Primary versus secondary source of data in observational studies and heterogeneity in meta-analyses of drug effects: a survey of major medical journals

Table 2 Reporting of the data source in the data extraction tables of the included meta-analyses

Meta-analysis (MA)	Exposure assessment							Outcome assessment
	Data source presented in MA	Cohort studies (n)			Case-control studies (n)			Data source presented in MA	Cohort studies (n)			Case-control studies (n)
	Data source presented in MA	1ry	2ry	NR	1ry	2ry	NR	Data source presented in MA	1ry	2ry	NR	1ry	2ry	NR
Weiss J [20] Harms outcomes	No	.	.	.	.	.	.	No	.	.	.	.	.	.
Bally M [21]	Yes	0	3^b	0	0	1	0	Yes	0	3^b	0	0	1	0
Sordo L [22]	No^a	.	.	.	.	.	.	No^a	.	.	.	.	.	.
Tariq R [23]	No^ac	.	.	.	.	.	.	No^a	.	.	.	.	.	.
Maruthur NM [24]	Yes^d	0	3	0	.	.	.	Yes^d	0	3	0	.	.	.
Paul S [25]	No^a	.	.	.	.	.	.	No^a	.	.	.	.	.	.
Li L [26] Heart failure	Yes	0	1	2	0	0	1	Yes	1	0	2	0	0	1
Li L [26] Hospital admissions for heart failure	Yes	0	0	6	0	0	2	Yes	3	0	3	0	0	2
Molnar AO [27]	No^a	.	.	.	.	.	.	No^a	.	.	.	.	.	.
Ziff OJ [28]	No^a	.	.	.	.	.	.	No^a	.	.	.	.	.	.
CGESOC [29]	No	.	.	.	.	.	.	No	.	.	.	.	.	.
Bellemain-Appaix A [30]	No^a	.	.	.	.	.	.	No^a	.	.	.	.	.	.
Grigoriadis S [31]	Yes	2	3	0	1	1	0	Yes	4	1	0	2	0	0
Li L [32]	Yes	0	1	2	0	1	1	Yes	1	2	0	0	0	2
Kalil AC [33]	No	.	.	.	.	.	.	No	.	.	.	.	.	.
Stegeman BH [34]	Yes	0	9	0	8	8	1	Yes	4	5	0	5	12	0
Maneiro JR [35]	No^a	.	.	.	.	.	.	No^a	.	.	.	.	.	.
Hartling L [36]	No	.	.	.	.	.	.	No	.	.	.	.	.	.
Hsu J [37]	No^a	.	.	.	.	.	.	No^a	.	.	.	.	.	.
Caldeira D [38]	Yes	2	2	7	0	7	1	Yes	0	1	10	3	1	4
MacArthur GJ [39]	No^a	.	.	.	.	.	.	No^a	.	.	.	.	.	.
Mantha S [40]	No	.	.	.	.	.	.	No	.	.	.	.	.	.
Silvain J [41]	Yes	0	7	0	.	.	.	Yes	0	7	0	.	.	.
McKnight RF [42]	No	.	.	.	.	.	.	No	.	.	.	.	.	.

Abbreviations: 1ry number of individual studies in each MA based on primary data sources, 2ry number of individual studies in each MA based on secondary data sources, NR number of individual studies in each MA with not reported data source
^aAlthough the meta-analysis shows the results of methodological quality assessment based on a standardized scale, it does not indicate the type of data source used for each individual observational study included in the meta-analysis
^bCohort with nested case-control analysis
^cThe meta-analysis reports that most of the included observational studies assessed medication exposure through a review of medical records
^dThe meta-analysis reports only data from high-quality observational studies

ISSN: 1471-2288