Abstract
Background: Although nursing intervention studies typically focus on testing hypothesized differences between intervention and control groups, moderator variables can reveal for whom or under what circumstances an intervention may be most effective.
Objectives: The aim of the study was to explain and illustrate moderator effects using data from a nursing intervention study to improve cognitive abilities in those with a chronic health condition.
Methods: The sample consisted of 178 individuals with multiple sclerosis participating in an experimental study of a cognitive intervention. General linear models were used for analyses. Interaction terms were created to represent moderator effects on three outcomes: self-reported cognitive abilities, use of memory strategies, and verbal memory performance.
Results: The Charlson comorbidity index significantly moderated the intervention effect on self-perceived cognitive abilities. Years of education significantly moderated the intervention effect on use of memory strategies. Scores on a general self-efficacy measure significantly moderated the intervention effect on the Controlled Verbal Learning Test-Second Edition.
Discussion: These analyses highlight the key role that moderator effects can play in nursing research. Although random assignment to groups can control potentially biasing effects of extraneous differences among individuals in intervention and control groups, those very differences may suggest fruitful avenues for hypothesis generating research about what works best for whom in intervention studies.