Three in 10 Doses of Metformin Not Taken by Patients With Type 2 Diabetes

Liam Davenport

December 22, 2017

Patients with type 2 diabetes are least likely to adhere to the most commonly prescribed medication for their disease, potentially because of its adverse-event profile, suggest UK researchers.

The review of almost 50 studies showed that patients did not take 30% of prescribed doses of metformin, compared with 23% of doses of sulfonylureas, such as gliclazide, and 20% of thiazolidinediones, such as pioglitazone.

The research, which was published online December 12 in Diabetes, Obesity and Metabolism, also showed that the newer dipeptidyl peptidase-4 (DPP-4) inhibitors, or gliptins, had higher adherence rates, with patients not taking just 10% to 20% of doses.

As well as the adverse events associated with the different classes of medication, the fact that some drugs have to be taken in multiple daily doses may also have an impact, they say.

"We have known for a long time that a lot of medication prescribed for chronic diseases never actually get taken," said lead author Andrew McGovern, BMBS, from the department of clinical and experimental medicine, University of Surrey, Guildford, United Kingdom, in a press release from his institution:

"What this latest research suggests is that patients find some of these medication classes much easier to take than others."

Adherence Is Key in Type 2 Diabetes; Medics Should Ask About Adherence

Dr McGovern emphasized that "the importance of diabetes patients taking their prescribed medication cannot be underestimated. A failure to do so can lead to complications in their condition, including eye disease and kidney damage."

"I urge anyone who is struggling to take their medication as prescribed, whether this is because of side effects or because the schedule is too complicated, to discuss this openly with their doctor or nurse."

He also encouraged doctors and nurses to ask their patients about medication adherence: "Fortunately for type 2 diabetes, we have lots of treatment options, and switching to a different medication class that is easier to take could provide an easy way to improve adherence."

To examine the contribution of specific medications to nonadherence in type 2 diabetes, Dr McGovern and his team searched the MEDLINE and Embase databases, the Cochrane Library, the Register of Controlled Trials, PsychINFO, and the Cumulative Index to Nursing and Allied Health Literature (CINAHL) and did bibliographic searches.

They looked for interventional and observational studies assessing medication adherence or persistence with medications from at least two distinct classes of glucose-lowering agent published between 2006 and 2015.

From over 4000 possible trials, 48 studies met the inclusion criteria and were included in the analysis, yielding a total of 1,696,939 individuals with type 2 diabetes.

Twenty-five studies compared oral therapies only, 19 compared only injectable therapies, and three compared oral and injectable therapies. The final study compared an oral with an inhaled agent.

The team notes that the definitions of adherence varied widely across the studies, with different parameters used, and there was even greater disparity in the definitions of medication persistence.

They also observe another limitation: the fact that most studies were funded by pharmaceutical companies. This funding source, they say, "may increase the risk of bias toward results favoring newer pharmacotherapies, although we were not able to assess this."

Switch Medication Class to Improve Poor Adherence

The authors found that compared with metformin, adherence was better with sulfonylureas, at a mean difference in adherence of 10.6% across five studies, and with thiazolidinediones, at a mean difference of 11.3% across six studies.

Adherence to thiazolidinediones was slightly better than that seen with sulfonylureas, at a mean difference across five studies of 1.5%.

DPP-4 inhibitors had better adherence and were taken for longer than both sulfonylureas and thiazolidinediones in almost all studies that compared the medications.

One study included a sodium-glucose cotransporter 2 inhibitor (SGLT2), canagliflozin (Invokana, Janssen), which had better persistence than did DPP-4 inhibitors.

Compared with long-acting insulin analogs, injectable glucagonlike peptide-1 (GLP-1) agonists were associated with an increased risk for discontinuation, at an odds ratio of 1.95 across six studies.

Long-acting insulin analogs were also taken for longer than human insulins, at a mean difference of 43.1 days across five studies.

Although metformin is the generally recommended first-line agent for type 2 diabetes worldwide, other medications can be considered if it is not well tolerated or contraindicated. The fact that metformin commonly results in gastrointestinal disturbance may contribute to the lower adherence with this agent, the researchers speculate.

The team writes that the most recent American Diabetes Association and European Association for the Study of Diabetes guidelines for diabetes management "give equipoise to a range of second-line therapies."

"Given this equipoise, medication class switching may be an effective method for improving adherence in people with poor adherence," they add, noting that this approach should be assessed in interventional studies.

"We suggest that class selection should consider future adherence and persistence. Where all else is equal, medication classes with better adherence …should be preferred."

Eli Lilly provided funding for this study. Dr McGovern has received research funding from AstraZeneca and Eli Lilly. Disclosures for the coauthors are listed in the paper.

Diabetes Obes Metab. Published online December 12, 2017. Abstract

For more diabetes and endocrinology news, please follow us on Twitter < and on Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.