The purpose of this study was to evaluate the effects of a revised wor
ksite health promotion program that featured an employee steering comm
ittee/menu approach to intervention. The ''Take Heart II'' program was
evaluated using a quasi-experimental matched-pair design with worksit
e as the unit of analysis. Experimental and control worksites did not
differ on baseline organizational or employee demographic variables or
on baseline levels of dependent variables. Outcome and process result
s revealed consistent, but modest effects favoring intervention worksi
tes on most measures. Cross-sectional analyses generally failed to pro
duce statistically significant intervention effects, but cohort analys
es revealed significant beneficial effects of the Take Heart II interv
ention on eating patterns, behavior change attempts, and perceived soc
ial support. Neither analysis detected a beneficial effect of interven
tion on cholesterol levels.