Objectives. To examine trends in pediatric residents’ training and job search experiences from 1997–2002.
Methods. Annual national random samples of 500 graduating pediatric residents were surveyed, and responses were compared across survey years to identify trends. The overall response rate was 71%.
Results. From 1997–2002, there were more female residents and US underrepresented minorities and fewer international medical graduates. Each successive group of residents rated higher their preparation for fellowship training, for child advocacy, and for assessing community needs. These increases paralleled an increase in resident exposure to community sites as part of their residency education. Educational debt (in 2002 dollars) for residents increased substantially across survey years from an average of $64 070 in 1997 to $87 539 in 2002. Meanwhile, starting salaries (in 2002 dollars) for residents entering general pediatrics actually decreased. Interest in general pediatrics among residents decreased, whereas interest in subspecialty practice increased during this time period. Fewer residents with general pediatrics as a career goal had a job when surveyed, and fewer obtained their first-choice positions across years.
Conclusions. Experiences of graduating residents over the past 6 years provide insights into changes in pediatric residency education and the pediatric workforce. Efforts by pediatric educators and academic leaders to increase community experiences and child advocacy and to encourage greater interest in pediatric subspecialty careers seem to be succeeding. Unfortunately, demand for general pediatricians is weakening, and residents are experiencing increasing debt burdens.
Pediatric residents can provide valuable information on the type and quality of their residency training and on their future career plans.1–17 The growth of managed care in the 1990s led to increased emphasis on primary care training in residency programs and increased resident interest in general pediatrics. The impact of this primary care emphasis on subspecialties and pediatric research has concerned some18,19 and suggests a need to monitor shifts in the career choices of pediatric residents.
Debate was ongoing throughout the 1990s concerning the larger workforce question of whether we are training too many,20,21 too few,22–25 or the right number of pediatricians.18,26 Residents’ job search experiences provide one method to examine whether the overall demand for pediatricians is tightening.27,28
The American Academy of Pediatrics (AAP) developed the Third Year Resident Survey to systematically collect information annually from a national sample of graduating pediatric residents. The Third Year Resident Survey has already provided insight into several pediatric workforce and training issues, including predictors of subspecialty career choice,29 interest in part-time employment,30 and research exposure during residency.31 The survey was begun in conjunction with the Future of Pediatric Education II project18 with the goals of longitudinally assessing both resident training experiences and residents’ job search experiences to gain insight into the market conditions that pediatricians face.
In this article, we present the first national trend report from the AAP Third Year Resident Survey regarding the job market for pediatric residency graduates, including starting salaries. In addition, we examined residents’ career intentions and their assessments of their training programs and their debt burden. These trends are important indicators of how well the current system is performing in the production and deployment of the next generation of skilled pediatricians.
We analyzed data from 6 years of the AAP Third Year Resident Survey, from 1997–2002. This survey was fielded annually to a random sample of 500 graduating pediatric residents from Accreditation Council for Graduate Medical Education-accredited categorical pediatrics programs. Residents from dual degree programs, such as Med/Peds programs, were not included in each sample. Up to 4 mailings of the survey were conducted from May through August of the summer that each group of residents completed and left their residencies. Thus, 2 mailings were conducted before and 2 mailings were conducted after residents had left their residencies. US Postal Service mail forwarding was relied on to reach residents once they left. Across survey years, the average return date was July 2. The standard deviation for return dates was 37 days, and 24%, 32%, 25%, and 19% of surveys were received in May, June, July, and August or later, respectively.
Response rates to the survey varied from 78% in 1997 to 64% in 2001. The combined response rate was 71%. Residents from 233 different residency programs responded to the survey across the 6-year period. This was higher than the current number of active categorical pediatrics programs as a result of program mergers and closures across the 6 years.
For assessing potential response bias, respondents were compared with nonrespondents for several demographic variables. For all survey years combined, no statistically significant differences were found between respondents and nonrespondents: percentage female (65% vs 62%; P = .103), percentage international medical graduate (24% vs 27%; P = .101), and age (mean: 31.5 vs 31.7; P = .152).
The current article is focused on the results from a core set of questions that are asked each year. Trend analyses comparing survey years were conducted for each variable. This was accomplished using linear regression analyses for continuous outcome variables and linear association χ2 tests for categorical outcome variables. Survey year was treated as a continuous predictor variable in each analysis to focus on linear trends. The number of cases varied slightly for each analysis on the basis of fluctuations in missing values. For variables expressed in dollars, adjustments for inflation were performed using the yearly Consumer Price Index to convert all values into 2002 dollars. Surveys for this study were approved by the AAP Institutional Review Board.
