Health insurance apps: background, status quo and challenges

More than half of all German citizens over the age of 14 (55 % and therefore 39 million German citizens) own a smartphone (BITKOM 2014), on which an average of up to 23 application programs (apps) are installed (BITKOM 2012). The popularity shown by these numbers offers – also for statutory health insurance companies (GKVn) – new opportunities for reaching target groups in a wide variety of areas, from primary prevention to diabetes management. But if GKVn want to use apps, for example, to fulfill their legal obligation to provide health information and advice (Section 1 SGB V), they have to take a few things into account in the highly sensitive health market.

>> According to a study by the Federal Association for Information Technology, Telecommunications and New Media (BITKOM), 83 percent of smartphone owners (21 million German citizens) have installed apps (BITKOM 2013). According to a Forsa survey commissioned by the AOK-Bundesverband, one in five smartphone owners uses health apps (AOK-Bundesverband 2013). The American market has similar numbers, with 19 % app users who primarily need fitness and training data (38 %), eating behavior (31 %), weight history (12 %), the menstrual cycle (7 %) or blood pressure data (5 %) document (Fox & Duggan 2012). The database “MyHealthApps.net” reveals extensive health app support options with currently 370 apps in 47 languages and 150 different areas of application, ranging from the finder of disabled toilets to birthmark scanners to blood pressure diaries (www.myhealthapps.net; as of April 1st. 2014). According to a representative study by the market research institute Nielsen, the length of time spent using apps has increased significantly since 2011: while it was 18 hours and 18 minutes at the end of 2011, the time factor rose to 30 hours and 15 minutes at the end of 2013 (Nielsen 2014).
Not all apps are the same
There is currently no consensus on a uniform definition of health apps, which is why terms such as health, medicine and health apps are often used interchangeably, even though they target different goals and target groups. Roughly defined, health apps, according to the World Health Organization (WHO) definition from 1946, can be described as mobile applications that are intended to influence physical, mental and social well-being positively, sustainably and based on scientific findings (Scherenberg & Kramer 2013). With regard to legal aspects, a differentiation is necessary based on the widespread classification into primary, secondary and tertiary prevention. If apps have a medical purpose and, for example, support the treatment of chronic diseases (tertiary prevention), they constitute medical devices according to Section 5 of the German Medical Devices Act and the European Medical Devices Directive 93/42/EEC. In practice, the question is whether an app (e.g. for... mobile rounds, for viewing x-ray images, to support diagnostics and treatment decisions) is often difficult to actually classify as a medical device and led to the EU Commission issuing a new MEDDEV guide (MEDcal DEVice) (EU Commission 2012). If apps are used for therapy and diagnostic purposes and have not gone through the approval process at the Federal Office for Risk Assessment and the Paul Ehrlich Institute, there is a risk that, in case of doubt, the manufacturer will bear the liability risk. The stores themselves only check technical defects before publication. Not without reason, because in February 2013 alone, 58,886 low-quality apps were subsequently removed from the stores by Google (Perez 2013).
Risks and side effects
Regardless of technical risks (e.g. data protection, sharing of personal data) that citizens are warned about (Consumer Center 2011), special attention must be paid to health aspects in health apps. This is also not without reason, as a study by the Department of Dermatology at the University of Pittsburgh revealed: three out of four skin cancer apps examined classified 30 of the 188 images that were diagnosed by dermatologists as harmless skin spots as not harmless. The hit rate fluctuated between 6.8 and 98.1 % (Wolf et al. 2010). Whether the damage is greater when patients gain apparent security with diagnostic apps or are given a disturbing misdiagnosis, an appeal to the responsibility of providers and patients with regard to the careful use of such media is necessary. In addition, there may be conflicts of interest on the part of the providers, so it must be ensured that users recognize when it is health information or advertising content. The disclosure of the advertising and financial policies of the respective providers, the existence of contact details and compliance with data protection guidelines should therefore of course be part of quality assurance standards. In addition, health information should always be tailored to the target group, factually correct and supported by sources, as recommended by the Health Information Systems Action Forum (agfis) (afgis 2010, Klemperer et al. 2010). Such quality-related transparency criteria and guidance have already proven successful for health-related websites; the HON seal from the Health on the Net Foundation (HON) or the afgis seal guarantee compliance with verifiable quality characteristics. There are also initial approaches for health apps based on these quality factors: Providers can voluntarily commit to complying with the transparent quality criteria of a HealthOnApp code of honor (www.healthon.de). Through voluntary commitments to certain values, which lead to automatic self-control, quality comes into focus, as providers distinguish themselves among app users and are rewarded with high download numbers and wide market penetration.
