Macarena

and 3 more

Original ArticleJournal of Clinical EpidemiologyTitle: Most systematic reviews of high methodological quality on psoriasis interventions are classified as high risk of bias using ROBIS toolRunning title: Risk of bias vs methodological quality analysis of reviews on psoriasisManuscript word count: XXX  Tables: XX  Figures:5  Authors: F. Gómez-García12#; J. Ruano12#*; J. Gay-Mimbrera2; M. Aguilar-Luque2; J.L.Sanz-Cabanillas1,2; P.Alcalde-Mellado2,3; B. Maestre-Lopez2,3; P.J. Carmona-Fernández2; M.González-Padilla1,2; A.Vélez García-Nieto1,2; B. Isla-Tejera2,4.1 Department of Dermatology, Hospital Universitario Reina Sofía, Córdoba, Spain.2 Instituto Maimónides de Investigación Biomédica de Córdoba (IMIBIC)/Hospital Universitario Reina Sofía/Universidad de Córdoba, Córdoba, Spain.3 School of Medicine, Universidad de Córdoba, Córdoba, Spain.4 Department of Pharmacy, Hospital Universitario Reina Sofía, Córdoba, Spain.*Corresponding author: Juan Ruano. Department of Dermatology. Hospital Universitario Reina Sofía, Menendez Pidal Ave, 14004, Córdoba, Spain. Telephone/Fax: +34-957-011214. ORCID ID: 0000-0002-0286-4107.E-mail address: juanruanoruiz@mac.com (J.Ruano).#Theseauthors contributed equally to this work.Keywords: Psoriasis; Systematic Reviews; Risk of Bias; Methodological Quality; AMSTAR; ROBIS. Funding sources: This work has been funded in part by project ICI1400136 to JR integrated into the National Plan of R+D+I 2008-2011 and co-financed by the ISCIII-Subdirección General de Evaluación and European Regional Development Fund (ERDF) and by grant PP13/009 of Plan Propio de movilidad para investigadores del Instituto Maimonides de Investigación Biomédica de Cordoba (IMIBIC). No funding was received from any pharmaceutical company.Conflict of Interest: FG-G has received honoraria for research from Pfizer and for lecturing from AbbVie Janssen-Cilag and Novartis; JR has received honoraria for lecturing and grants for research fromPfizer honoraria for lecturing from Janssen-Cilag and Novartis and other financial benefits from AbbVie and Novartis; AVG-N has received honoraria for lecturing from Pfizer Novartis AbbVie and Janssen-Cilag and other financial benefits from AbbVie Novartis and Janssen-Cilag; JG-M, MA-L, JLS-C, PA-M, BM-L, PJC, FMG-P, and BI-T have no disclosures.What is new??Psoriasis is a chronic immune-mediated inflammatory skin disease.Moderate-to-severe forms are associated to significant comorbidity impaired quality of life and high direct and indirect costs.This is the first study to compare the capacity of both AMSTAR and ROBIS instruments to assess the quality of published SRs about intervention in psoriasis.Más....Methodological bias can reduce the validity of conclusions and therefore impair the quality of decision making.Abstract (max 250 words)Background: There is no gold standard for the quality assessment of systematic reviews(SRs). Although AMSTAR is the most uniformly accepted tool for methodological quality analysis, the ROBIS instrument has recently been developed to evaluate the risk of bias of SRs. The objective of this study was to compare the capacity of both instruments to capture the quality of evidence summarized by SRs about interventions on psoriasis.Methods: Systematic literature searches were undertaken on relevant databases including EMBASE, MEDLINE, and Cochrane Database (up to January 2017). For each review the methodological quality and the risk of bias were evaluated using A Measurement Tool to Assess Systematic Reviews(AMSTAR) and the ROBIS tool (PROSPERO 2016:CRD42016053181). We conducted descriptive and principal component analysis to project the final classification based on the 11-AMSTAR and the 21-ROBIS items into a few orthogonal components describing similarities and discrepancies between both assessment tools. Results: We classified 139 intervention studies on psoriasis as high (22.3%), moderate (53.2%) or low (24.5%) methodological quality according to AMSTAR and as low (14.3%) or high (85.6%) risk of bias according to ROBIS. A high risk of bias was detected for most of those SRs classified as high (61.2%) or moderate (89.2%) methodological quality. When comparing ROBIS