Reporting

We evaluated 31 relevant items pertaining to the REporting of studies Conducted Observational Routinely-collected health Data (RECORD) checklist[17] as a method of evaluating the completeness of study reporting (see Supplementary Appendix B). We calculated the completeness score, based on RECORD checklist, as the percentage of items meeting the specified criteria in relation to the overall applicable items for each study, with scores closer to 100% indicating a more comprehensive reporting. Two reviewers (JOC, CB) independently scored a sample of 20% of studies and disagreements were resolved by discussion. Since we obtained a substantial Kappa agreement (k= 0.87),[18, 19] one reviewer (JOC) scored the remainder.