Trajectories of Long-term Care after Stroke in Sweden: Nationwide Study Exploring Patterns and Determinants

Stroke is a leading cause of mortality and disability worldwide. Although studies have primarily focused on health and functioning among stroke survivors, there is limited research on longitudinal patterns of long-term care use among older adults with stroke. This study explores long-term care trajectories among older men and women with stroke in the Swedish population. Nationwide prospective cohort study. All individuals aged ≥70 years hospitalized with a first stroke in 2015-2017 followed in the Swedish population registers for 3 years. Care trajectories among patients with strokes were visualized and compared to trajectories in 2 control groups: (1) same-aged peers randomly drawn from the general population and (2) older adults with health and sociodemographic characteristics similar to patients with strokes but without stroke identified through propensity score matching. Multivariable Cox regression and multistate models were used to identify determinants of mortality and care trajectories among patients with strokes. We identified 31,560 individuals with stroke (mean age 82.9 years). Already before the stroke, these individuals had a higher comorbidity burden and received more long-term care than their same-aged peers. Compared with both control groups, patients with strokes were also more likely to enter long-term care. However, 38% of patients with strokes survived for 3 years without taking up long-term care. Sex, income, cohabitation, and comorbidities were associated with care trajectories. Care status was a more robust predictor of mortality after stroke than the established Charlson comorbidity index. Experiencing a stroke increases both mortality and long-term care utilization, and once formal long-term care is provided, exceedingly few patients with strokes return to living without care. However, a considerable part of care utilization and mortality after stroke is concentrated among those with preexisting care needs. Prestroke care utilization should thus be considered in future studies exploring stroke prognosis.