Statistical methodology: VII. Q-methodology, a structural analytic approach to medical subjectivity.

Q-methodology is a relatively unknown tool to medical researchers, that uses a mixed quantitative/qualitative statistical technique known as by-person factor analysis to study subjectivity. Q-methodology allows the grouping of individuals according to their subjective feelings about a topic, without requiring preconceived ideas regarding the structure of those subjective feelings. Q-methodology is demonstrated using a study in which attitudes of emergency medicine residents toward computer education are explored systematically and classified in terms of statistically distinct factor viewpoints. In this example, Q-methodology identifies 4 attitude groups, as related to computers and computer education: 1) interested, eager to learn; 2) frustrated and interested, but with reservations; 3) interested mainly in benefits, willing to expend minimal effort; and 4) knowledgeable, independent learner. Q-methodology can be used to determine the structure of attitudes on a subjective topic, often yielding new insights.

Interpersonal styles of nurse practitioner students during simulated patient encounters.

Evaluation methods are needed to assess nurse practitioners' (NPs') interpersonal skills and provide students with systematic, qualitative feedback. The purpose of this study was to identify characteristics and styles of students' interpersonal behavior from patients' perspectives during simulated encounters. The 29-item Clinical Encounter Q-Set for NPs was generated pertaining to patients' perceptions of their interactions with NP students. Using Q-methodology, simulated patients (SPs) sorted the items immediately after each of their encounters with 45 NP students. Items were rank-ordered along a continuum, ranging from "most like my feelings regarding the encounter" to "least like my feelings." Three interpersonal styles were identified. "Nonjudgmental professionalism" characterized student behavior during the simulation portraying a patient with a sexually transmitted disease. "Competence/confidence" and "empathy/respect" were predominant styles exhibited during the hypertension simulation. The potential value of this method for teaching and evaluation is discussed.

Comparison of maternal absenteeism and infant illness rates among breast-feeding and formula-feeding women in two corporations.

PURPOSE: A comparison was made between breast-feeding and formula-feeding among employed mothers. Absenteeism directly related to child care was examined. DESIGN: This quasi-experimental study followed convenience samples of breast-feeding and formula-feeding mothers until their infants were weaned or reached 1 year of age. SETTING: Two corporations with established lactation programs were used. One had approximately 100 births annually among 2400 female employees, and the other had approximately 30 births annually among 1200 female employees. SUBJECTS: A sample of 101 participants, 59 feeding breast milk and 42 using commercial formula, was composed of employees returning from maternity leave for a medically uncomplicated birth. INTERVENTION: The programs provided counseling by a lactation professional for all participants and facilities to collect and store breast milk. MEASURES: Confidential participant diaries provided descriptive data on infant illnesses and related absenteeism that the lactation consultant verified with health care providers and through employer attendance records. ANALYSIS: Attribute counts of illnesses and absenteeism were reported as percentages. Single degree of freedom chi square tests were used to compare rates between nutrition groups. RESULTS: Approximately 28% of the infants in the study had no illnesses; 86% of these were breast-fed and 14% were formula-fed. When illnesses occurred, 25% of all 1-day maternal absences were among breast-fed babies and 75% were among the formula-fed group. CONCLUSIONS: In this study fewer and less severe infant illnesses and less maternal absenteeism was found in the breast-feeding group. This was not an experimental study. Participants were self-selected, and a comparison group was used rather than a true control group. Corroboration of these findings from larger experimental studies is needed to generalize beyond these groups.

Microbiological determination of drug partitioning. IV. Drug-protein interactions.

The protein binding characteristics of chloramphenicol, furazolium chloride, benzalkonium chloride, and phenylmercuric nitrate were described from their partitioning behavior in gelatin-acacia complex coacervate systems. Although the partitioning was determined by two different methods (microbiological and chemical), the microbiological method was more reliable for this type of investigation. Drug-protein parameters were calculated for the four antimicrobials. The advantages of the coacervate systems over other models for protein binding studies of drugs are discussed.