Background and Objectives: A school-wide curriculum renewal led to a new clerkship curriculum that teaches core family practice competencies by focusing on the process of care in generalist practice. The orga-nizing framework consists of five prototypic visits and their encounter tasks: 1) new problem visit, 2) checkup visit, 3) chronic illness visit, 4) psychosocial problem visit, and 5) behavioral change visit.

Methods: The seminars occur at the beginning of the rotation and use active learning techniques. Evaluation includes stu-dent perceptions of the seminars and teachers and student performance on a clinical performance exami-nation (CPX).

Results: Students rated the usefulness of the seminars and the seminar leaders’ teaching behaviors favorably. The CPX checklist scores showed that students could perform most of the behaviors expected for each prototypic visit. The students listed the appropriate encounter tasks nearly half of the time when describing what tasks they tried to accomplish during the CPX cases. The students listed concrete behaviors just over 50% of the time.

Conclusions: The students learned the material presented in the seminars and applied it during the CPX. Students can do most of the behaviors but do not seem to describe the tasks as abstractly as faculty. These results come from one class cohort in one medical school, so the generalizability is limited until further work, including other learners, confirms these findings.

Medical Student Education
(Fam Med 1999;31(6):391-7.)

Domestic Violence Education in Family Practice Residencies
Sue Rovi, PhD; Charles P. Mouton, MD, MS

Background and Objectives: his study evaluated the extent of domestic violence (DV) education in US family practice residency programs and compared the results to those of a prior study of the same topic.

Methods: We mailed a four-page survey to the directors of all US family practice residency programs. The survey asked the extent to which the topic of DV in particular and other areas of violence in general are included in the curriculum.

Results: Surveys were returned from 298 (65.9%) programs, of which 69.4% of respondents indicated that the extent to which violence education is a formal part of their curriculum is either somewhat or a great deal, and 79.9% responded similarly about DV education specifically. On average, programs provide 4–5 hours of training each year, mostly through didactic lectures. Compared to a previous study, our findings demonstrate an increase in violence education in these programs.

Conclusions: Our findings demonstrate that family medicine educators have increased the amount of residency curricular time devoted to training on DV.

Residency Education
(Fam Med 1999;31(6):398-403.)

Balint Group Observations: The White Knight and Other Heroic Physician Roles
Clive D. Brock, MD; Alan H. Johnson, PhD

Background: This article reports a typology of five roles that resident family physicians on occasion assume when relating to troubling patients presented in Balint group seminars. The five roles include the white knight (my way or no way), the Pogo look-alike (I feel your pain), the missing link (you made me do it), the surrogate (I can help), and the revolutionary (let me show you). Each role reflects a particular physician’s coping behavior in the context of a specific troubling relationship and is driven, in large part, by unrealistic professional expectations. The roles intend to perform a heroic function in rescuing or protecting the patient, the family, or the physician from a distressing medical situation. Balint group work provides participants with the opportunity to derive clinically useful meaning from their presentations. Residents begin to imagine a variety of therapeutic (helpful) roles to replace the ones they were induced to fill. This process has implications for practicing physicians and physician teachers for improving patient and doctor satisfaction and well-being.

Residency Education
(Fam Med 1999;31(6):404-8.)


Factors Influencing Satisfaction for Family Practice Residency Faculty
Lawrence E. Kay, MD; Frank D’Amico, PhD

Background and Objectives: Prior published family medicine faculty satisfaction survey results were performed in 1975, 1984, and 1989. The current survey identified specific factors that contribute to family medicine faculty satisfaction and career decision making.

Methods: We mailed a self-administered questionnaire to a proportionate random sample of family medicine faculty of residency programs identified by a pre-survey of programs. The eight-page survey explored 60 professional, scheduling, compensation, and regional factors as they related to overall satisfaction and career plans. The survey also explored 59 similar factors related to the initial decision to enter academic family medicine.

Results: Of 383 respondents (59.2% response rate), 93% felt satisfied overall with their faculty roles. Eighty-six percent felt appreciated in their current program, and 94% reported a positive sense of professional challenge. Satisfaction, appreciation, and challenge were strongly intercorrelated and were also positively related to whether the faculty surveyed planned to stay in their current position. The opportunity to mentor residents, the ability to keep more up to date with medical information, and income level stood out as being the most significant of all the factors in predicting overall satisfaction.

