Talking our way to better health: what’s next

I’ve been dictating papers, notes, emails for more than 15 years (and have a copy of Dragon Naturally Speaking v1 and a closet full of mics to prove it). Some of you know that I regularly dot my spoken langauge with dictation hyphen speak comma and frequently don’t know that I’m doing it exclamation mark.

So, I think there is great potential for the voice interface movement (see the Amazon Echo, Siri, Cortana, etc) to revolutionize the way that we interact with technology. We’re still in that early haphazard phase, in which companies are trying to inject voice into every box, app, tool, and small animal monitoring device — all to see what sticks. This is pretty common in the lifecycle of new technologies (see wearables — who exactly needs pulse ox), and I think we’ll soon see that voice interfaces have huge potential in digital health. Voice will help us improve accessibility, a much overlooked challenge for digital health apps. But just imagine the improvements we can make in hard-to-monitor factors like eating, activity, symptoms, and mood.

And with price points dropping on voice tools (like Amazon’s Echo Dot), there is potential to make voice entry ubiquitous.

(This is not an Amazon commercial — really — but) Amazon is making it easier than ever to make conversational voice (and text) agents with their new Lex framework.

I’ve played with several similar frameworks, but the sophistication in the language parsing, interoperability, flexibility (same logic for Messenger or Twilio), and cost efficiencies really makes Lex standout.

Next time, #thinktwicebeforeyouapp and go voice.

Healing the poor, digitally

(In case you’re looking for something to take your mind off the election, watch this).

I was thrilled to close out the Duke Forward road events in New York City, along with my graduate student colleague, Shelley Lanpher. Shelley and I talked about our work using digital health to improve obesity treatment in medically vulnerable populations.

Oh, and make sure you #waitforit — there’s a “surprise” reveal at the end.

When genetic tests are basically horoscopes

Fantastic piece in today’s Inside Stat.

The tips I got back were almost comically generic. One piece of advice from Kinetic Diagnostics on how to compensate for my increased risk of muscle cramping? “Do proper stretching and muscle warm ups before and after exercise.”

DNAFit’s recommendation to make up for a variant that predisposes me to to see fewer gains from endurance training? “Stay sufficiently hydrated.”

Kinetic Diagnostics said I was at elevated risk of high blood pressure; DNAFit said I was likely to experience fewer problems with blood pressure. They both offered the same advice, supposedly tailored to my genotype: exercise.

(When I later asked them about this recommendation, the companies acknowledged that such advice could benefit anyone but insisted that people with my genotype would find it especially useful.)

I suspect that this will mostly be interpreted as an indictment of the athletics genetic testing “industry.” And, they seem to deserve it. But there’s a bigger issue here: many similar companies enter the market with laughably limited evidence that their “personalized recommendations” are actually informed by science.

Then there were the interpretations that flat-out contradicted one another.

The tests each looked at different regions of my genome — which may have been necessary to distinguish themselves from their competitors, but which in and of itself suggests just how much this field is in its infancy. So it wasn’t possible to compare the complete results from each company head-to-head.

But among the scores of data points, I found 20 genetic variants that showed up on two or more test results. The companies all gave me the same genetic readout on those variants, so I have little doubt they correctly analyzed the cells in the cheek swab I’d sent them. In six cases, however, the interpretation I got from one company directly contradicted the interpretation from another.

I’m sensitive to the idea that [the long time it takes to generate] evidence frequently slows the process of bringing innovative tools to market. However, this is a helpful reminder that speed can also disadvantage consumers (while rewarding founders).

What’s next in digital health (vol 211)

Part of our ongoing, intermittent, and highly speculative series about what’s next in digital health

Invisible apps are what’s next in digital health. These are some of the best.

We’ve yet to see many invisible apps for digital health. But they’re clearly what’s next. As design focused as I am, it’s becoming increasingly clear that we can develop usable, sticky apps without graphical interfaces (there’s a trend to call these no-UI apps, but graphics do not an interface make).

Our studies show that we can achieve 12-month engagement of 84% engagement with a health app that’s delivered via interactive voice response and text. This means that 84% of participants will text or voice us at least weekly for a year.

Compare that to my rule of 70 which, in part says, most of your app’s users won’t come back for a second try.

Plus, these invisible apps have huge reach (91% of us text), lower cost (thanks Twilio), are easier to code (notifications APIs rock), and are easier on the MVP budget.

GirlTrek is what’s next in [digital] public health programs

The New York Times blog, Fixes, featured one of my favorite organizations today. Girl Trek is the best public health program you haven’t heard about [yet]. Look, I’m a scientist, a wonk, a tinkerer. I’m technically inclined, and quantitatively oriented. I’m hyperbolic and excitable, but I’m not easily inspired.

But GirlTrek inspires me [big time].

Here’s a gross simplification — recruit nearly 60k women nationally, women who are mostly sedentary, who lead busy lives and who don’t [yet] take enough time for themselves. Link them with groups, comprised of women, similar and dissimilar, of all ages and backgrounds. Then, motivate them to walk. And walk. And keep walking.

Physical inactivity is one of the most pressing public health crises of our time. And yet, many of our public health efforts haven’t gotten the population moving. This is especially true in high risk groups, like Black women.

