RT Journal Article SR Electronic T1 Circulating sex hormone binding globulin levels are modified with intensive lifestyle intervention, but their changes did not independently predict diabetes risk in the Diabetes Prevention Program JF BMJ Open Diabetes Research & Care JO BMJ Open Diab Res Care FD American Diabetes Association SP e001841 DO 10.1136/bmjdrc-2020-001841 VO 8 IS 2 A1 Vanita R Aroda A1 Costas A Christophi A1 Sharon L Edelstein A1 Leigh Perreault A1 Catherine Kim A1 Sherita H Golden A1 Edward Horton A1 Kieren J Mather A1 , YR 2020 UL http://drc.bmj.com/content/8/2/e001841.abstract AB Introduction Sex hormone binding globulin (SHBG) levels are reported to be inversely associated with diabetes risk. It is unknown whether diabetes prevention interventions increase SHBG and whether resultant changes in SHBG affect diabetes risk. The purpose of this analysis was to determine whether intensive lifestyle intervention (ILS) or metformin changed circulating SHBG and if resultant changes influenced diabetes risk in the Diabetes Prevention Program (DPP).Research design and methods This is a secondary analysis from the DPP (1996–2001), a randomized trial of ILS or metformin versus placebo on diabetes risk over a mean follow-up of 3.2 years. The DPP was conducted across 27 academic study centers in the USA. Men, premenopausal and postmenopausal women without hormone use in the DPP were evaluated. The DPP included overweight/obese persons with elevated fasting glucose and impaired glucose tolerance. Main outcomes measures were changes in SHBG levels at 1 year and risk of diabetes over 3 years.Results ILS resulted in significantly higher increases (postmenopausal women: p<0.01) or smaller decrements (men: p<0.05; premenopausal women: p<0.01) in SHBG compared with placebo or metformin. Changes in SHBG were primarily attributable to changes in adiposity. There were no consistent associations of change in SHBG with the risk of diabetes by treatment arm or participant group.Conclusions Lifestyle intervention may be associated with favorable changes in circulating SHBG, which is largely due to changes in adiposity. Changes in circulating SHBG do not independently predict reductions in diabetes incidence.