Rate of hypoglycemia in patients with type 2 diabetes receiving metformin plus saxagliptin versus metformin plus sulfonylurea: a retrospective observational cohort study using administrative claims data

Abstract

Objectives: Hypoglycemia is a limiting factor in the management of diabetes. Studies comparing oral antidiabetic medications are needed to identify treatment options that can help clinicians and patients minimize their associated hypoglycemia risk. The purpose of this study was to compare hypoglycemia rates in patients with type 2 diabetes on metformin who initiated treatment with saxagliptin versus sulfonylurea (SU). Methods: This retrospective analysis utilized US healthcare claims data from the Truven Health MarketScan Research Databases. Data were from adults on metformin monotherapy who added saxagliptin or SU between 1 August 2009 and 31 December 2010. Hypoglycemia event rates were compared during the 4 months after initiation of saxagliptin or SU. A hypoglycemia event was defined as a diagnosis of hypoglycemia on an outpatient or emergency room claim, a principal diagnosis on a hospital claim, or a glucagon injection in an outpatient setting. Patients taking SU were matched to patients taking saxagliptin (5 to 1) using propensity scores, and the rate ratio was further adjusted using multivariate regression. A total of 22,592 patients (1567 taking saxagliptin; 21,025 taking SU) qualified. Results: During 120 days of follow-up, there were 396 hypoglycemia events. Most of the hypoglycemia events (91.9%) occurred in the outpatient setting. There were no inpatient or emergency room hypoglycemia events in the saxagliptin cohort. The overall unadjusted rate of hypoglycemia was significantly lower in the saxagliptin cohort than in the SU cohort (1.74 vs 5.58 per 100 person-years; p < 0.001). The rate of hypoglycemia also was significantly lower in the saxagliptin cohort versus the propensity-matched SU cohort (1.74 vs 4.45 per 100 person-years; p = 0.005). Matching reduced the treatment effect by approximately 20%. The rate ratio comparing saxagliptin with the unmatched and propensity-matched SU cohorts was 0.31 (95% CI: 0.14–0.6) and 0.39 (95% CI: 0.17–0.77), respectively. The multivariate adjustment decreased the hypoglycemia rate ratio 0.37 (95% CI: 0.19–0.74). Conclusion: In a database reflective of real-world clinical practice, saxagliptin had a lower risk of hypoglycemia than SU in patients with type 2 diabetes receiving metformin. These results add confidence to similar findings from clinical trials.

Keywords::

• healthcare claims data
• hypoglycemia
• saxagliptin
• sulfonylurea
• type 2 diabetes mellitus

Author contributions

SM Curkendall was involved in conceptualization of the study, participated in its design and coordination and oversaw review and interpretation of results. B Zhang was involved in conceptualization of the study and participated in its design and coordination. G Lenhart participated in the design of the study and conducted the statistical analyses. E Thomson participated in conducting the analyses. SM Curkendall and E Thomson drafted the manuscript. KF Bell was involved in conceptualization of the study and review and interpretation of results. All authors read and approved the final manuscript.

Acknowledgements

The authors would like to thank M Shaw for independent writing assistance and J Kagan and L Lee, employees of Truven Health Analytics, for programming and editorial assistance, respectively.

Financial & competing interests disclosure

This study was conducted by Truven Health Analytics and funded by Bristol-Myers Squibb and Astra Zeneca LP. SM Curkendall was an employee of Truven at the time the work was conducted. B Zhang is an employee of Bristol-Myers Squibb. G Lenhart, E Thomson and KF Bell are employees of Truven. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Writing assistance was utilized in the production of this manuscript.

Key issues

• This study using insurance claims data compared saxagliptin with sulfonylureas (SUs) and confirmed clinical trial results that showed much lower rates of hypoglycemia for type 2 diabetes patients taking saxagliptin added to metformin compared with those taking glipizide (a sulfonylurea) added to metformin.

• In this study, hypoglycemia was defined using diagnoses recorded on insurance claims. In the clinical trial, hypoglycemia was defined as patient reports of signs or symptoms consistent with hypoglycemia or reported low glucose levels or hypoglycemia confirmed using a finger stick glucose test.

• This study counted hypoglycemia events that occurred in the community and were treated in one of three treatment settings: outpatient, emergency room or inpatient. The majority of hypoglycemia events were treated in the outpatient setting. Patients in the saxagliptin cohort had no hypoglycemia events that were treated in the emergency room (ER) or inpatient setting.

• The advantages of using claims data are that they are more reflective of what happens in real-world clinical practice than a clinical trial and the sample sizes are much larger. Also, because an insurance claim is required to identify hypoglycemia, a study in claims data identifies hypoglycemia that requires medical attention and the use of healthcare resources. However, claims data could underestimate hypoglycemia because patients who experience hypoglycemia and do not go to the doctor for treatment cannot be counted.

• One limitation of using claims data to study hypoglycemia is that the diagnosis used to identify most of the hypoglycemia events was ICD-9-CM code 250.8 ‘diabetes with other specified manifestations’. This code can also be used for other diabetes complications. This limitation was minimized by using a validated algorithm to remove claims with other complications.

Notes