O-19 Effectiveness of an inpatient stroke unit in a resource limited setting: Findings from the Barbados National Registry

Methods: We used data from the Barbados National Reg istry for Non-communicable Diseases (BNR), for all diag nosed cases of acute stroke. We described the characteristics of those admitted to the Queen Elizabeth Hospital (reach). We examined clinical effectiveness using in-hospital case fatality rates. We assessed the adoption of stroke unit thera pies by examining the factors associated with (occupational (OT), physiotherapy (PT) or speech therapy (ST)) using multivariable logistic regression models.

Results: Of the 2,491 strokes registered from 2016 to 2020, 38% were admitted to the stroke unit (SU). Admission rates ranged from 32% (2020) to 42% (p=0.012. Persons admit ted to the SU tended to be younger (by 5 years), male, and had similar proportions of ischemic stroke, hypertension and diabetes compared to those on other wards. After adjust ing for age, sex, Glasgow Coma Scale, co-morbidities, per sons admitted to stroke unit were more likely to receive OT (Odds Ratio: 11.1 95% CI (8.0, 15.4)), PT (OR: 5.2 95% CI (3.3, 8.2)) and ST (OR: 6.6 95% CI (4.8, 9.1)) compared to those admitted to other wards. There was a non-statisti cally significant reduction in hospital case fatality for those admitted to SU versus those on other wards (OR: 0.77 95% CI (0.56, 1.06)).

Conclusion: We found SU admission led to more rehabilita tive care and fewer deaths but less than half of eligible acute stroke patients were admitted. The admission reduction seen in 2020 was likely due to pandemic induced resource limitations