Background: Existing risk scores assessing geriatric vulnerability in the emergency department (ED) have shown limited predictive power, especially in diverse populations. We investigated the relationship of a quick and easy-to-administer geriatric vulnerability scoring system with functional decline and mortality in older patients admitted to multiple hospitals through the ED in the United States (US) and Brazil (BR).

Method: Federated, international, multicenter observational study of hospitalized ED patients aged ≥ 65 from US and BR. The six criteria from the PRO-AGE score (Physical impairment, Recent hospitalization, Older age [≥ 90], Acute mental alteration, Getting thinner, and Exhaustion; 0-8; higher scores = greater vulnerability) were assessed on admission. We used proportional hazards models to investigate the relationships between PRO-AGE score groups and 90-day mortality and functional decline, defined as new dependence in activities of daily living (ADL) and instrumental ADL (IADL), after adjusting for age, sex, race and ethnicity, education, Charlson comorbidity score, and study site. Death was considered a competing event for the functional decline outcome.

Results: A total of 1390 patients were included (US = 560; Brazil = 830). The 90-day risk of death was higher for the upper compared with the lower (reference) PRO-AGE group in both cohorts (US: HR = 11.76; 95% confidence interval [CI] = 2.56-54.04; BR: HR = 12.29; 95% CI = 3.54-42.59), whereas the risk of new 90-day ADL disability was higher for upper (HR = 2.08; 95% CI = 1.21-3.56) and middle groups (HR = 2.10; 95% CI = 1.35-3.27) in the US but only the upper group in BR (HR = 1.70; 95% CI = 1.02-2.85).

Conclusion: A higher PRO-AGE score was associated with mortality and functional decline in older ED patients admitted to hospitals in the US and BR, demonstrating its generalizability as a geriatric vulnerability risk score.