The community of all app users and developers monitors compliance with the code of honor. In this rapidly growing market, only they have the opportunity to detect and report violations in a timely manner and thus strengthen trust in the health app medium. Health insurance apps are currently available to users free of charge because the statutory health insurance companies are interested in saving medical costs by changing their behavior. Other institutions offer free light apps in the app stores in order to sell the actual full version of the app or app-supported additional products (so-called gadgets: blood glucose meters, scales, trackers, etc.). Sales of networkable personal scales have already almost doubled, while sales of mobile networked blood pressure monitors have increased by around a quarter (GfK 2014). The price of a health app is not a quality-determining factor, because whether apps are offered for a fee or for free depends primarily on the economic interest of the provider and the possibility of amortizing development costs. Based on anonymous reviews from other users, conclusions can be drawn about the perceived app quality. However, this should be done with caution, as 45% of users assume that digital reviews are manipulated (Fittkau & Maaß 2008).
Status quo, areas of application and
Possibilities
How many health apps are actually in the categories medicine, health and fitness or in health-relevant categories such as sports, lifestyle, food and drink, for example in the global Google Play Store (with around 1.5 million apps) (Anni App 2014), is not quantifiable. In an inventory of existing health insurance apps in the two largest app stores: 1.) Google Play (for Android apps) and 2.) Apple iTunes Store (for iPhone apps) (Canalys 2013) with insurance and health-related search terms (“keywords”) : health insurance, AOK, TK, DAK, Barmer, BKK, IKK, prevention, nutrition, exercise, stress, diabetes, etc.) and related synonyms, 61 (exclusively free) GKV apps were identified (as of August 1, 2014). 45 apps have a (more or less) clear preventative health focus, with nine of these apps being aimed at young families and pregnant women (6), children (1, dental health), caring relatives (1) and Turkish citizens (1, preventive care). while the majority is universally oriented. Apps that were not directly related to health (16) are also primarily aimed at universal use or are aimed at young people (2) and financial advisors (1).
There is a wide variety of digital support, ranging from apps in the classic preventative areas of nutrition, stress and exercise to precautionary planners and vaccination apps (see Figure 1). No app could be identified in the addiction area. In comparison to pharmaceutical apps, it becomes clear that the statutory health insurance companies are heavily involved in the area of universally oriented primary prevention, while the pharmaceutical industry sees itself as a partner for the management of certain clinical pictures (e.g. asthma, allergies) (Scherenberg & Kramer 2013) . As a result, the pharmaceutical industry seems to be more interested in directly addressing patient groups in their product environment and the health insurance companies in preventive cost savings and acquiring new customers. Two identified exercise or fitness apps from the GKVn (Barmer GEK; DAK) are linked to their bonus program. Insured persons are also rewarded for health-conscious behavior and monitoring of personal health data (nutrition, exercise, sleep profiles) beyond prevention courses. On the one hand, it can be assumed that health-conscious smartphone users and therefore more likely to take “good” risks will benefit here (“healthy user effect”). On the other hand, this means that low-income (non-smartphone owners) and therefore health-vulnerable population groups with the theoretically greatest preventive rationalization potential are at a disadvantage, while healthy insured people can derive monetary benefit from the new technologies (Scherenberg 2011). Although the statutory health insurance companies are not allowed to evaluate the personal health profiles of the apps of their insured persons, health-promoting behavior is automatically transparent when the bonus is granted.
