Table 2. Quality Assessment of included studies based on the Newcastle-Ottawa Scales.

Study	Howrepresentativewas the exposedcohort?	Selection ofnon-exposedcohort	Ascertainmentof exposure	Demonstrationthat outcome ofinterest was notpresent at startof study	Of cohorts onbasis of designor analysis		Assessmentof outcome		Follow uplong enoughfor outcomesto occur	Adequacy ofcohort follow-up
Appels (11)	Somewhatrepresentativeof males aged45–60	Drawn from the same community as exposed cohort	From structured interview	No	Study controls for multiple covariates		Independent assessment from secure records		Yes	No description
Benjamins (6)	Representativenational sampleof adults	Drawn from the same community as exposed cohort	From structured interview	Yes	Study controls for multiple covariates		Independent assessment from secure records		Yes	Over 99% followed
Bosworth (23)	Somewhatrepresentativeof CHD patients	Drawn from the same community as exposed cohort	From structured interview	Yes	Study controls for multiple covariates		Independent assessment from secure records		Yes	No description
Doğanay (24)	SomewhatRepresentativeof adults aged65–74	Drawn from the same community as exposed cohort	Not available	Yes	Study controls for multiple covariates		Independent assessment from secure records		Yes	Not available
Ernsten (25)	Representativeof nationalsample adultsover 70	Drawn from the same community as exposed cohort	Written self-report	Yes	Study controls for multiple covariates		Independent assessment from secure records		Yes	No description
Fang (26)	Somewhatrepresentativesample of adultsaged 55 and over	Drawn from the same community as exposed cohort	Written self-report	Yes	Study controls for multiple covariates		Independent assessment from secure records		Yes	No description
Fernandez-Ruiz(27)	Representativesample of adultsaged 65 and over	Drawn from the same community as exposed cohort		Yes	Study controls for multiple covariates		Independent assessment from secure records		Yes	No description
Heidrich (28)	Somewhatrepresentativesample of adultsaged 25–64	Drawn from the same community as exposed cohort	From structured interview	Yes	Study controls for multiple covariates		Independent assessment from secure records		Yes	Complete follow-up
Heistaro (29)	Representativenational sampleof adults	Drawn from the same community as exposed cohort	Written self-report	Yes	Study controls for multiple covariates		Independent assessment from secure records		Yes	No description
Idler (7)	Representative ofpersons with self-reported CVD andnewly diagnosedCVD	Drawn from the same community as exposed cohort	From structured interview	Yes	Study controls for multiple covariates in different subgroups		Independent assessment from secure records		Yes	90% follow-up
Kamphuis (30)	Somewhatrepresentativesample of middle-aged men	Drawn from the same community as exposed cohort	Written self-report	Yes	Study controls for multiple covariates		Independent assessment from secure records		Yes	No description
Kaplan (31)	Somewhatrepresentativesample of elderlymen	Drawn from the same community as exposed cohort	Written self-report	Yes	Study controls for multiple covariates		Independent assessment from secure records		Yes	No description
Kennedy (32)	Representative sample of community elderly	Drawn from the same community as exposed cohort	From structured interview	Yes	Study controls for multiple covariates		Independent assessment from secure records		Yes	Small number lost due to missing data
Kuper (33)	Representativesample of womenaged 30–50	Drawn from the same community as exposed cohort	Written self-report	Yes	Study controls for multiple covariates		Independent assessment from secure records		Yes	Complete follow-up
Pijls (9)	Somewhatrepresentativesample of elderlymen	Drawn from the same community as exposed cohort	Written self-report	Yes	Study controls for multiple covariates		Independent assessment from secure records		Yes	Complete follow-up
Rutledge (34)	Representative of women with cardiac symptoms	Drawn from the same community as exposed cohort	Written self-report	Yes	Study controlled for CAD severity scores		Telephone self report and from secure records		Yes	Complete follow-up
Tsuji (35)	Somewhatrepresentative ofadults 65 and over	Same community as exposed cohort	Written self-report	Yes	Study controls for multiple covariates		Independent assessment secure records		Yes	No description
van der Linde (36)	Populationrepresentative ofadults 39–74	Drawn from the same community as exposed cohort		Yes	Study controls for multiple covariates		Independent assessment from secure records		Yes	No description
Wannamethee (37)	Representativenational sampleof men aged40–59	Drawn from the same community as the exposed cohort	Structured interview	Yes	Study controls for multiple covariates	Independent assessment from secure records		Yes		No description