This article, titled “Theory of planned behavior and adherence in chronic illness: a meta-analysis,” by Antonia Rich and colleagues, presents the first meta-analysis specifically applying the Theory of Planned Behavior (TPB) to adherence behaviors in individuals suffering from chronic conditions. The authors emphasize that existing reviews typically do not differentiate adherence behavior from other health behaviors, nor do they specifically focus on adults with chronic illness. This study addresses a crucial gap, as adherence in chronic illness involves long-term behavior to manage a disease, with serious consequences for non-adherence, unlike general health behaviors.
The primary aim of the meta-analysis was to examine whether the Theory of Planned Behavior (TPB) can predict adherence in people diagnosed with a chronic condition. The researchers conducted a meta-analysis of 27 studies that met their inclusion criteria, which involved studies measuring TPB or Theory of Reasoned Action constructs with participants suffering from a chronic disease, examining adherence to treatment, and having adherence as an outcome measure. They used random-effects meta-analysis to compute averaged intercorrelations among theory variables, corrected for sampling error, and path analysis to test theory hypotheses and moderator effects.
Key findings from the study include:
- Predictive Power: The TPB explained 33% of the variance in intention and 9% of the variance in adherence behavior. This finding, while supporting theory predictions, indicates that the effect sizes were generally small, especially for the intention-behavior relationship.
- Relationships between Constructs: Consistent with the TPB, attitudes, subjective norms, and perceived behavioral control were statistically significant predictors of adherence intentions. Perceived behavioral control had the strongest relationship with intention (r+ = 0.51), followed by attitude (r+ = 0.41) and subjective norm (r+ = 0.32).
- Intention-Behavior Link: Intention was found to be a statistically significant predictor of behavior (b = 0.21). The analysis also supported the role of intention as a mediator for the effects of attitudes, subjective norms, and perceived behavioral control on behavior.
- Direct Effect of Perceived Behavioral Control: A statistically significant direct effect of perceived behavioral control on behavior was also observed (b = 0.13), contributing to its total effect on behavior.
- Moderation Analyses: The study investigated behavior type (medication, exercise, diet, self-care) and type of adherence measure (self-report vs. objective) as potential moderators. No significant differences were found in the strength of relationships across different adherence behaviors or measurement types, suggesting consistency in the TPB’s pattern of effects, although the magnitude might vary. Moderation by age, gender, and study design could not be comprehensively examined due to insufficient data.
- Comparison to Previous Research: The levels of prediction for both intention and behavior were lower than those found in previous TPB meta-analyses focusing on general health behaviors in mostly healthy, younger populations (e.g., 39–44% variance in intention and 19–27% in health behaviors). The authors suggest this difference may be due to the unique characteristics of chronically ill patients and the complex, long-term nature of adherence behaviors.
The article acknowledges the persistent “intention-behavior gap,” where the model is more effective in predicting intention than actual behavior, particularly given the prolonged nature of adherence behaviors in chronic illness. The study also highlights the considerable heterogeneity in effect sizes across studies, which limits the precision of the meta-analytic findings and indicates that other factors not accounted for may influence adherence. The authors suggest that future research could integrate additional factors known to influence adherence, such as patients’ beliefs about treatment necessity and concerns, or explore alternative theoretical models like the COM-B model. Despite limitations, the study concludes that while the TPB contributes to understanding adherence, focusing solely on its variables for interventions may be insufficient, underscoring the need for further robust experimental research to improve adherence outcomes.
Reference:Rich, A., Brandes, K., Mullan, B., & Hagger, M. S. (2015). Theory of planned behavior and adherence in chronic illness: a meta-analysis. Journal of Behavioral Medicine, 38(4), 673-688. https://doi.org/10.1007/s10865-015-9644-3
