Telephone surveys have been used for years to study a wide variety of topics ranging from public opinions to health information, and they will most likely be an important tool in the planning, conducting and evaluation of community-based health promotion programs designed to accomplish the broad public health goals set forth by the US Government for the year 2000. Many studies have compared the results from telephone and household surveys and found that, for some kinds of information, respondent characteristics and data quality of telephone surveys are similar to those of more time consuming and costly face-to-face household surveys. From March 1989 to May 1990, 1328 adults from Pawtucket, RI were interviewed either in person or by telephone about cardiovascular disease (CVD)-related risk factors, behaviours and knowledge, as well as selected demographic characteristics. Demographic characteristics of respondents to the two surveys were quite similar except for race, which differed significantly between the two surveys. Some self-reported CVD-related characteristics were similar between the two surveys (smoking, history of high blood pressure or cholesterol and self-rated blood pressure or cholesterol compared with others of similar age and sex), while others were not (CVD knowledge index, body mass index, prevalence of obesity, blood pressure, prevalence of hypertension and physical activity). With careful attention to the limitation of telephone surveys, this survey method can confidently be applied to the evaluation of other health promotion programs thus allowing more extensive data collection at less cost.

The processes of change model has been successful in predicting behavior change across a wide range of both addictive and nonaddictive problem behaviors. This study was designed to examine the application of the processes of change model to weight control. Study participants included 285 women and men enrolled in three community-based weight loss programs. Results based on structural equation analyses showed that the processes of change model fit the data better than several plausible alternative models. In addition, structural analyses revealed the existence of two general (higher order) processes of change for weight control, the experiential and behavioral processes. These results are similar to those previously reported for eight other problem behaviors. Limitations of the current work and future directions for this line of research are discussed.

Purpose. The purpose of the study was to evaluate the effectiveness of a community-based weight loss program. Design. A questionnaire was administered by telephone to subjects one to three years following participation in a weight loss program. Setting. The study took place in Pawtucket, Rhode Island. The Pawtucket Heart Health Program is part of this setting. Subjects. A stratified sample of 400 subjects was randomly selected from 2,186 people who participated in weight loss programs between 1985 and 1987. A total of 285 subjects completed the questionnaire; 229 subjects provided sufficient information to be included in the study. Intervention. The subjects participated in community-based and worksite-based weight loss programs. Measures. Self-reported heights and weights before and after intervention and demographic data were collected. Desirable body weight and Garrow's health risk classifications were calculated. Results. Overall, there was a 3.2% reduction in body weight between time of entrance into weight loss programs and time of interview. Eighty percent of the participants lost weight in the program; mean weight loss was 11 pounds. At time of follow-up interview one to three years later, 65% of subjects weighed less than at entrance into the community programs; on average, subjects weighed six pounds less. Sixty-nine percent of the sample was above 20% desirable body weight at entrance and 26% of this group lost enough weight to lower their health risk category. Conclusions. Weight loss in this community-based program compares favorably with those reported by more intense and expensive clinic-bound programs. The community approach may be an alternative to more expensive clinic-based programs.

Health promotion and intervention projects at State and community levels need computerized data bases to assist in making policy decisions and in operating the projects. Computer data base systems are used in entering, storing, retrieving, and analyzing information about health project activities and their participants in a timely and cost-effective manner. Computer support is essential for such labor-intensive tasks as post-screening followup of participants, identifying subpopulations, and evaluating recruitment efforts and behavior change programs. The Pawtucket Heart Health Program developed a microcomputer software package, FPbase, for community health project data base management. FPbase is described and is available for use by other organizations. FPbase incorporates formative and process interactive data base activities and is suitable for use in operating intervention and screening programs at State and local levels. The system accommodates management of data for social marketing, evaluation, followup, and promotional activities.

