eHealth Disparities Segmentation by Meaningful Access

Many of the same populations that suffer from health disparities also have lower Internet usage. How can we use the Web and mobile technologies to close the gap between the haves and have-nots, rather than increasing the gap?

As always, we need to start by understanding the people involved.

There are many ways to describe the people likely to get the short end of the stick in terms of health, healthcare, and technology. The most obvious are:

• Demographics (e.g., ethnicity, age, language, socioeconomic status)
• Psychographics (e.g., those deeply engaged in their health vs. those who don’t pay much attention to their health, or those who love the latest gadget vs. those who fear computers)
• Access to Technology (e.g., those with desktop broadband vs. those with dial-up, vs. those with smart phones, vs. those with cell phones & SMS, vs. those with none of these)

The problem I keep bumping into is that these factors overlap in very complex ways, and all simple approaches to segmentation seem to oversimplify way too much. For example, Hispanics are more likely to have adverse health outcomes than whites, and they’re also less likely to have broadband, but they’re more likely to access the Internet on their phones. Does this mean we can use smartphones to decrease health disparities for Hispanics? Not necessarily—I'd guess that the Hispanics suffering most from health disparities are those least likely to have smartphones.

Four years ago, in the report Expanding the Reach and Impact of Consumer eHealth Tools, Cynthia Bauer and colleagues at the Dept. of Health and Human Services did an impress job of researching, analyzing, and organizing the field of eHealth Disparities. One of their main conclusions was that we needed more data at the subpopulation level. That gap in our understanding has closed a little in the last four years, but we're still struggling to understand the individuals most at risk of being caught between health disparities and digital disparities.

That said, I think we're close to having a practical starting point.

As we think about strategies to address eHealth Disparities, we might find it helpful to start segmenting in terms of meaningful access to technology. “Meaningful Access” refers to the need to have more than just a computer. Meaningful access requires:

• hardware
• Internet connection
• skills to use them
• ongoing technical support
• relevant useful content and functionality
  

If we take the people most vulnerable to health disparities and subsegment them by meaningful access, then some high-level strategies start to emerge:

The “haves”

Those who already have meaningful access, or those who will gain meaningful access in the next few years with or without our efforts.

Strategies:

Promote existing content and functionality to them

Enhance current content and functionality to be more useful to them

Create new content and functionality for them

The “could haves”

Those who don’t have meaningful access, but who could gain meaningful access as a result of our efforts.

Strategies:

Use the same strategies as for the “haves” above, and also…

Support public access points (libraries, medical centers, shopping malls, etc.)

Support simple and inexpensive access on devices they already own, e.g., SMS texting, including paying the per-message fee

Support public policies and funding that increase access for the underserved (e.g., community-wide wi-fi, extend universal access programs to cover not just phone but also Internet)

The “won’t haves”

Those who don’t have meaningful access, and who still won’t have meaningful access 3 years from now regardless of our efforts.

Strategies:

Support “infomediaries” such as family members who use the Web and mobile devices on behalf of those who don’t

Maintain and enhance non-technology-based services

This might be a starting point. The next step would be to gather more information about each of these groups to understand whether these groups are homogeneous enough to have similar needs that can be addressed with similar efforts.

Mobile Health - the 2 big deals

I've been doing online consumer health for over 15 years, most of it with Kaiser Permanente. As I think back on some of the key capabilities that were originally visions on the far horizon and are now simply part of the landscape around us, I remember when each of these was "the next big thing."

• health information previously available only to professionals, made widely available to consumers
• health risk assessments with personalized feedback
  
• online appointment requests
• online prescription refills
• online appointments booked in real time
• select a physician online
  
• apply online for coverage
  
• email my doctor
• secure messaging with my doctor
• view my medical record
  

Each of these is now everyday reality to millions of Kaiser members. We've reached these horizons and moved on to the next. So what's next? When someone asks me, "What's the next big thing," I usually end up talking about two areas:

1. A better user experience
2. Broader reach

Despite lots of powerful and valuable possibilities for new functionality, from personalization to portable medical records to home monitoring, I think the biggest value to individuals and society will come from improving the user experience of the current functionality, and making that experience available to a broader audience.

