Trends in Emergent Interfacility Ischemic Stroke Transfers, 2016-2022.

Emergent ischemic stroke transfers increased from 2016 to 2019 and then declined through 2022, reversing a prior trend of continual increase.

• Estimated nationwide emergent transfers increased from 21,221 (12% of admissions) in 2016 to 26,720 (15%) in 2019.
• Transfers then decreased to 21,902 (12% of admissions) by 2022.
• The study period covered 2016 to 2022 using Medicare fee-for-service claims.

Among 824,551 Medicare FFS beneficiaries admitted for ischemic stroke, 12% underwent emergent interfacility transfer.

• Total cohort: 824,551 FFS beneficiaries (57% female), age 65 or older, admitted for ischemic stroke from 2016 to 2022.
• 99,751 (12%) underwent emergent transfer; 724,800 (88%) were not transferred.
• County-level Medicare Advantage penetration rates were used to extrapolate results to nationwide estimates.

Transferred patients were younger and received reperfusion therapy at substantially higher rates than nontransferred patients.

• Mean age of transferred patients was 78.2 years vs. 80.1 years for nontransferred patients.
• Thrombectomy rate: 15.9% in the transferred population vs. 2.8% in the nontransferred population.
• Thrombolysis rate: 24.1% in the transferred population vs. 10.0% in the nontransferred population.

Transferred patients were far more likely to present initially to rural, small, or critical access hospitals compared to nontransferred patients.

• Rural presenting facility: 55.3% of transferred vs. 22.6% of nontransferred patients.
• Presenting facility with fewer than 100 beds: 56.6% of transferred vs. 11.9% of nontransferred patients.
• Critical access hospital presentation: 24.9% of transferred vs. 2.8% of nontransferred patients.

Nontransferred patients more frequently presented to teaching facilities, stroke centers, and vertically integrated facilities.

• Teaching facility presentation: 66.2% of nontransferred vs. 32.3% of transferred patients.
• Stroke center presentation: 83.2% of nontransferred vs. 51.9% of transferred patients.
• Vertically integrated facility presentation: 93.5% of nontransferred vs. 76.5% of transferred patients.

Presentation to a critical access hospital was the strongest facility-level predictor of emergent transfer in multilevel logistic regression.

• Critical access hospital presentation: OR 2.8 (95% CI 2.4–3.3).
• Prestroke hemiplegia was also strongly associated with transfer: OR 2.3 (95% CI 2.3–2.4).
• Prestroke cerebrovascular disease was associated with transfer: OR 2.2 (95% CI 2.2–2.3).

Presentation to a large hospital (more than 400 beds) was the strongest factor associated with not being transferred.

• Presentation to a facility with >400 beds (reference ≤100 beds): OR 0.01 (95% CI 0.01–0.01).
• Presentation to a stroke center: OR 0.4 (95% CI 0.4–0.5).
• Older age: OR 0.6 (95% CI 0.6–0.6); higher affluence (4th quartile of ADI): OR 0.6 (95% CI 0.5–0.7).

Less affluent area of residence was associated with being in the transferred stroke population.

• The 4th quartile of the Area Deprivation Index (most affluent) was associated with lower odds of transfer: OR 0.6 (95% CI 0.5–0.7).
• This suggests that patients from less affluent areas are more likely to undergo emergent interfacility transfer for ischemic stroke.
• Multilevel logistic regression was used to identify these associations.