Objectives

We performed a randomized controlled study to evaluate the effects of caregiver training on the well-being of both people with dementia and their caregivers. Before the effect analysis, we conducted a process evaluation to estimate internal and external validity. This was anticipated to augment our understanding of the outcomes.

Methods

We focused on three questions. (a) Was the intervention performed as planned (internal validity)? (b) Can qualitative data be used to inform how the intervention evoked change? (c) Can the study outcomes be extrapolated to all caregivers living with people who have dementia (external validity)?

Results

Responses from participants assigned to the intervention group suggested that the intervention was feasible, could be performed as planned, and that modelling and discussions between participants were important. However, participant recruitment to the entire study was ultimately laborious because participants had issues with the study design (risk of being assigned to the control group) and referrers lacked familiarity with the training (new type of intervention). Participants were also younger and better educated compared with the general population. Some dropouts in the follow-up period occurred due to the number of questionnaires, and this was more pronounced in the control group.

Conclusions

Although we achieved high internal validity, we lack certainty about the external validity. We not only experienced general difficulty in recruiting participants but also tended to recruit a biased sample that was relatively young and well educated. These factors combine to limit our ability to extrapolate the results to the general population.

© 2020 The Authors. International Journal of Geriatric Psychiatry published by John Wiley & Sons Ltd.