A randomized controlled trial of the effect of the start (strategies for relatives) intervention: The effect on family caregiver’s potentially harmful behavior to people with dementia

Abstract

Elder abuse may be defined as violating a vulnerable older person's rights and includes behavior, such as shouting, which people may not realise is abusive. People with dementia are particularly at risk. Family caregivers are more likely to act harmfully if depressed or anxious. There are no evidence-based effective interventions. We hypothesized, that START (STrAtegies for RelaTives, a randomized controlled trial (RCT) of coping strategies for family caregivers of people with dementia, which reduced short and longer term affective symptoms (8 and 24 months), would also reduce harmful behavior towards care recipients who lived at home compared to treatment as usual (TAU). We interviewed caregivers at baseline, 4, 8, 12 and 24 months. Our outcome was the Modified Conflict Tactic Scale (MCTS) which has ten potentially harmful behaviors (scored as harmful behavior [≥2 on any item] or as a continuous scale) ranging from shouting, to threatening, to slapping. We used random effects models allowing for therapist clustering and repeated measures which adjusted for baseline MCTS, burden and neuropsychiatric symptoms; centre, patient age and sex, to consider differences in potentially abusive behavior between randomized groups. We randomized 173 caregivers to START and 87 to TAU. 38 (43.7%) from TAU and 82 (47.7%) START group scored ≥2 on MCTS at randomisation. There was no significant difference in the proportion with MCTS score ≥2 between randomization groups over two years (OR 0.65 (95% Confidence Interval 0.31,1.36) p=0.253). Short-term (4 and 8 months) those receiving the intervention reported less abuse (OR 0.47 (0.16,1.36), p=0.16) but not-significantly so. Our sensitivity analysis using repeated measures in those who were followed up at both 4 and 8 months showed those in the intervention group had lower scores (F= 5.723, p=.018). This is the first RCT with abusive behaviour in dementia as an outcome. Although in our main analysis there was no significant decrease in abusive behaviour, in our sensitivity analysis there was in those who remained in the study over 8months. The study was not powered on this outcome but these preliminary results suggest further development is warranted.