Heterogeneity in costs and prognosis for acute ischemic stroke treatment by comorbidities

Objective Comorbidities are prevalent among stroke patients. The current study assesses the variations in cost and stroke prognosis by concurrent comorbidities in patients with acute ischemic stroke. Methods The Charlson comorbidity index was used as the composite comorbidity level (0 none, 1 mild, 2 moderate, and ≥ 3 severe). Outcomes included modified Rankin Scale (mRS) at 3 months and 1-year mortality and stroke recurrence. We utilized a multivariate log-normal model for cost, a proportional Cox hazards model for outcomes, and a decision analytic model for the excess cost per unit change in outcome probability compared with the no-comorbidity group. Results A total of 3605 consecutive patients were enrolled. At 3 months, the severe comorbidity group was 0.32 times less likely to have mRS ≤ 2, and were 4.86 times more likely to die from stroke than the no-comorbidity group. Within 1 year, the severe comorbidity group showed 10.36 and 3.38 times higher likelihoods of death from stroke and stroke recurrence than the no-comorbidity group. The incremental cost was 4376 in 3 months and 7074 USD in 1 year for the severe comorbidity group, and 985 in 3 months and 1265 USD in 1 year for the mild comorbidity group compared to the no-comorbidity group. Conclusion The excess cost per unit increase of a short-term good prognosis was largest for the severe comorbidity group. Patients with severe comorbidities showed poor prognosis and large health expenditure. Assessing comorbidity level is crucial for better prediction of outcomes and excess cost.