Gender differences among smokers and quitters in the Working Well Trial.

BACKGROUND: Gender differences in smoking and smoking cessation among participants in the Working Well Trial are characterized. METHODS: A prospective randomized matched-pair evaluation was conducted among 90 predominantly blue-collar worksites. Cross-sectional surveys of employees' tobacco use behaviors were conducted at baseline and after a 2.5-year smoking cessation intervention. Respondents included 5,523 females and 12,313 males at baseline and 4,663 females and 10,919 males at follow-up. The main outcome measures included self-reported continuous smoking abstinence rates for 7 days and for 6 months. RESULTS: Smoking prevalence was significantly higher for women than for men at baseline, but not at follow-up. Variables believed to influence smoking cessation were compared at baseline. Significant gender differences were found for number of cigarettes smoked/day, number of previous quit attempts, job strain, stage of change, and behavioral processes of change. At follow-up, no gender differences in quit rates were observed; however, women in the intervention condition were more likely to quit than women in the control condition, whereas no differences were seen among men by treatment condition. CONCLUSIONS: Gender is not a strong predictor of smoking cessation in this population; however, women were more likely to quit with an intervention than without one.

Components of the working well trial intervention associated with adoption of healthful diets.

INTRODUCTION: This report examines whether variability in the type and amount of the nutrition intervention in a worksite-based intervention could explain dietary outcomes. METHODS: Data are from 55 intervention worksites in the Working Well Trial, a randomized controlled trial of worksite-based health promotion. The components of the nutrition intervention were kickoff event, direct education, interactive activities (e.g., food sampling), contests, printed information picked up by employees, and materials distributed to employees. We measured delivery of the nutrition intervention (i.e., dose) by determining the amount of workforce participation in each intervention component. Diet outcomes were changes in intakes of fat, fiber, and servings of fruits and vegetables (reported on food frequency questionnaires). All variables were aggregated to the worksite level. We correlated the dose variables with indices of receipt of the intervention and with the dietary outcomes. RESULTS: Contests were associated with employee awareness of and participation in the nutrition intervention (r = 0.49 and 0.28, respectively), and interactive activities were associated with intervention participation (r = 0.43). Contests were associated with increased fiber intake and fruit and vegetable consumption (r = 0.36 and 0.31, respectively), and direct education was associated with fruit and vegetable consumption (r = 0.38). All the above correlation coefficients were statistically significant (P < .05). Intervention dose was not associated with changes in fat intake. CONCLUSIONS: It appears that longer, interactive intervention efforts (contests and classes) resulted in more positive outcomes than did one-time activities (such as the kickoffs) or more passive efforts (use of printed materials). There is a need for studies designed to test worksite- and community-based nutrition intervention methods.

Stages of change in adopting healthy diets: fat, fiber, and correlates of nutrient intake.

The stages of change construct, which addresses the readiness to change, has only recently been applied to dietary behavior, such as fat consumption. This article describes the application of the stages of change construct to dietary fat and fiber consumption and examines the association of dietary stages to eating practices and related demographic and psychosocial factors in a large, geographically diverse population of workers. We present results from the baseline survey of 17,121 employees in the Working Well Trial. We assessed stage from an algorithm based on seven items and measured dietary intake with an 88-item food frequency questionnaire. Findings indicated that a greater proportion of the population has actively tried to reduce fat intake than to consume more fiber. Stage of change was associated with fat, fiber, and fruit and vegetable intake in a stepwise manner, as predicted. In multivariate analyses that controlled for demographic characteristics, stage of change predicted between 8 and 13% of the variance in dietary intake, and more than demographic variables. These findings have implications for the design of nutrition interventions and for the evaluation of intermediate outcomes.

Registries and administrative data: organization and accuracy.

In this article the organization and accuracy of the population registry and administrative data base in Manitoba, Canada are discussed. The overall data management strategy and a framework for analyzing the accuracy of such data are presented. The generally high quality of the Manitoba registry file (necessary to track individuals over time) and the hospital claims is supported by comparisons with other data sources. Hospital claims' main quality problems concern the reliability of certain secondary diagnoses and the level of aggregation necessary for reasonable agreement with other data collection methods (such as chart reviews). Finally, some of the research possibilities associated with population registries and administrative data are outlined.