We examined the effectiveness of a multicomponent group program for spouses of frail aging veterans that included support, education, problem solving, and stress reduction. Compared with caregivers who received no intervention, those in the group program showed significant increases in use of active behavioral coping strategies, knowledge of community resources, perceived independence in the marital relationship, and personal changes in the caregiving relationship. They also experienced significant decreases in subjective burden and the stress and severity of caregiving problems.

Long-term results of an evaluation of a multicomponent support-group program for spouses caring for frail elderly veterans indicated that participants experienced significant reductions in subjective burden. As compared to those of the control group, caregivers' perceptions of their husbands' health improved significantly from pretest to one year, while perceptions of their husbands' functioning on measures of instrumental daily activities showed significantly less deterioration in the period from pretest to posttest.

Physician behaviors were studied in 473 interactions between oncologists and adult cancer outpatients. Ninety-nine of these interactions occured when family members were present during the visit. Patients with family members present were likely to be sicker as demonstrated by a poorer performance status. Contrary to earlier reports, age of the patient did not predict whether the patient was likely to be accompanied by a family member. The physician behaviors were factor analyzed to produce six factors and a multivariate analysis of variance was conducted using the presence of family and performance status as independent variables. The time the physician spent in the patient's room, patient satisfaction and quality of life were also examined in separate analyses. The time the physician spent with the patient was greater when family were present. The results showed that, in general, physicians provide more information when patients are accompanied by family members, or if no family are present, when the patient has a worse performance status. Patient satisfaction and quality of life were rated lower for patients with a worse performance status and were not impacted by physician behaviors. Physicians' behavior was affected by both the presence of a family member, and the patient's performance status.

To examine potential predictors of cancer patient satisfaction with physician behavior, 366 cases were studied. Physician behavior was measured on morning rounds using the Physician Behavior Check List (PBCL). Patient satisfaction and perceptions were assessed after the visit. Patient characteristics were obtained from the chart and the physician. Results showed wide variation in physician behavior; no "standard" set of behaviors was seen in all interactions. Patient satisfaction was high (mean = 87.8 mm on a 100-mm scale). Path analysis showed four variables predicted 62% of the variance in patient satisfaction. The strongest predictor was the patient perception item, "perception of needs addressed that day." Other predictors were perception of emotional support provided by the physician, age (older), and one physician behavior, "discusses treatment." Patient perceptions of needs met or emotional support provided were predicted by perceptions of the occurrence of physician behaviors involving information such as the diagnosis and tests and treatment. Overall, patient perceptions of physician behaviors were stronger predictors of patient satisfaction than the actual occurrence or absence of those behaviors.

Participants in two types of professionally-led groups and one peer-led group reported significant changes in personal problems related to caregiving as compared to controls. No significant changes were found on measures of psychological status and caregiving burden.

Participants in both professionally led and peer-led groups experienced significant improvements in psychological functioning, increases in informal support networks, and positive personal changes in handling of the caregiving role when compared with control participants. Professionally led groups produced the greatest improvement in psychological functioning, and peer-led groups produced the greatest increases in informal support networks.

Adult women caring for frail older relatives were assigned to peer-led or professionalled support groups or to a control condition. Compared to control participants, group participants experienced significantly greater improvements in their ability to cope with the stresses of caregiving. As with past research, few differences were found between participants in the peer-led and professional-led conditions.

Physician behaviors and patient responses were studied in 439 interactions between hospitalized adult cancer patients and oncologists to investigate patient preferences for a participatory role in the interaction. Patients were asked their preference for information to be given (minimal; only if it is good news; or all information, good or bad) and their preference for participation in decision-making (prefer doctor makes therapeutic decisions or prefer to participate in decisions). The majority (92%) preferred all information be given, but only 69% preferred to participate in therapeutic decisions. Of those wanting all the information, 24.9% preferred the physician to make the therapeutic decisions. This group was comprised primarily of older, sicker males. Those who did not want to participate were also slightly more satisfied (P less than 0.05). These data suggest that, although most patients prefer all information to be given to them, almost one-fourth of them preferred a more authoritarian, rather than participatory, relationship with their oncologist.

To examine the impact of a subspecialty care unit on house staff behavior toward patients, fourteen internal medicine teams were observed in 1213 interactions with patients on morning rounds using a previously devised instrument, the Physican Behavior Check List. The eight teams assigned to the hematology/oncology unit (HO teams) saw significantly more cancer patients than did the six teams assigned to general medicine floors (GM teams). Patients seen by HO teams were sicker (P < 0.001). Hematology/oncology teams engaged in fewer support behaviors, particularly with the sickest cancer patients, than did the GM teams. Teams did not differ on time spent with patient or patient involvement but HO teams were rated significantly lower on addressing the patient's needs. The authors propose the concentration of sick cancer patients on a designated cancer unit diminishes house staff ability to engage in supportive behaviors and to address the needs of patients during morning rounds.

The present study examined the effects of asking subjects to keep their blood-pressure level low while in the presence of an arousing stimulus. It was hypothesized that, without assistance, subjects' attempts to keep their blood pressure low would actually produce increases in blood pressure, as compared to subjects simply asked to respond naturally to the arousing stimulus. Fifty male subjects watched a videotape containing a neutral (nonarousing) section and an erotic section while their blood pressure was recorded by means of an automated blood-pressure monitoring device. Some of the subjects were asked to relax and keep their blood pressure low during the erotic parts of the videotape; the rest of the subjects were asked to respond naturally. The results confirmed the hypothesis, suggesting that urging people to relax can be counterproductive if they do not also receive systematic instruction on how to relax or control blood pressure. Implications for health messages in the mass media and physician-patient interactions are discussed.