作者: A. Singh-Manoux , M. Kivimaki , M. M. Glymour , A. Elbaz , C. Berr
DOI: 10.1136/BMJ.D7622
关键词:
摘要: OBJECTIVES: To estimate 10 year decline in cognitive function from longitudinal data a middle aged cohort and to examine whether age cohorts can be compared with cross sectional infer the effect of on decline. DESIGN: Prospective study. At study inception 1985-8, there were 10,308 participants, representing recruitment rate 73%. SETTING: Civil service departments London, United Kingdom. PARTICIPANTS: 5198 men 2192 women, 45-70 at beginning testing 1997-9. MAIN OUTCOME MEASURE: Tests memory, reasoning, vocabulary, phonemic semantic fluency, assessed three times over years. RESULTS: All scores, except declined all five categories (age 45-49, 50-54, 55-59, 60-64, 65-70 baseline), evidence faster older people. In men, decline, shown as change/range test × 100, reasoning was -3.6% (95% confidence interval -4.1% -3.0%) those 45-49 baseline -9.6% (-10.6% -8.6%) 65-70. corresponding (-4.6% -2.7%) -7.4% (-9.1% -5.7%). Comparisons effects suggest that latter overestimate women because differences education. For example, analysis showed have by (-4.5% -2.8%) but suggested -11.4% (-14.0% -8.9%). CONCLUSIONS: Cognitive is already evident 45-49).