Do ward changes affect outcomes differently in people living with dementia?

BACKGROUND: Ward changes during hospital admissions are associated with poorer outcomes, but their impact on people living with dementia is unknown. OBJECTIVE: To examine whether individuals with dementia are more adversely affected by ward changes. SETTING: Four hospitals within a UK NHS Trust. SUBJECTS: Individuals aged ≥65 years. METHODS: This service evaluation included data from 01/2023-02/2024. Generalised estimating equations were used to fit linear and logistic regression models. Associations between ward changes, dementia status and their interaction on three outcomes-length of stay (LOS), inpatient mortality and discharged to a care home-were examined. Fully adjusted models accounted for demographic, socioeconomic factors and hospital site. RESULTS: 27,140 admissions, 19,392 unique individuals (2760 with dementia) were included. Mean age at first admission 79.0 (SD 8.1). In the fully adjusted LOS model, both ward changes (β = 5.2, P < .001) and the interaction with dementia (β = 1.7, P < .001) were associated with longer LOS. In the fully adjusted mortality model, dementia was associated with increased risk of mortality (OR = 1.4, P = .013) but there was no interaction effect of ward changes and dementia. In the fully adjusted care home admission model, dementia (OR = 4.4, P < .001) and ward changes (OR = 1.3, P < .001) were associated with increased risk, without evidence of interaction. CONCLUSIONS: Our results suggest that ward changes disproportionately affect LOS, but not mortality or discharge destination, in people living with dementia. Minimising ward transfers may improve outcomes for all older adults but is particularly important in dementia care to reduce the risk of extended LOS and potential associated inpatient harm.