41-OR: The Health Effects of a Behavior Change Program for Prediabetes—Quasi-Experimental Evidence from 2.1 Million Primary Care Patients in England
JULIA M. LEMP,CHRISTIAN BOMMER,MIN XIE,JUSTINE DAVIES,ANANT JANI,SEBASTIAN VOLLMER,TILL BARNIGHAUSEN,PASCAL GELDSETZER
DOI: https://doi.org/10.2337/db23-41-or
IF: 7.7
2023-06-20
Diabetes
Abstract:Introduction: There is an urgent need to implement population-based measures that prevent diabetes, enhance its detection, and avert its progression to complications. However, there remains widespread doubt among clinicians that lifestyle counseling provided in routine care can improve health. This study aims to investigate whether the largest behavior change program for pre-diabetes globally (the English Diabetes Prevention Programme) leads to improvements in key cardiovascular risk factors. Methods: We exploited the threshold in HbA1c used to decide on program eligibility by applying a regression discontinuity design, one of the most credible quasi-experimental strategies for causal inference, to electronic health data from the Clinical Practice Research Datalink, covering nearly one fifth of all GP practices in England. Using local linear regression, we compared patients lying closely on either side of the threshold for program entry (HbA1c of 42 mmol/mol [6.0%]). The primary outcome was change in HbA1c. Secondary outcomes were body mass index (BMI), weight, blood pressure, and serum lipids. Results: 2 106 469 adult patients met inclusion criteria in 2017-2018 with follow-up ending June 2020. Program referral led to significant improvement in HbA1c (-0.82 mmol/mol, 95%CI -1.42, -0.21), BMI (-1.34, 95%CI -1.86, -0.81), weight (-2.94 kg, 95%CI -4.33, -1.56), and triglycerides (-0.32 mmol/l, 95%CI -0.53, -0.10). There was suggestive evidence that program referral improved HDL (0.04 mmol/l, 95%CI 0, 0.08). Blood pressure and LDL did not significantly improve. Sensitivity analyses showed that improvements in HbA1c were not attributable to being prescribed medication. Conclusion: By leveraging a threshold rule induced by public health guidelines, this analysis provides causal, rather than associational, evidence that lifestyle counseling implemented in a national health system can achieve important health improvements. Disclosure J.M.Lemp: None. C.Bommer: None. M.Xie: None. J.Davies: None. A.Jani: Advisory Panel; Allergan. S.Vollmer: None. T.Barnighausen: None. P.Geldsetzer: None. Funding Chan Zuckerberg Biohub Investigator Award (to P.G.); Alexander von Humboldt Foundation (to T.B.); Ministry of Science, Research, and the Arts Baden-Wuerttemberg; German Research Foundation (INST 35/1314-1 FUGG); State of Baden-Wuerttemberg, Germany
endocrinology & metabolism