In a recent blog post, I talked about the new book that Paul Cerrato and I just completed, The Transformative Power of Mobile Medicine. In that post, we shared the Preface to the book in the hope that it might pique readers’ interest in mobile health. What follows is an excerpt from Chapter 3, “Exploring the Strengths and Weaknesses of Mobile Health Apps.”
Choosing Effective, Sticky Health Apps
Even healthcare providers who see the need for innovative mobile apps still face numerous obstacles. Given the human tendency to seek the path of least resistance, identifying the most effective, “stickiest” mobile apps becomes a real challenge. In Realizing the Promise of Precision Medicine, we discussed the need to individualize medical care and the importance of improving patient engagement. When choosing mobile health apps to meet patients’ needs, it is critical to keep both goals in mind. Each patient is at a different stage in their journey, with some lacking basic knowledge about their disorder and others almost as well informed as their providers. With that in mind, the prescription of health apps should be geared to an individual’s level of patient engagement.
Mobile apps can be divided into several broad categories based on the level of engagement that each patient has reached. Patients will likely lose interest in a health app if it is not consistent with their level of engagement.  Among the categories that can meet patients’ needs are apps that:
'Provide educational information
Alert patients to take some specific action
Track their health or medical data
Present patients with data that they have put into their mobile device
Offer advice based on the data that patients input into their device
Allow patients to send information to their family or healthcare provider
Provide social network support
Reward patients for changing their behavior.’ 
An activated, fully engaged patient will likely know most of the basics that would be provided in a mobile app that only offers educational information and will lose interest in the digital tool quickly. Conversely, a patient who is only modestly interested in managing a chronic condition may not benefit from a more in-depth app that tracks their medical data or physiological parameters. They must learn to “crawl before they walk.”
The second category on the list, namely alerting patients to take some action, requires a closer look as well. No doubt many patients have benefited from mobile apps that remind them to take their medication on time or to make an appointment for their periodic mammogram or colonoscopy. Forgetfulness is a normal human failing and these apps can address that. But to be realistic, most non-adherence is not driven by poor memory. It’s driven by far more complex and entrenched motives, and the reason many patients fail to heed their provider’s advice is because it is just not that important to them, or because in their minds the risks outweigh the benefits, or because they can’t afford the prescribed intervention, or because they didn’t fully understand the advice offered or…. The list is long.
Addressing the first issue, Ira Wilson, an authority on patient adherence, points out that ‘we don’t forget to pick up our kids from day care or to make dinner or anything else that’s really important.’  With that reality in mind, it’s not surprising that reminder apps that send patients alerts frequently fall short. This once again emphasizes the point we have made elsewhere in this book: Mobile tools can only supplement medical care, not replace it. And for clinicians to motivate such uncooperative patients will require time, a precious commodity in today’s healthcare environment.
Time is required to ask patients about why they don’t want to follow a prescribed course of action. Time is required to query patients about possible obstacles to adherence: “Can you afford this medication?” “Does it cause unbearable GI reactions?” “Do you have a way to get to your next appointment or would it mean losing a day’s pay and possibly termination?” “Do you think your hypertension requires medication even though it’s not causing you any pain or discomfort?” We obviously can’t solve all our patients’ problems, but knowing what’s behind their noncompliance is the first step toward resolving it.
Ira Wilson takes this type of deeper probing to heart when he works with patients:
Wilson doesn’t push reluctant patients to take their medications. During a visit with a man with poorly controlled hypertension, for example, Wilson began by asking, “What does hypertension mean to you?” The man replied, “I’m kind of a hyper guy. And sometimes I get tense.” He explained that he takes his medications only when he feels both hyper and tense. In such situations, I [the author of a New England Journal of Medicine editorial] would probably reply, “That’s not how it works,” but Wilson gently asks, “May I share a different perspective?” And patients usually say, “Of course, that’s why I’m here.”
People like Wilson don’t need a digital reminder to have these conversations or to abandon the “doctor knows best” dynamic. For those of us who struggle, the most effective adherence booster may be giving doctors and patients the time to explore the beliefs and attributions informing medication behaviors. These conversations can’t happen in a 15-minute visit. 
Singh K, Drounin K, Newmark LP et al. Many Mobile Health Apps Target High-Need, High-Cost Populations, But Gaps Remain. Health Affairs. 2016; 35:2310-2318.
Rosenbaum L. Swallowing a spy—The potential uses of digital adherence monitoring. N Engl J Med. 2018;378:101-103.