台灣老年人自覺健康與認知功能下降之關聯性研究

Abstract

研究背景: 隨著人口快速老化，全世界失智症盛行率也快速增加。自覺健康為一種主觀的測量，在過去被用來預測失智症或認知功能，然而過去研究經常只探討單題自覺健康與整體或特定範疇認知功能，且並沒有考慮失智症重要的危險因子[載脂蛋白E (APOE ε4) 狀態]。因此，本研究目的為探討自覺健康與預測追蹤兩年後整體和特定範疇認知功能下降之關聯。 材料與方法: 本研究為前瞻性世代研究，招募492位於2011-2013年(基線)參加臺大醫院老年健康檢查年齡65歲或以上的長者，並於兩年後進行追蹤(2013-2015)。本研究的依變項為兩年後認知功能下降，以兩年中整體與特定範疇認知功能(邏輯記憶、專注力、語言流暢及執行功能)的變化來評估，而自變項包含三種自覺健康的測量(現在、與過去一年和同年紀相比的健康狀況)，為了使三種自覺健康有足夠的樣本數進行分析，將長者的五組答案進一步合併，分為好(包含非常好與好)、普通(差不多)與不好(包含不好與非常不好)。共變量的資料則來自基線的問卷(包含年齡、性別、受教育年份、身體功能與共病等)，也收集血液樣本進行基因分型。使用多元線性回歸與羅吉斯回歸模型進行統計分析，在調整共變量後，估計自覺健康與認知功能下降的關聯，並進行重要因子(包含年齡、性別、教育年份、載脂蛋白E狀態)的分層分析。 研究結果: 長者的未參考自覺健康較差者比起較好者，其兩年後的專注力顯著下降(順向數字的調整勝算比=1.86)。在進行重要因子的分層分析後，女性(順向數字的迴歸係數= -0.35，順向數字的調整勝算比= 2.70)及教育程度較低(≦12年，順向數字的迴歸係數= -0.39)的長者中，其未參考自覺健康較差者比起較好者，兩年後專注力有顯著的下降。 結論: 我們的研究發現，自覺健康可以預測老年人在失智症臨床前期的認知功能下降，特別是在女性或教育程度較低(≦12年)的老年人中。

Background: As population aging quickly, the prevalence of dementia increases rapidly worldwide. Self-rated health (SRH), a subjective measure, has been used to predict dementia or cognitive function in the past. However, previous studies tend to relate a single question of SRH to global or domain-specific cognition, and did not include the important risk factor of dementia, apolipoprotein E (APOE) ε4 status. Therefore, this study aimed to explore the association between SRH and the decline of global and domain-specific cognition after two-year follow-up. Materials and Methods: This is a prospective cohort study and 492 participants aged ≧65 years were recruited from the Elderly Health Checkup (EHC) at National Taiwan University Hospital (NTUH) during 2011-2013 (baseline) and followed up from 2013-2015. The outcome of this study was cognitive decline, which assessed by the change of global and domain-specific (logical memory, attention, verbal fluency, and executive function) cognition over two years. The independent variable was SRH, which include the three types (unreferenced, time-referenced, and age-referenced). For each SRH, five answers were further grouped into “good” (very good or good), “average” (not too good), and “poor” (poor or very poor) SRH in order to have sufficient sample size for statistical analysis. Covariates were collected by a questionnaire, which included age, gender, years of education, physical function, and comorbidities, etc. We also collected blood samples for genotyping purpose at baseline. Statistical analysis included multiple linear regression and logistic regression models to estimate the association between SRH and cognitive decline after adjusting for covariates. Stratified analyses were performed by important covariates, which included age groups, gender, years of education, comorbidities, and APOE ε4 status. Results: Compared with good unreferenced SRH, elders with poor unreferenced SRH showed attention decline after 2 years [digit span-forward: adjusted odds ratio (AOR) = 1.86]. After stratification by important covariates, the associations between poor unreferenced SRH and attention decline remained among women (digit span-forward: β = -0.35, AOR =2.70) or elders with lower education (≦12 years, digit span-forward: β = -0.39). Conclusion: Our findings suggested that SRH may be useful for prediction of dementia in the elders in the preclinical phase, especially in women or elderly with lower education (≦12 years).