Table 3

Results of methodological quality assessment of included studies
Domain 1 2 3 4 5 6 No. biases
Criterion A B C D E F G H I J K L M N O P
Barnes et al., 2000 + + ? + + ± ± + + ± + + ± + + ± 0
Danielsson et al., 2011 + + + + + NA + + + ± + + + + + + 0
Mogro-Wilson, 2008 + + ? + + NA + + ± + + + ± + + + 0
Kuntsche et al., 2009 + + ? + + + + + + + ± + ± + + + 0
Latendresse et al., 2008 + + + ± + + + + + ± + + + + + + 0
Shelton & Van den Bree, 2010 + + ? + + NA + + + + + + + + + + 0
Simons-Morton, 2004 ± + + + + NA + + ± + ± NA ± NA + + 0
Van der Vorst et al., 2006 + ± + + + + + + + + ± + + + + + 0
Wu et al., 2006 + + + + + + + + + + + NA ± NA + + 0
Aseltine et al., 2000 + + ? + ? ± + + ± + + + ± + + + 1
Chuang et al., 2005 + + - + ± ± + + + + ± + - + + + 1
Cookston & Finlay, 2006 + + ? ± ? NA + + + + ± + + + + + 1
Droomers et al., 2003 + + + ? + + + + + ± ± + ± + + + 1
Eisenberg et al., 2008 + + + + ± NA + + + + ± + - + + + 1
Ennett et al., 2001 ± + - + + NA + + + ± ± + ± + + + 1
Flory et al., 2004 + + + + + - + + + + ± + + + ± + 1
Hung et al., 2009 + + ? + ± NA + + + + ± + + + + + 1
Kosterman et al., 2000 + + ? + + + ± + - + ± + ± + + ± 1
Paschall et al., 2004 + + ? + ? NA + + ± + ± + + + + + 1
Andrews et al., 1997 ± - NA - ? ± + + + ± + + ± + + + 2
Branstetter et al., 2011 ± ± ? + ? NA + + + + ± + - + + + 2
Crawford & Novak, 2002 - - ? + ? NA + + ± + ± + + + ? + 2
Donohew et al., 1999 + + + + + - + + ± + ± ± + + + + 2
Guilamo-Ramos et al., 2004 + + ? ± ? NA + + ± + ± + - ? + - 2
Gutman et al., 2011 + ± - + ± ± + + + + - + + + + ± 2
Cohen et al., 1994 ± ± ? + ? ± ± + + ± ± + + + ± - 3
Horton & Gil, 2008 ± + ? - + - + + ± ± ? + - + + + 3
Adrados, 1995 + + ? - - NA - ? ? - ? ? - + ? - 6

+ = “yes” (2), ± = “partly” (1), - = “no” (0), ? = “unsure” (0), NA = “not applicable”.

Domain: 1= study participation; 2= study attrition; 3= predictor measurement; 4= outcome measurement, 5=confounding measurement, 6= analysis

Criteria: A to P as in Table 1.

If ≤50% of the maximum score for a possible bias was obtained, the bias was scored as 1 for the domain concerned.

High quality is defined as the number of biases is 0.

Visser et al.

Visser et al. BMC Public Health 2012 12:886   doi:10.1186/1471-2458-12-886

Open Data