EUVIMEDEUVIMED entdecken
Crossref

Abstract 9293: Modeling Contemporary Age Related Stroke Risk in a Community Based Population

Malak Elokour, Sarah Dias, Ada Tadeo, Alexis Simpkins

Circulation · 02.11.2026 · 154 (Suppl_1)

DOI: 10.1161/circ.154.suppl_1.9293 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Circulation, erschienen am 02.11.2026. Verfasst von Malak Elokour und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Acute ischemic stroke (AIS), the primary cause of permanent disability in adults, is rising amongst young adults. To assess the possible impact of newly available classes of stroke prevention medications and guidelines across age groups, particularly young AIS, we conducted cluster-based modeling using Latent Class Analysis (LCA) in a large, contemporary, community-based population of adults. We hypothesized that LCA would cluster AIS patients more strongly by stroke risk factor (SRF) burden and access to healthcare and may identify a distinct SRF phenotype in younger individuals.

Methods

This cohort study included de-identified electronic health records of adults discharged with an ICD code for AIS between 2009 - 2026. Patients were stratified into age groups (18-45, 46-55, 56-65, 66-75, 76+). Variables included demographic variables, discharge disposition, pre-existing SRFs, and measures of quantitative stroke risk (blood pressure, peak A1C, peak LDL, body mass index (BMI), lipoprotein a (Lp(a)), primary care utilization prior to AIS admission, and healthcare access (insurance type and the Centers for Disease Control and Prevention's social vulnerability index (SVI)). Python was used to conduct bivariate analysis and LCA, using the Bayesian Information Criterion (BIC) and Silhouette score to identify distant SRF clusters/classes.

Results

Of the 15,969 patients included in the cohort, 49% were women, 21% aged 18-45, 19% aged 46-55, 24% aged 56-65, 20% aged 66-75, and 16% aged 76+. The median LDL was 89 mg/dL (IQR 65-118), peak A1c found in each medical record 5.8% (IQR 5.5%-6.6%), BMI 25.8 kg/m2 (IQR 22.6-29.7), Lp(a) 38 mg/dL (IQR 15 - 91), highest recorded systolic blood pressure prior to the AIS admission 169 mmHg (IQR 149 mmHg - 192 mmHg). The median NIHSS score was 2 (IQR 0-6), and 44% were discharged home after AIS admission. Most (74%) were insured by Medicare or private insurance, and 9.8% resided in high SIV areas. SRFs were prominent in all cohorts, including those aged 18-45, but there were significant variances in phenotype amongst age brackets. LCA identified 7 distinct clusters/classes with an optimal model showing a BIC of -83369.2 and a Silhouette Score of 0.702. Cluster/class characteristics varied by primary care utilization, demographics, SIV, SRF phenotype, and quantitative SRF measures. Notably, class 6 had the youngest median age, 48 (IQR 37-59), and a high prevalence of SRF related to metabolic disease, smoking, and obesity. However, Lp(a) was not elevated in ages 18-45, or any other group in our cohort, and was the only non-significant SRF in the analysis (p =0.64).

Conclusion

Using probabilistic modeling, our results suggest that modifiable SRFs are highly prevalent across all age groups, with a subgroup of younger patients likely to benefit from early and aggressive behavioral and pharmacologic interventions targeting obesity and metabolic disease.

Zitieren

APA

Elokour, M., Dias, S., Tadeo, A., & Simpkins, A. (2026). Abstract 9293: Modeling Contemporary Age Related Stroke Risk in a Community Based Population. Circulation, 154(Suppl_1). https://doi.org/10.1161/circ.154.suppl_1.9293

Vancouver

Elokour M, Dias S, Tadeo A, Simpkins A. Abstract 9293: Modeling Contemporary Age Related Stroke Risk in a Community Based Population. Circulation. 2026;154(Suppl_1). doi:10.1161/circ.154.suppl_1.9293

BibTeX
@article{Elokour2026,
  author = {Elokour, Malak and Dias, Sarah and Tadeo, Ada and Simpkins, Alexis},
  title = {Abstract 9293: Modeling Contemporary Age Related Stroke Risk in a Community Based Population},
  journal = {Circulation},
  year = {2026},
  volume = {154},
  number = {Suppl_1},
  doi = {10.1161/circ.154.suppl_1.9293}
}
Kostenlos starten →