Study Evaluates Whether Diabetes Self-Management Apps are Effective

While hundreds of apps for diabetes self-management are commercially available, diabetes patients lack information on which apps improve diabetes-related outcomes, according to a study.

The study, published in the Journal of General Internal Medicine, conducted a rapid evidence review to examine features, clinical efficacy, and usability of apps for self-management of type 1 and type 2 diabetes in adults.

Fifteen articles evaluating 11 apps were identified: six apps for type 1 and five apps for type 2 diabetes. Of these 11 apps, only five were associated with clinically significant improvements in HbA1c, an important clinical test for monitoring diabetes.

Lead study author Stephanie Veazie, MPH, Research Associate, Scientific Resource Center, Portland VA Research Foundation, and colleagues searched both recently published studies and systematic reviews for data on commercially available apps. They evaluated these primary studies for risk of bias (i.e., quality) to determine if the results could be trusted. The researchers then looked for the apps online and evaluated their features, privacy policies, and usability. They then organized the information by app to help inform patient, provider, and health system decisions on whether to use a diabetes app, and which app to use. The report was reviewed by experts in both diabetes and mobile health.

The study found:

• Of the 11 apps identified in the literature search, two had two tiers of access, for a total of 13 unique apps. Some of these apps were not available for download or required a prescription for use; therefore only 8 were reviewed for usability, privacy, and security.
• Of the eight apps the researchers were only able to download, only two were rated as having “acceptable” usability. The other six were rated as having “marginal” or “not acceptable” usability. This suggests patients may have a difficult time using these apps.
• Of the eight apps available for review, only five had clear privacy and security policies.

“None of the studies showed patient improvements in quality of life, blood pressure, weight, or body mass index,” says Veazie. “The studies had methodological limitations. More rigorous and longer-term research studies could determine whether apps help people manage their diabetes and reduce complications.”

Patients within health systems should be aware that most commercially apps for diabetes self-management are not supported by evidence, according to Veazie. “Most of the studies were short-term and it’s unclear if app use can lead to sustained improvements. In the studies we found, apps were combined with support from a provider or study staff member. Therefore, it is not clear whether the app or the additional interaction with a provider or study staff led to the improvements.”

Based on the study, here are three things Veazie wants healthcare executives to know about mHealth for diabetes care management:

• Although there are hundreds of commercially available apps for diabetes self-management, only a few have been rigorously evaluated. If you are contemplating using an app in your healthcare system, consider whether there is evidence it improves patient outcomes.
• Apps for diabetes self-management are variable in the amount of privacy and security they offer. This may determine the extent to which patient’s health information is being collected and used by the app company or others.
• App usability was also variable. Although there may not be a “one-size-fits-all” approach to app selection, be aware that patients may have difficulty using some of these apps.

The study was funded by the Agency for Healthcare Research and Quality (AHRQ).