Objective. To examine 6 years of practice characteristics data of Fellows of the American Academy of Pediatrics (AAP), focusing on sex differences for specialty area, primary activity, practice setting, and practice location.
Methods. We analyzed data from 19 Periodic Surveys that were fielded between 1987 and 1992. The Periodic Survey is used to survey AAP members regularly about current issues in pediatric practice. There are no duplicate respondents in these analyses of the first 19 Periodic Surveys. We collapsed the 19 surveys into the years in which they were fielded, and analyzed sex differences for each of the 6 years. In addition, we ran logistic regressions on several questions, including all 16 868 respondents, to examine how the characteristics of the specialty have been affected by the increase in the number of female pediatricians, controlling for survey year, age of respondents, and specialty area practiced.
Results. The proportion of nonresident AAP members who are female has grown throughout the 6 years; in 1987, 26.9% were female, and in 1992, 36.4% were female. For 5 of the 6 years there were sex differences in specialty area, usually concerning pediatric subspecialties. Substantial sex differences occurred in primary activity, in which each year women were more likely than men to be salaried. Men were more often in group practices, whereas women were generally more likely to practice in hospitals or clinics. Logistic regression demonstrated that there are sex differences in practice characteristics across time, but there is also a substantial change in practice characteristics accountable to survey year, eg, a pediatrician of either sex was 75% more likely to be salaried in 1992 than in 1987.
Conclusions. Throughout the 6-year period, AAP members became increasingly more likely to practice general pediatrics, to be salaried, and to be younger—all effects independent of sex, all effects stronger for females. Rapid transformations in the health care system will likely reduce current sex differences in practice characteristics of the future.
One of the more obvious recent trends in medicine has been the dramatic increase in the number of female physicians. Pediatrics has always attracted disproportionately more women than most other specialties, and this has remained true even as other, traditionally more male-dominated specialties became more attractive to female entrants.1
It has been projected that by 2010 more than 50% of pediatricians will be female.2 Because there are so many women entering pediatrics, it is important to examine how male/female practice differences may be shaping the characteristics of the specialty overall. Practice differences between male and female physicians have been well documented; however, most of these studies have examined these differences only through the use of cross-sectional data3,4 or, if longitudinal in design, have focused on one particular practice characteristic, such as income5 or specialty choice.6 How the influx of women and the corresponding effect of their generally younger ages have influenced the specialty is considered in this study by both examining several years of cross-sectional data and combining those years of data.
We analyzed data from 19 different Periodic Surveys that were fielded between 1987 and 1992. The Periodic Survey was instituted to survey American Academy of Pediatrics (AAP) members on a regular basis on current issues in pediatric practice and on topics of concern to AAP policy makers. Each questionnaire is administered to a random sample of AAP members. The sampling frame from the AAP membership file is designed so that members would not be resurveyed for a period of approximately 4 years; therefore, there are no duplicate respondents in these analyses of the first 19 Periodic Surveys.
The first four Periodic Surveys were sent to samples of approximately 1000 pediatricians each. After that, a more typical sample size was 1600. Through four, and sometimes five mailings, the Periodic Surveys typically achieve high response rates. For the first 19 surveys, the lowest response rate was 69.2% (Survey #8) and the highest was 80.7% (Survey #17). The average response rate for the first 19 surveys is 74.9%.
The focus of these analyses are the questions that have been asked with consistency on the Periodic Survey. Questions on gender, primary professional activity, and area of pediatrics that the respondent spends most of his or her time in (eg, general pediatrics or a subspecialty), have been asked on all Periodic Surveys. Less frequently asked, but analyzed for the respondents who answered them, were questions on year of birth, practice type (eg, solo, group), and a population description of the respondent's community (eg, rural, urban). These six questions are analyzed for the 16 868 respondents who indicated they were not currently in a pediatric residency program.
Two different types of analyses are presented. The first examines sex differences for each question, grouped by year. The second set of analyses uses logistic regression to examine how the characteristics of the specialty of pediatrics—at least as measured by these questions—have been affected by the increase in the number of female pediatricians.
Male Versus Female Practice Characteristics, 1987–1992
Overall, 69.6% of respondents to the 19 surveys were men and 30.4% were women. The proportion of AAP members who are female has grown over time. In 1987, 26.9% of the respondents to the combined questionnaires were women; by 1992 36.4% were women (see Table1).
