Digital therapeutics are being touted as having the potential to transform health care by improving people’s experience, increasing effectiveness, and reducing costs. A few digital aids are recommended and integral to health services, but much of the e-health field depends on overselling [1]. This business plan seems to be working as the digital health field was estimated to be worth $25 billion globally in 2017 [2,3]. One US survey found that 58% of smartphone users have downloaded at least one health app [4]. The overselling of health apps needs the health community’s immediate attention as many of these apps may provide little benefit and some apps may cause harm. With little formal regulation, alternative light-touch approaches to consumer protection are necessary to give customers a modicum of information as a basis for choosing from the vast array of so-called health apps.

We believe that simple, but informative, evidence should be available at the point of downloading and propose 4 succinct aspects that represent the critical information required for responsible health app marketplaces. We refer to these 4 principles—privacy and data security, development practices, feasibility, and health benefits—as the Transparency for Trust (T4T) principles. The goal of these principles is to operationalize efforts that can be made by app marketplaces to answer calls for better oversight and backing up products with data and research [3,5,6]. These T4T principles draw from several sources including patient and regulatory perspectives, recent systematic reviews, and experimental studies (eg [4,7-14]). We have used the definition of a health app from Innovate UK as those apps that contribute to the physical, mental, or social well-being of the user [15]. Our principles are designed to be applicable to the whole panoply of health apps from sleep apps to diabetes apps, symptom trackers to mindfulness interventions eg[16-18]. The fastest growing segment of digital therapeutics is health apps. Current estimates suggest that over 300,000 health apps exist with a recent yearly growth rate of 25% [17]. The rapid growth of products and lack of growth of information and regulation have resulted in very little information to separate quality health apps from those that are at best useless and at worst harmful. As a result, consumers are left to navigate the app stores alone. For one fast-growing sector, digital mental health, the likelihood of using an app is also affected by the relative lack of access or choice of mental health services as well as the stigma and discrimination experienced by sufferers and the interrelationship with physical health. We know, for instance, that people with depression face a higher risk of developing heart disease than individuals without depression and that following a heart attack, each additional depressive symptom that develops increases the risk of another heart attack by 15% [17]. Mental health problems also affect morbidity in other disorders such as rheumatoid arthritis and asthma [16]. We have used examples from mental health, but our principles are intended to inform and empower all health app users.

Most proposed evaluations (eg, Mobile App Rating Scale, MARS [30], Enlight assessment tools [31]) assess usability, aesthetics, content, user engagement, and available research evidence, and others have been adding to this list [32]. These systems are useful because they facilitate multifaceted and thorough evaluations of apps, but they fall short of allowing clear recommendations. In fact, more recent evidence from Canada involving service users demonstrated that a high MARS rating would not on its own provide enough information to allow service users to form a decision on whether to download an app [33]. Advisory bodies such as NICE in the United Kingdom make a determination of what is likely to be cost-effective (effective, cost relative to benefit, and other comparison treatments), before they recommend its use in the UK NHS. Consumers, however, want to make choices based on simpler information. One for-profit company, ORCHA [34], provides reviews based on current standards, regulation and good practice, but their overall score does not allow a consumer to decide which components are important to them. PsyberGuide [35], a nonprofit organization, also provides reviews that include a service user focus but does not receive data directly from app developers. But apart from their lack of fulfilling all users’ expectations, no method provides clear information at the point of sale, and a potential consumer would have to search in 2 places for the information they need, to make a choice. We propose only 4 aspects of apps that represent the critical information required for responsible health app marketplaces. These 4 principles, deemed the T4T principles, are privacy and data security, development characteristics, feasibility data, and benefits.

We have inserted the information from 4 health apps, one of which was named the app of the year in the iTunes Store in 2017 (Calm) (see Table 1). We have extracted, where possible, the information on each of our principles from the information provided with the app. The differences are very clear, especially in privacy and health benefits. There could, of course, be more data available on benefits held elsewhere, but this was not available at the point of download. However, what these simple principles also provide is the ability for a consumer to trade off the attributes. Some may want to know that their data are totally secure, whereas others might want to allow some encrypted anonymized data to be transmitted if the effectiveness of the app is proven. Indeed, in a recent survey of participants recruited from a mood and anxiety disorder clinic, many respondents were willing to allow an app to collect data directly from one’s phone, including global positioning system motion sensors, and screen state [37].

As we have said, T4T principles were based on those suggested by others in recent years. However, we more clearly operationalize our principles into concrete questions that could be answered and made available to potential users. To do so, we considered information important to regulators, developers, and health services, as well as integrating patient viewpoints taken from a number of different studies [37,66]. Patients are, after all, the consumer group of interest. Their views do not necessarily coincide with the expert groups’ views, as shown in the Delphi exercise by Zelmer et al [33]. Privacy and security feature in every assessment system and in regulations and are high on the list for patients, especially those with a mental health problem who may be more sensitive about information about them being shared [11]; So, it is included here, but in the simplest terms and not buried in an incomprehensible privacy statement. Our principle for a fit-for-purpose app includes development with patients. This principle is often suggested [19] but rarely incorporated into app assessment. As we know that some commercial apps are complex and hard to use by the patient group they were intended for, we have valued this section highly. Effectiveness is often mentioned in many assessment systems, but the promised effects are also dependent on the dose of the app and how intensively it is used. Patients need to consider this time constraint when deciding to make a purchase. Patients also want to know not just how effective it is, but also whether anyone does not receive any benefit. This is also important to clinicians, as patients who receive no benefit may view themselves as hopeless cases and not, as in the BlueIce exemplar, just part of the quarter of patients who report no advantage from following the app.

Our approach has, therefore, been to provide information to patients at the point of the download that allows them to make an informed decision, and which they can refer to later as part of their self-management plan.

If these simple T4T principles are followed, then we will have gone some way toward protecting patients. Whose job is it to monitor the T4T principles? Our view is that formal regulation is not needed. We just need the information to allow patients (and patient groups) to make informed choices. Information that is not true can be picked up by advertising standards authorities. Recent examples of this process are the US Federal Trade Commission fining of Lumosity for deceiving consumers with unfounded claims about cognitive benefits [67] and Carrot Neurotechnology for claiming that their app, Ultimeyes, can improve users’ vision [68]. Health apps are not a passing fad, and the low barrier of entry into current app marketplaces has resulted in an environment that at best confuses and at worst delays effective treatment. The problems have been highlighted but rarely have clear solutions like ours been proposed. Developers may be encouraged to produce these answers by commercial advantages, as apps with T4T principles might increase consumer comfort and produce unique revenue streams through increased adoption, not only from direct-to-patient markets, but also from health systems. They will also enjoy increased legitimacy among patient groups.

We also note that the contributions of these principles are that they are a small, yet informative, set of questions that could be adopted relatively simply. We suggest these principles as the first, but important step. Further steps could attempt to explore if these principles could be defined with more structure. However, this structure would likely require further empirical work and coordination between different stakeholders in the health app space, particularly, developers and purveyors.

Confidence in the efficacy and safety of these health apps is the least that patients should expect in making a choice to buy or use them. It is now time that existing commercial app stores, specifically the Google Play and Apple iTunes stores, step back from their libertarian ideology and adopt some rules for health app marketing. They should tighten up the definition of health apps and adopt a system, ours hopefully, to allow patients to understand what to expect from a health app. Although some might believe that this proposal is other worldly, starting somewhere is important. Health app marketplaces have a duty to, and health app developers a commercial advantage, from following our suggestions – we should not need to wait for another scandal or disaster before the Google Play or Apple iTunes Stores step up to the plate and help prevent worthless products being pressed on those with health needs.