Physical activity interventions.docVIP

Physical activity interventions In choosing interventions for analysis, we have aimed to include a broad array of intervention approaches that could be implemented in Australia. Intervention studies, both published and unpublished, were identified by the first author through a search of PUBMED, EMBASE, The Cochrane Library, and Web of Knowledge (Web of Science and ISI) databases, the grey literature, and bibliographies of relevant studies and reviews, and through information supplied by relevant experts. We evaluated the strength of evidence of each study using an extended six-level version of the standard evidence classification system (Table I). Ambiguities in the evaluation of evidence were discussed by the first two authors, but a Technical Advisory Panel, consisting of three Australian experts in physical activity research, was also formed to provide additional advice where uncertainties could not be resolved by the authors. Studies were excluded from the evaluation if: the intervention did not involve primary prevention in the adult population; the study did not have a ‘usual care’ or ‘current practice’ comparator for the cost-effectiveness analysis; the evidence of effectiveness was inconclusive or did not exist (Level 5-6 in Table I); or the evidence of effectiveness was weak or limited (Level 3-4 in Table I) AND the evidence of effectiveness could not be generalised to another setting or population. Where there were multiple studies of the same type of intervention, studies were combined in a meta-analysis. However, where multiple studies were too heterogeneous to enable a precise definition of the intervention and comparator (i.e. who did what, to whom, when, and where) and accurate measurement of resources used, a single study was chosen for economic evaluation. This selection was based on the strength of evidence of effectiveness and generalisability of the setting and population to the Australian context (i.e. giving regard to both the internal and ext