Table 1 presents the distribution of specialty areas for male and female pediatricians for the years 1987 through 1992. The question of which area the pediatrician respondent spent most of his or her time was not asked in the same way on each questionnaire. Categories of smaller, newer specialty areas were occasionally added to the question; on surveys in which these areas were not provided respondents would have written in an “other” response. The result of combining questions within a year does not significantly alter the basic distribution of pediatricians into general pediatrics and the major pediatric subspecialties.
In each year, except 1987, there are significant sex differences in specialty areas chosen by men and women. For 3 of the remaining 5 years, these differences are seen mostly in the choice of different subspecialty areas; during 1988 through 1990 there was approximately equal representation in general pediatrics. In 1991 and 1992, there were higher percentages of women in general pediatrics than of men. Each year men were more likely than women to be in pediatric allergy, pediatric cardiology, and pediatric surgery.
Women were more likely to be in adolescent medicine than men in each of the 6 years, and for 5 out of the 6 years, were more likely to practice developmental/behavioral pediatrics, and neonatal/perinatal medicine. There was no consistent pattern for the smaller, newer subspecialties, and indeed some of the percentages are too small to allow for reliable estimates of differences in participation between men and women in these areas.
Pediatricians were asked to indicate what their primary activity was from an array of choices; direct patient care, administration, academic medicine (medical teaching), research, public health, fellowship, temporarily not in practice, or other. For all but one questionnaire the direct patient care choice was divided into “self-employed” and “not self-employed.” In 1987, only one of two questionnaires offered these subdivisions as answer choices. Therefore, these two response categories in this questionnaire were collapsed to combine the responses for the two questionnaires in that year.
Table 2 presents responses for 1987 through 1992 for men and women. Significant sex differences are found in almost every year. Male pediatricians were consistently more likely to report that their primary activity was in direct patient care (combining both forms of direct patient care). From 1988 through 1992, male pediatricians were much more likely to be self-employed than female pediatricians. Men were always more likely to be self-employed than not self-employed (hereafter, for ease of discussion, this category will be referred to as salaried, although not all those who are not self-employed are salaried). Male pediatricians also were more likely to be in administrative positions than female pediatricians, whereas a greater proportion of women than men were in academic positions, fellowships, or public health.
Primary Practice Setting
Respondents to the Periodic Survey have been asked to describe their primary practice setting in a variety of ways during the years. Table 3 presents results from 1989 through 1992. For both 1990 and 1991, this question was asked once during the course of the year. For 3 of the 4 years that included this question, male respondents were significantly more likely than female respondents to state they worked in a solo or two physician practice. Men were also more likely to be in group practice than were women, although group practice was typically the preferred setting for both sexes.
Women were consistently more likely than men to report working in a hospital or clinic setting. The hospital and clinic response categories were collapsed for 1992 to combine the total responses of two questionnaires. However, the trend continues, in that proportionately more women than men reported working in either of these settings. Women were more likely to report working in other patient care settings as well.
Women are more likely than men to be located in the most densely populated areas of the country, and interestingly enough, the least populated, although this latter finding is based on few respondents (see Table 4). Men, on the other hand, are more likely than women to practice in large suburban and central city areas. Over time, the latter area has become more popular with both male and female pediatricians. The area that has lost relative appeal is that of communities with 2500 to 50 000 residents.
Given the recent increase in the number of women entering the field of medicine, and pediatrics in particular, it is no surprise that female members of the Academy are, on average, significantly younger than male members. Table 5 presents the average age for male and female pediatricians for the years 1988 through 1992, as well as the median age, and age at the 75th and 25th percentiles. The age difference between men and women has grown over time (approximately 6 years) even though the average age of men has decreased. The age of female pediatricians has decreased even more dramatically; this is particularly evident when examining the age at the 75th percentile. In 1988, 75% of all female pediatricians were 48 or younger. By 1992, three-quarters of female pediatricians were 44 or younger. During this same period, the age at the 75th percentile for men decreased from age 55 to 53. It is important to remember that none of these respondents were pediatric residents at the time they were surveyed.
Has the Increase in the Number of Female Pediatricians Affected the Characteristics
of the Specialty?
There are many ways in which the increasing number and proportion of female pediatricians are undoubtedly altering characteristics of the pediatrics specialty. One of the ways we will examine is how specialty and practice area preferences of all pediatricians are affected by the preferences of female pediatricians. For example, if there are proportionately more pediatric generalists versus subspecialists now than before, is it because of the preferences of women, or are there other factors that are affecting the specialty and practice distribution of all pediatricians?
