A hypothetical model of the effect of medical education on specialty choice.

BACKGROUND AND OBJECTIVES: Using the Theory of Reasoned Action, we propose a model that diagrams medical school characteristics known or hypothesized to influence the process of specialty choice. The medical school characteristics we consider are administrative support, special programs, primary care funding, number and quality of primary care faculty, faculty influence, primary care residencies, committee representation, primary care environment, required time, and student contact. This model provides explicit hypotheses to be tested in future research on specialty choice.

A systematic analysis of how medical school characteristics relate to graduates' choices of primary care specialties.

PURPOSE: To examine medical school characteristics, in particular federal funding for biomedical research, as they relate to the graduates' choices of family medicine, general internal medicine, general pediatrics, or all three specialties. METHOD: Data were collected for 121 U.S. medical schools, including information on funding, faculty, curricula, and other school characteristics. In addition, a questionnaire was mailed to the schools requesting information about non-federal funding for primary care, primary care department characteristics, and primary care representation on the admission, curriculum, and promotion and tenure committees. Analyses were carried out separately for each specialty and for all three combined. The first multiple regression analysis was done to predict specialty choice (proximate predictors), the second to predict the predictors of specialty choice (intermediate predictors), and the third to predict those predictors (distal predictors). RESULTS: Prediction was best for family medicine practice. Interest at matriculation and required third-year and fourth-year time in primary care were the two best proximate predictors. The best predictors of initial interest were the percentage of rural students and special programs for primary care, while the best predictors of required time in primary care were funding for family medicine and the percentage of faculty in family medicine (intermediate predictors). The best predictor of the percentage of faculty in family medicine was funding for family medicine (distal predictor). CONCLUSION: The results suggest that the most effective way to increase the number of physicians with generalist practices is to increase the number of students interested in a family medicine career at matriculation.

The effect of a required third-year family medicine clerkship on medical students' attitudes: value indoctrination and value clarification.

BACKGROUND: The effect of a required six-week third-year family medicine clerkship was examined within a framework of professional socialization. Socialization was considered to consist of an institutional process, i.e., value indoctrination, and a learner process, i.e., value clarification. METHOD: Pre- and postclerkship data from 1,095 students (classes of 1981-1993) at the University of Arizona College of Medicine were analyzed. In addition, specialty match data were obtained. Factor analysis of 19 items on the pre- and postclerkship questionnaires was used to derive four scales measuring attitudes related to family medicine. The students were first grouped into four groups: those who preferred family medicine before and after the clerkship, those who preferred other specialties both times, those who switched to family medicine, and those who switched away from family medicine. Then the students were grouped into eight groups by dividing each of the specialty-preference groups into two sections: those who matched to family medicine and those who did not. Statistical comparisons involved the use of the t and F statistics. RESULTS: Usable data were available for a maximum of 997 students (91%). The students' attitudes about family medicine changed during the clerkship to become more consistent with their postclerkship specialty preferences. In addition, more students preferred family medicine after the clerkship than before it. When each group was further divided into those matching and not matching into family medicine, no significant difference in attitudes was found between those matching and those not matching. CONCLUSION: These results reflect both a value clarification process and a value indoctrination effect. The discrepancy between postclerkship specialty preferences and later match data indicates that the indoctrination effect and clarification process continue into the fourth year.

Introducing a procedure using videotape instruction: the case of the lateral birth position.

BACKGROUND: Videotapes have been variably successful in teaching physical examination skills, interviewing strategies, and clinically relevant didactic material to medical students, residents, and physicians. The literature does not include evaluations of their success in introducing procedural techniques into clinical practice. OBJECTIVES: This study assessed whether the presentation of an instructional videotape influenced the adoption of a specific technique, assisting birth in the lateral position. METHODS: Family practice residents, faculty, and physicians participated in educational presentations that reviewed how to assist childbirth in the lateral position. Three different formats were used for the presentations; an instructional videotape formed the core element of each presentation. Immediately following the presentations, and on one other occasion 4-6 months later, the participating physicians received questionnaires surveying their use of the lateral birth position. RESULTS: Approximately 30% of the respondents who had not previously used the position used the lateral birth technique. Of the remaining respondents who had not used the technique, 64% considered using the technique with their laboring patients. Physicians who viewed the videotape in concert with other educational interventions were more likely to have adopted the new birth technique (P = .03). CONCLUSIONS: An instructional videotape, as part of a structured, interactive presentation, can be used successfully to introduce a procedure into the clinical practices of family physicians.

