Sociodemographic structure and health care-related outcomes of community-dwelling dementia caregiving dyads: a latent class replication study

Background: The identification of dyadic subgroups of individuals living with dementia and their informal caregivers can help to design effective tailored support. In a previous German study, we identified six dementia dyad subgroups by applying Latent Class Analysis (LCA). Results showed sociodemographic heterogeneity as well as differences in health care outcomes (i.e., quality of life, health status, caregiver burden) between subgroups. The objective of this study is to determine if the dyad subgroups from the previous analysis can be replicated in a similar but distinct Dutch sample. Methods: A LCA 3-step procedure was applied to baseline data from the COMPAS study, a prospective cohort study. LCA is a statistical approach used to identify heterogeneous subgroups within populations based on their pattern of answers on a set of categorical variables. Data comprises 509 community-living individuals with predominantly mild to moderate dementia and their informal caregivers. A narrative analysis was used to compare latent class structures of the original versus the replication study. Results: Six distinct dementia dyad subgroups were identified: A subgroup of “adult–child-parent relation with younger informal caregiver” (31.8%), a “couple with female informal caregiver of older age” group (23.1%), an “adult–child-parent relation with middle-aged informal caregiver” group (14.2%), a “couple with middle-aged female informal caregiver” group (12.4%), a “couple with older male informal caregiver” group (11.2%) and a “couple with middle-aged male informal caregiver” group (7.4%). Quality of life of individuals with dementia was rated better in couples than in adult–child-relationships. Worst health for caregivers was reported by subgroups with female informal caregivers living together with male individuals with dementia in couple relationships. A subgroup with older female informal caregivers in couple relationships report the most severe burden on physical and mental health. In both studies, a model with six subgroups fitted the data best. Although substantive similarities between the subgroups of both studies appeared, considerable differences are also evident. Conclusion: This replication study confirmed the existence of informal dementia dyad subgroups. The observed differences between the subgroups provide useful contributions for a more tailored health care services for informal caregivers and individuals living with dementia. Furthermore, it underlines the relevance of dyadic perspectives. To facilitate replication studies and increase the validity of evidence, a standardization of collected data across studies would be beneficial.