Intensity of religious belief moderates the effects of workplace health promotion programs among older workers

Objective: To evaluate whether intensity of religious belief moderates the effects of workplace health promotion programs on behavioral stage of change among older adults.

Design: University employees ≥40 years of age (n = 423) were assigned to an intervention involving periodic interaction with a health educator (COACH), a self-regulated Internet-based intervention (RealAge), or control and followed for up to 12 months.

Main Outcome Measures: Religious beliefs were measured prior to randomization. Stage of change for six health behaviors was measured at baseline and 6 and 12 months thereafter. Ordinal logistic regression was used to model improvement in stage of change for each behavior as a function of interactions involving the assigned intervention, time, and intensity of religious belief.

Results: Among participants who reported religion to be somewhat important, those assigned to COACH were more likely to experience 12-month improvements in stage of change for fat intake (OR = 2.73, p = 0.028) and fruit/vegetable intake (OR = 2.85, p = 0.037) than those assigned to RealAge. Among participants who reported moderate use of religious beliefs to cope with stress, those assigned to COACH were more likely to experience 12-month improvements in stage of change for weight control (OR = 2.63, p = 0.020), fat intake (OR = 2.10, p = 0.042), fruit/vegetable intake (OR = 2.79, p = 0.014), and overall diet (OR = 2.51, p = 0.022) than those assigned to RealAge. Similar results were obtained for other religious belief measures. Adjustment for potential confounders did not alter these findings.

Conclusions: Older workers with moderately strong religious beliefs appear to benefit more from workplace health promotion programs that are interactive than those that are self-regulated. Our findings may reflect the complex interplay of unmeasured correlates of religious beliefs. Individuals who are highly religious may be externally motivated, which would favor interactive programs, but also unreceptive to information provided by non-believers, which would counter the effects of interactive programs that are not faith specific. Conversely, individuals who espouse minimal or no religious beliefs may be internally motivated, which would favor self-regulated programs, but also open to information provided by social group outsiders, which would favor interactive programs. Persons who are moderately religious may exhibit high levels of openness and externality, which could explain their differential response to interactive programs.