There was a parallel increase in the percentage of general pediatricians and subspecialists who are female until 1990 (see Fig1). At this point, there was a sharp increase in female general pediatricians, and a dip in the percentage of female subspecialists. Figure 2 presents the proportion of pediatricians who are female in each of the major subspecialties. Adolescent medicine, developmental/behavioral medicine, and hematology/oncology have disproportionately more women than pediatrics overall, or general pediatrics. Pediatric allergy and pediatric cardiology appear to be more attractive to male pediatricians. Neonatology/perinatology more closely resembles the gender distribution of the specialty overall. The small number of respondents practicing in cardiology, allergy, and hematology/oncology results in substantial variation from year to year; hence, the dramatic peaks and valleys on Fig 2.
In conjunction with the increasing number of female pediatricians, two other trends are revealed by examining the 19 surveys; from 1987 to 1992, respondents were increasingly younger, and more likely to practice general pediatrics. These concurring trends suggest the possibility that whether a pediatrician practices in general pediatrics or in a subspecialty may be affected by any combination of the following factors: age, sex, and the year in which the pediatrician was queried. To examine how each of these three factors may be related to a pediatrician's choice of specialty area, primary activity, and location of practice area, logistic regressions were run on the combined dataset of 16 868 pediatricians. These regressions assess how the odds of certain practice characteristics change because of respondent age, sex, and survey year. The odds of an event occurring are defined as the ratio of the probability that it will occur to the probability that it will not. For example, if the odds for women practicing general pediatrics was three that would mean for every three female pediatricians in general pediatrics, there was only one female pediatrician in a subspecialty. How much these odds change because of respondent characteristics can be estimated using logistic coefficients. A main effect of age or sex, for instance, may reveal how a pediatrician's age or sex may be associated with certain practice characteristics. A main effect of time, as indicated by survey year, may indicate a trend or overall gravitation toward certain patterns of pediatric practice during the time period under study, regardless of individual age and gender.
Two interactions were also included. If there are consistent practice characteristics differences between men and women of the same age, then an age by sex interaction is present. The interaction of survey year by sex illustrates whether different trends are more or less strongly associated with women compared with men during the time period. Characteristics of the dataset suggested that there are two clusters of pediatricians; those aged 45 or younger at the time of the survey, and those more than age 45. Logically these two groups should be different; by age 45 most pediatricians have completed all the training they are going to undertake, and are fairly settled into their careers. The following analyses are separate for these two age groups, and all reported results are statistically significant.
Table 6 presents the factors by which the odds of practicing general pediatrics versus the major subspecialties change because of respondent age, sex, survey year, or the interactions of these variables. For example, a factor of 2.5 for women in general pediatrics would imply that when the effects of age and survey year were controlled for, women were 250% as likely as men to be in general pediatrics; thus, being female increased the odds of practicing in general pediatrics by 150%. Please note that the values in Table 5represent the odds of practicing, for example, adolescent medicine compared with general pediatrics or any other subspecialty, controlling for age, sex, and survey year.
Gender by itself did not affect the likelihood of those younger than 45 practicing general pediatrics. Within the younger group, the likelihood of practicing general pediatrics decreased slightly with each year of age, meaning that for any given survey year or gender, the younger the pediatrician the greater the likelihood of practicing general pediatrics. The odds increased 9% percent for each survey year of a younger pediatrician practicing general pediatrics, with the odds for women increasing slightly more each year (10%). The former means that, for example, a younger pediatrician of either sex was 27% more likely to be in general pediatrics in 1992 than in 1989. The latter suggests that this trend was stronger for women than for men. Overall, two trends are present for younger pediatricians and general pediatrics during this time period; younger age was significantly related to practicing general pediatrics, and for every survey year after 1987 until 1992, an increasing proportion of pediatricians practiced general pediatrics.
The results for the older age group were different on many counts. The odds of practicing general pediatrics were increased by 661% for female pediatricians compared with male pediatricians. Each year of age increased the odds of practicing general pediatrics by a slight amount. Following the trend of the younger pediatricians, the odds of a pediatrician in this age group practicing general pediatrics increased 14% for each year between 1987 and 1992. Each year of age for a woman decreased the odds of her practicing general pediatrics by 4%, compared with similarly aged men.
