Authors cited in relevant reports were followed with citation

tracking. The reference lists of relevant articles were hand searched for additional relevant papers. Search results were imported into bibliographic management software and duplicates Selleckchem Apoptosis Compound Library discarded. The titles and abstracts were screened against the inclusion criteria (Box 1) by one author (JH). The full text of potentially relevant papers was obtained and assessed against the same criteria. Non-English language publications were excluded. Design • Prospective cohort studies Participants • Aged 18 years or younger Outcomes • Risk of subsequent onset

of low back pain associated with a previously measured factor, where an episode of low back pain and any recall period are clearly defined, and where the low back pain does not develop as a result of serious pathology, as defined by red flags (Rosen 1994). Quality: There is no ‘gold standard’ for assessing the quality of the methods used in studies of risk factors. Bias, confounding, and chance can distort the validity of epidemiological studies ( Zaccai 2004) and studies of predictive VE-822 concentration utility. Quality assessment criteria were therefore developed to identify sources of bias that might affect the credibility of conclusions about the relationships between possible risk factors and the first episode of low back pain. Nine quality assessment criteria

were chosen, based on arguments made in the MOOSE Statement ( Stroup et al 2000) and by Hoogendorn and colleagues (2000). The criteria were grouped under three Dipeptidyl peptidase questions related to the representativeness of the study population, the definition of an episode of low back pain, and the data collection and analysis. Included studies were awarded a ‘yes’ for each of the quality criteria that were clearly met and a ‘no’ for criteria that were not met or that could not be determined from the methods reported. The maximum quality score that could be achieved was 9. Box 2 Questions and criteria used to assess the methodological quality of included studies.