Table 4

Assessment of methodological quality of the included studies

Study ID	Data collected prospectively	Sample representative	Important covariates measured	Important covariates incorporated in analysis	Method of dealing with confounding adequate*
Aronson 20087	Yes	Unclear	Yes	Yes	Yes
Bernard 200931	No	Yes	Yes	Yes	Yes
Bochicchio 20088	Yes	Yes	No	No	No
Charles 200732	No	Yes	No	No	No
Edgren 201036	No	Yes	No	No	No
Engoren 20099	No	Yes	No	No	No
Engoren 200910	No	Yes	No	No	No
Garty 200911	No	Yes	Yes	Yes	Yes
Jani 200712	No	No (>60 years)	Yes	Yes	Yes
Johnson 200613	No	Unclear	No	No	No
Khorana 200814	No	Yes	No	No	No
Koch 200615 16	Yes	Yes	Yes	Yes	Yes
Koch 200837	No	Yes	Yes	Yes	Yes
Murphy 200717	No	Yes	Yes	Yes	Yes
Nikolsky 200918	Yes	Yes	Yes	Yes	Yes
O'Keeffe 201033	Yes	Yes	No	No	No
Pederson 200919	No	Yes	No	No	No
Rogers 200620	No	No (>65 years)	No	No	No
Rogers 200921	No	No (> 65 years)	No	No	No
Ruttinger 200734	No	Yes	No	No	No
Salehiomran 200922	No	Yes	Yes	Yes	Yes
Stone 200823	No	No (< 16 years))	No	No	No
Surgenor 200924	Yes	Yes	No	No	No
Taylor 200625	Yes	Yes	No	No	No
van de Watering 200638	No	Yes	No	No	No
van Straten 201026	Yes	Yes	No	No	No
Vincent 200827	Yes	Yes	No	No	No
Weightman 200935	Yes	Yes	No	No	No
Weinburg 200828	No	Yes	No	No	No
Weinburg 200839	No	Yes	No	No	No
Wu 201029	No	No (> 65 years)	Yes	Yes	Yes
Zilberberg 200830	No	Yes	No	No	No

*The method of dealing with confounding was deemed adequate if important covariates were measured and adjusted for in the analysis.