Claims‐based studies of oral glucose‐lowering medications can achieve balance in critical clinical variables only observed in electronic health records

Abstract

Background: Healthcare claims databases can provide information on the effects of type 2 diabetes (T2DM) medications as used in routine care, but often do not contain data on important clinical characteristics, which may be captured in electronic health records (EHR). Objectives: To evaluate the extent to which balance in unmeasured patient characteristics was achieved in claims data, by comparing against more detailed information from linked EHR data. Methods: Within a large US commercial insurance database and using a cohort design, we identified T2DM patients initiating linagliptin or a comparator agent within class (i.e., other DPP-4 inhibitors) or outside class (i.e., (pioglitazone or sulfonylureas) between 05/2011-12/2012. We focused on comparators used at a similar stage of diabetes as linagliptin. For each comparison, 1:1 propensity score (PS) matching was used to balance over 100 baseline claims-based characteristics, including proxies of diabetes severity and duration. Additional clinical data from EHRs was available for a subset of patients. We assessed representativeness of the claims-EHR linked subset, evaluated the balance of claims- and EHR-based covariates before and after PS-matching via standardized differences (SD), and quantified the potential bias associated with observed imbalances. Results: From a claims-based study population of 166,613 T2DM patients, 7,219 (4.3%) patients were linked to their EHR data. Claims-based characteristics between the EHR-linked and EHR-unlinked patients were comparable (SD<0.1), confirming representativeness of the EHR-linked subset. The balance of claims-based and EHR-based patient characteristics appeared to be reasonable before PS-matching and generally improved in the PS-matched population, to be SD<0.1 for most patient characteristics and SD<0.2 for select laboratory results and BMI categories, not large enough to cause meaningful confounding. Conclusion: In the context of pharmacoepidemiologic research on diabetes therapy, choosing appropriate comparison groups paired with a new user design and 1:1 PS matching on many proxies of diabetes severity and duration improves balance in covariates typically unmeasured in administrative claims datasets, to an extent that residual confounding is unlikely.

Comparative risk of genital infections associated with sodium‐glucose co‐transporter‐2 inhibitors

“…However, it is unlikely that the risk of genital infections would vary among patients with different insurance types. Fourth, because this was a claim‐based study, the analysis did not control for some important variables such as duration of diabetes or body mass index; however, claims‐based proxies have been shown to be good surrogates for these characteristics …”

Cardiovascular safety of linagliptin compared with other oral glucose‐lowering agents in patients with type 2 diabetes: A sequential monitoring programme in routine care

“…The comparators for linagliptin, that is, other DPP‐4 inhibitors (alogliptin, saxagliptin or sitagliptin), pioglitazone or second‐generation sulphonylureas (glimepiride, glipizide or glyburide), were chosen as they represent common therapeutic strategies used at comparable stages of diabetes progression. This approach would be expected to improve clinical equipoise across treatment groups and to reduce confounding …”

“…Covariates of interest included demographics, comorbidities, use of medications and indicators of healthcare utilization as proxy for overall disease state and care intensity (Tables and S2). As previously reported, emphasis was placed on the identification of claims‐measured indicators of diabetes severity, for example, the number of glucose‐lowering medications at index date and specific past or concurrent diabetes therapy, diabetic nephropathy, neuropathy, retinopathy, diabetic foot, number of glycated haemoglobin (HbA1c) or glucose tests ordered . Comorbidities were defined using ICD‐9 or ICD‐10 diagnosis codes and Current Procedural Terminology‐4 codes.…”

Emulation Differences vs. Biases When Calibrating Real‐World Evidence Findings Against Randomized Controlled Trials

“…If adjustment for additional covariates or use of an adjustment approach that better balances covariates can result in RWE effect estimates that are closer to the effect estimate from the RCT, then this exploration can provide evidence that initial estimates may have been biased due to confounding. Sensitivity analyses exploring unmeasured confounding, such as evaluation of control outcomes or evaluation of confounders measured in a subset of the RWE study population can also substantiate some concerns about bias . Similarly, sensitivity analyses that consider a somewhat more or less restrictive implementation of eligibility/exclusion criteria of the RWE study compared with the trial will illuminate the emulation success.…”