Physician Practice Characteristics Research Articles

Delivery of smoking prevention and cessation services to adolescents.

To describe the delivery of smoking prevention and cessation screening and counseling practices to adolescents and to examine the effect of physician specialty, sex, practice characteristics, and familiarity with preventive care guidelines on the delivery of smoking cessation counseling services. Cross-sectional self-reported survey of pediatricians and family physicians in 3 New York metropolitan statistical areas who had seen 1 or more adolescents for well care within the past 6 months. Of 564 eligible physicians, 371 (66%) responded. Physicians reported asking most adolescents about smoking (91%) but were less likely to ask about peer smoking use (41%) or smokeless tobacco use (32%). Similarly, they reported assessing motivation to quit for 81% of smokers, but less often helped set quit dates (34%) or scheduled follow-up visits (28%). Family physicians were more likely to provide more effective smoking cessation interactions than pediatricians (mean smoking counseling performance score, 61 vs. 53; P<.001). Family physicians were also more likely to be familiar with National Cancer Institute guidelines than pediatricians (48% vs. 27%; P<.001). Female physicians reported having spent more time with their last adolescent patient (mean, 26 vs 21 minutes; P<.001) and more often spent time alone with adolescent patients (85% vs. 76% of visits; P<.001) than did male physicians. In multivariate modeling, specialty, familiarity with National Cancer Institute guidelines, time spent, and confidentiality factors were associated with better smoking counseling performance. Familiarity with smoking cessation guidelines and physician's specialty and practice style with adolescents are associated with better delivery of tobacco cessation counseling to adolescents.

Physician-patient communication in HIV disease: the importance of patient, physician, and visit characteristics.

Although previous work that considered a variety of chronic conditions has shown that higher quality physician-patient communication care is related to better health outcomes, the quality of physician-patient communication itself for patients with HIV disease has not been well studied. To determine the relationship of patient, visit, physician, and physician practice characteristics to two measures of physician-patient communication for patients with HIV disease. Cross-sectional survey of physicians and patients. Cohort study enrolling patients from throughout eastern Massachusetts. 264 patients with HIV disease and their their primary HIV physicians (n = 69). Two measures of physician-patient communication were used, a five-item general communication measure (Cronbach's alpha = 0.93), and a four-item HIV-specific communication measure that included items about alcohol, drug use, and sexual behaviors (Cronbach's alpha = 0.92). The mean age of patients was 39. 5 years, 24% patients were women, 31.1% were nonwhite, and 52% indicated same-sex contact as their principal HIV risk factor. The mean age of physicians was 39.1 years, 33.3% were female, 39.7% were specialists, and 25.0% self-identified as gay, lesbian, or bisexual. In multivariable models relating patient and visit characteristics to general communication, longer reported visit length (p<.0001), longer duration of the physician-patient relationship (p =.02), and female gender (p =.04) were significantly associated with better communication. The interaction of patient gender and visit length was also significant (p =.02); longer visit length was more strongly associated with better general communication for male than female patients. In similar models relating patient and visit characteristics to HIV-specific communication, longer visit length (p <.0001) and less advanced disease stage (p =.009) were associated with better communication. In multivariable models relating physician and practice characteristics to general communication no variables were significant. However, both female physician gender (p =.002) and gay/lesbian/bisexual sexual preference (p =.003) were significantly associated with better HIV-specific communication. In this study, female and homosexual physicians provided higher quality HIV-specific communication than male and heterosexual physicians. Better understanding the processes by which female and homosexual physicians achieve higher quality communication may help other physicians communicate more effectively. Health care providers and third-party payers should be aware that shorter visits may compromise physician-patient communication, and that this effect may be more consequential for male patients.