Over time, the percentage of pediatric allergists has been decreasing (Table 1). The proportion of women in pediatric allergy has not been substantial, but has been rising (Fig 2). Nonetheless, logistic regression demonstrates that, overall, the odds of a young female pediatrician practicing pediatric allergy, versus any other area, are 43% less than the odds for a male pediatrician; for an older female pediatrician, the odds are 49% less. Each year of age for the older group increased the odds of practicing pediatric allergy by 2%, whereas the likelihood of any pediatrician in this age group practicing pediatric allergy decreased 14% for each year between 1987 and 1992.
Pediatric cardiology has traditionally not attracted many women subspecialists, and indeed, we find that younger women were 67% less likely to be pediatric cardiologists than were men. The odds of any pediatrician in this age group of being in this subspecialty increased 15% for each survey year. There were no significant findings for the older age group of pediatric cardiologists.
Overall, younger men and women were equally likely to be in adolescent medicine. Each subsequent survey year after 1987 decreased the overall probability of a pediatrician in either age group of practicing adolescent medicine by 30%. However, relative to men, the odds increased for women by 28% for each survey year. Unlike the younger group, among the older pediatricians sex had a significant effect; women were 69% more likely to be in this subspecialty than men.
The logistic regression results for developmental/behavioral medicine show that being a woman increased the odds of practicing developmental/behavioral medicine by 45% for younger pediatricians, and 163% for older pediatricians. Each survey year decreased the odds for the younger group by 17%, and 19% for the older group. Each year of age increased the probability by 7% for any pediatrician in the younger age group, whereas age had no effect in the older age group.
The likelihood of a woman practicing hematology/oncology is 99% less than the likelihood of a man practicing the subspecialty, but only for the older group. On the whole, the odds of practicing this area decreased by 6% for each year of age more than 45 year, albeit women's odds of being in this subspecialty increased by 11% for each year of age compared with similarly aged men. There were no significant relationships for the younger age group.
We did not find a relationship between sex and the odds of being a neonatologist or perinatologist. Every year of age slightly increased the odds for male or female pediatricians of being in neonatology/perinatology for younger pediatricians, yet every year of age decreased by 9% the likelihood for older pediatricians.
We used the same statistical technique to examine the affect of the rising proportion of female pediatricians on the distribution overall of pediatricians into different types of primary activities. We examined 1) self-employed direct patient care, 2) salaried direct patient care, and 3) academic medicine, as these are the three areas that attract the most pediatricians. Figure 3 presents the proportion of women in these three activities over time (for year 1987, direct patient care is not divided into self-employed and salaried). Figure 3 demonstrates the disproportionate presence of women in salaried direct patient care and academic medicine, as well as the increase over time of their number. As in the analyses above, we included sex and age in the logistic regression equations, and also included whether the pediatrician practiced general pediatrics or not. Again, we separated the sample into those aged 45 or younger, and those older than age 45. Please keep in mind that each comparison looks at one activity versus all others.
Self-employed Direct Patient Care
The single most important factor that affects the odds of a pediatrician being in self-employed direct patient care is practicing general pediatrics (see Table 7). Both older and younger pediatricians who practiced general pediatrics were four times more likely than pediatric subspecialists to be in self-employed direct patient care. Being female decreased the odds of being self-employed by 48% among the younger group and by 55% among the older group. Among the younger group, the odds of being in this activity increased by 10% for each year of age; the odds decreased by 11% for every survey year after 1987 until 1992.
Salaried Direct Patient Care
As the majority of pediatricians are in direct patient care, either self-employed or salaried, it is not surprising that the variables we examine here had opposing effects on these two activities. Women 45 or younger were 65% more likely than men to be in salaried direct patient care than in another activity. The odds of a woman more than the age of 45 of being salaried were 142% more than those of a similarly aged man. Both younger and older pediatricians who practiced general pediatrics were less likely than subspecialists to be in salaried direct patient care. For both age groups, every year of age decreased the odds of being a salaried provider by 3 to 4%. The odds of being in this practice activity increased 16% for each survey year for the younger group of pediatricians, and increased 14% for the older group; however, the odds were decreased by 12% for each survey year for women in the older age group, relative to men.
Being a general pediatrician decreased the odds of being in academic medicine by 74% for pediatricians aged 45 or younger, and by 78% for pediatricians older than 45. The odds of younger pediatricians being in academics were decreased by 26% for each survey year after 1987; the odds for older pediatricians were decreased by 10%. When all other variables are held constant, men and women are equally likely to be in academics.