PURPOSE:This study examined the use of the stages of change model to design an exercise intervention for community volunteers.DESIGN:The "Imagine Action" campaign was a community-wide event incorporating the involvement of local worksites and community agencies. Community members registering for the campaign were enrolled in a six-week intervention program designed to encourage participation in physical activity.SUBJECTS:Six hundred and ten adults aged 18 to 82 years old enrolled in the program. Seventy-seven percent of the participants were female and the average age was 41.8 years (SD = 13.8).SETTING:The campaign was conducted in a city with a population of approximately 72,000 and was promoted throughout community worksites, area schools, organizations, and local media channels.MEASURES:One question designed to assess current stage of exercise adoption was included on the campaign registration form as were questions about subject name, address, telephone number, birthdate, and gender.INTERVENTION:The intervention included written materials designed to encourage participants to initiate or increase physical activity, a resource manual describing activity options in the community, and weekly "fun walks" and "activity nights."RESULTS:A Stuart-Maxwell test for correlated proportions revealed that subjects were significantly more active after the six-week intervention. Sixty-two percent of participants in Contemplation became more active while 61% in Preparation became more active.CONCLUSIONS:Most participants increased their stage of exercise adoption during the six-week intervention. This study provides preliminary support for use of the stages of change model in designing exercise interventions.

Self-reported weights and heights of 82 adults were compared with measured weights and heights 1 to 3 years after participation in community weight loss programs. The mean self-reported weight was 2.3 +/- 1.9 kg lower than measured weight (P < .05). Differences in underreporting were not significant for gender or age group. Heavier individuals misreported their weight to a greater extent (P < .05) than lighter persons, and individuals who had not recently weighed themselves underreported their weight to a greater extent than those who had weighed recently (P < .05). On the average, height was overreported by a mean of 1.8 +/- 2.7 cm. Overreporting increased with increasing height, and men overestimated their height to a greater extent than women (P < .05). Younger subjects reported their height more accurately than those older than 60 years. Results of our study are similar to those of previous investigations that examined self-reporting bias in subjects enrolled in weight loss programs. The mean discrepancy in body weight, however, was greater than that reported in samples drawn from the general population. Our findings indicate that self-reported weight and height values in overweight populations should be interpreted with caution.

Adherence to referral recommendations given to participants at blood cholesterol screening programs is a critical aspect of these efforts to help detect and control high blood cholesterol in the US adult population. In this study, 386 participants who had received two consecutive blood cholesterol measurements above 240 mg/dl (6.21 mmol/l) were interviewed by telephone 3 months after their second measurement (May 1987 - May 1988). Approximately 40% of respondents had seen a physician by the time of the interview; another 30% reported having scheduled an appointment. There was no significant difference in adherence behavior between participants who received a letter reiterating the referral and those who did not. However, participants who received the letter reported greater physician attention to the evaluation and treatment of their high blood cholesterol, primarily because these participants stated that they visited their physician for their high blood cholesterol. Significantly higher rates of further blood tests, cholesterol education material distribution, cholesterol-lowering medication prescription and patient-physician discussions about cholesterol were the result. These findings suggest that consumers can be effective in stimulating and reinforcing physician practice behaviors related to cholesterol control. However, strategies must be crafted so that consumers are aware of, and appreciate, the necessity of seeking physician care when they become aware of a high blood cholesterol level.

Point-of-purchase nutrition education in supermarkets is one intervention strategy of the Pawtucket Heart Health Program, a community cardiovascular disease prevention program in Pawtucket, Rhode Island. Using consumer intercept interviews, awareness of shelf labels and their effect on purchase behavior have been continuously evaluated. Between 1984 and 1988, the percent of shoppers who could identify correct labels increased from 11 percent to 24 percent (95% confidence intervals of difference: 7.17). The percent who reported they were encouraged to purchase the identified foods increased from 36 percent to 54 percent (95% CI of difference: 5.41).

Current physician attitudes and behavior concerning elevated blood cholesterol, recent changes, and reasons for change were measured in a survey of physicians in two cities. Those in a community with both continuing medical and public education programs reported changing their practice significantly over the past two years, more so than those in a comparison community. The physicians did not identify specific elements of a continuing medical education program as important in these differences. Physicians in Pawtucket, Rhode Island, identified requests from the public as important and statistically significant factors in their changed behavior, suggesting that the public education program has become an important influence on physician behavior.