Helping caregivers of persons with dementia: which interventions work and how large are their effects?

Study Aim/Purpose: This meta-analysis sought to improve upon findings from extant evaluations of various interventions serving dementia caregivers by pooling the samples and analyses of intervention effects from studies that generally use small sample sizes and measuring more reliable mean effects. 

Summary of Methods: Using a keyword search of electronic databases (Psycinfo, Medline, Ageline, and Psyndex) and strict inclusion criteria, 127 articles were identified on interventions helping caregivers of persons with dementia. The inclusion criteria were as follows: 1) the care receiver being studied had dementia; 2) the intervention was compared to a control condition that did not receive the intervention; 3) one of the following outcome domains were reported: caregiver burden, depressive symptoms, indicators of positive subjective well-being (e.g. life-satisfaction, happiness), knowledge and/or coping abilities of the caregiver, symptoms or outcomes of the care receiver, and/or institutionalization; 4) statistics could be converted into effect sizes; and 5) the study was written in English or German (or in two cases another language for which the authors could get translation). 

The authors compared interventions that had been evaluated in five or more controlled studies and categorized according to the dominant component. If no dominant component was identified, the category for these interventions was labeled “multicomponent.” The dominant components identified were psychoeducational interventions, cognitive-behavioral therapy (cbt), respite, counseling/case management, general support, and training of the care receiver. 

The authors reviewed 127 studies conducting statistical integration of the studies’ findings using random-effects models to compute average effects for most of the outcomes and a log odds ratio was used to compute the relative odds of the care receiver being institutionalized after exposure to the intervention. These analyses were conducted for each outcome across the types of interventions studied and for each type of intervention with a separate analysis for multi-component interventions. Analyses were also conducted of the studies’ characteristics to see how that may have affected findings. 

Summary of Key Results: The authors’ outcome specific analysis across all interventions found that the interventions had statistically significant, but small immediate effects on caregiver burden, depressive symptoms, subjective well-being, abilities/knowledge of the caregiver, and symptoms of the care receiver. Studies with follow-ups with an average time lapse of 11 months showed significant positive effect on caregiver burden, depressive symptoms and ability/ knowledge. On the other hand, no significant effect of intervention was found on the risk of institutionalization. When the effect sizes were analyzed for specific kinds of interventions, the analysis showed that psychoeducational intervention had significant effects on all outcomes except institutionalization, with those requiring active participation of the caregivers (e.g. to practice or role-play what they learned) having the broadest effects. Cognitive behavioral therapy (CBT) had a small to moderate positive impact on caregiver burden and a large positive effect on caregiver depression, though the authors note that only one CBT study was available on the impact of CBT on caregiver depression. Counseling showed significant and moderately sized effects on burden, while supportive interventions (e.g. with group support groups) improved subjective well-being. Training (e.g., on improving memory) of the care receiver had small significant effects on the symptoms of the care receiver. Studies on respite interventions also had small significant effects, specifically on caregiver burden, depression and subjective well-being. “Structured” multi-component interventions did not show significant effects on most of the outcomes, but were significantly related to delayed institutionalization for the care receiver. The authors also found that longer interventions were more likely to improve depression and to decrease the risk of institutionalization. 

Comparing effect sizes by study characteristics, the authors found that studies with a higher percentage of caregiver women in the sample demonstrated greater improvements in knowledge/abilities, depression, and a greater delay of institutionalization, but less improvement in subjective well-being.

Limitations of Studies Reviewed: The authors note that there is insufficient research on sustained or long-term consequences of caregiver interventions. Additionally, they note that few studies focus on the positive aspects of caregiving, such as caregiver well-being or finding benefits in the caregiving role. They also note the lack of studies on mediators that may be causing the observed effects of the interventions, such as caregivers’ beliefs, coping strategies, or a reduction in the amount of care they need to provide. 

Authors Discussion/Conclusions: Overall, the authors conclude that given their findings of small but meaningful effects of caregiver interventions, “there is a continued need for improvement in the quality of the interventions.” They recommend future long-term impact research as well as studies addressing the methodological limitations of the studies reviewed. The authors also stress the need for future research on individual differences in caregivers’ response to particular interventions. Finally, they suggest that more research is needed in multi-component interventions. Specifically, they recommend research to test whether a combination of individual interventions that their meta-analysis found effective, such as education or CBT with respite, would produce stronger positive effects on caregiver and care receiver outcomes or whether interventions with fewer components are better received by caregivers and therefore more effective.