Quality of trials in Australian Journal of Physiotherapy.

233 Editorial The Editorial Board of Australian Journal of Physiotherapy strives to publish an informative and interesting journal featuring high quality research that has implications for the practice of physiotherapy. One research design that has particular importance as a guide to clinical practice is the randomised trial, because it provides the least biased estimate of the effect of intervention. In order to publish the highest quality randomised trials, Australian Journal of Physiotherapy employs several complementary processes. The first is to actively seek high quality trials. We search clinical trial registers and conference proceedings to identify well conducted trials and invite the investigators to publish with AJP. Next, the AJP Author Guidelines require the submission of information, such as trial registration and declaration of conflicts of interest, which can improve the believability of the results. In the final manuscript, we ensure that authors report on the use of features in the design, conduct, and analysis of the trial that affect trial quality, whether favourable or not. For example, Australian Journal of Physiotherapy works closely with submitting authors to ensure they have clearly reported whether features such as concealed allocation and intention-to-treat analysis were carried out. To assist with standardisation of this process, we use the Consolidated Standards of Reporting Trials (CONSORT) checklist of recommended standards for reporting randomised trials, including its various extensions (Vaarbakken et al 2008). We also work with authors to ensure their data are presented to maximise ease of interpretation. We use a consistent format for tables of results to ensure CONSORT recommendations for reporting are met and to facilitate comprehension of the data by our readers. We require authors to report the between-group difference with 95% CIs instead of p values. This is because the 95% CI provides equivalent information about statistical significance and greater information about the magnitude of the effect of intervention (Gardner and Altman 1986). We also encourage authors to make their data freely accessible via our eAddenda (Herbert 2008). This provision of individual data allows it to be scrutinised and re-analysed, not only as power calculations for future trials but also for inclusion in meta-analyses of individual participant data (Higgins and Green 2009). Given all these efforts, it is worth examining the current quality of physiotherapy trials in Australian Journal of Physiotherapy. In order to do this, we used the Physiotherapy Evidence Database (PEDro) scale as a measure of methodological quality of the …