Evaluation of methodology and quality characteristics of systematic reviews in orthodontics.

Systematic reviews (SRs) are published with an increasing rate in many fields of biomedical literature, including orthodontics. Although SRs should consolidate the evidence-based characteristics of contemporary orthodontic practice, doubts on the validity of their conclusions have been frequently expressed. The aim of this study was to evaluate the methodology and quality characteristics of orthodontic SRs as well as to assess their quality of reporting during the last years. Electronic databases were searched for SRs (without any meta-analytical data synthesis) in the field of orthodontics, indexed up to the start of 2010. The Assessment of Multiple Systematic Reviews (AMSTAR) tool was used for quality assessment of the included articles. Data were analyzed with Student's t-test, one-way ANOVA, and linear regression. Risk ratios (RR) with 95% confidence intervals were calculated to represent changes during the years in reporting of key items associated with quality. A total of 110 SRs were included in this evaluation. About half of the SRs (46.4%) were published in orthodontic journals, while few (5.5%) were updates of previously published reviews. Using the AMSTAR tool, thirty (27.3%) of the SRs were found to be of low quality, 63 (57.3%) of medium quality, and 17 (15.5%) of high quality. No significant trend for quality improvement was observed during the last years. The overall quality of orthodontic SRs may be considered as medium. Although the number of orthodontic SRs has increased over the last decade, their quality characteristics can be characterized as moderate.