Online Program

Estimating the Reference Range from a Meta-Analysis Using Aggregate or Individual Participant Data (305254)

Fateh Bazerbachi, St. Cloud Hospital 
Haitao Chu, University of Minnesota, Twin Cities 
Mohammad Hassan Murad, Mayo Clinic 
Richard D. Riley, Keele University School of Medicine 
*Lianne Kay Siegel, University of Minnesota 
Zhen Wang, Mayo Clinic 

Keywords: meta-analysis, reference range, individual participant data, prediction interval

Clinicians frequently must decide whether a patient’s measurement reflects that of a healthy “normal” individual. Thus, the reference range is defined as the interval in which some proportion (frequently 95%) of measurements from a healthy population is expected to fall. One can estimate it from a single study, or preferably from a meta-analysis of multiple studies to increase generalizability. We introduce how to calculate the reference range from a meta-analysis and outline how it differs from the confidence interval for the pooled mean and the prediction interval for the mean of a new study. We provide an overview of three recently proposed methods that incorporate both within and between-study variation: a frequentist, a Bayesian, and an empirical method, and apply them to a systematic review and meta-analysis of studies measuring normative liver stiffness in adults. Identification of the target population and exploration of potential subgroup heterogeneity will also be discussed. We consider using aggregate data from publications, but also extend these methods to the case where individual participant data are available.