Quality, Safety of Health Apps Remains Murky

We have come to a place in mobile health (mHealth) where the problem is no longer a lack of available apps. Patients and healthcare providers are using health-related apps on their smartphones. Research studies have shown promising evidence that certain disease outcomes can be improved with the implementation of a mobile app.

The issue is now ensuring the quality and safety of patient-facing apps. It is unclear if apps can cause harm for certain users. There is a need for more research on the validity of health app reviews. Few healthcare providers are involved in the design of health apps, and no consensus exists on how medical professionals should evaluate and recommend an app to patients. Without guidance from the medical establishment, consumers can be led into using "digital snake oil."

The term was used by James Madara, MD, CEO of the American Medical Association (AMA), during the 2016 Annual AMA meeting in Chicago. Madara was talking about the current landscape of medical IT innovation, and what he saw as the lack of evidence and medical oversight in a new area of healthcare that could present potential dangers for patients.

Last fall, during the 2016 Health 2.0 conference in Silicon Valley, the mHealth community invited Madara on center stage to explain his thinking. As a physician attending this conference with mainly health technology innovators, I agreed among much contention with Madara about the need for more evidence and involvement from the medical establishment.  Researchers from Brigham and Women's Hospital and the University of Michigan School of Medicine addressed these issues with a publication in Health Affairs. Titled "Many Mobile Health Apps Target High-Need, High-Cost Populations, But Gaps Remain," the article reviewed apps marketed to high-need, high-cost (HNHC) patients. The authors used specific criteria to determine the utility, usability, and safety of these apps. This paper provides the groundwork needed for key players in healthcare to become more involved in mHealth.

In their study, the authors conducted a search for patient-facing mHealth apps in iTunes and Google Play during February 2015. Keywords for chronic disease and HNHC risk factors were used as search terms. Recommendations from medical professional societies and expert interviews were also used to narrow down to a final sample of 137 apps that specifically targeted HNHC populations.

The apps were evaluated by both clinician and non-clinician reviewers. Qualitative assessments were used to analyze each app. Clinician reviewers rated apps on a 0-10 scale on whether they would recommend the app to a friend or colleague. They also determined whether appropriate action was taken when users input dangerous health information. Non-clinician reviewers evaluated the usability of an app using a validated scale. All reviewers characterized the type of patient engagement used, app store rating, privacy policy, cost model, and mechanism of data sharing.

This study provides interesting data on the current landscape of mHealth. Consumer app store reviews correlated poorly with clinical utility and usability scores recorded by the study reviewers.

Overall, 121 apps in the sample size of 137 had a function to record health information. Only 28 of these responded appropriately to dangerous user input. The majority of apps also shared health information by insecure methods such as e-mail or text messaging. Only 64% of apps had a privacy policy. The cost was essentially free for the vast majority of apps, and a minority of apps used what the reviewers considered an advanced form of user engagement.

With the prevalence of mHealth applications being marketed to consumers, there is potential for innovation to improve outcomes in a segment of the population that has been refractory to conventional methods. This also carries a potential for harm, especially when important factors like health information security and patient safety are inadequate.  It is important to remember that consumer reviews are just as important as a clinician's.  Both offer their own perspective on what they prioritize as important features. But medical professionals have the training and knowledge base to understand the safety and regulatory aspects meant to protect consumers from potential harms. Reliance on anecdotal ratings is not enough for consumers to determine the quality of an app. It is concerning that the majority of apps in this study did not have a safe mechanism for handling a dangerous user input such as suicidality.

This article also reviews the multitude of regulations that apps are already subject to: The FDA enforces safety and effectiveness, the FTC protects against misleading claims, and the Office for Civil Rights, a branch of the Department of Health and Human Services, enforces HIPAA regulations. Adding more rules and oversight may not be the right strategy, as this could be stifling to innovation.

Instead, this paper recommends that medical professional societies directly create initiatives to help patients find and understand content on apps. This suggestion is similar to one brought up by Madara. One interesting idea postulated by the authors was to create "app labels" similar to ones found on food packaging to help consumers understand the content of the product they are purchasing.

The stage is set for future researchers and developers to address the areas of improvement outlined in this study. Recent innovations in mHealth applications have created a diverse consumer market, but relatively little data exist to validate their safety and effectiveness. This paper shows that there are many gaps that need to be addressed.

Medical professionals and patient advocates need to begin working with innovators to ensure development continues in areas of safety and usability. Healthcare providers need to start addressing the use of mHealth with their patients. Consumer ratings should be supported by an established resource for objective analysis. If these gaps can be closed, the mHealth community can prevent digital snake oil from pervading the future of healthcare. 

By: David Tseng, MD

Original posting:  iMedical Apps - MedPage Today, 2017/05/18