Keller H H, Østbye T
Department of Family Relations and Applied Nutrition, University of Guelph, Guelph, Ontario N1G 2W1, Canada.
J Nutr Health Aging. 2003;7(4):274-9.
Undernutrition in community-living seniors is common and has the potential to adversely influence health outcomes. Nutritional risk screening tools can help identify seniors at risk, but few have predicted health outcomes.
Seniors were recruited from 23 community service providers. The 8-item abbreviated version SCREEN (Seniors in the Community Risk Evaluation for Eating and Nutrition) was used to identify nutritional risk in 367 seniors; demographics, health, activities of daily living, and psychosocial variables were included in a baseline assessment. The seniors were followed-up by telephone for 18 months to determine the occurrence of health outcomes, including death. Cox regression was used to identify predictors of survival time.
During the 18-month follow-up there were 27 deaths (approximately 7%). Using the abbreviated tool, nutritional risk was common (42.2%). This low rate of death limited the modeling to only a few key covariates, which were based on bivariate analyses. Nutritional risk was significantly associated with time to death. Gender was also associated with time to death, with men more likely to die sooner than women. Increasing age was also significantly associated with shorter survival times.
Nutritional risk as measured by SCREEN was predictive of time to death. This simple tool may be useful for future epidemiological research on health outcomes of seniors. Further work should confirm these results, as the low event rate influenced the modeling strategy.
社区老年人营养不良情况普遍，且可能对健康结果产生不利影响。营养风险筛查工具有助于识别有风险的老年人，但很少有工具能预测健康结果。
从23个社区服务机构招募老年人。采用8项简版SCREEN（社区老年人饮食与营养风险评估）工具对367名老年人进行营养风险识别；基线评估纳入人口统计学、健康状况、日常生活活动及心理社会变量。通过电话随访老年人18个月，以确定包括死亡在内的健康结果的发生情况。采用Cox回归分析确定生存时间的预测因素。
在18个月的随访期间，有27人死亡（约7%）。使用简版工具时，营养风险情况普遍（42.2%）。这种低死亡率限制了建模只能纳入基于双变量分析的少数关键协变量。营养风险与死亡时间显著相关。性别也与死亡时间相关，男性比女性更早死亡的可能性更大。年龄增长也与较短的生存时间显著相关。
SCREEN所衡量的营养风险可预测死亡时间。这个简单的工具可能对未来关于老年人健康结果的流行病学研究有用。由于低事件发生率影响了建模策略，后续工作应证实这些结果。
