“…The design of seven studies (50%) was cross-sectional, but two of the six longitudinal studies reported relevant statistics only for the cross-sectional associations (Benore et al, 2008;Luberto et al, 2012) and one study did not clearly specify the time points used (Lyon et al, 2014). Thus, only four studies clearly reported on longitudinal associations, spanning a one-year (Zehnder et al, 2006), two-year (Reynolds, Mrug, Hensler, et al, 2014), and five-year (Reynolds, Mrug, Britton, et al, 2014) prospective periods and a three-year retrospective period (Grossoehme et al, 2013).…”