Achieving glycemic goal with initial versus sequential combination therapy using metformin and pioglitazone in type 2 diabetes mellitus.

OBJECTIVE

To compare glycemic goal achievement (HbA(1c) < 7%) in type 2 diabetes patients receiving initial metformin plus pioglitazone combination therapy and initial metformin monotherapy augmented with pioglitazone in a cohort follow-up study.

RESEARCH DESIGN AND METHODS

Adult patients were identified from the Ingenix Impact database (01/01/99-03/31/07). Qualified patients had a baseline HbA(1c) ≥ 7%; a second laboratory value within 9 months; no other anti-diabetic prescriptions 6 months before or 30 days after treatment initiation; and continuous enrollment during baseline. The index date was the date on which the second medication was initiated. Goal achievement was compared independently at 6, 12 and 18 months using a chi-square test. Logistic regression was used to control for baseline differences. Last observation carried forward was used to impute missing HbA(1c) values. Sub-group analysis was conducted on patients with baseline HbA(1c) values between 7% and 9%, and >9%.

MAIN OUTCOME MEASURES

The proportion of patients achieving glycemic goal at each specified time point.

RESULTS

A total of 179 patients received initial combination therapy and 347 patients received sequential therapy. A greater proportion of initial combination patients achieved the glycemic goal compared to sequential patients at months 6, 12 and 18 (66.5 vs. 49.6%; 65.9 vs. 48.1%; 65.9 vs. 48.4%, respectively; p < 0.001 for all). Logistic regression confirmed these findings (odds ratios [OR]: 3.18-3.31). Sub-group analysis showed a more pronounced advantage for aggressive initial combination treatment among patients with HbA(1c) > 9% (OR: 5.39-6.04) than among patients with HbA(1c) between 7% and 9% (OR: 2.28-2.79).

CONCLUSIONS

Initial combination therapy patients are more likely to achieve glycemic control than sequential therapy patients, especially for patients with baseline HbA(1c) > 9%. This study is limited by the relatively small sample size and the frequency of HbA(1c) reporting. Future research could examine goal achievement using a larger sample and more complete laboratory data to confirm these findings.