Correlates of compliance with measurement protocols in a Latino nutrition-intervention study.

It is well known that attrition and noncompliance in longitudinal health intervention studies contribute to bias in both internal and external validity. However, little is known about rates and correlates of compliance with measurement protocols among Latinos. This article investigates correlates of compliance with follow-up physical measurement protocols among Latino subjects in a nutrition-oriented cardiovascular disease prevention program targeting low-English literate adults. Correlates of compliance, measured at baseline, included four classes of variables: demographic characteristics, physical measures, health behaviors, and nutrition-related psychological variables. Subjects were categorized into one of three compliance groups: on-time compliers, late or "reluctant" compliers, and noncompliers. Approximately 36% of subjects complied on time, 25% complied late, and 39% did not comply. Analyses showed that, relative to on-time and late compliers, noncompliers tended to be male, younger, of lower income and Spanish-literacy level, to drink more alcohol, to be less physically active, and to have lower dietary fat avoidance scores. No significant differences were found for other factors considered (e.g., physical health measures). These results generally reflect those of non-Hispanics that indicate that individuals at greater "risk" are less likely to comply with study protocols. Such results may be useful for designing culturally appropriate cohort maintenance strategies for longitudinal studies with Latinos.