Figure 4 shows the percentage of pediatricians who are female in different practice locations. Our final set of multivariate analyses looks at practice location, namely, comparing metropolitan areas (population greater than 50 000) to all other smaller areas (Table 8). Practicing general pediatrics versus a subspecialty decreased the odds of practicing in a metropolitan area by 69% for the younger group of pediatricians, and by 57% for the older group. In addition, being in self-employed direct patient care decreased the odds of practicing in a metropolitan area by 49% for the younger group, and by 57% for older pediatricians. The odds of practicing in a metropolitan area increased for younger pediatricians 54% for each survey year, and by 48% for older pediatricians. Sex did not enter the equation for either age group.
Because of the large dataset of pediatricians and the number of years during which it was collected, we believe these findings provide an accurate picture of trends in pediatrics for these variables. There are limitations to the generalizability of these findings; namely, the results are based on board certified pediatricians who belong to the AAP. However, most pediatricians (64.4%) become board certified,1 and it is estimated that 73.5% of eligible pediatricians do join the AAP (AAP Department of Membership, 1996). Another potential limitation concerns estimates of specialty area, primary activity, and practice setting that were principally based on respondents reporting how, or where, they spend most of their time. A significant minority of time could be spent in other activities, or locations, which could dramatically alter a particular pediatrician's practice profile. Given those caveats, the major findings of this study, focusing principally on the multivariate analyses, are as follows:
1. No different than any other specialty, pediatrics is affected by the changing practice climate. The increase over time in the proportion of pediatricians practicing general pediatrics, and most likely primary care pediatrics, is borne out by both bivariate and multivariate statistics. Between 1987 and 1992, the proportion of male and female pediatricians practicing general pediatrics increased by 13.2% and 19.1%, respectively, with the odds increasing yearly by 9% for younger pediatricians and 14% for older pediatricians. This translates to a total percent increase between the years 1987 and 1992 in the odds of a younger pediatrician practicing general pediatrics, versus a subspecialty, of 45%. The total percent increase in odds for an older pediatrician during this time period is even greater, at 70%. This increase comes mostly at the expense of adolescent medicine and developmental/behavioral pediatrics, the two areas that experienced the greatest yearly decline in odds. As both areas are often considered to be components of general pediatrics,7,8 this change could be a result of greater incorporation of these specialty areas into the practice of general pediatrics, rather than a strict decline in their popularity.
Also increasing over time is the proportion of pediatricians in salaried direct patient care. The likelihood of being a salaried direct patient care pediatrician increased by 14% to 16% with each year in the study; in other words, all else being equal, a pediatrician was approximately 75% more likely to be salaried in 1992 than in 1987. At the same time the likelihood of pediatricians reporting being in medical teaching decreased each survey year by 10% to 26%. Another additional decrease was apparent for younger pediatricians in the likelihood of self-employment. Trying to separate out the effects of life-stage versus health care trends is difficult. Previous research in the 1980s reporting the greater tendency for younger physicians to be employees suggested that this was a life-cycle phenomenon; that with time, these younger physicians would move into self-employment.9 Indeed, others have found that for young physicians, each year of experience increases the probability of a career change from employee to self-employment.10 Current market conditions in medicine suggest that such moves, from employee to self-employment, are becoming rarer, and that the opposite move is more frequent than ever before.11 That we found pediatricians of both age groups increasingly likely to be salaried with each survey year points to a trend that is weakening any former life- or career-stage sequence.
Like much of the United States population, pediatricians are becoming increasingly more urban. Our research supports the practice location projections of pediatricians for the years 1988 to 2000 suggesting there would be greater growth in metropolitan regions than in nonmetropolitan areas.12
2. Another major effect on how pediatricians practice is the age of the pediatrician. Although the effects of age seem small, each effect is for each year of age; an increase in odds of 2% means an overall difference in odds of 20%, for example, between a 35 year old and a 45 year old. By grouping the pediatricians according to age we can also see that practice characteristics are associated with the life-stage of a pediatrician's career. Neonatology/perinatology is a good case in point. For the younger pediatricians, each year of age is related to an increase in the likelihood of practicing in this area; yet for older pediatricians, each year of age decreases the likelihood of practicing neonatology/perinatology. This pattern suggests a possible age burn-out within this subspecialty,13 although it is also likely that there are fewer older neonatologists simply because of fewer opportunities for neonatology training when these pediatricians were younger. The trend for practicing general pediatrics is in the opposite direction; for younger pediatricians, each year of age is related to a decrease of 4% in the likelihood of practicing general pediatrics, whereas for older pediatricians there is a slight increase in this likelihood associated with age. This is probably the combined result of the length of fellowship training, in the case of the younger pediatricians completing their training and concentrating on a subspecialty, and the availability of such training for the older pediatricians for whom training opportunities may have been rarer.