Brief Reports

Background: Although previous work that considered a variety of chronic conditions has shown that higher quality physician-patient communication care is related to better health outcomes, the quality of physician-patient communication itself for patients with HIV disease has not been well studied. Objective: To determine the relationship of patient, visit, physician, and physician practice characteristics to two measures of physician-patient communication for patients with HIV disease. Design: Cross-sectional survey of physicians and patients. Setting: Cohort study enrolling patients from throughout eastern Massachusetts. Study subjects: 264 patients with HIV disease and their their primary HIV physicians (n = 69). Measurements: Two measures of physician-patient communication were used, a five-item general communication measure (Cronbach's α = 0.93), and a four-item HIV-specific communication measure that included items about alcohol, drug use, and sexual behaviors (Cronbach's α = 0.92). Results: The mean age of patients was 39.5 years, 24% patients were women, 31.1% were nonwhite, and 52% indicated same-sex contact as their principal HIV risk factor. The mean age of physicians was 39.1 years, 33.3% were female, 39.7% were specialists, and 25.0% self-identified as gay, lesbian, or bisexual. In multivariable models relating patient and visit characteristics to general communication, longer reported visit length (p < .0001), longer duration of the physician-patient relationship (p = .02), and female gender (p = .04) were significantly associated with better communication. The interaction of patient gender and visit length was also significant (p = .02); longer visit length was more strongly associated with better general communication for male than female patients. In similar models relating patient and visit characteristics to HIV-specific communication, longer visit length (p < .0001) and less advanced disease stage (p = .009) were associated with better communication. In multivariable models relating physician and practice characteristics to general communication no variables were significant. However, both female physician gender (p = .002) and gay/lesbian/bisexual sexual preference (p = .003) were significantly associated with better HIV-specific communication. Conclusions: In this study, female and homosexual physicians provided higher quality HIV-specific communication than male and heterosexual physicians. Better understanding the processes by which female and homosexual physicians achieve higher quality communication may help other physicians communicate more effectively. Health care providers and third-party payers should be aware that shorter visits may compromise physician-patient communication, and that this effect may be more consequential for male patients.

Medical malpractice among physicians: who will be sued and who will pay?

This paper examines whether a physician's future claims of medical malpractice are predictable from information on the physician's recent claims history, training credentials, practice characteristics, and demographics. Data on the medical malpractice experience of 8,733 Michigan physicians between 1980 and 1989 is analyzed. We find strong evidence of repetition over time regarding who was sued and who paid claims. The worse a physician's malpractice litigation record during 1980-1984, the worse was his record during 1985-1989. Training credentials were also highly predictive of future malpractice experience. Physicians trained at lower ranked medical schools or who went through lower-ranked residency programs faced higher odds of developing adverse malpractice records, even after controlling for their previous litigation record. Growing internet access to information on these characteristics will help inform prospective patients if they wish to avoid physicians likely to be sued and likely to make payments in the future for malpractice.

Primary care physician perceptions of the nurse practitioner in the 1990s.

To evaluate factors associated with primary care physician attitudes toward nurse practitioners (NPs) providing primary care. A mailed survey of primary care physicians in Iowa. Half (N = 616) of the non-institutional-based, full-time, primary care physicians in Iowa in spring 1994. Although 360 (58.4%) responded, only physicians with complete data on all items in the model were used in these analyses (n = 259 [42.0%]). There were 2 principal dependent measures: physician attitudes toward NPs providing primary care (an 11-item instrument) and physician experience with NPs in this role. Bivariate relationships between physician demographic and practice characteristics were evaluated by chi 2 tests, as were both dependent variables. Ordinary least-squares regression was used to determine factors related to physician attitudes toward NPs. In bivariate analyses, physicians were significantly more likely to have had experience with an NP providing primary care if they were in pediatrics or obstetrics-gynecology (78.3% and 70.0%, respectively; P < .001), had been in practice for fewer than 20 years (P = .045), or were in practices with 5 or more physicians. The ordinary least-squares regression indicated that physicians with previous experience working with NPs providing primary care (P = .01), physicians practicing in urban areas with populations greater than 20,000 but far from a metropolitan area (P = .03), and general practice physicians (P = .04) had significantly more favorable attitudes toward NPs than did other primary care physicians. The association between previous experience with a primary care NP and a more positive attitude toward NPs has important implications for the training of primary care physicians, particularly in community-based, multidisciplinary settings.

