Send to

Choose Destination
Curr Med Res Opin. 2015;31(7):1283-96. doi: 10.1185/03007995.2015.1053048. Epub 2015 Jun 18.

Meta-analysis of studies examining medication adherence, persistence, and discontinuation of oral antihyperglycemic agents in type 2 diabetes.

Author information

Merck & Co., Inc. , Kenilworth, NJ , USA.



To estimate overall rates of adherence, persistence, and discontinuation for patients with type 2 diabetes mellitus (T2DM) prescribed oral antihyperglycemic agents (OAHAs) by combining results of published studies.


A systematic literature review was conducted to identify articles published in English over the last 10 years evaluating the use of OAHAs for the treatment of T2DM. Databases searched included PubMed/MEDLINE, EMBASE, and the Cochrane Library. Seventy studies reporting adherence, persistence or discontinuation were identified by two independent reviewers and 40 reported relevant endpoints for the analysis. Outcomes included: (1) mean adherence defined as the average medication possession ratio (MPR); (2) proportion of adherent patients (MPR ≥ 80%); (3) discontinuation; and (4) persistence. Adherence and persistence were reported in observational studies only. Discontinuation was examined separately in randomized controlled trials (RCTs) and observational studies. Meta-analyses were conducted using both fixed and random effects models. When meta-analysis was not appropriate for a given outcome, descriptive statistics were provided.


The pooled mean MPR (95% confidence interval [CI]) was 75.3% (68.8%-81.7%; n = 13) and the proportion of adherent patients (95% CI) was 67.9% (59.6%-76.3%; n = 12). The discontinuation rate (95% CI) in RCTs was 31.8% (17.0%-46.7%; n = 7). Persistence and discontinuation were not assessed via meta-analysis for observational studies due to the limited number of available studies and differences in outcome definitions. In these studies, persistence estimates ranged from 41.0% to 81.1%, with a mean (95% CI) of 56.2% (46.1%-66.3%; n = 6), while discontinuation estimates ranged from 9.9% to 60.1%, with a mean (95% CI) of 31.4% (17.6%-45.3%; n = 6).


Limitations include (1) the use of MPR as a proxy for adherence, (2) limited number of studies available, and (3) observed heterogeneity.


The results of the analysis demonstrate that medication adherence, persistence, and discontinuation rates are suboptimal in patients with T2DM prescribed OAHAs.


Adherence; Discontinuation; Oral antihyperglycemic agents; Persistence; Type 2 diabetes

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Taylor & Francis
Loading ...
Support Center