Years of age were significant in predicting some practice characteristics for one age group of pediatricians, and not for another. Other variables differed in the magnitude of the effect on the odds, depending on the age group; for example, the decrease in the odds of being a medical educator by survey year is substantially greater for the group of younger pediatricians (26%) than for older pediatricians (10%). Research examining separation and retirement trends of medical school faculty has noted a decrease in job mobility in the 1980s, which may make it difficult for younger pediatricians to gain entry into academics.14 Other factors, such as a declining interest among medical school graduates in pursuing research careers, may also have an effect.15
3. Confirming other research, female pediatricians in many circumstances, but not all, have different practice characteristics compared with male pediatricians. The most dramatic sex difference occurs when measuring the likelihood of practicing general pediatrics versus a pediatric subspecialty. This difference is dramatic in two senses; first of all, women in the older age group are seven times more likely to be in general pediatrics compared with men in the same age group. Given this large difference, what is equally striking is the fact that there is no sex difference in predicting general pediatrics practice for pediatricians younger than age 45.
There are sex differences in other areas, but none of the magnitude we find for the odds of practicing general pediatrics. Older women were 69% more likely to practice adolescent medicine compared with similarly aged men. Women in both age groups were more likely to be developmental/behavioral pediatricians than were men, whereas younger women were less likely to be pediatric cardiologists. The trend over time toward practicing general pediatrics, for the younger pediatricians at any rate, is heightened slightly for women. Other strong effects of gender are found for primary activity in which women are more likely to be salaried and less likely to be self-employed than men. This is over and above the countervailing relationship between practicing general pediatrics and primary activity (practicing general pediatrics is strongly predictive of self-employment). This effect is greater among the older pediatricians, again pointing to diminishing practice differences between younger male and female pediatricians.16 Overall, the proportional increase in the percentage of male pediatricians being in salaried direct patient care between 1988 and 1992 grew faster than the increase for women (61.0% vs 45.0%).
Clearly, practice characteristics of pediatrics are changing. The most conspicuous stimulus of this change is the influx of women into the specialty. Difficult to tease out and separate are the changes that the increase in the number of women bring to the specialty, eg, as recent entrants, their age is younger, and therefore, on average, their practice characteristics will be different from older, and mostly, male pediatricians—and the simultaneous and overwhelming changes occurring in the health care system, eg, a greater emphasis on primary care and the substantial influence of managed care and other, formerly nontraditional employment mechanisms. As, in general, male and female pediatricians are more similar to each other than they are different, their adaptations to the rapid transformations in the health care system will likely reduce current sex differences in practice characteristics of the future.
We would like to thank Sanford Sharp, Survey Assistant, who was responsible for data collection and data entry of the Periodic Survey, as well as the helpful comments of two anonymous reviewers.
- Received July 8, 1996.
- Accepted December 3, 1996.
Reprint requests to (S.E.B.) Research Associate, Department of Research, American Academy of Pediatrics, 141 Northwest Point Blvd, Elk Grove Village, IL 60007–1098.
- AAP =
- American Academy of Pediatrics
- ↵Randolph L, Seidman B, Pasko T. Physician Characteristics and Distribution in the US. Chicago, IL: American Medical Association; 1996
- Silberger AB,
- Marder WD,
- Willke RJ
- Hojat M,
- Robeson MR,
- Veloski JJ,
- et al.
- Strain JE
- Marder WD,
- Emmons DW,
- Kletke PR,
- et al.
- ↵Kletke PR, Marder WD, Willke RJ. A projection of the primary care physician population in metropolitan and nonmetropolitan areas. In: Grady ML, eds. Conference Proceedings. Primary Care Research. Theory and Methods. Rockville, MD: USDHHS, PHS, ACHPR; 1991
- Clarke TA,
- Maniscalco WM,
- Taylor-Brown S,
- et al.
- Copyright © 1997 American Academy of Pediatrics