Young adults, alcohol and condom use: what is the connection?

PURPOSE: Concern about HIV in adolescents and young adults has led a number of investigators to examine the link between drinking and risky behavior, including failure to use condoms. Most of the studies that conclude that this link exists have used global measures of both variables and have not investigated drinking and use of condoms on the same occasion. If drinking leads to low condom use, both behaviors must occur together and the failure to find that they do weakens the hypothesis that they are causally related. METHODS: Data from three separate studies on different adolescent and young adult populations are reported. In each study the relationship between drinking behavior and condom use at last intercourse was examined. RESULTS: Drinking was not related to lower condom use in any of the populations studied. CONCLUSIONS: This negative result implies that the dis-inhibiting effect of alcohol is not a major factor in the failure of young adults to use condoms.

PIP: The use of alcohol is often cited as a potential precursor to risky behavior including intercourse with casual partners and unprotected sexual intercourse. Most studies which have concluded that a link exists between adolescent and young adult drinking and risky behavior, however, have used global measures of both variables and have not investigated drinking and condom use on the same occasion. If drinking leads to low levels of condom use, both behaviors must occur together. The failure to find that they do simply weakens the hypothesis that they are causally related. The authors report data from three studies on different adolescent and young adult populations, examining the reported relationship between drinking behavior and condom use at last intercourse. One study surveyed a computer-generated random sample of 1287 university undergraduates in the Fall of 1991. Responses were received from 708 students aged 16-62 of mean age 22.5 years. The same questionnaire was administered over the period February-October 1992 to a convenience sample of 369 students aged 16-60 years of mean age 29.9 years at technical and community colleges. Final study results are derived from responses to a questionnaire about HIV and condom use distributed Spring 1992 to 172 people aged 14-35 of mean age 17.7 attending an urban street fair. Each sample was overwhelmingly white and self-reported heterosexual. 58%, 81%, and 65% of the first, second, and third studies, respectively, were female. Survey responses indicate that drinking was not related to lower condom use in any of the populations studied. This finding implies that the inhibiting effect of alcohol is not a major factor in the failure of young adults to use condoms.

Patient satisfaction survey instrument for use in health maintenance organizations.

The objective of this research was to devise a survey instrument specifically applicable to prepaid health care plans that could accurately predict whether patients would disenroll from their current plan because of dissatisfaction when given the opportunity to do so. A "prequestionnaire" was sent to all employees at a southwestern university whose employee benefit package included the option of selecting one of several health maintenance organizations (HMOs) as a source of health care. The prequestionnaire included 90 variables reported in the literature as related to patient satisfaction. The prequestionnaire was mailed two months before "open enrollment," the time at which subjects would have the opportunity, if desired, to change HMOs. After open enrollment, a "postquestionnaire" was sent to the same subjects, asking whether or not they did change plans during open enrollment. There were 2,365 respondents enrolled in HMOs who formed the study population. Of these, 189 (8.0%) changed HMOs during open enrollment. Discriminant function analysis was used to identify prequestionnaire variables which were predictive that subjects had changed plans; 10 variables were identified. They were combined into a survey instrument, which can be scored to predict an individual subject's probability of changing plans.

The current status of family medicine geriatric fellowships.

A questionnaire was sent to directors of all family medicine-affiliated geriatric fellowship programs to characterize trends and changes since institution of ACGME accreditation of fellowships. The number of fellowships has more than doubled since 1986, but few fellows graduate from these programs. There is currently a surplus of fellowship positions; over half of all programs did not recruit a first-year fellow for the 1989 academic year. In 1986, most family medicine-affiliated programs had independent administration. Almost half of the programs are now jointly sponsored with an internal medicine department. These programs are over six times more likely to train internists than family physicians. Although there has been a growth in geriatric training programs over the past three years, the number of family physicians seeking such training remains negligible.

No relationship between pregnancy complications and variations in blood glucose levels among nondiabetic women.

Research using high-risk pregnant women suggests that postprandial blood glucose levels at the high end of normal are associated with greater risk of adverse pregnancy outcomes than low-normal levels. The objective of this study was to determine the relationships between pregnancy complications and glucose levels in low-risk pregnant women. Based on 2-hour postprandial glucose testing at 27 to 33 weeks, 337 women with normal reproductive histories were divided into three groups: group A, glucose less than 5.6 mmol/L (100 mg/dL), group B, 5.6 to 6.6 mmol/L (100 to 119 mg/dL), and group C, 6.7 to 9.1 mmol/L (120 to 164 mg/dL). Women with glucose levels greater than or equal to 9.2 mmol/L (165 mg/dL) were excluded. The groups were compared to detect differences in rates of various maternal outcomes (preeclampsia, cesarean delivery, forceps delivery) and neonatal outcomes (macrosomia, Apgar scores, prematurity, fetal death, infant death, congenital anomalies). No significant differences were found. These data indicate that variations in maternal glucose tolerance (within the normal range) are not associated with adverse outcomes in normal pregnant women.