Commitment to the Underserved: Evaluating the Effect of an Extracurricular Medical Student Program on Career Choice

Background: This study was conducted to compare the practice locations and characteristics of physicians who participated as medical students in an extracurricular program to foster interest in careers of service to medically underserved populations with those of their classmates who did not participate in the program. Methods: Using a mailed questionnaire, we conducted a cross-sectional study of graduates from the classes of 1983-1987 at one southwestern, public medical school. All Commitment to Underserved People (CUP) participants (n = 94)and a random sample of nonparticipating classmates (n = 188)were surveyed. CUP is an extracurricular project with components in each of 4 years of medical school that provides peer and faculty support, curriculum enrichment, and direct service to medically underserved populations. Outcome measures included the size of community of practice, practice type, and practice patient characteristics. Results: Sixty seven (71%) of CUP participants and 126 (67%) of nonparticipants responded. CUP participants were more likely to be women, to specialize in family practice, to practice in the Indian Health Service (IHS) or overseas, to be located in a community of 25,000 or less, and to have participated in the state, service-payback loan program. In multiple regression, the specialty of family practice was associated with practice in a small community, the IHS, and a community health center; CUP participation was correlated with practice in small communities, the IHS, and a foreign country. Conclusions: Participation in the CUP program was associated with the specialty choice of family practice and with practice in settings associated with medically underserved populations. The CUP program has been successful in sustaining entering medical students' interests in underserved practice.

Human immunodeficiency virus infection in patients older than 50 years. A survey of primary care physicians' beliefs, practices, and knowledge.

To assess primary care physicians' attitudes, knowledge, and practices with respect to the human immunodeficiency virus (HIV) in older patients, a prospective survey of a representative cohort of primary care physicians was conducted in Dallas County, Texas, a large metropolitan area. Three hundred thirty primary care physicians participated in the survey. Questions were asked regarding physician demographics, practice characteristics, and knowledge and practices with respect to HIV and the acquired immunodeficiency syndrome (AIDS) in patients older than 50 years. The responses of the following groups were compared: family practitioners vs internists, physicians younger than 40 years vs those aged 40 years and older, those who saw 5 or less vs more than 5 patients with HIV or AIDS per year, and those in private vs nonprivate practice. Most respondents (85.5%) reported having seen 10 or fewer patients with HIV or AIDS in the previous year. Most physicians (69.7%) reported that patients older than 50 years rarely or never asked questions concerning HIV or AIDS. Most physicians rarely or never discussed HIV or AIDS with patients older than 50 years (60.8%) and rarely or never discussed risk factor reduction (67.5%). Physicians were more likely to rarely or never ask patients older than 50 years compared with those younger than 30 years about HIV risk factors (40.0% vs 6.8%, P < .001). Physicians incorrectly rank ordered the most prevalent risk factors in patients older than 50 years. The correct order is (1) male-male sex, (2) intravenous drug use, (3) blood transfusion, and (4) heterosexual sex. Physicians aged 40 years and older were more likely to correctly identify the most prevalent risk factor (P = .03). Family practitioners were more likely to rarely or never ask older patients about risk factors for HIV (54.9% vs 28.9%, P = .007). Primary care physicians have inadequate knowledge concerning HIV and AIDS risk factors in older patients and insufficiently discuss HIV and AIDS with older patients. Physicians should counsel patients of all ages about HIV and AIDS.

Use of hematopoietic colony-stimulating factors: the American Society of Clinical Oncology survey. The Health Services Research Committee of the American Society of Clinical Oncology.

