FRACTURE RISK AMONG TAIWANESE ELDERLY: A STUDY OF THE ELDERLY NUTRITION AND HEALTH SURVEY. (30th June 2017)

• Record Type: Journal Article
• Authors: Bann, M.; Gau, J.; Nakazawa, M.; Pan, W.
• Is Part Of: Innovation in aging. Volume 1(2017)Supplement 1
• Page Start: 558
• Page End: 558
• Publication Date: 2017-06-30
• Subjects: Aging -- Periodicals; Gerontology -- Periodicals; 612.67
• DOI: 10.1093/geroni/igx004.1964
• Languages: English
• ISSNs: 2399-5300

Abstract

Abstract: The Elderly Nutrition and Health Survey in Taiwan 1999 dataset was used for analyzing risk factors of self-reported fracture. Patient survey data was available for 2038 patients, including: 290 self-reported fractures, demographic, smoking & alcohol use status, past medical history, current medication use, and dietary recall data. Variables, including dietary intakes, medication or supplement, were examined for a significant association with PUD. Association between PUD and fracture was used to build a multi-variable logistic regression model. Variables included in the logistic regression models were age, sex, use of supplements, PUD, GI medication use, total calorie intake, folic acid intake, magnesium intake, and phosphate intake. The results show PUD is associated with self-reported fracture with crude odds ratio(OR) of 1.71 (95% CI, 1.22–2.40, p=0.0017). Similarly, GI medication use is associated with self-reported fracture with crude OR=1.96 (95% CI, 1.41–2.72, p<0.001). Multivariate logistic regressions were created step-wise to demonstrate the relationship between PUD, GI medication use and fracture. Age, gender and supplement use remain significant at p<0.05; all dietary intake variable became insignificant. Adjusted OR for PUD becomes insignificant at 1.37, (95% CI: 0.934–2.004, p=0.11) while GI Medication use remains significant with adjusted OR=1.68 (95% CI: 1.163–2.424, p=0.0057). Overall, the model has a p-value=0.9206 for goodness of fit criteria. The step-wise approach for building the multivariate logistic regression model demonstrated that GI medication use is more heavily weighted as a covariate with PUD to explain the self-reported fracture in the Taiwanese Elderly Nutrition and Health Survey Data.