Digital Health Innovations, Part 2: Foundation & Hypothesis Formation
This 5-part blog series is about designing evidence-based patient-facing digital health interventions for vulnerable populations that are efficacious, scalable, and cost-effective. Think it’s a tall order? IT IS! But it’s not impossible. We have some insights we’d love to share with you. These insights come from our combined 20 years of experience designing, testing, and disseminating effective digital health interventions in medically vulnerable populations.
If you want to start from the beginning, check out Part 1.
So you want to build an app? Confused about where to start? Look no further. Let’s start with the basics. Any technology that you build needs to solve a meaningful problem that your patients have. This problem may not be the most important problem but could be the easiest, fastest, or cheapest problem to solve and may lead to future opportunities such as funding, partnerships, and innovation.
If you’re anything like other institutions and communities we have worked with, your first problem is that there are too many problems to choose from! Read below to learn a few considerations identifying and prioritizing problems.
Is it measurable?
You can’t fix what you can’t measure. Maybe it starts with just a “feeling” that something could be better. “My patients don’t seem to be taking their medication as they should.” How do you support that observation? By finding data and evidence! Not only will measuring the problem let you know if you’re correct about the scope of the problem, but now you’ll have tools to find out if your innovation has actually solved your problem, and if continued investment is merited.
A measurable problem typically has the following components:
- Target population(s) affected by the issue
- Definition of current coping mechanisms or workarounds
- Quantification of losses incurred as a result of the missing or inefficient workaround (eg. time, expenses, adverse consequences etc.)
Don’t have robust, reliable data internally? There are many sources of secondary data and analysis available either at no or low cost. Also consider surveying some of your patients to confirm or deny your hypothesis (more on this below).
Secondary Data Sources
- PLOS-One offers free access to peer-reviewed, open access scientific studies.
- The Journal of Medical Internet Research is a peer-reviewed open-access medical journal covering eHealth and “healthcare in the Internet age” with limited free access.
- Healthdata.gov is a federally funded portal of government health data made available through the Health Data Initiative.
You may be tempted to rely on your internal IT department for all of your data and technology needs but this might not be the best option, depending on what you want to do. Health IT is not the same thing as data science, but they do work well in concert. Data science is a discipline that uses processes and systems to harness insights from data. Bring on data science expertise early in the innovation process to better ensure success. No in-house data scientists? Look into partnership opportunities with local academic research institutions for access to data as well as talent including researchers and data scientists. Here are a few suggestions of nonprofit organizations that offer data science services for the nonprofit and public sectors.
- Bayes Social Impact is a nonprofit that uses data science to solve the world’s biggest problems.
- DataKind brings data scientists together with mission-driven organizations to use data science in the service of humanity.
- Open Data Nation creates productivity solutions that combine open, public data with data science techniques to increase the transparency and productivity of public agencies.
Is it beneficial?
There’s more to problem understanding than data and stats. You need to understand your customers as well as the broader context of the problem. It is important that your innovation be informed by evidence as well as user behavior, motivation, values, attitudes and norms. These qualitative insights provide a more complete picture so that your innovation will be more likely to make patients’ lives easier and better.
As a starting point, examine your patient complaints. What are the issues that you hear the most about? Which issues have been the most persistent over time? If you don’t have a treasure trove of complaints readily available, there are a number of free and low-cost alternatives to find out your patients’ biggest complaints.
- Informal, quick surveys, even 5 minute waiting room surveys, can inform your problem definition. For example, your organization may want an app for patients but your patients may not have smartphones or they may not use them frequently enough for the app to be useful. A couple of common survey questions include:
- What types of technology do you use? How frequently?
- What are your major barriers to achieving your goals?
- Focus groups are guided group discussions of 8-15 people for 1-2 hours to discuss a particular issue. They can be a powerful tool for understanding an issue through different perspectives. Any university near you that has a graduate school of public health, sociology, or anthropology might have faculty or graduate students who can help guide you towards designing a great focus group.
- Workflow mapping produces a visual representation of the actions, decisions and tasks required to attain a certain goal. Completing patient flows and provider workflows will help to identify bottlenecks and gaps.
- Patient advisory boards are groups of patients who volunteer their insight and expertise to improve care for all patients. These boards also provide a natural testing ground for assessing the feasibility and acceptability of potential innovations.
- Community-based organizations and collaboratives can be helpful given their trusted relationships with the communities they serve. To learn more about the power of health collaboratives, look at the BUILD Health Challenge which supports “bold, upstream, integrated, local, and data-driven” (BUILD) community health interventions in low-income, urban neighborhoods.
Is it impactful?
