Diseño de una escala para la evaluación de calidad metodológica de estudios de pronóstico

Abstract

Background: One of the main classification systems to describe the quality of scientific publications is the level of evidence. However, for our research team, the methodological quality is multidimensional. Aim: To design a scale to determine the methodological quality of studies dealing with prognosis. Material and Method: On a first stage, five professionals with experience in research were interviewed about the items that, in their opinion, must be considered to assess the methodological quality of a manuscript. On a second stage, 121 prognostic studies were critically reviewed by four professionals with research experience In addition, the number of publications of each author was determined and the evidence level of each article was described according to the Oxford Centre for Evidence-based Medicine. A subsequent factorial analysis was performed and the fields and items of the scale were generated. Results: Sixty six percent of reviewed articles had a cohort design (prospective or retrospective), 45% were located at an evidence level 1b and 39% are at an evidence level of 4. After reviewing the articles, a factorial analysis was performed choosing an orthogonal matrix. A scale (MinCir Pr) was generated with a total of 25 items, 6 fields and a Cronbach coefficient alpha of 0.89. Conclusions: A scale to evaluate the methodological quality of prognosis studies (MinCir Pr) was designed, fulfilling rigorous methodological steps and the criterion of multidimensionality for methodological quality.