Dissemination of use of the hematopoietic colony-stimulating factors (CSFs) is unprecedented in oncology, with almost all physicians having experience with granulocyte colony-stimulating factor (G-CSF) or granulocyte-macrophage colony-stimulating factor (GM-CSF) shortly after the drugs received Food and Drug Administration (FDA) approval in 1991. The American Society of Clinical Oncology (ASCO) Health Services Research Committee sought to assess patterns of use of CSFs before dissemination of its first-ever publication of ASCO guidelines. A questionnaire describing clinical scenarios was mailed to American oncologists and hematologists who practice medical oncology. In each scenario, the physician was asked whether he would prefer to use a CSF to prevent or treat neutropenia. The response rate to the mailed survey was 49% (N = 475). Most physicians preferred to use CSFs for secondary prophylaxis in patients receiving chemotherapy at rates of 44% to 85%, rather than reduce doses. Patterns of use did not differ for palliative, curative, or adjuvant chemotherapy. While the majority of CSF patterns of care were similar to those recommended in the ASCO guidelines, more than half of the physicians chose to use CSFs in the treatment of febrile neutropenia, an area not supported in the subsequent guidelines. In general, physicians at academic medical centers and in Health Maintenance Organization (HMO) practices were more likely to prefer dose-reduction strategies over addition of CSFs, while fee-for-service physicians preferred the opposite strategies. Variations in CSF preferences for use were related to differences in clinical characteristics (history of afebrile v febrile neutropenia), drug characteristics (G-CSF or GM-CSF), and physician practice characteristics (HMO or fee-for-service setting). However, before dissemination of the guidelines, the majority of American oncologists preferred strategies that were subsequently included in the ASCO CSF guidelines. CSF guidelines would be most likely to reduce CSF use for treatment of afebrile and uncomplicated febrile neutropenia.

Attitudes and Practices of Primary Care Physicians for Prostate Cancer Screening

Prostate cancer screening with digital rectal examination (DRE) and prostate-specific antigen (PSA) is recommended by several professional organizations. Our objective was to assess the prostate cancer screening practices and attitudes reported by primary care physicians. We randomly surveyed 454 Arizona primary care physicians, subsequently excluding 124 ineligible subjects. Overall, 141 of 329 eligible physicians completed the survey (42.9%). Survey data included physician demographics, practice characteristics, screening and follow-up strategies, and attitudes toward screening. One hundred thirty-one physicians (93%) reported screening asymptomatic men with DRE or PSA. Respondents generally agreed that screening tests were accurate and that early detection was beneficial. Screening began at an average patient age of 45 years, though 7.8% of respondents began screening men younger than 40 years and 7.0% began screening men older than 50 years. PSA levels ranging from 3.9 to 40 ng/mL were considered abnormal, and 11.6% of respondents used a cutpoint higher than 10 ng/mL. Primary care physicians report a high rate of screening for prostate cancer and consider PSA and DRE accurate and useful tests. Screening practices, however, varied considerably between physicians. The screening of younger men reported by practitioners would tend to increase the rate of false-positive tests, while using a high cutpoint for PSA and delaying screening beyond age 50 years would decrease the chance for early detection. These screening practices may increase health care costs without necessarily leading to improved health outcomes. Medical Subject Headings (MeSH): prostatic neoplasms, prostate-specific antigen, primary health care, screening.

Physicians, pharmaceutical sales representatives, and the cost of prescribing.

To evaluate the influence of primary care physicians' attitudes toward and use of information provided by pharmaceutical representatives on prescribing costs in ambulatory practice. A mailed questionnaire collected information about physician demographic and practice characteristics and attitudes toward and use of information provided by pharmaceutical representatives. Kentucky physicians practicing primary care adult medicine (family medicine, general practice, general medicine; n = 1603). Relative cost of prescribing, based on physician responses to treatment choices for ambulatory clinical scenarios in primary care. A multivariable regression model assessed predictive relationships between independent variables and prescription costs. Four hundred forty-six returned questionnaires were suitable for analysis. No significant differences were noted in age, gender, days worked per week, or years since graduation between responders and a sample of nonresponders. A significant positive correlation was found between physician cost of prescribing and perceived credibility, availability, applicability, and use of information provided by pharmaceutical representatives (P < .01, Pearson's Product-Moment Correlation Coefficient). Physicians in academic or hospital-based practice settings had significantly lower prescribing costs than physicians in nonacademic and nonhospital practices (P = .001, analysis of variance). Frequency of use of information provided by pharmaceutical representatives (P = .01, multiple linear regression) and the group practice setting (P = .02, multiple linear regression) remained significant, independent positive predictors of cost in the multivariable regression model. Frequency of use of information provided by pharmaceutical representatives and the group practice, nonacademic and nonhospital setting may be associated with increased primary care physician prescribing costs.

