Published ahead of print on January 9, 2003, doi:10.1164/rccm.200203-186OC
© 2003 American Thoracic Society A Snapshot of Pulmonary Medicine at the Turn of the CenturyThe American Thoracic Society MembershipDivision of Pulmonary and Critical Care Medicine, Harborview Medical Center, University of Washington, Seattle, Washington; Membership Services and Marketing, American Thoracic Society, New York, New York; Department of Clinical Investigation and Pulmonary Critical Care Service, William Beaumont Army Medical Center, El Paso, Texas; Pulmonary Division, Denver Veterans Affairs Medical Center; and University of Colorado Health Sciences Center, Denver, Colorado Correspondence and requests for reprints should be addressed to Lynn M. Schnapp, M.D., Box 359640, 325 Ninth Avenue, Harborview Medical Center, University of Washington, Seattle, WA 98104. E-mail: lschnapp{at}u.washington.edu
To describe the characteristics of the American Thoracic Society, the Membership Committee developed a survey to assess demographics, training, professional activities, and needs of a diverse membership with a growing international segment. It also provided an opportunity to determine how the Society reflects the current state of pulmonary medicine in the United States. A self-administered survey was mailed to active members. Of responding members, 80% reside in the United States or Canada; the remainder come from 90 different countries. The majority of North American respondents (79%) were white, non-Hispanic. Seventeen percent of respondents were female. Female respondents were younger, with a mean age of 42 years, compared with 47 years for males. Sixty-five percent of respondents identified clinical practice, 20% research, and 5% teaching as their major activity. More women (33%) than men (22%) identified themselves as researchers. The majority of respondents (69%) have a medical school faculty affiliation. The American Thoracic Society represents a global organization with diverse clinical expertise and scientific interests. The majority of respondents are clinicians; however, the membership has a strong academic bent with most reporting academic affiliation, and describing teaching as a secondary activity.
Key Words: career choice • medical faculty • pulmonary disease (specialty) • questionnaires A century ago, practice in pulmonary medicine was primarily that of diagnosing and monitoring infection. Tuberculosis was rampant, and infectious diseases were common killers. The American Thoracic Society, originally named the American Sanatorium Society, was formed at the turn of the nineteenth century, in 1905, as a division of the American Lung Association and focused on the medical aspects of tuberculosis (1). In 1960, the name of the society was changed to its current one to better reflect clinical practice (2). In more recent years, the Society has expanded its scope of activities to meet the needs of its growing international membership and to encompass areas including critical care, sleep, nursing, and behavioral science. In 2000, the American Thoracic Society became an independently incorporated society. At that time, the American Thoracic Society conducted a survey of its membership to understand better the changing demographics and activities of the membership. The survey was designed to obtain information about the demographics, work practices, and areas of specialization of members, as well as to elicit responses regarding satisfaction with the Society activities. This represents the first comprehensive survey of pulmonary physician practices, and includes both U.S. and international members.
Survey Development The ATS Membership Committee was charged with developing a survey to address issues related to demographics, training and professional activities, type of practice, and faculty affiliation of its membership. In November 1998, Phase I of the ATS Membership Survey was initiated. A six-page, self-administered survey was sent to 13,598 members (3,113 international, and 10,485 United States and Canada). A reminder postcard was sent 3 weeks after the initial mailing, and a second mailing was sent to all nonresponders 1 month later. Survey replies were accepted through April 1999. At the completion of Phase I, a preliminary data analysis was conducted. From this initial analysis, 13 items were deleted from the questionnaire in an effort to increase the response rate and the revised questionnaire was mailed out to nonresponders (Phase II). Results from the Phase II data collection were similar to the Phase I data and thus the data were pooled. The questionnaire was divided into the following issue areas: Training and Professional Activities (eight questions), Member Benefits (five questions), Postgraduate and Continuing Medical Education (two questions), Annual International Conference (four questions), Journals (two questions), Technology (five questions), Overall Satisfaction (three questions), Demographic Information (four questions), and a section with questions specific only to the international members (seven questions). For the complete set of survey questions see Figure E1 in the online supplement.
Statistical Analysis
From combined Phase I and Phase II mailings sent to the 13,598 active members in the organization at the time of the survey, 126 were returned undelivered, and 6,973 responses were received for the final analyses. An initial 5,660 surveys were returned (42.0% response rate) with Phase II yielding an additional 1,313 responses out of 7,938 surveys sent (16.5%), for a total response rate of 51.8% (6,973/13,472). See the online supplement for a complete set of responses.
Demographics
The survey asked members to categorize their principal professional activity, defined as more than 50% of time in that activity. Sixty-five percent of respondents listed clinical practice as their major activity (Figure 2) . Twenty percent identified themselves as primarily researchers, either in basic science or clinical research. When analyzed in terms of sex, women were more likely to identify themselves as researchers than were men (p < 0.001) (Table 2) . The principal activities of international members were not significantly different from those of U.S./Canadian members (see Table E25 in the online supplement). Teaching was considered a primary or secondary activity for 69% of respondents. Thus, although the majority regard themselves as clinicians, Society members retain an educational and academic orientation in their careers.
