Initial nonadherence, primary failure and therapeutic success of metformin monotherapy in clinical practice

Curr Med Res Opin. 2010 Sep;26(9):2127-35. doi: 10.1185/03007995.2010.504396.


Objective: To document the extent to which drug naïve patients with diabetes continued newly initiated metformin monotherapy, and who subsequently experienced primary failure or therapeutic success.

Research design and methods: Using an observational longitudinal cohort design, we identified all 3116 type 2 diabetes patients who initiated metformin monotherapy as their first-ever anti-hyperglycemic drug in 2004-2006. We defined initial nonadherence as having occurred in the first 6 months if a patient (1) received only a single metformin dispense, (2) received less than a 90 day supply, or (3) switched to a second anti-hyperglycemic agent. Among those patients who continued metformin monotherapy for at least 6 months, we determined the proportion attaining A1C < 7% by examining A1Cs recorded prior to adding a second anti-hyperglycemic agent or December 31, 2008, whichever occurred first. We identified factors associated with achieving A1C < 7% with logistic regression.

Results: Initial nonadherence occurred in 518 (16.6%) of the 3116 patients studied. Of those who continued metformin for at least 6 months, a majority (71.7% overall) achieved an A1C < 7%, even when pre-metformin A1C was high (59.6% among those with A1C > 9%). Initiating metformin within 3 months of diagnosis increased the probability of success nearly three-fold compared to initiating after 12-23 months (OR 2.85, 95% CI 2.04-3.98). Independent of duration, lower A1C at initiation increased the probability of success.

Conclusions: Initiating metformin immediately at diagnosis and while A1C is still low substantially increases the probability of attaining glycemic success. Our findings support the current guidelines recommending metformin therapy as soon as type 2 diabetes is diagnosed. This study was limited by its observational design, the inability to assess reasons for metformin nonadherence, and the highly automated study setting.

Publication types

  • Evaluation Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Aged
  • Algorithms
  • Cohort Studies
  • Databases as Topic
  • Diabetes Mellitus, Type 2 / drug therapy*
  • Diabetes Mellitus, Type 2 / epidemiology*
  • Diabetes Mellitus, Type 2 / psychology
  • Female
  • Humans
  • Hypoglycemic Agents / adverse effects
  • Hypoglycemic Agents / therapeutic use
  • Longitudinal Studies
  • Male
  • Metformin / adverse effects
  • Metformin / therapeutic use*
  • Middle Aged
  • Oregon
  • Patient Compliance / statistics & numerical data*
  • Professional Practice
  • Treatment Failure
  • Treatment Outcome
  • Washington


  • Hypoglycemic Agents
  • Metformin