Racial Disparities in the Development of Dysphagia After Stroke: Analysis of the California (MIRCal) and New York (SPARCS) Inpatient Databases

Objectives: To determine whether the proportion of patients with stroke experiencing dysphagia differs among racial groups and whether this relation can be explained by stroke type or severity.

Design: Case-control study using California's Medical Information Reporting and New York's Statewide Planning and Research Cooperative System databases for 2002. Cases had primary diagnosis of cerebrovascular disease (International Classification of Disease, 9th Revision [ICD-9] codes 430-438.9, excluding transient [435-435.9] and late-effects [438-438.9]), and self-identified race was white, black, or Asian. Two comparison groups were selected: (1) Parkinson's disease (ICD-9 codes 332-332.1) and (2) oral cancer (ICD-9 codes 141-149).

Setting: Inpatient admissions in the respective states.

Participants: Cases with primary diagnosis of cerebrovascular disease whose self-identified race was white, black, or Asian.

Interventions: Not applicable.

Main Outcome Measure: Dysphagia, defined by ICD-9 codes 787.2 (dysphagia), 507.0 (aspiration pneumonia), or presence of a feeding tube in the absence of a diagnosis of coma (Current Procedural Terminology codes 432.46 or 437.50 without ICD-9 code 780.01).

Results: In the stroke group, the adjusted odds ratio (OR) with 95% confidence interval (CI) for dysphagia was significantly higher for Asians than whites in New York (OR=1.64; 95% CI, 1.50-1.79) and California (OR=1.69; 95% CI, 1.34-2.13). The adjusted OR was slightly but significantly higher for blacks than whites in New York (OR=1.15; 95% CI, 1.03-1.28), but not in California (OR=1.08; 95% CI, 0.97-1.19). No statistically significant differences among racial groups were found in patients with Parkinson's disease or oral cancer. Other factors strongly associated with dysphagia included hemiplegia (OR=2.19; 95% CI, 2.07-2.32) and aphasia (OR=1.97; 95% CI, 1.83-2.11).

Conclusions: Asians were more likely to have dysphagia after stroke. This association was statistically significant after adjusting for age, sex, stroke severity indicators, comorbidities, and stroke type.