Abstract
Background and Aims
The CREDENCE trial with canagliflozin demonstrated definitive evidence of renal benefits to slow the progression of end stage renal disease (ESRD) in type 2 diabetes (T2D) patients with chronic kidney disease (CKD). This real-world study was undertaken to better understand the prevalence of CKD among T2D patients in Ontario using the CREDENCE trial criteria. Ontario is the largest province, accounting for 38.8% of Canada’s population in 20191. Patients were identified in the following cohorts: T2D-CKD, T2D-CKD+cardiovascular disease (CVD), T2D-CKD+stroke, and T2D-CKD+CVD or stroke.
Method
This population-based retrospective cohort study was conducted in partnership with IQVIA and the Institute for Clinical Evaluative Sciences (ICES). The ICES data repository contains publicly funded administrative health service records for the Ontario population eligible for universal health coverage since 1986. Patients’ eligibility for this study was aligned with enrolment criteria in the CREDENCE trial. Patients were ≥30 years of age at index and were identified as having both T2D and CKD. Diabetes patients were identified using the validated Ontario Diabetes Database (ODD), patients <19 years of age when first diagnosed with diabetes were excluded due to suspected type 1 diabetes (T1D). Additionally, patients with a T1D diagnosis at any time-point were excluded. CKD patients were identified through diagnosis/billing codes or estimated glomerular filtration rate(eGFR)<90 ml/min/1.73ml2 derived from serum creatinine laboratory values. Diagnosis/billing codes are expected to have poor sensitivity2,3 when used as the sole method to identify CKD. To account for anticipated missing information in each dataset, the capture-recapture method was used to obtain a more accurate estimate for the total prevalence. Capture-recapture accounts for incomplete ascertainment of administrative datasets by using the overlap between the datasets to derive an estimate of the total population4,5 (Figure 1). This method was used in Manitoba to estimate the total CKD population, using administrative and laboratory datasets4. Therefore, each cohort (T2D-CKD, T2D-CKD+CVD, T2D-CKD+stroke, and T2D-CKD+CVD or stroke) has utilized each of the following four methods to identify CKD patients: diagnosis/billing codes, eGFR, diagnosis/billing codes or eGFR, and capture-recapture method. Yearly point prevalence of CKD among T2D patients is reported for the five fiscal years (FY) between 2011/12 and 2015/16.
Results
The prevalence of T2D patients ≥30 years of age in Ontario has increased from 959,850 in FY2011/12 to 1,169,759 in FY2015/16 (Figure 2). The prevalence of CKD among T2D patients ≥30 years of age in Ontario has increased across all methods from FY2011/12 to FY2015/16: from 21% to 28% based on diagnosis/billing codes, 47% to 63% based on eGFR, 55% to 70% based on diagnosis/billing codes or eGFR, and 76% to 84% based on capture-recapture. Similarly, prevalence of T2D-CKD+CVD, T2D-CKD+stroke, and T2D-CKD+CVD or stroke has increased in most cases (Figure 3).
Conclusion
CKD is a common comorbidity amongst T2D patients ≥30 years of age. The study provides estimates of the prevalence of CKD in four cohorts of T2D patients with defined co-morbidities and shows that the use of diagnosis/billing codes alone may underestimate the prevalence of CKD in T2D patients.
Furthermore, this real-world analysis highlights a significant, increasing prevalence of CKD among T2D patients ≥30 years of age in Ontario with all methods. On-going research aims to assess the burden of illness of patients with both T2D and CKD who are incident to T2D-related outcomes (CKD or CVD related death, kidney transplant, kidney dialysis, doubling of serum creatinine).
Use of real-world evidence data to evaluate the comparative effectiveness of second-line type 2 diabetes medications on chronic kidney disease