1. Better user experience
We need to take all the capabilities we've already implemented, and make them...

• easier to use
• more integrated
• better adapted to real-life scenarios and tasks of our users

2. Broader reach
Over 3 million Kaiser members use the powerful tools we've provided. That's not nearly enough. In addition to increasing the number of web-using Kaiser members who use this stuff, we need to expand these tools to...

• people who are traditionally underserved by the healthcare system
• people who don't have easy access to PCs with broadband connections
• people whose physicians aren't currently part of an integrated group practice

The Mobile Factor
Cell phones won't take us all the way to these horizons. But they can certainly help us get there. In terms of user experience, mobile devices can make simple transactions ridiculously easy, and they can fill in the gaps between in-person, telephone, and desktop web interactions. If we do it right, mobile interactions will become a lynch pin of ubiquitous, integrated, cross-channel experiences.

Not only can mobile devices support much better experiences, it's getting clearer all the time that they can help us extend these services to people who are traditionally left behind by the latest technology. If we do it wrong, our mobile efforts will just exacerbate the already shameful chasm between the haves and have-nots. But if we do it right, and I think we can, we can use mobile technologies as a powerful tool in shrinking that gap.

In-person or virtual?

As it becomes easier to interact digitally, there will be plenty of opportunities to see your doctor without the two of you being in the same room at the same time. In many instances this will be more convenient, with equal quality, and with potentially lower costs to the health care system.

On the one hand, a virtual visit can be a fabulous thing all around, but there are certainly times when an in-person visit is more appropriate. So the question I'm asking myself lately is,

How do we decide between an in-person visit and a virtual visit?

The easiest, and possibly best, answer is, "Let the patient choose." This initially looks like the patient-centered approach. And we essentially let the patient choose today in most situations, as patients choose whether to call on the phone, email, or schedule an in-person appointment. But I'm guessing that approach oversimplifies the situation.

There will certainly be situations in which the health care provider should strongly recommend an in-person visit, even when the patient initially prefers a virtual visit. Possible criteria for an in-person visit include:

• a physical examination is needed
• an emotionally intense decision needs to be addressed
• the patient and their primary physician don't yet have a solid relationship, and an in-person visit could help establish rapport that could then be carried over into future virtual visits

Conversely, what are the indications that a virtual visit is more appropriate? When should the health care provider strongly recommend a virtual visit? Here are a few possibilities:

• the ordeal of traveling to the clinic would be unhealthy
• biometrics are needed that can be more accurately measured when the patient hasn't just spent two hours walking, riding buses, and being quizzed by people in white coats
• time is of the essence and a virtual visit could take place sooner than an in-person visit (this criteria could be applied, e.g., in rural areas or anywhere that has long distances between patients and providers)
  

These are some starting points, hopefully raising interesting and useful questions. Here are a few:

• Who has done work on this problem? I'm particularly interested in any efforts to quantify the analysis.
• What metrics can be used to assess the relative value of an in-person or virtual visit? The obvious starting places are quality, cost, and satisfaction, but we would need some way to measure these across a wide variety of contexts, and I fear the measurement would quickly get too complex.
• How can we apply the tools of human-centered design to these questions? E.g., are there opportunities for rapid prototyping of a decision tool to choose between in-person and virtual?
• How can we ensure patient safety while experimenting in this space?
• What are the implications for pricing of care? Should all virtual visits and all in-person visits be covered, or only those deemed "appropriate"?
• Has anyone attempted to bake this decision into a clinical guideline?
• Telephone nurse triage systems currently have similar decisions built into their protocols--can we take care of this person over the phone, or do we need to see them? Care models that use telephone MD visits are also relevant. How can we expand what we've learned from these experiences to address a new type of interaction that is typically a less rich than an in-person visit, and more rich than a phone call?

I can't help but come back to the original, simplest answer: an in-person visit is appropriate when the patient wants an in-person visit. The challenge for health care providers is the same challenge as usual: how can we apply our knowledge, expertise, and compassion to help people make sound judgments in the face of uncertainty? Despite the complexities of reimbursement, diagnostics, relationships, and technology, the patient should get to decide. How can we best help them with that decision?