Estimating the effects of second-line therapy for type 2 diabetes mellitus: retrospective cohort study

Abstract

Objective: Metformin is the recommended initial drug treatment in type 2 diabetes mellitus, but there is no clearly preferred choice for an additional drug when indicated. We compare the counterfactual drug effectiveness in lowering glycated hemoglobin (HbA1c) levels and effect on body mass index (BMI) of four diabetes second-line drug classes using electronic health records.

Study design and setting: Retrospective analysis of electronic health records of US-based patients in the Explorys database using causal inference methodology to adjust for patient censoring and confounders.

Participants and exposures: Our cohort consisted of more than 40 000 patients with type 2 diabetes, prescribed metformin along with a drug out of four second-line drug classes-sulfonylureas, thiazolidinediones, dipeptidyl peptidase 4 (DPP-4) inhibitors and glucagon-like peptide-1 agonists-during the years 2000-2015. Roughly, 17 000 of these patients were followed for 12 months after being prescribed a second-line drug.

Main outcome measures: HbA1c and BMI of these patients after 6 and 12 months following treatment.

Results: We demonstrate that all four drug classes reduce HbA1c levels, but the effect of sulfonylureas after 6 and 12 months of treatment is less pronounced compared with other classes. We also estimate that DPP-4 inhibitors decrease body weight significantly more than sulfonylureas and thiazolidinediones.

Conclusion: Our results are in line with current knowledge on second-line drug effectiveness and effect on BMI. They demonstrate that causal inference from electronic health records is an effective way for conducting multitreatment causal inference studies.

Keywords: anti-diabetic drugs; electronic medical records; treatment efficacy; type 2 diabetes.

Figure 1

Illustration of the causal inference scheme. (A) Index date is the first prescription of diabetes second-line drug after use of metformin. (B) Potential censoring events include switching to another second-line drug, missing glycated hemoglobin (HbA1c) or body mass index (BMI) measurement, or undergoing bariatric surgery. (C) Outcomes (HbA1c and BMI) are checked after 6 and 12 months from index date. (D) Follow-up ends after 15 months.

Figure 2

Predicted and observed HbA1c levels using doubly robust estimation adjusting for either a comprehensive set of confounders (left panel) or a set of confounders provided by a domain expert (right panel). Red dots indicate the actual measurements of patients at baseline (before second-line treatment), after 6 and 12 months. Black dots (with error bars) represent the counterfactual predictions and 95% CIs, supposing all patients were treated with that drug class. The results of the Bayesian mixed-treatment comparison (MTC) meta-analysis by McIntosh et al are marked MTC. DPP-4, dipeptidyl peptidase 4; GLP-1, glucagon-like peptide-1 receptor agonists; HbA1c, glycated hemoglobin; SU, sulfonylurea; TZD, thiazolidinedione.

Figure 3

Predicted and observed BMI levels using doubly robust estimation adjusting for either a comprehensive set of confounders (left panel) or a set of confounders provided by a domain expert (right panel). Red dots indicate the actual measurements of patients at baseline (before second-line treatment), after 6 and 12 months. Black dots (with error bars) represent the counterfactual predictions and 95% CIs, supposing all patients were treated with that drug class. BMI, body mass index; DPP-4, dipeptidyl peptidase 4; GLP-1, glucagon-like peptide-1 receptor agonists; SU, sulfonylurea; TZD, thiazolidinedione.