The long-term effects of insulin use in incident cystic fibrosis-related diabetes: a target trial emulated using longitudinal national registry data.

Abstract

Cystic fibrosis-related diabetes (CFRD) is a common complication of cystic fibrosis and is associated with deleterious clinical outcomes. Insulin is recommended as a treatment by international guidelines. However, there are scarce clinical trial data to support the use of insulin, and little is known about the long-term outcomes of treatment. The aim of this study was to compare the long-term impacts of insulin use versus non-use in CFRD. We used data from the national UK Cystic Fibrosis Registry and adopted a target trial framework. Eligible individuals included those 12 years and older with a new diagnosis of CFRD. Outcomes were change in % predicted forced expiratory volume in 1 s (FEV1 %) and body mass index z-scores (BMI) over a 5-year follow-up period. Treatment strategies were to receive insulin or not for the duration of follow-up. Treatment effect estimates were obtained using two methods to control for confounding: inverse-probability-of-treatment weighted estimation of marginal structural models and the G-formula. We identified 1613 individuals diagnosed with CFRD between 2008 and 2016 and included 1196 and 1192 in the FEV1 % and BMI outcome analyses respectively. We found no evidence of an effect of insulin on FEV1 % over the 5-year study period. Similarly, we found no overall effect of insulin on BMI; however, there was some evidence for a positive treatment effect in patients with lower baseline BMI. Using well-established national registry data, we found no evidence of long-term treatment effects for insulin on FEV1 % or BMI in people with incident CFRD.