We next assessed whether age was associated with career choice differently for men and women. The percentage of men identifying themselves as clinicians is the same for those younger than 45 years (78.1%) compared with men at least 45 years old (78.5%) (p = 0.743). In contrast, for women younger than 45 years, 69.6% identify themselves as clinicians, whereas for women older than 45 years old, a smaller proportion, 62.5%, identify themselves as clinicians (p = 0.039) (Table 2). Thus, younger women are more likely to identify themselves as clinicians than are their older counterparts.
Training and Certification To understand the prior training of our physician members, we asked about primary and secondary board certification. The primary specialties listed for the majority of physician members are internal medicine (74%) and pediatrics (14%). Other primary specialties include surgery (1.6%), anesthesiology (1.5%), pathology (1.1%), preventive medicine (1.0%), family practice (0.5%), physical medicine and rehabilitation (0.3%), and radiology (0.3%). The secondary (subspecialty) certifications for members are listed in Table 3 . Members have subspecialty certification in numerous areas, with many members having subspecialty certification in more than one field. As expected, physician members are most likely to identify their area of practice as pulmonary medicine, including critical care (78%). Critical care medicine and sleep medicine are relatively new areas of subspecialization. To assess the impact of these areas on physicians' clinical activities, we asked what percentage of time physicians spent in those areas. The majority of physician members (77%) spend some time in critical care medicine, although only 14% spend more than half of their time in critical care (Table 4) . The time currently spent in sleep medicine is more limited, with 90% of clinicians spending less than 25% of their time in sleep medicine and only 2.3% spending more than half of their time in sleep medicine (Table 4).
Work Setting Work setting was assessed for members, and analyzed according to geographical and sex differences (Table 5) . Significant differences in practice setting were noted for both criteria. International members were more likely to practice in a university setting than were North American members (39.6 versus 31.5%, p < 0.0001). Female members were also more likely to practice in a university setting than were male members (41.5 versus 31.6%, p < 0.0001).
Consistent with the majority of respondents reporting teaching as a primary or secondary activity, 69% of respondents report faculty affiliation with a medical school. This number includes full and part-time salaried faculty as well as volunteer faculty. Overall faculty affiliation rates are higher for international members than for North American respondents and there are differences with respect to faculty rank. Specifically, for North American respondents, there are proportionately fewer full professors compared with international respondents (25 versus 30%, p < 0.0001). For North American and international respondents, fewer women than men have reached the full professor level (12 versus 28.6%, p < 0.0001) (see Table E24 in the online supplement). For both groups, there is a sex discrepancy at the assistant professor level, with 41.1% of women and only 29.9% of men holding this rank (p < 0.0001). Similarly, proportionately more women are at an academic level junior to assistant professor (20.5 versus 14.6%, p < 0.0001).
Technology
The response generated from this survey, the first survey of practitioners of pulmonary and critical care medicine, provided the American Thoracic Society with valuable insight into the needs of Society members. Not only were opinions on Society services, publications, and meetings compiled but the survey has also provided detailed demographics, training, and practice information as presented here. A strength of this survey compared with similar surveys of other physician groups is the inclusion of a sampling of both North American and international members. Other surveys of professional organizations have examined U.S. members, Canadian members, or European members, but such surveys have not compared data across nations (4–10); this survey is unique in this regard. Our findings illustrate that ethnicities other than Caucasian are underrepresented in the North American membership. In particular, African American membership is sparse, as African Americans comprise about 11% of the U.S. population, but only 1.5% of the ATS respondents. This low percentage of minorities is not unique to pulmonary medicine. Overall, African Americans represent 2.6% of all physicians in the United States. Only 3.6% of physicians in internal medicine, 2.2% in pulmonary diseases, and 1.5% in neurology are African American (11). The lack of minority physicians has important ramifications because minority physicians are more likely to provide medical care for minority patients and underserved populations (12–15). Current data indicate that minority students are selecting careers other than medicine (16, 17). If so, the paucity of minority members in the ATS will persist for a long time. To improve this, recruitment strategies such as targeting prospective students with an interest in medicine at a high school, college, and medical school level, focusing on mentorship support during pulmonary training, and promotion of monetary support/scholarship programs are needed. The mean age of the membership is young, which may reflect this as an organization with a younger age than the majority of physicians in practice. In particular, there is a preponderance of females in the younger age groups, consistent with the younger age of women physicians in the United States. According to the American Medical Association (AMA), females currently comprise 22.8% of all U.S. physicians (177,030 of 777,859) (12). Per the AMA database, 45% of U.S. physicians are older than 45 years (mean age, 47.5 years), but 65% of female physicians are younger than 45 years and only 39% of male physicians are younger than 45 years (11) Women comprise a small proportion of pulmonary physicians. As of 1999, 11% of board-certified pulmonary physicians in the United States were women (direct communication, American Board of Internal Medicine). American Thoracic Society respondents, however, show a higher percentage of women (17%). There are several possible explanations for this: first, our membership may reflect younger physicians, among whom the percentage of women is higher. Second, women may have been more likely than men to respond to the survey. However, the percentage of women respondents is identical to the percentage of women in the ATS membership database. Third, nonphysicians within the Society may skew these proportions. However, nonphysicians account