Recently, we met with a group of potential collaborators to discuss a health care research app they had built. The app was intended to support a year-long investigation into the physical and mental health of retired professional athletes. It did this by prescribing physical activity to them, which they reported weekly through the app. The study was well funded, the app was visually appealing, and the study rollout and promotion strategy were well executed. Yet, by the time the research team spoke to us a few months in, nearly all participants had dropped out. The research team was mystified. Everyone loves apps! Why not this one?
From our perspective, the team made an all-too-common mistake: prioritizing their own interests over those of their users. The app gave the athletes no feedback, and thus no reward for using it, save for the altruistic feeling of helping to promote research. It’s no wonder they lost interest.
The reality is that everyone loves a select few apps, while millions more die an obscure death. Designing health apps that are truly engaging is even more challenging than creating the next Instagram or Snapchat. Despite what many health care observers may think, digital solutions are not inherently engaging, and most successful health apps must account for the needs and interests of multiple stakeholders.
At Partners Connected Health, we pursue a Human-Centered Design approach that is rooted in empathy, multidisciplinary perspectives, visual learning, and experimentation, to develop a deep understanding of users’ motivations and values.
We find that, too often, health app design relies on the scientific approach, prioritizing empirically derived data to solve one problem based on known facts. Human-Centered Design is less linear. While it incorporates elements of scientific reasoning, it also seeks to understand the problem from multiple points of view, and in the context of the chaos and complexity of everyday life.
Human-Centered Design is particularly suited to solving problems in health care, which rarely exist in isolation but generally involve multiple stakeholders — including patients (who may have multiple health problems), physicians, nurses, hospitals, laboratories, payers, researchers, and the patients’ families or support networks. The perspectives and motivations of these stakeholders can differ and may even conflict. Any solution likely has to identify and address multiple factors to achieve success — but the more factors involved, the more complex the design is likely to become.
In designing successful and engaging products, we adhere to a number of design principles, two of which we discuss below:
- Design for Context. All the stakeholders who contribute to the issue, as well as those participating in the solution, need to be identified and successfully engaged. Their incentive to engage must be rooted in the value generated from meeting users’ needs, but it must also be sensitive to the needs of stakeholders who are not direct users.
- Design for Trust. We, as designers, have a responsibility to ensure stakeholders feel that our process is informed, reliable, and rigorously conducted, and that it will result in a product that addresses users’ needs. Further, the product resulting from this process must continually prove its worth to maintain this trust.
Design Principles Applied to Our Work: ePAL Case Study
These design principles can be seen in our work with ePAL, a cancer pain management app developed for the palliative care clinic at Massachusetts General Hospital. Cancer pain, whether caused by the disease itself, by diagnostic procedures, or by the treatment provided, is a major source of distress for cancer patients. Alleviating this pain can dramatically improve their quality of life.
Yet, many treating physicians report feeling overloaded by the volume of work, conflicted by demands on their time, and, ultimately, burnt out. Too often they concentrate on the information necessary to make technical decisions related to cancer treatment. At the same time, patients report not feeling ”heard” or ”seen” as individuals living with cancer and suffering pain. They receive care in discrete, infrequently scheduled 10-minute consultations, during which many feel self-conscious about sharing their true experiences of pain. Our own research corroborates this, as many patients reported they were uncomfortable sharing the true severity of their pain for fear of placing undue burden on their care team. This inadequate exchange of information and misalignment of priorities presents a problem for both physicians and patients.
To design an appropriate solution for this problem, we first needed to identify the web of actors whose involvement was critical to its success. We then needed to design appropriate incentives to engage them. The two most important actors in the case of ePAL were patients and providers.
We needed to create a shared awareness and understanding of the patient’s pain. To do this, ePAL prompts patients to assess and rate their pain three times a week, incorporates alerts for improved communication with their care team, and provides them with algorithm-based decision support and educational resources. If a patient reports a pain score of 8 or higher (on a scale of 0–10), reports new pain, or has any other significant treatment issue, the patient’s care team is immediately notified. Upon receiving an alert, the care team responds to the patient within an hour.
While the app creates a powerful self-management incentive for the patient, it also risks increasing the workload for the care team. The design, therefore, had to balance the perceived threat of an increased workload with the prospect of more effective patient care. From the care team’s perspective, if patients could better manage less complex issues via the app, additional clinic time could be redirected toward patients with more complex care needs.
The success of this solution depended on establishing a shared trust among the stakeholders, as well as between them and the app. Physicians played the initial role in instilling trust by recommending ePAL to their patients. This trust is further reinforced by ensuring the user feels empowered. Recognizing this, ePAL was designed to allow patients the flexibility to report severe pain without initiating a care team response should they wish not to.
While ePAL’s clinical trial is still in progress, early results suggest that patients are reporting their pain at least twice per week and that its communication feature is not being abused, but utilized in response to real need. Patients are also reporting fewer instances of pain spikes, and they tell us that they feel more comfortable reporting pain accurately. Clinicians are also reporting reduced demands on their time. The clinical trial was initially designed to run over an 8-week period, after which patients were scheduled to stop using the app. Surprisingly, however, patients petitioned for continued access to the app beyond the study period, demonstrating that ePAL fulfills a real unmet need. Given this response, it is our intention to roll ePAL out for broader use.
To conclude, there is no single method for achieving effective user engagement. Creating successful and engaging experiences involves many variables acting in concert toward a shared goal. There are, however, key considerations that make success more likely. Chief among these is a thorough understanding of the issue at hand and an appreciation for the context involved, as well as genuine curiosity about the human experience, tolerance for ambiguity, and the exercise of a truly reflective design approach.
On a practical level, it also requires direct access to end users (such as patients and doctors) to identify their needs. Here we call on health care providers and administrators to reflect on their role as gatekeepers of research and innovation — as in the case of ePAL, they are an essential component of any positive innovation effort. Finally, more effective design can be fostered through a culture of shared learning and a willingness to be open about the challenges faced and lessons learned.
Acknowledgements: The authors would like to thank the following contributors for their hard work on this formative research and design: UX researchers Meghan Searl, Barbara Barry, Clare Flanagan, and Lara Traeger, UX designer Saket Mishra, clinical investigator Mihir Kamdar, and principal investigator Kamal Jethwani.
This article originally appeared in NEJM Catalyst on May 15, 2017.