Demographic characteristics of respondents were tracked across the 6-year study period. The percentage of female residents significantly increased across survey years to a high of 69% in both 2001 and 2002 (P = .002). The percentage of underrepresented minorities (black, Hispanic, or American Indian) from US medical schools also increased significantly from 6% in 1997 to 12% in 2002 (P < .001). At the same time, the percentage of residents graduating from international medical schools decreased from 30% in 1998 to 15% in 2002 (P < .001). The percentage of residents who were married or living with a partner remained stable across survey years (71% overall; P = .982).
Residents were asked to provide overall ratings of how well their residency prepared them for various professional activities using a 5-level scale with poor, fair, good, very good, and excellent as anchors. The majority of residents each year reported that their residency preparation for primary care pediatric practice and for pediatric fellowship training was very good or excellent (Table 1). Self-rated preparation for pediatric fellowship training significantly increased over time (P = .001). However, residents’ rating of their preparation for general pediatric practice did not increase (P = .152), mostly as a result of a sharp drop-off in ratings in 2002. A lower percentage of residents reported very good or excellent preparation for using technology in practice, assessing the needs of their community, and child advocacy, but residents’ ratings of their preparation for assessing community needs (17% in 1997, 34% in 2002; P < .001) and for child advocacy increased over time (16% in 1997, 45% in 2002; P < .001).
This increase in preparation ratings of assessing community needs corresponds with an increase in resident exposure to community sites as part of their formal residency education during the same time period (Table 2). From 1997–2002, residents were less likely to have continuity clinic in hospital-based clinics (P < .001) and more likely to practice at community health centers (P = .001) or private pediatrician offices (P < .001). Furthermore, more residents reported experiences in schools (P < .001), child care centers (P < .001), public health departments (P < .001), and community health centers (P < .001) over time as part of their residency education.
From 1997–2002, residents reported increasing total educational debt (Fig 1). Educational debt included college, medical school, and spouse or partner educational debt. The percentage of residents who reported educational debt significantly increased across survey years (69% in 1997, 78% in 2002; P < .001). The average resident debt (in 2002 dollars) for all respondents increased 37% from $64 070 in 1997 to $87 539 in 2002 (P < .001). Among residents who reported any educational debt, the average debt (in 2002 dollars) increased 22% from $92 656 in 1997 to $113 476 in 2002 (P < .001). In contrast to financial debt, there was no increase in the percentage of residents who reported service obligations (9% overall) over time (P = .215).
Significant changes occurred in the future clinical goals of graduating residents (Table 3). Although general pediatric practice remained the most common practice goal, the percentage of residents who were interested in general pediatric practice declined across survey years (P = .026), and the percentage who were interested in subspecialty practice increased (P < .001). There was also a decline in those who were interested in combining primary care and subspecialty practice (P < .001).
The percentage of residents who entered directly into a general pediatrics position after residency also decreased significantly from 60% in 1997 to 43% in 2002 (P < .001). Since 2000, fewer than half of the residents accepted general pediatrics positions on graduation, leaving roughly 18% of residents who anticipated that they would enter primary care at a later time.
The lower portion of Table 3 describes characteristics of the general pediatrics positions that were accepted by residents. Across survey years, significantly fewer residents each year accepted positions in rural areas (P < .001). By 2001, this percentage had diminished to only 1% of graduating residents. Each year, more than half of the residents who accepted a general practice position stayed in the same state in which they trained. This rate was consistent across years (P = .711), ranging from 59% to 67%.
Starting salaries for those who headed directly into general pediatric practice (Fig 2) significantly declined (in 2002 dollars) across survey years from $103 161 in 1997 to $99 123 in 2002 (P = .001). At the same time, no significant changes were apparent in the number of patients that residents expected to see per day (P = .936) or in the number of hours that residents expected to work in practice per week (P = .569; Table 3).
In addition to general pediatricians, Fig 2 shows starting salaries for hospitalists, chief residents, and subspecialty fellows. Salaries for these other job types were not collected in the first 3 survey years. Across the last 3 years, starting salaries did not significantly change for subspecialty fellows (P = .414) and chief residents (P = .710). A significant increase was apparent in salaries for hospitalists (P = .041). The percentage of residents who accepted hospitalist positions decreased slightly from 15% in 2000 to 14% in 2001 to 12% in 2002, but this was not a significant trend (P = .172).
Job Search Experience
Almost all residents did have a job (91%) when surveyed, and most had obtained their first-choice position (85%; Table 4). Market trends distinctly diverged, however, for those with general pediatrics as their future clinical goal and those with subspecialty or with combined subspecialty and primary care as their goal. Across survey years, residents with a general pediatrics goal were increasingly more likely to have been without a job when surveyed (P = .005) and less likely to have reported obtaining their first-choice position (P = .017). Meanwhile, those with a subspecialty or combined subspecialty and primary care goal were less likely to have been without a job when surveyed (P = .016) and more likely to have obtained their first-choice fellowship position (P = .035) across years. Across survey years, the number of applications that residents sent (P = .305) and the number of residents who reported that their spouse/partner’s career plans or their family situation limited their selection of positions (P = .071) did not significantly vary.
