Systematic review of intervention design and delivery in pragmatic and explanatory surgical randomized clinical trials.
Blencowe NS., Boddy AP., Harris A., Hanna T., Whiting P., Cook JA., Blazeby JM.
BACKGROUND: Surgical interventions are complex, with multiple components that require consideration in trial reporting. This review examines the reporting of details of surgical interventions in randomized clinical trials (RCTs) within the context of explanatory and pragmatic study designs. METHODS: Systematic searches identified RCTs of surgical interventions published in 2010 and 2011. Included studies were categorized as predominantly explanatory or pragmatic. The extent of intervention details in the reports were compared with the CONSORT statement for reporting trials of non-pharmacological treatments (CONSORT-NPT). CONSORT-NPT recommends reporting the descriptions of surgical interventions, whether they were standardized and adhered to (items 4a, 4b and 4c). Reporting of the context of intervention delivery (items 3 and 15) and operator expertise (item 15) were assessed. RESULTS: Of 4541 abstracts and 131 full-text articles, 80 were included (of which 39 were classified as predominantly pragmatic), reporting 160 interventions. Descriptions of 129 interventions (80.6 per cent) were provided. Standardization was mentioned for 47 (29.4 per cent) of the 160 interventions, and 22 articles (28 per cent) reported measurement of adherence to at least one aspect of the intervention. Seventy-one papers (89 per cent) provided some information about context. For one-third of interventions (55, 34.4 per cent), some data were provided regarding the expertise of personnel involved. Reporting standards were similar in trials classified as pragmatic or explanatory. CONCLUSION: The lack of detail in trial reports about surgical interventions creates difficulties in understanding which operations were actually evaluated. Methods for designing and reporting surgical interventions in RCTs, contributing to the quality of the overall study design, are required. This should allow better implementation of trial results into practice.