Characteristics of medical schools related to the choice of family medicine as a specialty.

Previous research has identified five characteristics of medical schools that are related to the choice of family medicine as a specialty: (1) the amount of time devoted to required training in family medicine, (2) the timing of the required family medicine training, (3) the type of ownership of the school (public or private), (4) the geographic location of the school, and (5) the administrative structure of family medicine within the school. These five characteristics of U.S. medical schools during the mid-1980s, together with the school tuition levels, were examined with both univariate and multivariate analysis to observe their relationships to the percentage of U.S. medical graduates entering family medicine between July 1986 and December 1987. With univariate analysis, each characteristic was significantly related to the percentage of graduates entering family medicine. Using multivariate analysis, only the number of weeks required and the type of ownership of the school were significantly related to the percentage of graduates entering family medicine, with the higher percentages related to greater numbers of required weeks of family medicine training and to public ownership of the school.

The postpartum Papanicolaou smear.

This study was performed to evaluate the importance of obtaining postpartum Papanicolaou (Pap) smears routinely. Four hundred eighty-nine patients receiving pregnancy care had a normal prenatal Pap smear and a repeat Pap smear at their postpartum visit. Twenty-four (4.9 percent) had an abnormal postpartum Pap smear (95 percent confidence interval: 3.1-6.9 percent). Twenty-one (87.5 percent) of the abnormal smears showed squamous dysplasia; three (12.5 percent) showed squamous atypia. No specific risk factors were identified that predicted the occurrence of an abnormal postpartum Pap following a normal prenatal Pap except for age. Women more than 30 years of age were less likely to have an abnormal postpartum Pap smear (P = 0.008). The results of this study support the practice of performing Pap smears during prenatal care and again at postpartum examination, even when the prenatal Pap smear is normal.

Cost containment: the effect of educating family practice residents and their patients.

A study was carried out to see if medical costs could be lowered by educating twenty six family practice residents and their patients. Four study conditions were were created: both resident and patient educated, resident only educated, patient only educated, and neither educated. Encounter data were collected before and after the resident education phase and a follow-up phone call was made to patients to obtain information on their current health. Cost measures included 1) physician's fee, 2) laboratory fees and 3) total ambulatory costs. Costs were lowered only in the both-educated group for laboratory costs. It was not sufficient to educate only residents, the group generally targeted for education about cost. This finding suggests that the most effective educational programs may involve patients as well as physicians.

Confidentiality expectations of patients in teaching hospital clinics versus private practice offices.

Two hundred and fifty patients at a university hospital teaching clinic and 200 patients in a small private medical office were questioned about their expectations for confidentiality. It was found that in general, subjects in both groups had similar expectations of how physicians handle confidential information. This included physician behavior in situations such as discussing cases with other physicians for second opinions, submitting cases to medical journals, and discussing cases at parties or with spouses and friends. The similarity existed despite demographic differences between the two groups and despite marked differences in the two practice settings. A few differences between the groups were noted, however. University clinic patients were more likely than private office patients (54.5% vs 38.5%) to expect that cases are discussed at large physicians' meetings. Private office patients were more likely (77.8% vs 67%) to expect physicians to discuss cases with nurses. Men in both practice settings were more likely than women to believe that physicians commonly submit cases to medical journals (75.8% vs 59.9%), or discuss cases with nurses (81% vs 67%), non-medical friends (10.9% vs 4.4%), spouses (53% vs 36.4%), and at parties with physicians (56.4% vs 39.6%).

KIE: Confidentiality expectations were solicited from 250 patients of a University of Arizona outpatient teaching clinic and from 200 private office practice patients of two of the authors. Despite demographic and practice setting differences, both groups of patients in general had similar impressions of how physicians handle confidential information with other health professionals, spouses, and friends, at medical meetings, and for publication. The authors conclude that physicians' behavior regarding confidentiality is not of critical importance in patient choice of a health care setting. They note, however, that their study evaluated perceptions of confidentiality and not preferences about confidentiality.