Background: Little is known about mortality among nonagenarians after an earthquake.
Methods: Using secondary data analyses from the 2005 study called the Project of Longevity and Aging in Dujiangyan(n = 870), 1-year mortality rates were compared among a pre-earthquake group and a post-earthquake group of nonagenarians. All participants were from Dujiangyan, 50 km from the epicenter of the May 12, 2008 earthquake, in China. The pre-earthquake group was a subset of the 870 Project of Longevity and Aging in Dujiangyan participants, ages 93-95 years at the beginning of "Time Frame 1" (July 2005 through June 2006; n = 228). The post-earthquake group was a different subset of the 870 Project of Longevity and Aging in Dujiangyan participants, ages 93-95 years and alive at the beginning of Time Frame 2 (July 2008 through June 2009; n = 235). Time Frame 2 excluded a 7-week period following the earthquake in order to account for deaths due to trauma. Pre-earthquake health assessment data from the 2005 Project of Longevity and Aging in Dujiangyan study were used to calculate unadjusted/adjusted hazard ratios (HRs) for mortality.
Results: One-year mortality rates were 8.3% (19/228) and 16.2% (38/235) in the pre-earthquake group and the post-earthquake group, respectively (p =.01). In unadjusted analyses, only "being in the post-earthquake group" was associated with death (HR = 2.04; 95% confidence interval [CI], 1.17-3.53; p = .011). In the multivariable Cox regression model, being in the post-earthquake group continued to be the strongest risk factor associated with mortality (HR = 2.47; 95% CI, 1.39-4.40; p = .002). Other significant risk factors included impaired cognition (HR = 1.97; 95% CI, 1.10-3.53; p = .024), serum albumin (HR = 0.90; 95% CI, 0.82-0.98; p < .015), and serum triglycerides (HR = 1.51; 95% CI, 1.15-1.99; p = .003).
Conclusion: The May 12, 2008 earthquake in Wenchuan, China, was associated with a twofold increase in the 1-year mortality among a group of nonagenarians who lived nearby.