Clinical studies/trials are experiments or observations on human subjects considered by the scientific community the most appropriate instrument to answer specific research questions on interventions on health outcomes. The time-line of the observations might be focused on a single time point or to follow time, backward or forward, in the so called, respectively, retrospective and prospective study design. Since the retrospective approach has been criticized for the possible sources of errors due to bias and confounding, we aimed this study to assess if there is a prevalence of retrospective vs. prospective design in the clinical studies/trials by querying MEDLINE. Our results on a sample of 1,438,872 studies/trials, (yrs 1960–2017), support a prevalence of retrospective, respectively 55% vs. 45%. To explain this result, a random sub-sample of studies where the country of origin was reported (n = 1,576) was categorized in high and low-income based onthe nominal Gross Domestic Product (GDP) and matched with the topic of the research. As expected, the absolute majority of studies/trials are carried on by high-income countries, respectively 86% vs. 14%; even if a slight prevalence of retrospective was recorded in both income groups, for the most part prospective studies are carried out by high-GDP countries, 85% vs. 15%. Finally, the differences in the design of the study are understandable when considering the topic of the research.
Time arrow in published clinical studies/trials indexed in MEDLINE : a systematic analysis of retrospective vs. prospective study design, from 1960 to 2017 / M. Ciulla, P. Vivona. - In: PEERJ. - ISSN 2167-8359. - 7(2019 Feb 01), pp. e6363.1-e6363.8.
|Titolo:||Time arrow in published clinical studies/trials indexed in MEDLINE : a systematic analysis of retrospective vs. prospective study design, from 1960 to 2017|
CIULLA, MICHELE MARIO (Primo) (Corresponding)
VIVONA, PATRIZIA (Secondo)
|Parole Chiave:||Systematic review; Bioinformatics; Clinical Trials; Epidemiology; Statistics; Data Mining; Machine Learning|
|Settore Scientifico Disciplinare:||Settore MED/09 - Medicina Interna|
Settore MED/01 - Statistica Medica
|Data di pubblicazione:||1-feb-2019|
|Digital Object Identifier (DOI):||http://dx.doi.org/10.7717/peerj.6363|
|Appare nelle tipologie:||01 - Articolo su periodico|