Most digital health companies treat patient engagement as a product design problem. Build a better interface, add a streak mechanic, send a smarter push notification, and patients will keep showing up. The assumption is reasonable. But the clinical data does not support it.
A systematic review and meta-analysis of app-based interventions for chronic disease found that up to 80% of users engage only minimally, defined as logging in fewer than twice. Around 2% showed the kind of sustained, continuous use the evidence links to improved clinical outcomes. These figures do not describe a handful of poorly designed products. They describe the category.
For a clinic director or insurance CMO commissioning a chronic disease programme, this number carries a specific implication. A patient who disengages at week six has not received an intervention. They have received an onboarding experience. The clinical outcomes, the avoided hospitalisations, the ROI on the programme investment, none of that materialises from a patient who stopped engaging before behaviour change had a chance to take hold.
The wrong diagnostic
The digital health industry has spent considerable effort trying to solve a problem it has consistently misdiagnosed. Engagement metrics borrowed from consumer technology; daily active users, session length, feature adoption, measure interaction with a product, not progress toward a clinical outcome. A patient who opens an app every morning to log their breakfast and then abandons the programme at month two has excellent DAU numbers and no improved HbA1c. The metric and the outcome are disconnected.
The science of sustained behaviour change has been building for decades and produces a more precise account of what is actually happening. The COM-B model, developed by researchers at University College London, identifies three conditions that must coexist for behaviour change to occur and persist: “capability, opportunity, and motivation”.
- Capability means having the knowledge and skills.
- Opportunity means having the circumstances and environment that allow the behaviour.
- Motivation means having the internal drive to choose the behaviour, repeatedly, over time.
Most digital health tools are designed around the first two. The third receives almost no systematic attention.
Why motivation is harder than it looks
Motivation in chronic disease management is not a personality trait patients either have or lack. Research on motivational interviewing, the structured clinical communication approach developed by Miller and Rollnick and studied extensively across chronic conditions, shows that motivation fluctuates in response to setbacks, life circumstances, and the presence or absence of the right kind of clinical conversation. A trained coach who can identify ambivalence, work with a patient's own stated reasons for change, and adapt when the original plan stops fitting the patient's reality produces measurably different outcomes than a notification that says "don't forget your daily check-in."
This is not a philosophical distinction. Research on dropout prediction in digital health programmes found that one of the strongest predictors of whether a patient stays enrolled is how frequently they interact with a health coach. More than the programme content. More than the app design. More than baseline health literacy. The human contact is the retention mechanism. Vendors building engagement strategies around interface design are optimising the wrong variable.
What a properly designed programme looks like
The clinical evidence points toward a specific model: technology handles the parts of chronic care that scale easily, which includes monitoring, data capture, programme delivery, and habit tracking. Human coaching handles the parts that do not scale without the right infrastructure, which is the ongoing motivational work that determines whether a patient is still in the programme at month six or month twelve.
Separating these two functions changes the economics of chronic care delivery significantly. A health coach or dietician focused entirely on behaviour change conversations can support a much larger caseload than one also managing administrative tasks and data logging. The technology extends the reach. The coach determines whether that reach translates into outcomes.
How we approach this
Liva's programmes are built around this evidence base. The coaching methodology draws from the COM-B framework and motivational interviewing practice. Health coaches and dieticians are trained to treat motivation as the specific, person-dependent variable the research shows it to be, adjusting their approach when circumstances change rather than following a fixed script. The digital platform carries the longitudinal layer. The coaching carries the work that actually drives retention.
For clinical teams and insurers evaluating chronic disease programmes, the question worth asking any vendor is straightforward: what is the mechanism by which your patients stay engaged past month three? If the answer involves app design, that is worth probing further.
Talk to our team to understand how our approach addresses this.
