Implementing clinical practice guidelines: social influence strategies and practitioner behavior change.

Clinical practice guidelines offer potentially valuable tools for rationalizing health care delivery and improving patient outcomes. Currently, major efforts are under way to develop, test, and refine guidelines for a wide variety of medical conditions and procedures. Although methods for producing guidelines are fairly well understood and continue to improve, experience suggests that guidelines rarely translate directly into changes in practice. For various reasons, including physician resistance or incomplete understanding of the need for guidelines, they have proven difficult to implement. This article describes a broad range of strategies for implementing clinical practice guidelines based on the social influences that affect health practitioner behavior. The article also examines issues surrounding the selection and use of individual strategies or combinations of strategies in specific efforts to improve the quality of health care.

Two ways to assess problems of evaluating third- and fourth-year medical students: a case study.

Clinical faculty commonly acknowledge that they have difficulty evaluating third- and fourth-year medical students. The faculty and administration of the Louisiana State University (LSU) School of Medicine in New Orleans conducted pilot tests of two types of materials to assess these problems as part of the Clinical Evaluation Program undertaken by the Association of American Medical Colleges in 1984-1985. The two types of materials--questionnaires and the Problem-Case Analysis--were developed by the staff of the AAMC program and used by the LSU Department of Medicine and the Department of Obstetrics and Gynecology to identify problems in the process of evaluating clerks. Furthermore, the two methods proved complementary: both methods identified some of the same problems and each identified problems not identified by the other. Handling difficult students was the single problem identified by both methods in both departments. The questionnaires permit a simple method of collecting data about a broad range of problems. The Problem-Case Analysis, on the other hand, requires more time but evokes possible causes and alternative solutions, as well as identifying the problems. Improvement plans were developed by each department to address the problems that surfaced.

Types of problem students encountered by clinical teachers on clerkships.

The teachers who play the all-important role of enabling students to learn on clinical clerkships must balance the two essential skills of being a good role model and maintaining objectivity in order to identify students with a variety of problems. This study describes the findings of a survey that identifies both the type of the problems that most bother teachers and the relative frequency of those problems. Non-cognitive problems (poor interpersonal skills and non-assertive, shy students) were identified by teachers as being seen at the same relative frequency but posing greater difficulty than cognitive problems (poor integration skills, disorganization, poor fund of knowledge, etc.). A variety of the types of interventions to these problems are discussed.

An AAMC pilot study by 10 medical schools of clinical evaluation of students.

In 1979 the Association of American Medical Colleges (AAMC) began a Clinical Evaluation Program. In the first phase of the program, AAMC staff members interviewed more than 500 persons in approximately 40 medical schools and solicited responses from clerkship coordinators in six specialties on the problems encountered in evaluating students' clinical performance. These results were published in 1983. The second phase of the program, which is reported on in the present paper, consisted of developing a set of self-assessment instruments in clinical evaluation and testing these instruments by 10 pilot schools. Clinical faculty members and clerkship coordinators in the pilot schools and residents at affiliated hospitals who worked with the students were asked to identify the problems they faced in evaluating students' clinical performance, and these responses were analyzed in terms of the respondents' specialties. The problems identified by the highest percentages of clinical faculty members and residents were: inadequate guidelines for handling problem students, failure to act on negative evaluations, lack of information about problems that students bring with them into clerkships, and faculty members' unwillingness to record negative evaluations. A rank-ordering of these responses on problems showed that different specialties considered different problems to be severe. In rating difficulties in evaluating problem students, the respondents reported that the type of problem student most frequently encountered was one who cannot focus on what is important. Rank-ordering of these responses showed that each type of problem student poses different evaluation difficulties in each specialty. The respondents also identified the problems they had with the evaluation forms used in their departments.

The house officer as a teacher: what schools expect and measure.

Teaching is generally regarded as an important function of the house officer not only from a practical viewpoint but also as an integral part of the training of a professional. A survey by the Association of American Medical Colleges shows, however, that unlike other competencies, the house officer's teaching is not explicitly evaluated in most graduate training programs. Furthermore, it appears that the teaching role itself is not always well defined. It is urged that, if the teaching role is regarded as an important element in graduate medical education, earlier and more gradual preparation of the medical student for that role is needed.

Comparison Of Item-Level And Score-Level Typological Analysis: A Simulation Study.

A method of typological analysis was applied to computer-generated l 96-item "questionnaire" data for 100 cases, under a variety of conditions as to: a. response pattern differentiation in the data; b. similarity index used; c. criteria for the formation of clusters-types. 'Under each combination of conditions, analyses were carried out at both item-level and score-level (i.e., records reduced to 12 eight-item "scores"). The results showed a considerable advantage of score-level approach in the number, size, and replicability of clusters recovered.