GirlTrek is different. They reach, engage, motivate, and inspire with an approach that’s organic, culturally resonant, and technologically sophisticated. My take?

“We’ve spent an enormous amount of money on research-based approaches to obesity prevention and treatment, and almost none of them have worked with black women,” says Gary G. Bennett, a professor at Duke University and a leading researcher on obesity. “One of the key predictors of positive treatment outcomes is really high levels of engagement. I’ve been doing work on obesity as it affects medically vulnerable populations for 15 years, and I don’t know of anything in the scientific community or any public health campaigns that have been able to produce and sustain engagement around physical activity for black women like GirlTrek does. Not even close.”

And, it’s working.

Their secret? Focusing on what matters to women today. Not the health benefits that might accrue in the far future.

“It wasn’t about looking good or weight loss or fitting into a certain type of clothing,” she recalled. “It wasn’t, ‘Hey, you fat person, you need to do this or you’re going to die.’ It was, ‘I love you and I want you to love yourself enough to invest in 30 minutes a day, to walk yourself to freedom like Harriet Tubman did.’ And that spoke deeply for me because my life work is showing up for other people, but I wasn’t showing up for myself.”

We researchers can occasionally have a bit of hubris (!) about what it takes to improve public health. But the data don’t lie. For some of these issues, we need bright, creative, and novel ideas that can work — at scale.

Look no further than GirlTrek.

[Old] data is the new oil

 

By now, you’re probably tired of hearing that “data is (sic) the new oil.”

It’s true, but unlike oil, data’s value is on the rise.

Apparently that’s also true for old data. There’s word today that, Viant, the parent company of Myspace (yes, that one) has been acquired by Time, Inc.

Because, data.

In buying MySpace, Viant acquired the data of more than 1 billion registered users. While not all of those people may have kept the same email address from their MySpace days, it still has an enviable database of first-party data….First-party data is considered the holy grail when it comes to advertising online because it means marketers know they are serving ads to the actual consumer they want to be targeting, rather than making probabilistic bets based on browsing behaviour…this gives Time an immediate leg-up … and provides a first-party data set that, in Time’s own words, “rivals industry leaders Facebook and Google.”

Two thoughts here:

  1. Wow

  2. Our data are the gifts that keep on giving. We’re increasingly less likely to churn email addresses and social media credentials, allowing even old data to be linked to what we do today. We have growing comfort with data sharing. Concerns about data privacy are virtually nonexistent — particularly if you grew up in the Facebook age. And there’s revenue to be gained in selling our data (who reads terms of service anyway). Many will be surprised that Myspace still exists, but we shouldn’t be surprised that our data still exists to provide value, long after we’ve moved on.

Is digital health bad news for academic medicine?

Here’s an interesting piece about the emigration of medical students from top-tier West Coast medical schools and into digital health startups.

Bay Area-based medical students from Stanford and UCSF have among the very lowest rates of pursuing residency programs after graduation compared to the rest of the country. Stanford ranked 117th among 123 U.S. medical schools with just 65 percent of students going on to residencies in 2011…UCSF is 98th on the list, with 79 percent of its graduating students going on to residency…“We’ve seen that many of these Bay Area-based medical students are drawn to startup opportunities,” said Jeff Tangney, CEO of Doximity. “It used to be biotech, and now it’s more often digital health.”

This is tough news for medicine – both clinical and academic.

It will be tempting for some to de-trend these findings, questioning whether these emigrees should’ve ever entered medical school in the first place. Others might argue that these departures are actually good news for future patients. But these perspectives miss the underlying trend.

It used to be the case that if you were interested in improving patient care and creating better, more efficient treatments, you went into academic medicine. If you found innovation more compelling than full days of patient care, you could find an academic position, secure a more limited clinical role, and start creating. Today, given funding restrictions, beauracracy, and the long [long] road to impact, the startup economy is a more attractive option.

In short, I wonder if people are running to digital health, or running away from academic medicine.

What do we need from digital health science?

Not another app. There I said it [again]. The market is crowded and science can’t (and shouldn’t) compete on design, updates, integrations, marketing, etc.

So what is there for a behavioral digital health scientist to do?

Answer the tricky, confounding, unanswerable questions that are constraining the growth and utility of digital health.

Like this.

Vanessa Friedman wrote great piece in the Times last week about breaking up with her Apple Watch (I’ll take it). Amidst her litany of concerns was this:

Likewise…the fitness-app aspect — the tracking of my steps, the measuring of my heart rate, the telling me to stand up when I am in the middle of an article — seems more like a burden than freedom…I have worked hard to wean myself from a reliance on exercise machines telling me how hard I had worked…because I knew I was cheating pretty much all the time anyway and thus could not trust the results, and in part because it became an excuse to modify, or not, my ensuing behavior…But the truth is, I know when I am in shape… The watch threatened to drag me back into a numbers-driven neurosis, and that’s a temptation I would rather not have.

Her comments are no surprise for anyone who’s helped a patient to change her behavior (especially the masses who aren’t interested in the quantified self approach). It’s dangerous to treat anecdotes as data, but I suspect her experience is widely shared. I’ve yet to meet a patient who was motivated by the reams of data that we scientists like to provide.

So, how do we fix this? That’s the question.