It is striking that the statutory health insurance providers for common diseases do not offer any app-supported everyday aids (diabetes management), despite possible savings potential, for example for around 7 % diabetics (Heidemann et al. 2013). Diabetics do not seem to be an interesting target group, or it is assumed that they do not have a smartphone. This certainly does not apply to type 1 diabetics, primarily adolescents and young adults. It should also be noted that it is no longer just teenagers and young adults who use smartphones. The strongest increase in smartphone owners is among those over 65 years of age, which increased from 7 to 17 % within just one year (2013/2014) and has therefore more than doubled (BITKOM 2014). Mobile helpers could also not be identified for other insured groups with risk factors, such as non-smoking apps. This is surprising since, according to a study by the Institute for Health Economics and Management in Health Care at the Helmholtz Center Munich, the costs of smoking are estimated at approximately 31.3 billion euros per year (2008) (Wacker et al. 2013). It should be noted that, from the perspective of the statutory health insurance companies, chronic apps still run the risk of attracting a clientele that appears unattractive, despite morbidity-oriented risk structure compensation (Morbi-RSA). This indicates that for systemic reasons there are still too few economic incentives for prevention (Scherenberg 2011).
Only long-term studies will be able to prove whether health apps are effective and save costs. Initial studies, however, reveal that apps are certainly able to have an influence on motivation. A study on the mobile application of a nutrition record showed that young people in particular use apps to control their nutritional behavior compared to adults (Daugherty et al. 2012). Followers of the so-called movement “The Quantified Self,” who regularly record all bodily functions and needs, are continually increasing not only in the USA, but also in Germany (see http://quantifiedself.com). Measuring physical activities (running, walking, cycling, etc.) and body data (e.g. body mass index) is particularly popular (Gimpel, Nißen & Görlitz 2013). Respectively in technical
Tracking and health apps seem to trigger motivational effects among male users. In a Forsa survey commissioned by the AOK Federal Association, 42 % men (and 30 % women) stated that their health behavior had improved positively through health apps (AOK Federal Association 2013). It remains to be seen whether this (technology) enthusiasm will continue, because according to the “Mobile Health Report”, women under 50 years of age, with a high level of education and net household income are more likely to have experienced a change in their health status within the last 12 months (Fox & Duggan 2012) who use health apps. Accordingly, personal suffering and benefit seem to be a personal motive for whether apps are used or not. Self-tracking technologies, such as digital pedometers or apps, should be able to increase the number of steps per day by up to 43 % (Wang 2013). It should be noted that only active self-trackers were included in this internal provider investigation. Health risks such as the potential for addiction and the repression of one's own body image are still largely ignored and should be researched more. They are particularly the case when the new possibilities of self-control and self-optimization tempt you to constantly set your own records and to relate your own achievements to those of other users via social media (e.g. Facebook) and thus create an additional comparison -, regulation and control level arises. Users must also be aware that regular use of tracking apps, even with minimal use, provides a detailed profile of the user's usage and lifestyle habits, which can be evaluated and used for marketing purposes (Schollas 2014). It remains to be seen whether apps will be used to increase motivation in the long term. Studies have shown that around a third of apps are no longer used after 3 months (Purcell et al. 2010), around half of all tracking apps are no longer used after 24 months (Endeavor Partner 2013) or apps are only installed directly for testing purposes be deleted again (Eimeren & Frees 2012).
Conclusion
Whether health apps, in comparison to conventional methods of health education as well as the behavioral motivation and stabilization of specific high-risk groups or chronically ill people (diabetics, asthmatics, etc.), are able to exploit rationalization potential in the area of primary, secondary and tertiary prevention in the long term In order to reach particularly those who have not yet shown good health behavior (keyword: prevention dilemma), there is a lack of scientific long-term studies on this. Only when there is evidence of health economic potential and the health insurance companies have a stronger incentive to focus on health effectiveness can the dynamics in the market be further stimulated. The acceptance of health apps is measured by the quality and usefulness of the applications. User education plays an important role in this quality process. Since any layperson can theoretically place health apps in stores, there is a risk that even good health apps from health insurance companies (e.g. whose content is based on scientifically reputable sources) can sometimes be overshadowed by bad apps, which can sometimes do more harm than good Benefits (see example of skin cancer apps), whose health side effects and risks in terms of costs sometimes have to be borne by the health insurance companies. <<