Because the timing of survey completion could affect resident responses concerning having a job when surveyed or having a first-choice position, we performed additional analyses focused on survey timing. The time a survey was completed was not significantly associated with residents’ likelihood of reporting having a job (P = .112) or obtaining their first-choice position (P = .110).
Our study indicates that significant changes in pediatric training, resident career goals, and the pediatric job market have occurred over the past 6 years. Many of these trends indicate the impact that recent policies have had on resident education and career choices. These trends are discussed within several topics of importance within the field of pediatrics.
Residents’ Financial Situation
Unfortunately, the financial situation of new pediatricians is becoming less desirable. More residents are carrying debt, and the amount of debt increased by 37% since 1997. At the same time, starting salaries for those entering practice decreased 4%. Increasing educational debt is common for all recently trained physicians, but the impact of this debt may be felt more acutely by pediatricians, who are among the lowest-paid specialists.32 Although previous studies have shown only minimal impact of educational debt on career choice,33,34 our results suggest that loan repayment initiatives that would provide new opportunities for residents tied to desirable policy goals, such as improved geographic distribution of physicians, may be timely and effective.
Our results on community exposure during residency suggest that the changes to Residency Review Committee requirements in the late 1990s were successful.3 The increased emphasis on community experiences and child advocacy is reflected by greater numbers of residents at nonhospital continuity clinic sites, increased experiences in community sites, and improved self-rated preparation in assessing community needs and in child advocacy at the end of residency. These trends demonstrate the increased recognition within the field of the necessity for pediatricians to work with communities to make improvements in child health.
Geographic and Racial/Ethnic Maldistribution
Our data show that, among respondents, the percentage of underrepresented minorities from US medical schools doubled to 12% in the past 6 years. However, this level is still well beneath the 33% percent of US children who are black, Hispanic, or Native American.35,36
Geographic maldistribution remains a problem, and our data indicate that it may be growing. Across survey years, the number of residents who entered general pediatrics positions in rural areas dwindled. Although the definition of rural used (<2500 people) may be too limiting, in the sense that these areas may be of questionable viability to operate a practice, it is still concerning that there is downward movement for these areas. Barriers to practice in these areas for recent graduates may also include issues such as career opportunities for spouses, educational opportunities for children, and potential professional isolation. Overall, the majority of residents who headed into general pediatrics accepted positions in the same state as their residency, and a substantial number stayed in the same city/area. This may reflect the importance of family issues among the many factors involved in residents’ career decisions.29,30
Primary Care Interest and Job Market
The growth of managed care in the 1990s led to increased interest in primary care practice. In pediatrics, this interest seems to have peaked in 1998, and the pendulum has since swung back toward subspecialty practice. General pediatric practice remains the most common future clinical goal of residents, but in 2002, only 43% of residents entered directly into a general pediatric practice position on graduation. The number of residents who accept traditional general pediatric practice positions has been reduced by greater interest in subspecialties and by many residents who accept hospitalist positions.
In the early 1990s, there was also concern about an imminent oversupply of physicians as a result of declining demand for physician services by managed care.20,21 One indicator that has been considered in the workforce debate is the job search success of newly trained physicians. Using this measure, our results suggest that the job market for general pediatricians is tightening. More research is needed to examine which specific areas of the country are driving this trend. The percentage of residents who are interested in general pediatrics without a job when surveyed tripled from 5% in 1997 to 15% in 2002. Similarly, the percentage of residents who were interested in general pediatrics and were not able to obtain their first-choice positions increased from 14% in 1997 to 20% in 2002. Thus, although roughly 8 of every 10 residents still obtained their first-choice position, the trends are concerning.
In 2002, fewer residents rated their preparation for primary care practice as excellent or very good. This cohort of graduating residents may represent an aberration, but this decline needs to be monitored to ensure that primary care practice is not excessively de-emphasized by residency training programs.
Our results show that across the last 6 years, interest in a subspecialty career doubled from 13% to 27%. This may reflect increased emphasis on subspecialty practice by residency faculty, which is also consistent with the finding that residents rated their preparation for pediatric fellowship training higher across survey years.
The increase in subspecialty interest occurred despite a drop in the percentage of international medical graduates, who historically have shown more interest in subspecialties than US medical graduates.29 Recent efforts to increase subspecialty interest among US medical graduates have included loan repayment for residents involved in research and increased flexibility of fellowship requirements by the American Board of Pediatrics.
In addition, residents are reporting a declining market in primary care positions and an improving market for fellowship positions as reflected by increasing numbers of residents’ obtaining their first-choice fellowship positions across survey years. A concomitant increase in the number of pediatric fellowship programs and fellowship slots37 may be part of the reason for the increase.
