Taming the Wild West: regulating digital health apps

O

ver the past few years, health-related smartphone apps have become a force to be reckoned with. Whether you’re looking for a diabetes app for monitoring blood glucose, a brain training app to help beat brain fog, or nutrition advice based on your own genetic profile, there’s an app for that. Especially since the start of the pandemic, digital health tools of this kind have surged in popularity.

On the one hand, the proliferation of apps has many benefits. Apps can be empowering for users, allowing them to take aspects of their healthcare into their own hands and improving the accessibility of services. Mental health apps, for instance, could play a key role in mitigating our present mental health crisis, at a time when existing provisions are struggling to meet demand.

Apps for specific conditions could also improve patient outcomes, for instance by reminding users to take their medications or nudging them to adjust their behaviours. They can monitor patients’ vital signs and flag up any troubling results that emerge.

“Apps hold great potential to extend the capabilities of traditional healthcare systems, for instance by supporting us in our self-care or health-promoting behaviours,” says Anna Essén, an associate professor at the Stockholm School of Economics. “In some cases, these activities can reduce the need for healthcare or make healthcare encounters more effective. In short, apps can be a tool for improving health and the effectiveness of healthcare systems.”

On the other hand, there is little to say a given app will actually provide these benefits. There are thousands of poor quality apps in circulation, which may contain inaccurate information, compromise user privacy and even jeopardize patient safety.

“Studies repeatedly show that the quality available in commercial app stores, as well as curated lists, often fail to fulfil basic quality requirements,” says Essén. “For instance, some apps do not comply with the latest medical guidelines and therefore do not represent evidence-based medicine. Even in cases where apps have a ‘data policy’ they sometimes do not comply with it if we look more closely at how data is shared.”

To run through a few particularly egregious examples: one app, for bipolar disorder, advised users to ‘take a shot of hard liquor’ to help them fall asleep. Another app claimed to measure your unborn baby’s heartbeat using a smartphone camera or sensors, but had not performed any testing to back up its claims. One popular blood pressure app measured blood pressure inaccurately 77.5% of the time. 

On the privacy front, a 2021 analysis found that health apps had ‘serious problems’ with patient privacy. While only 4% actually transmitted data to third parties, 88% included code that could potentially share that data – a real issue when it comes to sensitive health-related information.
 
Another 2021 study found room for improvement on four key metrics – not just accuracy and privacy, but also usability and accessibility.

“Does the app take into account the fact that users may not be digital natives, or may have hearing and visual impairments?” says study author Cara Tenenbaum, a health policy advocate and principal at Strathmore Health Strategy. “How do you make sure that your user – whether they are a migrant worker in California with limited WiFi access, or an underserved Medicaid beneficiary in Chicago – is actually able to use the app?”

Up till now, regulatory oversight in this field has proven challenging, to the point that the current app landscape is often compared to the Wild West. For sure, it’s a fast-emerging and exciting field, but there is little to stop an app developer from selling snake oil.

In the US, apps that provide treatment advice are considered medical devices, requiring them to be regulated by the FDA. In practice, though, only a tiny fraction of health apps are actually overseen this way. Many are able to bypass the paperwork by listing themselves as ‘health and fitness’ apps, a category that doesn’t require such stringent oversight.

Given the sheer numbers hitting the app store, and the frequency with which they update, asking the FDA to intervene further may not be feasible.

“I feel like we don't want to leave app oversight to postmarket review and recalls, because it's so, so important,” says Tenenbaum. “On the other hand, we don't have a regulatory system right now that can do the premarket approvals. The FDA doesn't have the resources even if it did have the perfect framework.”

In the UK, some (but not all) health apps need to be registered with the national regulator, while those offering medical advice require a CE mark. Up until recently, the NHS offered an apps library – a list that had been assessed against NHS standards – but this has now been decommissioned, in favour of linking to apps throughout its website.

Other countries are feeling their way towards their own frameworks. Essén recently worked on a study comparing health app policy within nine countries. She found that Belgium and Germany had come the furthest. Although there are many uncertainties at play, both have put processes in place for market and reimbursement approval.

“The question is, is the regulation of apps a question for consumer goods agencies or for healthcare regulators?” says Essén. “Typically, different agencies are responsible for regulations in these two areas. That is why many countries are trying to develop additional standards allowing a better filtering of apps. Who should enforce these guidelines, however, is still an open question in many countries.”

Essén believes that most countries will eventually settle on a dual approach, in which some apps are formally evaluated and others receive a more basic quality check.

“Several countries are employing centralised approaches with one major approving unit,” she says. “However, there is also a discussion about more decentralised approaches where third party actors can be accredited to be ‘app assessors’. Overall, most expect that a combined approach will emerge, where some features of some apps are assessed centrally while others continue to be assessed by the market.”

Tenenbaum says that while we are a long way off finding ‘the magic wand solution’, there are many new suggestions being floated, with the topic attracting more attention from all quarters.

“Could there be a way that provider groups could work with patient groups or other outside experts to review these apps?” she says. “Maybe there’s a kind of building block model, where third parties review the app as it iterates, and then the regulatory agency performs a more streamlined review.”

She cautions, however, that any regulatory framework that’s developed needs to be careful to protect innovation.

“The idea that developers might have to go through numerous layers of red tape, which could be costly and stifle innovation, is hard to swallow,” she says. “But I think it’s more important to protect the patients. So how do you balance that? That question is going to be crucial for everyone involved.”

In a sense, what we’re seeing now is a culture clash between the fast-moving world of app development, and the cautious world of medtech regulation. With patient safety at stake, it’s clear that we need some good ideas around how these two worlds can be bridged.

Main image: BigTunaOnline / Shutterstock.com