Specialty Practice of Family Practice Residency Graduates, 1969 Through 1993

To describe demographic and practice characteristics of physicians who graduated from family practice residency training programs. Information was obtained from the databases of the American Academy of Family Physicians, the American Board of Family Practice (ABFP), and the American Medical Association. Data sets were cross-referenced, and quantitative data from medical membership organizations and the certifying board were summarized. Information regarding medical school graduation and family practice residency completion was reported by the participating residency programs, and data on practice specialty were self-designated by the participating physicians. Physician graduates of Accreditation Council for Graduate Medical Education-accredited family practice residency programs from 1969 through 1993. Practice specialty designation by graduates of family practice residency programs. Of the 38659 physicians who graduated from family practice residency programs from 1969 through 1993, 8780 (23%) were women, 2181 (6%) were graduates of colleges of osteopathic medicine, 4777 (12%) were international medical graduates, and 33484 (87%) were certified by the ABFP. Of the 36088 family practice residency graduates who designated a practice specialty, 32764 (91%) identified their specialty as family practice, including 32608 physicians who identified family practice, 129 general practice, 21 family practice-geriatric medicine, and six family practice-sports medicine. Among the 3324 family practice residency graduates (9%) who identified themselves as practicing in another specialty, the most common specialties were emergency medicine (1446 physicians [4%]) and preventive medicine (386 physicians [1%]). No other practice specialty was designated by more that 0.5% of family practice residency graduates. The vast majority of family practice residency graduates are self-designated as practicing family medicine. These data suggest that family practice residency programs are an effective mechanism for the production of generalist physicians.

Physician-reported determinants of screening mammography in older women: the impact of physician and practice characteristics.

Despite having markedly higher breast cancer risk, compliance of older women with screening mammography has been poor. This study was undertaken to determine which physician and patient practice characteristics were associated with high self-reported mammography referral rates for older women. Primary care physicians (n = 129) from three socioeconomically diverse communities in Los Angeles were surveyed. Agreement with annual screening and self-reported referral rates were assessed for two groups of women, those 65 to 74 years of age and those 75 years and older. Screening outcomes were compared with physician and patient practice characteristics using bivariate and multivariate techniques. Although 73% of physicians agreed with annual screening of women aged 65 to 74 years, only 24% of physicians reported actually screening most women seen in this age group. Similarly, 57% of physicians agreed that women age 75 years and older should be screened annually, but only 21% reported recommending mammograms for most women seen in this age group. In multivariate analysis, white physicians (adjusted OR = 9.1), younger physicians (adjusted OR = 3.85), and those who used the American Cancer Society's low cost mammography projects (adjusted OR = 5.01) were more likely to report screening the majority of women seen. This study suggests that although physicians' intentions to screen older women may be relatively high, a gap exists between intentions and what is reported to be accomplished in practice. Race/ethnicity and physician specialty were the two strongest predictors of high self-reported referral rates, suggesting that targeted interventions may be useful.

Preliminary evidence on retention rates of primary care physicians in rural and urban areas.

