American Journal of Health Promotion

To describe a 4-year intervention targeting fruit/vegetable consumption by high school students.

This is a cohort study involving six pairs of schools (n = 12) matched on gender, race, enrollment, and location with schools randomly assigned within pairs to intervention or control conditions.

Twelve Archdiocese of New Orleans high schools.

Cohort was defined as students (n = 2339) who were ninth-graders in the 1993–94 school year who provided baseline data.

Four components of the intervention are: (1) school-wide media-marketing campaign, (2) school-wide meal and snack modification, (3) classroom workshops and supplementary subject matter activities, and (4) parental involvement.

Focus groups were conducted for target population input and program development. Process evaluation included student feedback on media-marketing intervention materials and activities reported here. Process measures also included school meal participation, student characteristics, and verification of intervention activities.

Focus groups indentified barriers to increased consumption of fruit and vegetables as lack of availability, variety, and inconsistency in taste. Student attitudes were favorable regarding a school program to improve diet and parental involvement. Low consumption of fruits/vegetables was reported. After a 2-month school-wide program introduction utilizing various media-marketing materials and activities, 93% of students were aware of the program and 96% could identify the healthy eating message.

Program development can be guided and enriched by student input via focus groups. Media-marketing activities effectively delivered health messages and attracted students' attention. Materials and activities used were acceptable channels for increasing awareness, positive attitudes, and knowledge about fruits/vegetables.

Despite high rates of spontaneous and assisted smoking cessation during pregnancy, postpartum maintenance is disappointingly low. Predictors of return to smoking remain unclear, thus limiting the development of interventions that could protect the health of women and their children. This study followed women who had participated in a prenatal smoking cessation intervention trial and successfully stopped smoking to address two aims: (1) describe the probability of relapse in confirmed quitters during the first 6 months after the birth, and (2) identify factors that increase relapse.

Prospective design during pregnancy and retrospective report at 6 months postpartum used survival analysis with return to smoking as the dependent variable and the Cox proportional hazards regression technique.

Questionnaires were used at the first prenatal visit and telephone interviews at the 26th week of pregnancy and 6 months postpartum. Nonsmoking after the 20th week was measured by urine cotinine tests (m = 3); nonsmoking postpartum was measured by retrospective self-report.

The setting was an HMO-based group practice in Los Angeles.

Subjects were white, black, and Hispanic women (n = 127) available for follow-up.

The proportion of the sample classified as relapsers at 6 months postpartum was 62.9%. The final model identified the following risk factors for smoking: taking puffs in late pregnancy and having friends who smoke at the first prenatal visit, less confidence in mid-pregnancy that they could maintain nonsmoking postpartum, and a partner who smokes postpartum.

Intervention should begin in late pregnancy, and smoking networks, including partner smoking, should be addressed.

The study was designed to investigate differences in predisposing factors, enabling factors, reinforcing factors, and smoking behavior among middle school students living in the city, suburbs, and rural areas of Shanghai.

A one-time cross-sectional survey design was applied in this study.

The study was conducted in 11 schools in three areas.

A total number of 1116 students from these three areas participated in this study. The response rates were 92, 88, and 84% for city, suburbs, and rural areas, respectively.

A PRECEDE instrument including predisposing, enabling, and reinforcing factors was employed to collect data. A Multivariate Analysis of Variance (MANOVA) and Multiple Regression Analysis were used to analyze the data.

Smoking was reported by 39.8% of boys and 5.14% of girls. Students in the city of Shanghai had the best (against smoking) predisposing factors, but students in rural areas had the best (against smoking) enabling and reinforcing factors and consumed the fewest cigarettes.

Results indicated that smoking behavior in students was affected strongly by enabling factors and reinforcing factors that were related to socioeconomic, cultural, and environmental influences in China.

In this study, the type and size of participating and nonparticipating worksites in a health promotion research trial were examined.

In-person and telephone contacts were made with representatives from a random sample of eligible worksites recruited to participate in a research trial to increase fruit and vegetable consumption.

Worksites that had 250 to 2000 employees and a worksite cafeteria were recruited.

Eighty-one worksites were contacted.

A census of eligible worksites (N = 109) formed the recruitment base, and a random selection of worksites (N = 81) was personally contacted and asked to participate in the research. Information on size, type of worksite, and a worksite representative was obtained from a privately compiled list covering the greater Puget Sound area. Eligibility for the project included the presence of an onsite cafeteria at the worksite as assessed by a brief telephone call.

Thirty-six of the contacted worksites (44.5%) agreed to participate. There were no significant differences in participation by type or size of worksite.

Worksites can be recruited to participate in research trials. No particular type or size of worksite is more likely to participate. Instituting pre-randomization participation conditions may reduce post-randomization dropout by ascertaining compliance with research requirements before randomization.