As an industry, healthcare is undergoing a paradigm shift lead by the concept of the Triple Aim: health care that delivers (1) greater patient satisfaction and (2) better population health outcomes at (3) lower costs. Now that you have found measurable and potentially beneficial issues to address with your potential innovation, filter the list further by finding problems that affect your financial bottom line and incur the largest sources of costs for your organization. Common low hanging fruit that other safety net clinics and providers address include appointment reminders, medication adherence, and screening appointments.
If you have the data, identify which specific patient populations cost the most money for your organization and which conditions incur the highest costs. When choosing a health outcome, consider short-term behavior change milestones that can be achieved that contribute to long-term health outcome improvement.
- For example, obesity-related conditions may incur some of your highest costs so your long-term health outcome of interest is weight loss. A more short-term behavior change could be increasing fruit and vegetable consumption.
- In another example, hypertension control is a long-term health outcome. A smaller, more easily measurable outcome that supports this outcome could be increasing medication adherence or supporting timely prescription refills.
Congratulations! You should have a thorough understanding of a meaningful problem. Now it’s time to form your hypothesis of how your innovation will address it. This general format of a hypothesis if this: If I implement this particular solution, I will affect my outcome of interest. For example, a hypothesis could be: “If I send regular text message reminders to check blood glucose levels, my diabetes patients will be more controlled.
Common Behavior Change Models in Digital Health
There are evidence-based frameworks that you can use to serve as a cheat sheet for hypothesis formation. We will briefly discuss three common models: Michie Behavior Change Wheel, Transtheoretical Model, Social Cognitive Theory.
Michie Behavior Change Wheel
The Behavior Change Wheel (BCW) emerged from a systematic literature review of 19 frameworks of behavior change. The wheel consists of three layers: (1) sources of behavior; (2) intervention functions; (3) policy categories.
The innermost layer, based on the COM-B (‘capability’, ‘opportunity’, ‘motivation’ and ‘behavior’) model, outlines potential targets for your innovation. Your innovation does not have to address all sources of behavior but it does not to affect at least one source significantly such that the new behavior is sustainable.
The next layer consists of nine intervention functions. The functions that your innovation address will depend on the research you did to define your problem as well as the source(s) of behavior that you target. The outer layer, the rim of the wheel, outlines seven policy categories that can support the delivery of the intervention functions.
Transtheoretical Model of Behavior Change
The Transtheoretical Model is describes the process of intentional behavior change. This behavior change involves progressing through five stages: Precontemplation, Contemplation, Preparation, Action, and Maintenance.
The Transtheoretical Model provides a number of advantages for recruitment and retention for technological interventions.
- Because the model accounts for the fact that different individuals will be in different stages, it compels you to develop diverse recruitment strategies for everyone.
- Because the model requires you to individualize approaches to the specific needs of individuals, the Transtheoretical Model tends drive high retention rates because people drop out less frequently.
- The Transtheoretical Model includes a set of outcome measures that can recognize and capture smaller steps toward behavior change that other more traditional action-oriented approaches miss.
Social Cognitive Theory
Social Cognitive Theory posits that learning for an individual is driven by observing others within the context of cognitive factors such as beliefs, self-perceptions, and expectations and environmental factors including social interactions and outside media influences. The theory states that when people observe a particular model perform a behavior and experience the consequences of that behavior, the observer remembers the sequence of events and uses this memory to shape future behaviors. Depending on whether the model is rewarded or punished for the behavior and whether the outcome of the behavior is positive or negative, the observer may choose to replicate behavior modeled.
Make a list of potential solutions for your selected problem that achieve your stated outcome and rank based on the following considerations and framework of choice:
- Give a higher ranking to hypotheses that allow you to leverage in-house resources (eg. work within your current workflow, or utilize easily leveraged volunteers)
- Give a lower ranking to hypotheses where you need to purchase many resources
- Give a higher ranking to hypotheses that can be easily and quickly tested
Can’t wait to move this conversation forward! Share your experiences, thoughts and questions on Twitter with the hashtag #safetynettech.
Erica Levine is the Programs Director at the Duke Global Digital Health Science Center. She has over 8 years of experience translating evidence-based behavior change interventions for delivery on various technology platforms (SMS, IVR, web). She was leveraging technology for health in medically vulnerable populations way before it was cool. She has worked on projects in rural North Carolina, Boston, and Beijing.
Vanessa Mason is the co-founder of P2Health, an initiative that supports innovation that fosters the protection of population health and promotion of disease prevention and founder and CEO of Riveted Partners, a consultancy that sparks behavior change through accelerating data-driven innovation. She has over a decade of experience in healthcare innovation and consumer engagement in both the United States and developing countries. Her experience in global health has shaped the way that she sees the role of technology and design in health for vulnerable populations: innovate and integrate rather than break and disrupt.