The primary study objectives were to 1) determine how many physicians entered primary care practice in rural and urban counties of North Carolina in the 1981 to 1989 period and 2) estimate their length of tenure in these areas. The secondary objective was to identify the physician's demographic, training, and practice characteristics that influence geographic location of practice and length of tenure. A cohort of 1,947 physicians was identified from the North Carolina Board of Medical Examiners database, which included all active, nonfederal primary care physicians who began their initial practice in North Carolina in 1981 or later. The primary outcome was time in practice in a given rural or urban county. Selected data on physician demographic, training and practice characteristics were also available in the database. Approximately one third of physicians beginning their initial North Carolina practice selected a rural county for the location. Almost half of these primary care physicians were still in the county of their initial practice in 1989. An additional 20% of these physicians had changed practice location within the State, of which half chose a similar type of county to that of their initial practice. Length of tenure was similar across geographic locations of the medical practice, with the average length of tenure being 4.6 and 4.4 years among physicians in rural and urban counties, respectively. The strongest predictors of tenure were practice organizational characteristics with physicians in either an office-based solo practice or partnership having longer tenures.(ABSTRACT TRUNCATED AT 250 WORDS)

Demographic characteristics of full-time emergency physicians in Canada.

The objective of this study was to describe individual and practice characteristics of Canadian emergency physicians, and to determine if these characteristics varied with geographic region, type of certification, or method of reimbursement. The study took place in emergency departments across Canada, and all full-time physicians with certification in emergency medicine, and a random sample of Canadian Association of Emergency Physician members without certification were surveyed with a 23-item questionnaire. Descriptive statistics on individual and practice characteristics were outcome measures. Comparative statistics evaluating demographic characteristics by region, type of certification, and method of reimbursement were used. There were significant differences between responders and nonresponders based on certification (P < .001) and region of residence (P < .005). Most physicians are male (87.3%), married (83.3%), reimbursed on a fee for service basis (61%), work in teaching hospitals (65.8%), and participate in medical education (82.3%). On average physicians are 38 years of age, work of 14 shifts/month, 9 hours/shift, 11 nonclinical emergency related hours/week, and have 5 vacation weeks/year. More physicians with FRCP(EM) certification work in teaching hospitals (P < .00001), do nonclinical emergency related work (P < .0001), participate in medical education (P < .0001), and are involved in research (P < .001). There are significant differences by geographic region in gross annual salary (P < .00001), method of reimbursement (P < .00001), and independent hospital emergency department status (P < .00001). Fee-for-service reimbursement for clinical service provides a higher gross annual income than other methods (P = .028). Emergency medicine is an emerging clinical discipline with a strong academic focus.(ABSTRACT TRUNCATED AT 250 WORDS)

Physician practice characteristics as a context for primary care treatment decision making

Physician practice characteristics as a context for primary care treatment decision making

Recruitment and retention of rural physicians: issues for the 1990s.

This paper briefly describes a number of structural and economic changes in the profession of medicine and in the rural medical care delivery system that have occurred since about 1970. Changes in the national physician supply; in the training, work, and practice characteristics of physicians; in the demographic characteristics of physicians; in the medical resources available in rural communities; and in federal and state support for the provision of medical services are noted. Four conceptual models that underlie physician recruitment and retention programs for small towns and rural communities are described. These include affinity models, which attempt to recruit rural persons into training or foster interest in rural practice among trainees; economic incentive models, which address reimbursement or payment mechanisms to increase economic rewards for rural practice; practice characteristics models, which address technical, collegial, referral, and other structural barriers to rural practice; and indenture models, which recruit temporary providers in exchange for scholarship support, loan forgiveness, or licensure. Examples of applications of each model are provided and the effects of changes in the medical care system on the effectiveness of each model are assessed. Finally, it is argued that elements of an optimal model for the recruitment of physicians to rural practice include the promotion of medical careers among rural high school students, the provision of financial and cultural support for their training, the development of technical and collegial support systems, and the limited use of indenture mechanisms to meet the needs of the most impoverished or isolated rural settings.

