Highlights (what is new)
- •An analysis of 67,308 Cochrane meta-analyses suggests that unrestricted weighted least squares (UWLS) usually represent medical research better than random effects, often substantially so.
- •The better performance of UWLS is evident across meta-analyses regardless of heterogeneity, number of studies and type of outcome.
- •UWLS should be routinely reported in meta-analyses.
Study design and setting
Purchase one-time access:Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online access
One-time access price info
- For academic or personal research use, select 'Academic and Personal'
- For corporate R&D use, select 'Corporate R&D Professionals'
Subscribe:Subscribe to Journal of Clinical Epidemiology
Publication stageIn Press Journal Pre-Proof
T. D. Stanley: Conceptualization; Formal analysis; Investigation; Methodology; Validation; Writing – original draft; Writing – review & editing.
John P. A. Ioannidis: Conceptualization; Writing – review & editing.
Maximilian Maier: Conceptualization; Writing –review & editing.
Hristos Doucouliagos: Conceptualization; Writing –review & editing.
Willem M. Otte: Software; Validation; Writing – review & editing.
František Bartoš: Conceptualization; Data curation; Formal analysis; Methodology; Validation; Software; Validation; Writing – review & editing.