result profiles of high methodological quality SRs with low vs.high risk of bias domain 4 items showed the greatest achievement reduction, while domain 2 items showed the smallest changes.Conclusions: When considering SRs published about psoriasis the methodological quality remains suboptimal and the risk of bias even for those of the best quality is mostly high. The AMSTAR and ROBIS maybe considered as complementary instruments to assess the quality of SRs.IntroductionPSORIASISPsoriasis is a chronic disease whose moderate and severe forms are associated with significant comorbility impaired quality of life and high direct and indirect costs (cita). New therapies have been developed during last decade increasingly the effectivity but with potentially significant adverse effects and higher costs that puts at risk the patients and the sustainability of health systems(1)⁠. Therefore therapeutic decision-making processes about the interventions in psoriasis should be based on the best evidence.SYSTEMATIC REVIEWSSystematic reviews are the standard for the synthesis of the evidence. Their conclusions are often used as a starting point for developing clinical practice guidelines laying down the recommendations of diagnostic prognostic or therapeutic interventions. Taking decisions based on them can improve health outcomes(2)⁠. However, some research groups have found discrepancies on SRs that respond to the same research questions(3)⁠⁠. This is one risk for performances both health authorities and doctors which also affects the enhancement of the preferences of patients. This risk is greater given the proliferation in recent years of RS is not intended to improve the evidence but to use the prestige of these publications to launch concerned messages or RS made within competence.CALIDAD DE LAS REVISIONES SISTEMÁTICAS (traducir)Por todo lo anterior resulta fundamentar elegir de entre todas aquellas de mayor calidad. Dado que el objetivo de la conducción de revisiones sistemáticas es minimizar los sesgos la medición de la calidad debería estar relacionada con la del riesgo de sesgo. En este sentido, Mohers (1995) definió la calidad como la probabilidad de que el diseño de una revisión sistemática genere resultados no sesgados. La evaluación de la calidad tiene dos complicaciones importantes la primera es la subjetividad y la segunda es que únicamente se puede evaluar el trabajo publicado. Se han desarrollado múltiples herramientas de evaluación de la calidad sin que ninguna este universalmente aceptada. La más empleada es AMSTAR que ha demostrado ser fiable y válida. AMSTAR se puede emplear individualmente (componentes) y como lista deverificación sumando el resultado de los items (puntuación general). Sin embargo presenta limitaciones por un lado la importancia de cada componente es variable en función de cada situación particular (naturaleza de la RS) y por otro aunque la puntuación total es significativa no se establece una relación entre ésta y el riesgo de sesgo de las revisiones sistemáticas.Hasta el momento las publicaciones que han empleado AMSTAR clasifican por consenso el resultado de las puntuaciones de las RS en alta media o baja calidad. Si bien esto es orientativo es posible por ejemplo que algún ítem de AMSTAR no tenga relación con el riesgo de sesgo de la RS ( ocurre por ejemplo con el cálculo del tamaño muestral y el riesgo de sesgo en ensayos clínicos que es un aspecto metodológico que no influye en el RR) para algún tipo específico de RS o que exista un límite de puntuación a partir del cual se pueda demostrar un riesgo elevado de sesgo de la RS. Reciente se ha publicado ROBIS con la intención de medir el riesgo de sesgo de las revisiones sistemáticas. Está basado en la evaluación de dominios mediante items y en cómo las limitaciones metodológicas son tenidas en cuenta por los autores para la redacción de las conclusiones. Ofrece como resultado un nivel de riesgo de sesgo alto/bajo pero no tieneen cuenta el conflicto de intereses o la financiación cuyainfluencia sobre la calidad y la dirección de las conclusiones hasido previamente demostrada ni se ha establecido la importanciarelativa de los dominios para tipos específicos de