Primary Care Physicians’ Assessment and Prevention of HIV Infection

The degree and depth to which primary care physicians counsel patients at risk for human immunodeficiency virus (HIV) infection is a major concern. To determine which factors influence whether physicians counsel patients at risk for HIV, primary care physicians's clinical experience, knowledge, attitudes, and preventive counseling advice in hypothetical case scenarios were assessed. Ninety-nine adult primary care physicians in the Washington, D.C., metropolitan area were interviewed by telephone from May through November 1987. Ninety-one physicians had tested or referred patients for HIV antibody tests. However, 58% could not name the ELISA or Western blot as the tests. The most frequent HIV prevention recommendations were using condoms (67.7%), abstaining from sexual activity (36.4%), getting tested for HIV (30.3%), and safe sex (23.2%). Naming the HIV antibody tests was the most significant predictor of preventive counseling advice; other significant predictors included physicians' personal comfort with counseling homosexual patients and various physician practice and demographic characteristics. Previous studies showed that homophobia was the main inhibitor of effective AIDS counseling. These results suggest that physicians' lack of knowledge and general discomfort in counseling patients about sexual risk factors, rather than homophobia alone, are important barriers to preventive counseling about HIV infection.

Trends in health personnel.

Each quarter, Health Affairs reports on new developments in one of four sectors of the health sphere: health personnel in Winter, health status and health care utilization in Spring, health care innovation in Summer, and national health care spending in Fall In this issue, we offer an update on trends in physician supply, health personnel, and medical education. In a related DataWatch, a group of American Medical Association researchers compares the practice characteristics of male and female physicians. Next, Dennis Andrulis and colleagues of the National Association of Public Hospitals and the Council of Teaching Hospitals provide data on Medicaid policies and AIDS. Finally, Glen Misek and Lucy Hynds Karnell of the American College of Surgeons look at surgical residents.

Practice characteristics of male and female physicians.

During the 1970s, more and more women chose medicine as a profession. In 1970–1971, only 9.2 percent of all U.S. medical graduates were women. Ten years later, the percentage of women graduating from medical school had increased to 25.0 percent, and in the most recent graduating class, 1985–1986, the percentage of female medical graduates rose to 30.7 percent. Despite increases in the number of female physicians, a number of differences persist in the socioeconomic practice patterns of male and female physicians. One of the most striking is a continued difference in net income. Previous studies, using data from the 1970s, have attempted to explain the higher earnings of male physicians. These studies found a significant earnings differential, even after controlling for differences in the personal characteristics of male and female physicians. A more recent paper, based on 1982 and 1983 American Medical Association (AMA) data, reported an unadjusted 30 percent difference in the hourly earnings of male and female physicians. After controlling for a number of differences in the personal characteristics of male and female physicians, this study pointed to an unexplained earnings differential of 12 to 13 percent. In effect, this percentage acts as “an upper-bound estimate of the impact of market discrimination on the earnings of female physicians.” Even today, the hourly earnings differential persists. An analysis of 1986 data shows that male physicians continue to make about 30 percent more than their female colleagues make, either per hour or per patient visit. This DataWatch examines the most recent data on the earnings of male and female physicians, paying particular attention to those socioeconomic characteristics that may explain their income differences. The data used in this DataWatch come from the AMA’s Socioeconomic Monitoring System (SMS), a telephone survey program designed to provide information on all nonfederal patient-care physicians, exclud-

Physician Utilization of Computers:

This study demonstrates how social network analysis can be used to identify the underlying communication and influence structure that affects the diffusion of the use of a computer-based medical information system among physicians. Interviews, hospital records, and computer system tapes were used to collect data on referrals, consultations, and practice characteristics from 24 physicians who comprise a private group practice. Multidimensional scaling was used to spatially represent the referral and consultation network among the physicans. Also, several indices that measure structural characteristics of the network and practice characteristics of physicians were derived. Four groups of physicians were identified who have similar computer utilization patterns and who perform similar roles within the network. The results indicate that the use of network analytic techniques to study complex physician networks may provide new insights into the diffusion process.