Utilizing a novel approach of empirically derived hospital regions from Medicare claims data and a risk adjustment model calibrated using CARES registry data, we identified significant regional disparities for survival to discharge rates for OHCA. Across all regions, we identified a mean SIR for survival to hospital discharge of 0.91 ± 0.48, indicating that on average, regions had slightly lower survival rates than expected. We also noted considerable variability in substantial standard deviation among regions (SIRs ranging from 0.40 to 4.00). Some regions clustered near SIR = 1.0 demonstrated disproportionately wide confidence intervals. This likely reflects statistical noise from low case volumes and sparse predicted survival counts. This highlights the challenge of interpreting regional performance in low-volume settings. In the multiple logistic regression model examining hospital-level predictors of regional overperformance, we note that larger hospital bed capacity was strongly associated with higher odds of being in an overperforming region. Compared to hospitals with fewer than 100 beds, those with 100–399 beds had nearly three times the odds of overperformance, and hospitals with 400 or more beds had almost four times the odds. Conversely, hospitals with cardiac catheterization capability had significantly lower odds of being in overperforming regions, suggesting that while such capabilities are critical for post-arrest care, they may not alone drive regional performance. Minor teaching hospitals also showed lower odds of overperformance, though this finding did not reach statistical significance. Collectively, these results indicate that hospital capacity may be a key structural factor in regional OHCA outcomes, while the presence of specific interventions, such as catheterization labs, may not be sufficient on their own to improve performance at the regional level.
By performance metrics, we noted half of the analyzed regions (101 out of 205) significantly underperformed in terms of survival predictions, while only 4.0% (9 regions) overperformed. By population and demographic factors, we noted that overperforming regions tended to have smaller populations (mean 220,054 vs. 2,880,034 in underperforming regions) and a higher proportion of residents aged 65 and older (17.0% vs. 14.0%). Notably, racial and ethnic composition differed between high and low-performing regions, with overperforming areas having a higher proportion of White residents (83.9% vs. 72.9%) and lower proportions of Black or African American residents (10.8% vs. 17.3%) and Hispanic (5.8% vs. 18.2%) residents. By hospital characteristics, we noted that regions with better OHCA outcomes were associated with a higher proportion of larger hospitals (400+ beds) and major teaching hospitals. These findings highlight the complex interplay of demographic, socioeconomic, and healthcare system factors influencing OHCA outcomes. The marked regional disparities underscore the need for a multifaceted approach to improving OHCA survival rates across the nation through regionalized care approaches, further aligning with proposed AHA strategies.
A comprehensive body of research has previously identified significant regional variation in OHCA survival and outcomes. A landmark study by Nichol et al. from 2008 using the Resuscitation Outcomes Consortium (ROC) Epistry-Cardiac Arrest dataset from 8 US and 3 Canadian sites found significant variation in survival to discharge, ranging from 3.0 to 16.2%13. The study further noted substantial disparities in survival rates, OHCA incidence, bystander CPR rates, and EMS response times across regions. Even after adjusting for patient and clinical factors, these site-to-site variations persisted, highlighting the complex interplay of factors influencing OHCA outcomes. In a more contemporary study from 2016, Girotra et al., using the Cardiac Arrest Registry to Enhance Survival (CARES) database, analyzed regional variations in OHCA outcomes across the United States that included 132 U.S. counties from 2011 to 20143. Their findings demonstrated significant disparities in county-level survival rates to discharge, ranging from 3.4% to 22.0%. This nearly sevenfold difference in survival rates across counties was striking and persisted even after adjusting for patient and community characteristics. There were also significant variations in rates of bystander CPR and AED use, ranging from 10% to 64% and 2% to 15%, respectively, across counties. Importantly, Girotra et al. identified that county-level factors accounted for 40.4% of the observed variation in survival, highlighting the crucial role of local and regional factors in determining OHCA outcomes. These factors included not only differences in emergency medical services but also broader community characteristics, such as population demographics, education levels, and socioeconomic status. This study further emphasized the importance of understanding local and regional factors that contribute to OHCA outcomes and suggested that targeted interventions at the county level could potentially improve survival rates.
Our examination of weighted socioeconomic metrics underscores additional nuances. While underperforming regions had marginally lower poverty rates (11.86 vs. 12.31%) and a slightly higher median household income ($59,259 USD vs. $46,430 USD), overperforming regions had a higher proportion of publicly insured residents (34.9 vs. 31.9%), lower unemployment (3.9 vs. 5.4%), and a greater share of individuals with at least a high-school education (24.3 vs. 21.0%). These patterns suggest that income alone may not fully account for differences in OHCA outcomes, and that public insurance coverage, employment, and educational attainment could be important drivers of regional performance.
