“…In the studies in which ITT analysis was performed, nine studies (Campbell et al, 2012;Jain et al, 2007;Leichsenring et al, 2009;Robinson et al, 2010;Steinmetz et al, 2011;Titov et al, 2010;Vollestad et al, 2011;Watkins et al, 2011;Watkins et al, 2012) employed the LOCF method for data imputation; and five of these studies (Campbell et al, 2012;Jain et al, 2007;Titov et al, 2010;Vollestad et al, 2011;Watkins et al, 2012) had an attrition rate of greater than 10 percent. However, Jain et al (2007) and Vollestad et al (2011) evidenced best practice by reporting results for both 'completer' and ITT samples providing assurance that the attrition rate did not compromise reported results; and Watkins et al (2011) and Watkins et al (2012) conducted sensitivity analysis to ensure that drop-outs did not adversely affect their results.…”