The medical record was used to obtain patient age, whether patients were new or established, and the extent to which a family medical history was cited in the chart for the observed visit or during any visit over the past year. The presence of a genogram (family tree) was noted, as was whether the medical record contained sufficient information to assess a family history of breast cancer, colon cancer, or alcohol abuse.
The patient exit questionnaire was used to assess education level and health status as measured by the Medical Outcomes Study (MOS) General Health Survey.7 These items used a 5-point Likert-type scale for responses to questions about global health status, health limitations in everyday physical activities, emotional problems, limitations in work because of physical or emotional problems, and body pain during the 4 weeks before the visit. Patient satisfaction was assessed with the MOS 9-Item Visit Rating Scale.8 The questionnaire also assessed the patient’s report of a family history that was taken during the visit or within the past year and whether other family members were patients of the physician. Patients also were asked to rate on a 5-point Likert-type scale the degree to which they agreed with the following statement: “This doctor knows a lot about the rest of my family.” The patient exit questionnaire also assessed 4 components of the quality of primary care, using the previously validated Components of Primary Care Instrument (CPCI).9 The CPCI assesses the amount of interpersonal communication, coordination of care, physician’s in-depth knowledge of the patient, and how much the patient values continuity of care. Its domains have been found to be internally reliable,9 associated with patient satisfaction9 and preventive services delivery,5 and diminished by forced disruption in continuity of care.10
The physician questionnaire measured physicians’ age, sex, and residency training, and also asked them to rate how much they “focus on the family as the unit of care,” assessed with a 5-point Likert-type scale. Physicians also were asked to estimate the percentage of patients that they periodically counsel about familial or genetic diseases. The outcome measures of preventive services delivery, patient satisfaction with the visit, and patient perception of components of primary care were measured using the arithmetic mean and standard deviation of patient scores for each physician.
Development of Family Factors
A factor analysis was performed with physicians as the unit of analysis to determine whether a limited number of physician styles could be ascertained from the family items collected for this study. Data on all patients seen by a physician were represented by a mean for each physician. A detailed description of the resulting 3-factor solution has been published previously.1 Factor scores were computed for each physician on the basis of a sum of the standardized items defining each factor [Table 1].
Two of the 3 factors identified relate to the physician’s degree of focus on the family; the third assessed whether the physician performed prenatal care or deliveries. The first factor, the family-orientation factor, described those physicians who focused on the family as the unit of care. The second factor, the family-history factor, indicated an approach in which the physician obtained a considerable amount of contextual information about the patient and family and used that information to care for the individual patient. The specific items defining each of these factors have been described previously and are also shown in Table 1.1 Scores on these 2 factors were not found to be correlated (r =tion was found between the family-history style and this scale measuring in-depth knowledge of the patient and the family.
Time Use
Physicians who scored high on the family-orientation factor were significantly more likely than those who scored low to spend a greater percent of the visit time gathering family information and counseling and spend less time structuring the visit interaction (Figure 5). Physicians with a high score on the family-history factor had significantly longer visit times (average: ~ 1.5 minutes longer) and devoted a greater percentage of time to preventive services delivery, health promotion, exercise and nutrition advice, counseling, family information gathering, and history taking while spending less time on treatment planning.
After adjusting for multiple testing, only 2 of the clinical behaviors measured by the DOC showed a difference in magnitude of correlation between the 2 family focus styles. The family-orientation style was associated with a significantly greater percentage of time spent obtaining family information; the family-history style was associated with a greater percentage of time spent on preventive services delivery. The latter finding, based on direct observation of time spent for a specific visit, further validates the associations shown in Table 4.