Decision Making In A World Of Comparative Effectiveness Research

Author : Sara Riaz
Publisher :
ISBN 13 :
Total Pages : pages
Book Rating : 4.:/5 (129 download)

DOWNLOAD NOW!

Download or read book Publication Rates of Randomized Comparative Effectiveness Research Relevant to Primary Care written by Sara Riaz and published by . This book was released on 2020 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: "Introduction: Pragmatic studies such as the comparative effectiveness research (CER) trial, are increasingly being used to support evidence-based medicine (EBM). The CER trial aims to generate evidence by directly comparing two or more alternative interventions that are used in routine practice; benefit to primary care is substantial. As a result, there is need to monitor publication of completed CER trials that are valuable to primary healthcare. The publication patterns of these trials and characteristics associated with the likelihood of publication have yet to be evaluated. There is also little known about time to publication after study completion for published studies. Objective: This thesis sought to evaluate what proportion of primary care CER trials registered on clinicaltrials.gov (CT.gov) were published and the time to publication following primary study completion as well as associated factors. Methods: The CT.gov database was searched for completed randomized CER trials relevant to primary care with a study start date no earlier than September 27, 2007 and a primary completion date no later than August 01, 2015, to ensure a minimum of four-years of follow-up time for publication. PubMed, Embase and for NIH-funded studies, the RePORTER online portal, were searched for matching publications. Statistical Analysis: Logistic regression was used to identify factors associated with published trials compared to unpublished ones. For published studies, descriptive statistics were used to identify characteristics of all studies, including reporting of significant primary endpoints and lastly, linear regression was used to explore factors associated with time lag from primary study completion to publication. Results: The study cohort included 122 trials. 56 (46%) trials had no matching full-text publication within four years and 39 (32%) trials had not been published in any form. Factors associated with published trials compared to unpublished trials on univariate logistic regression were government-funding and trials studying cardiovascular conditions seen in primary care. On multivariable analysis, the only significant factor associated with publication was a cardiovascular study condition, as compared to an infectious disease study condition (odds ratio [4.4], 95% confidence interval 0.04-2.9, p=0.04). Out of the 54% of studies that were published, median time lag between primary study completion and publication was 25 months (IQR, 16-34). Trials comparing interventions for infectious disease and psychiatry conditions were associated with a longer time lag in publishing compared to cardiovascular study conditions (14.9 months, p=0.001 and 11.1 months, p=0.019, respectively). More than half (54%) of published studies reported a significant primary endpoint. Conclusion: Timely publishing of randomized CER trial results is unsatisfactory. Primary care physicians are often involved in the care of patients with cardiovascular disease, and timely publication of evidence that can inform real-world medical decision-making for these patients is promising. However, without timely reporting of all CER trials, potential evidence is left undiscovered and as a result, little to no valuable contribution is made to the overall field of primary healthcare. Consequences of non-publication include lack of regard for the primary care patients who volunteered in these studies, a breach in society's trust in the biomedical research enterprise and a growing skepticism around the substantial efforts toward CER and EBM, given that the purpose of conducting comparative effectiveness studies is undermined when results fail to be reported"--