There are several limitations of our study. First, all study data were self-reported by residents and are subject to some recall bias. Second, although the response rate was good for each survey year, variation in response rates across survey years may have had somewhat of an impact on the comparability of the samples. Third, questions particularly related to residents’ job search could have been affected by variability in the timing of survey responses. However, the amount of time before a survey was completed was not found to be significantly associated with residents’ likelihood of having a job or having their first-choice position. Fourth, because we surveyed only third-year residents, we were unable to examine further the training and job market for graduating pediatric subspecialty fellows. Finally, the correlational nature of this study does not allow us to determine the causal mechanisms driving each of the historical trends.
Our study provides new insight into a time of change in pediatric residency education and the pediatric workforce. Efforts in the past 6 years by pediatric educators and academic leaders to increase community experiences and child advocacy and to encourage greater interest in pediatric subspecialty careers seem to be succeeding. Unfortunately, demand for general pediatricians is weakening while residents are experiencing increasing debt burdens. These trends need to be monitored continually to assess the impact that policy changes will have on residency training and the career choices of new pediatricians.
The American Academy of Pediatrics and the Future of Pediatric Education II Project supported this study.
We acknowledge helpful feedback from the American Academy of Pediatrics Resident Section and the American Academy of Pediatrics, Committee on Pediatric Workforce in the development of the AAP Third Year Resident Survey.
- ↵Steinbach WJ, Sectish TC. Pediatric resident training in the diagnosis and treatment of acute otitis media. Pediatrics.2002;109 :404– 488
- ↵Mulvey HJ, Ogle-Jewett EA, Cheng TL, Johnson RL. Pediatric residency education. Pediatrics.2000;106 :323– 399
- Emans SJ, Bravender T, Knight J, et al. Adolescent medicine training in pediatric residency programs: are we doing a good job? Pediatrics.1998;102 :588– 595
- Serwint JR. Multisite survey of pediatric residents’ continuity experiences: their perceptions of the clinical and educational opportunities. Pediatrics.2001;107(5) . Available at: http://www.pediatrics.org/cgi/content/full/107/5/e78
- Roberts KB, Starr S, DeWitt TG. The University of Massachusetts Medical Center office-based continuity experience: are we preparing pediatrics residents for primary care practice? Pediatrics.1997;100(4) . Available at: http://www.pediatrics.org/cgi/content/full/100/4/e2
- ↵The Future of Pediatric Education II: organizing pediatric education to meet the needs of infants, children, adolescents, and young adults in the 21st century. Pediatrics.2000;105(suppl) :161– 212
- ↵American Academy of Pediatrics, Committee on Pediatric Research. Promoting education, mentorship, and support for pediatric research. Pediatrics.2001;107 :1447– 1450
- ↵Cooper RA, Getzen TE, McKee HJ, Laud P. Economic and demographic trends signal an impending physician shortage. Health Aff.2002;21 :140– 154
- Grumbach K. The ramifications of specialty-dominated medicine. Health Aff.2002;21 :155– 157
- Weiner JP. A shortage of physicians or a surplus of assumptions? Health Aff.2002;21 :160– 162
- ↵Barer M. New opportunities for old mistakes. Health Aff.2002;21 :169– 171
- ↵American Academy of Pediatrics, Committee on Pediatric Workforce. Pediatric workforce statement. Pediatrics.1998;102 :418– 427
- ↵Larson WL, Holloway RG, Keran CM. Employment-seeking experiences of residents in 1996: a window into the neurology marketplace. Neurology.2000;54 :214– 288
- ↵Pan RJ, Cull WL, Brotherton SE. Pediatric residents’ career intentions: data from the leading edge of the pediatrician workforce. Pediatrics.2002;109 :182– 188
- ↵Cull WL, Mulvey HJ, O’Connor KG, Sowell DR, Berkowitz CD, Britton CV. Pediatricians working part-time: past, present, and future. Pediatrics.2002;109 :1015– 1020
- ↵Cull WL, Yudkowsky BK, Schonfeld DJ, Berkowitz CD, Pan RJ. Research exposure during pediatric residency: influence on career expectations. J Pediatr. In press
- ↵American Medical Association. Physician Socioeconomic Statistics 2000–2002 Edition. Chicago, IL: American Medical Association; 2001
- ↵Bazzoli GJ. Does educational indebtedness affect physician specialty choice? J Health Econ.1985;4 :11– 19
- ↵American Academy of Pediatrics, Committee on Pediatric Workforce. Enhancing the racial and ethnic diversity of the pediatric workforce. Pediatrics.2000;105 :129– 131
- ↵US Census Bureau. Census 2000 Summary File 1 (SF1) 100-Percent Data. Washington, DC: US Census Bureau; 2000
- Copyright © 2003 by the American Academy of Pediatrics