Conclusions: The faculty surveyed indicated high levels of satisfaction, feeling appreciated, and professional challenge. The results of this cross-sectional survey identify factors most related to satisfaction and the initial decision to enter academic family medicine.

Faculty Development
(Fam Med 1999;31(6):409-14.)

Clinical Factors Affecting Physicians’ Management Decisions in Cases of Female Partner Abuse
Lorraine E. Ferris, PhD, CPsych; Peter Norton, PhD, MD; Earl V. Dunn, MD; Elaine H. Gort, MSc

Background and Objectives: This study determined which clinical factors influence Canadian primary care physicians’ management decisions in cases of female partner abuse.

Methods: We used a cross-sectional survey design and randomly sampled (n=2,014) English-speaking Canadian physicians with a primary interest in family or general practice who were practicing in any of the 12 provinces and territories in Canada and who were active in private practice and registered to prescribe. Respondents completed a questionnaire that required them to score management decision plans in response to case scenarios illustrating typical office-based situations that might involve domestic violence.

Results: The response rate was 50.7% (n=1,022). Using forward stepwise regression analysis, the strongest predictor of whether a physician endorsed a management plan in response to violence was whether the woman acknowledged or revealed the abuse. Male physicians were more likely than females to endorse talking with the suspected abuser if he was known to them, regardless of the quality of this patient-physician relationship with the abuser.

Conclusions: Decisions about whether to deal with the abuse or the selection of a management plan are not dependent on the severity of the physical abuse and the emotional consequences. Whether a woman acknowledges or reveals the abuse, as well as whether both the male and female patients are in the physician’s practice, are predictive of whether a physician’s response to a case scenario involves dealing with spousal abuse and how he/she will address it.

Clinical Research and Methods
(Fam Med 1999;31(6):415-25.)

Medical Investigations Requested by Patients: How Do Primary Care Physicians React?
Orly Cohen, MD; Ernesto Kahan, MD; Simon Zalewski, MD; Eliezer Kitai, MD

Background and Objectives: We investigated the characteristics of patients who request medical investigations and the type of tests requested to study the manner in which primary care physicians react to these requests.

Methods: The study was conducted within the framework of a national health insurance system. Twelve primary care practices from three randomly chosen clinics with different population characteristics participated in the study. The attending physicians were instructed to ask all patients who presented to the clinics within a 7-month period and requested a medical test to complete a questionnaire, indicating the type of test(s) requested and the reason. The physicians were asked to rate the manner in which the patient made the request, their own reaction to the request, and whether they ordered the tests that were requested.

Results: During the survey period, 12,322 patients visited the clinics, of whom 295 (2.4%) were reported by a physician to have requested a medical investigation. More-educated patients were more likely to request tests for disease prevention. The types of tests requested were imaging scans, laboratory (blood) tests, and others. The main reason for the request was symptoms (60%), followed by disease prevention (25%). More than 30% of the requests generated self-reported negative feelings in the physician. Physician compliance with patient requests was not significantly correlated with the reason for the request. Laboratory tests were ordered significantly more often than other types. There was a strong correlation between physicians’ compliance with the request and physicians’ feelings about the request.

Conclusions: Our findings raise questions about the frequency with which physicians order tests solely in response to patients’ requests and provide information about circumstances in which patients make requests for medical investigations.

Clinical Research and Methods
(Fam Med 1999;31(6):426-31.)

How to Use and Interpret Interval Likelihood Ratios
Jeffrey Sonis, MD, MPH

Background: Likelihood ratios offer important advantages over sensitivity and specificity for characterizing diagnostic tests. They can capture the magnitude of abnormality of test results, whereas sensitivity and specificity require that the test results be dichotomized into positive or negative. This is an important advantage because many diagnostic tests are measured on continuous or ordinal scales. Posttest probabilities calculated from interval likelihood ratios may be different than those calculated from sensitivity and specificity; clinical decisions derived from the use of likelihood ratios may therefore be different from decisions derived from test results characterized by sensitivity and specificity. This article demonstrates the advantages, use, and interpretation of interval likelihood ratios using the clinical scenario of a young child with a high fever.

Research Series
(Fam Med 1999;31(5):432-7.)