midwest	7 (8, 11)
Northeast	11 (4, 17)
South	42 (29, 52)
West	74 (61, 86)

Satisfaction with Residency Choice

When asked to reflect on their experiences and decisions, respondents indicated strong overall satisfaction with their residency choice. Seventy-nine percent (95% CI = 76% to 82%) of residents surveyed in their final year of residency identified themselves as “highly satisfied.” Sixty percent (95% CI = 53% to 65%) of participating residents surveyed in their final year identified their EM program as very much like what they wanted when selecting a residency program. Ninety percent (95% CI = 86% to 93%) of residents surveyed in their final year answered that the residency program met or exceeded expectations.

Discussion

Recruiting residents is an expensive and time-consuming endeavor. Understanding the variables associated with residency choice can assist residency directors in focusing recruiting efforts and allocating resources. The ABEM LSEMR indicates that the most influential variables in residency choice are institutional and residency program director reputation and hospital facilities. Personal issues, such as community size, recreational opportunities, and spousal opinion, are also important but less influential and vary with geographic location. In general, both men and women report similar variables as influential in decision-making about residency, with minor variations in the influence of program director reputation, recreational opportunities, community size, and personal ties to an area. Overall, satisfaction with residency choice among residents is high.

Although residency program director and institution reputation play a dominant role in affecting program choice, the LSEMR did not explore how reputations are developed and maintained. As national rankings of EM residency programs are not published, this is an area that needs further study. It may benefit programs to explore their reputations among applicants and determine issues that contribute to their reputation to improve recruitment efforts.

Of the variables that play an important role in choosing an EM program, some are potentially mutable, such as spousal influence and reputation, but others are relatively fixed, such as community size and facilities. The trend toward choosing specialties that allow for a more controllable lifestyle has made EM a popular specialty choice. It is not surprising, given this trend, that availability of recreational opportunities ranks as important for many applicants. Salary and financial considerations were not considered important. Salary and benefits across residency programs are similar enough that the small differences among them appear to have little impact on applicant choice of programs.

Regardless of gender or location, residents report high levels of satisfaction with EM and their individual residency programs. While statistical differences between men and women were noted with regard to recreation options, weather, and program director reputation, the practical differences were quite small and probably do not warrant gender specific recruiting techniques. Weather and recreation are more important to residents who are training in the West, where weather is generally sunnier and more conducive to year-round outdoor activity than other regions. Westerners also indicated a desire to serve specific populations, which may reflect the diversity of the immigrant populations in the western states. Personal ties to the area ranked highly among residents who matched in the Northeast. Assessing associations between variables considered important in the selection of residency and satisfaction with residency program was beyond the scope of this project.

Limitations

The sample size of the LSEMR is small. Although the participants are randomly selected and the response rate is excellent, the sample may not be representative of the entire sample of U.S. EM residents. The LSEMR study did not address the effect of friendliness, environment, interview day, academics, positive resident opinions of their program, the content of a program’s website, or availability of specialized study in ultrasound or international medicine, factors that were previously shown to be important in residency selection.^7.

Conclusions

A majority of residents were highly satisfied overall with their residency choices. A number of factors influence medical students’ ranking of residency programs. The most significant factors include residency director reputation and hospital facilities. Personal issues, such as recreational opportunities and spousal opinion, are also important, but less influential. Program directors may wish to use this information to further concentrate their efforts and allocate recruiting resources to highlight the issues and strengths of their programs that matter most to applicants.