for a small proportion of respondents (11%). Although women comprise 17% of American Thoracic Society membership, they appear to be entering pulmonary specialization from internal medicine at a lower rate than women completing internal medicine residency programs. In 1998–1999, 23% of the first-year fellows in pulmonary/critical care were women, whereas in 1997–1998, 35% of internal medicine graduates were women (18). Investigation into subspecialty choice of women residents may be informative. In terms of academic position, results of the survey show that a lower percentage of women are full professors than men, despite the fact that older women were more likely than comparably aged men to identify themselves as researchers, particularly for the North American members. The lower success rates of women in scaling the academic ladder are similar to other reports of women physicians and women scientists (6, 19–23). A cohort study of medical school graduates showed that women pursue an academic career more often than do men; however, the number of women who advanced to associate and full professors was significantly lower than expected (19). In 1987, the first added qualification was offered in critical care medicine. In 1994, as an indication of the increasing involvement of pulmonary physicians in critical care medicine, the official journal of the American Thoracic Society changed its name from the American Review of Respiratory Disease (1959–1993) to the American Journal of Respiratory and Critical Care Medicine. Although the majority of our members spend some time in the critical care field, few spend the majority of their time doing critical care. This may represent self-selection of our membership: pulmonary physicians who spend the majority of their time in critical care medicine may choose membership in other professional organizations such as the Society of Critical Care Medicine. The same may hold true with physicians involved in sleep medicine; physicians with a strong interest and concentration in sleep may select other professional organizations. Sleep medicine is also a relatively new field of study, with a board certification first offered in 1978. Of interest, in 1990, only 54 of 320 (17%) professional recruitment advertisements in the then American Review of Respiratory Disease requested sleep expertise; in 2000, 187 of 397 (47%) advertisements in the American Journal of Respiratory and Critical Care Medicine requested sleep expertise. It will be of interest to track practice activities over time to determine whether the areas of sleep and critical care medicine become the domains of a select group of pulmonary physicians or whether these areas will be integrated into a general pulmonary medicine practice. The rapid growth of computer-based communication and electronic transfer of information is evident in access to these technologies by the respondents. The use of technology by respondents parallels increasing use throughout the United States. As of 2001, 56.4% of all U.S. households owned a computer and 50.4% had Internet access. For households with incomes greater than $75,000, 89% owned computers, and 85.4% had Internet access (24). Despite the prevalence of Internet and e-mail access, the majority of respondents preferred mail as the method of communication. As Internet use continues to grow at a record pace (25), it will be important to determine whether electronic communication is embraced by more members over time. There are a number of potential biases to these data. First, surveys employ self-reporting, which may be less accurate than observational studies. The response rate for the survey at 52% was comparable to other large sample surveys looking at a minimum of 1,000 physicians (26–28). However, nonresponders may have different characteristics from responders. We attempted to validate our results by comparing responses from Phase I with responses obtained from Phase II. Identical results were obtained from both phases, which suggests that nonresponders may be similar to responders. Furthermore, other studies of physicians have shown that survey responders and nonresponders share similar demographic profiles, perhaps because physicians are a more homogeneous group than the general population (26–28). We also compared the survey demographic data with the ATS membership database and found similar breakdown of sex, age, ethnicity, work settings, principal activities, and board certification, suggesting that the survey responders are representative of the current ATS membership. For North American members, American Thoracic Society membership might be considered representative of board-certified pulmonary physicians. Within the United States this is a reasonable assumption, as the American Thoracic Society membership represents 85% of the 9,102 board-certified adult pulmonary physicians (direct communication with American Board of Internal Medicine, 1999 data). There have been many changes in medicine during the last century, including the development of the specialty of pulmonary medicine and the formation of the American Thoracic Society. The number of pulmonary physicians has dramatically increased in the past century, and the focus of clinical activities has continued to expand and evolve. The survey has provided a snapshot of the current activities and practices and demographics of pulmonary physicians. Some of the findings, such as the lack of ethnic diversity, small numbers of women choosing pulmonary or critical care medicine compared with internal medicine, and slow academic progression of women, are similar to those reported by other professional societies (9, 10, 21, 29). Other findings, such as the strong educational ties of the membership, were gratifying, suggesting that clinical and academic endeavors are important to ATS membership. With the large number of international members and members in numerous subspecialties, we are an increasingly diverse group. Knowledge of the membership facilitates strategic planning for the Society. The organization can be strengthened by focusing on the clinical interests of its membership, and by improving representation of minorities, women, and international members. In addition to identifying the current demographics and activities of our members, and by extension, pulmonary physicians, the survey results provide a benchmark to measure changes in the profession as we continue into the twenty-first century.
The authors thank Drs. Beth Kolko and J. Randall Curtis for advice and review of the manuscript, and Chris Keron for statistical expertise.
This article has an online supplement, which is accessible from this issue's table of contents online at www.atsjournals.org Received in original form March 8, 2002; accepted in final form January 7, 2003
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
