Table 1. Characteristics of Included Studies.

Study (location)	Period	Setting	Recruitment			Follow-up		Dementia			Mortality
			Sample	Response rate, %	Included or excluded individuals	Source of outcome data	Assessment	Maximum follow-up time, y	Diagnosis	Lost to follow-up, %	Maximum follow-up time, y	Verification
ZARADEMP (Zaragoza, Spain)16	1994-1999	Community dwelling and institutionalized	Random census sample stratified by age and sex	79	Included: aged ≥55 y	Follow-up cognitive screening and municipal death registry	Every 2 y	4	Psychiatrist-confirmed DSM-IV diagnosisa	10	13	Verified by death records
SNAC-K (Stockholm, Sweden)17	2001-2020	Community dwelling and institutionalized	Random population sample stratified by age; multiple cohorts	73	Included: aged ≥60 y	Follow-up cognitive screening, medical registries, informant interviews, and death certificates	Every 6 y (for those aged <78 y) 3-yearly (for aged ≥78 y)	12	2nd independent physician–confirmed DSM-III-R diagnosis	11	16	Verified by death records
Kungsholmen Project (Stockholm, Sweden)18	1987-1996	Community dwelling and institutionalized	Kungsholmen district (75+)	76	Included: aged ≥75 y	Follow-up cognitive screening, medical registries, informant interviews, and death certificates	Every 3 y	9	2nd independent physician–confirmed DSM-III-R diagnosis	12	11	Verified by death records
LEILA 75+ (Leipzig, Germany)19	1997-2014	Community dwelling and institutionalized	Representative population sample	75	Included: aged ≥75 y	Follow-up cognitive screening, death certificates, and relative interviews	Every 1.5 y	16	Expert panel–confirmed DSM-III-R or DSM-IV diagnosis	9	16	Verified by death records
H70 (Gothenburg, Sweden)20	2000-2012	Community dwelling and institutionalized	Representative population samples; 1930 cohorts	72	Included: aged 70, 75, and 79 y	Follow-up cognitive screening, medical registries, informant interviews, and death records	Every 5 y	12	DSM-III-R diagnosis	0	12	Verified by death records
PreDIVA (Amsterdam, the Netherlands)21	2006-2015	Community dwelling only	General practice populations (>98% Dutch population is registered with a general practitioner)	53	Included: aged 70-78 y Excluded: with (possible) dementia and limited life expectancy (eg, terminal disease)	Follow-up cognitive screening, EHRs, and municipal death records	Every 2 y	9	Expert panel–confirmed DSM-IV diagnosis	2	9	Verified by death records
ACT (Seattle, Washington)22	1994-2020	Community dwelling only	Random health insurance sample (representative of local population)	48	Included: aged ≥65 y Excluded: with (possible) dementia	Follow-up cognitive screening, medical records, and death certificates	Every 2 y	25	Expert panel–confirmed DSM-IV diagnosis	3	25	Verified by death records

Abbreviations: ACT, Adult Changes in Thought; DSM-III-R, Diagnostic and Statistical Manual of Mental Disorders (Third Edition Revised); DSM-IV, Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition); EHR, electronic health record; H70, Gothenburg H70 Birth Cohort Study; LEILA 75+, Leipzig Longitudinal Study of the Aged; PreDIVA, Prevention of Dementia by Intensive Vascular Care; SNAC-K, Swedish National Study of Aging and Care in Kungsholmen; ZARADEMP, Zaragoza Dementia Depression Project.

^a No dementia information was available for individuals between last assessment and death or study dropout.