Systematic reviews and meta-analyses on psoriasis: role of funding sources, conflict of interest and bibliometric indices as predictors of methodological quality

Francisco Gómez‐García, Juan Ruano, Macarena Aguilar-Luque, Jesús Gay-Mimbrera, Beatriz Maestre-López, Juan Luís Sanz-Cabanillas, Pedro J. Carmona-Fernández, Marcelino González-Padilla, Antonio Vélez García‐Nieto, Beatriz Isla-Tejera

British Journal of Dermatology · 2017 · 44 citations · 29 references

Concepts

Abstract

BACKGROUND: The quality of systematic reviews and meta-analyses on psoriasis, a chronic inflammatory skin disease that severely impairs quality of life and is associated with high costs, remains unknown. OBJECTIVES: To assess the methodological quality of systematic reviews published on psoriasis. METHODS: After a comprehensive search in MEDLINE, Embase and the Cochrane Database (PROSPERO: CDR42016041611), the quality of studies was assessed by two raters using the Assessment of Multiple Systematic Reviews (AMSTAR) tool. Article metadata and journal-related bibliometric indices were also obtained. Systematic reviews were classified as low (0-4), moderate (5-8) or high (9-11) quality. A prediction model for methodological quality was fitted using principal component and multivariate ordinal logistic regression analyses. RESULTS: We classified 220 studies as high (17·2%), moderate (55·0%) or low (27·8%) quality. Lower compliance rates were found for AMSTAR question (Q)5 (list of studies provided, 11·4%), Q10 (publication bias assessed, 27·7%), Q4 (status of publication included, 39·5%) and Q1 (a priori design provided, 40·9%). Factors such as meta-analysis inclusion [odds ratio (OR) 6·22; 95% confidence interval (CI) 2·78-14·86], funding by academic institutions (OR 2·90, 95% CI 1·11-7·89), Article Influence score (OR 2·14, 95% CI 1·05-6·67), 5-year impact factor (OR 1·34, 95% CI 1·02-1·40) and article page count (OR 1·08, 95% CI 1·02-1·15) significantly predicted higher quality. A high number of authors with a conflict of interest (OR 0·90, 95% CI 0·82-0·99) was significantly associated with lower quality. CONCLUSIONS: The methodological quality of systematic reviews published about psoriasis remains suboptimal. The type of funding sources and author conflicts may compromise study quality, increasing the risk of bias.

References

29