In the context of this prior research, our study provides a novel and important contribution to understanding regional variations in OHCA outcomes. Using SIRs derived from Medicare claims data and a risk-adjustment model based on registry data, we offer a more comprehensive analysis of regional disparities in OHCA survival that captures all US geographies. We analyzed 202,406 unique beneficiary-level OHCA claims across 205 empirically derived hospital regions, finding SIRs ranging from 0.40 to 4.00. This suggests that regional variations in OHCA outcomes may be even more pronounced than previously recognized, especially when focusing on the Medicare population. This variability in SIRs underscores the critical need for targeted interventions and policy changes to address these persistent disparities. Notably, 49.3% of regions (101 out of 205) significantly underperformed in survival predictions, while only 4.4% (9 regions) overperformed, highlighting the need to further explore the factors contributing to these differences. Furthermore, our study’s focus on Medicare beneficiaries provides crucial insights into OHCA outcomes among older adults, a more vulnerable population that is often underrepresented in prior research. By examining hospital characteristics, such as bed size and teaching status in relation to OHCA outcomes, our study also provides insight into the role of hospital-level factors in regional variations, an aspect not fully explored in previous research. We found that overperforming regions were associated with a higher proportion of larger hospitals (400+ beds) and major teaching hospitals, suggesting that access and timing to larger health systems and hospitals that can provide specialized care may play a crucial role in OHCA survival rates. Cardiac catheterization lab availability however did not show significant differences between overperforming and underperforming regions (p = 0.996). This suggests that while cardiac catheterization is a key intervention for specific post-arrest patients, its availability may not be the primary driver of regional differences in overall OHCA survival rates among Medicare beneficiaries. Additionally, our analysis of demographic factors revealed important disparities. Overperforming regions had a higher proportion of White residents (83.9 vs. 72.9% in underperforming regions) and lower proportions of Black/African American (10.8 vs. 17.3%) and Hispanic (5.8 vs. 18.2%) residents. These findings align with and extend previous research on racial and ethnic disparities in OHCA outcomes, emphasizing the need for targeted interventions in diverse communities.
A growing body of evidence supports the concept of CRCs, where specialized post–cardiac arrest care is coordinated regionally. Several studies and a meta-analysis have reported associations between CRC treatment and improved survival or neurological outcomes19. The AHA similarly endorses the benefits of CRCs for OHCA, noting that both direct transport and early interfacility transfer to specialized centers can reduce mortality and enhance neurological recovery20,21. Our results—demonstrating regions with higher-performing hospitals and more robust care structures tend to have better survival—reinforce calls for regionalization efforts that build on the CRC model. Fostering collaborative networks, promoting interfacility transfer when appropriate, and aligning financial or regulatory incentives could drive improvements in OHCA care at the regional and national levels.
Our study provides a comprehensive, nationwide assessment of OHCA outcomes among Medicare beneficiaries, offering valuable insights into regional disparities, demographic factors, and hospital characteristics that influence survival rates. Unlike prior studies that relied on registry data from selected hospital systems, our study leverages nationally representative Medicare claims data to examine risk-adjusted OHCA outcomes across empirically derived regions—offering a novel, more generalizable approach to evaluating geographic disparities in survival. These findings can inform policy decisions, resource allocation, and future research aimed at improving OHCA outcomes and reducing disparities across different regions in the United States.
This study’s findings have several important limitations. Medicare administrative claims data provide a broad national perspective but lack the sensitivity and specificity of registry data for identifying OHCA cases. Our findings are limited to Medicare beneficiaries aged 65 and older, limiting generalizability to younger populations with potentially different risk factors and outcomes. Patients included in our analysis must survive to reach the emergency department, possibly inflating our reported survival rate (14.9%) compared to registry cohorts. Furthermore, the predominantly White beneficiary composition of our sample (77.6%) restricts generalizability to more diverse populations. Administrative claims also lack detailed clinical and pre-hospital information, preventing deeper insights into care quality and specific interventions. Our hospital region definitions, though empirically derived, may not exactly match real-world care delivery patterns for OHCA. While the study relies on data from 2013–2015, national guidelines regarding OHCA regionalization have remained constant, thus preserving the relevance of our findings to current system planning. Additionally, due to constraints imposed by probabilistic matching without hospital identifiers, only a subset of registry data could be confidently linked to CMS claims. This necessary exclusion of unmatched encounters may limit the representativeness of the matched cohort and introduce potential selection bias. Despite these limitations, this study’s nationwide scope and large Medicare cohort offer valuable insights into regional variability in OHCA outcomes and highlight potential disparities. Using Medicare claims data in the absence of a comprehensive national OHCA registry remains essential for understanding regional variations in survival.
In conclusion, our study provides a comprehensive nationwide assessment of OHCA outcomes among Medicare beneficiaries, offering valuable insights into regional disparities, demographic factors, and hospital characteristics that influence survival rates. Our findings reveal significant variations in OHCA outcomes across the United States, underscoring the critical need for targeted interventions and policy changes to address these persistent disparities. Moving forward, a multifaceted approach that addresses pre-hospital care, hospital capabilities, and community-level factors will be crucial to narrowing the gap in OHCA outcomes and improving patient outcomes.
