dr. gary bennett

digital health wonk and [coffee] geek

I’m pretty open-minded when it comes to food. And I’m first in line to try a new obesity treatment, but…haggis?

Turns out that a British MP wants the U.S. to drop its import ban on haggis. Why? Because, public health.

He claimed haggis “satisfied hunger very much more than the junk food which Americans consume”. The surgeon and former shadow health minister…claimed that haggis consumption would help deal with the “obesity epidemic” in the US.

So there you have it. Haggis – a lovely delicacy containing, “sheep’s pluck (heart, liver and lungs); minced with onion, oatmeal, suet, spices, and salt, mixed with stock, and traditionally encased in the animal’s stomach and simmered for approximately three hours” – is the answer. Clearly, the metabolic benefits of the pluck combined with the brown fat promoted by the suet, and the oatmeal-induced GI motility produces rapid weight loss. This must be why the Scots have some of the lowest rates of obesity globally…uhh, maybe not.

For those of us who spend time thinking, building, testing, and refining personalization algorithms, it’s all too easy to ignore their key limitation: they’re impersonal. We spend enormous amounts of time adapting algorithms to our user’s particular characteristics and context, but they will [likely] always lack empathy, sympathy, and basic kindness.

This post from Eric Meyer captures the challenge well. If you’ve used Facebook this holiday season, you’ve undoubtedly experienced its “Year in Review” posts hogging your timeline. For Meyer, the algorithmic “feature” reminded him of events that he will not soon forget:”

A picture of my daughter, who is dead.  Who died this year.

Yes, my year looked like that.  True enough.  My year looked like the now-absent face of my little girl.  It was still unkind to remind me so forcefully.

And I know, of course, that this is not a deliberate assault.  This inadvertent algorithmic cruelty is the result of code that works in the overwhelming majority of cases, reminding people of the awesomeness of their years, showing them selfies at a party or whale spouts from sailing boats or the marina outside their vacation house.

But for those of us who lived through the death of loved ones, or spent extended time in the hospital, or were hit by divorce or losing a job or any one of a hundred crises, we might not want another look at this past year.

My experience was similar to Meyer’s. When I received the prompt to see my “Year in Review,” I worked hard to ignore it. After a challenging year, I’m thankful for that decision.

For me, this experience is an important reminder: personalization can be impersonal. When designing personalization algorithms, we have to both predict what our users want to experience, and what experiences they want to avoid.

Update: Facebook apparently agrees.

A few weeks back I had the pleasure of joining colleagues Dr. Krishna Udayakumar, Dr. Ricky Bloomfield, and Dr. Bernard Fuemmeler on a panel discussing, Should there be an app for that?

Mid-rant that day, I found a tune floating (Ally McBeal-style) through my mind. Although I [uncharacteristically] repressed it then, I can do so no longer. Here goes:

“It’s hard out here for an app.”

Look, I am utterly bullish on the digital health market in general and apps specifically. But you can’t market enthusiasm [unless you’re Lil’ Jon]. And it’s the marketing of mHealth apps that presents some challenges to developers, particularly we academic ones.

The challenge? I call it The Rule of 70%.
TL; DR: People don’t download new apps, and when they do, they don’t use, and when they use, it’s not for long.

1. Most users don’t download new apps. Nearly 70% of users don’t download any apps in an average month. As a recent Quartz piece noted, “The top 7% of smartphone owners account for ‘nearly half of all download activity in a given month…'”

1352055-1-eng-US/Number_of_App_Downloads_Month.png

2. When they download new apps, most don’t use them very often. Nearly 70% use a new app 10 or fewer times; more than a quarter only open an app once.

localytics-app-user-retention-data-11.jpg

3. When they do use an app, they don’t do it for long. Nearly 70% of time spent in the top 3 apps (and the biggest time leeching app is [“duh”-alert] Facebook).

share_of_time_spent_by_rank.png

There is your challenge. So what’s left for those of us who are slogging away in those under-loved categories, like health/fitness (and weather for that matter)?

statistic_id270291_most-popular-apple-app-store-categories-in-september-2014

Here are a few thoughts:

Consider whether you need an app. We’re careful about building apps because in some cases, you can achieve broader reach and engagement using other digital health approaches (e.g., text messaging, interactive voice response, mobile web, email). These approaches can also be developed more rapidly and less expensively than [well designed] apps. Digital health often depends on engagement to produce optimal outcomes and non-app digital health approaches perform better than most apps great.

Make it stick, because people stop using apps pretty quickly. We’ve had a lot of success with this in research settings — our last study had 76% retention at 1-year. But many commercial apps struggle with stickiness. Digital health apps have a head start, as people are less likely to stop using relative to other categories (e.g. games). But if your app is opened only once in the first week, you have a 60% chance that they’ll stop using.. (BTW — wondering about our secret sauce? Ask for regular interaction and then use behavior change theory to figure out what feedback to provide).

Find a new distribution model. If you’re marketing to consumers, it’s unlikely that you’ll get many downloads, particularly after your initial launch. I know you have the next Facebook, but it’s unlikely that anyone outside of The Valley or The City will find you. This piece from Unread developer, Jared Sinclair recent went viral because it offered a rare glimpse inside the books of an independent app developer. And the view isn’t pretty:

Half of the lifetime sales of Unread were generated in the first five days. It would take another 170 days (24 weeks) to generate that same amount again… I conclude from all this that anyone who wants to make a satisfying living as an independent app developer should seriously consider only building apps based on sustainable revenue models.

This is a challenge in today’s digital health, because many users are young and [snacks, soda, and ETOH aside] pretty healthy. However, we need business models that target niche conditions and varying market segments (health systems, EHR vendors, HRA vendors, providers, consumer products companies, pharma). Researchers — you can’t ignore this one. If you’re deciding between making a standalone app, or one that connects with an EHR decision support tool, the latter will be a better dissemination bet.

We conduct digital health research trials almost exclusively in community health centers. These primary care settings are described as part of the nation’s “safety net,” but let’s be clear: for most of these patients, there is no catcher on the trapeze. Health centers are the primary, primary care option for our nation’s poor. Without them, we’d be certain to see [even more] rampant rates of suffering, poverty, and health care costs.

Health center docs are some of the most amazing physicians you’ll ever encounter. While I’m somewhat prone to hyperbole, I’m probably understating this particular point. Need evidence? This episode from David Plotz’s wonderful new podcast series aptly captures the amazing gifts that health center physicians provide to their patients [and our nation].

Interesting piece in Slate today about the demise of the mobile app.

… apps are very clearly not going to be around forever. Certainly not in their current, bulky square form. There isn’t enough mobile homepage real estate for each of the web’s 500-million-plus active websites to have its own app…

(The comments section — this time — is worth checking out)

This may seem like an extreme [read: click baiting] position, but the author has a point. With change in hardware configs (big phones with big screens, watches with small ones), we’re likely to see evolution in app interface designs as well.

My prediction? Hate to say it, but I think iOS 8’s “new” widget functionality portends where we might be going. I suspect apps will live in the background and/or cloud, with minimal interfaces, push notifications that are triggered by inputs, context, and biology, with as-needed full “screen” experiences.

The days of icons arrayed in rows and columns will be over sooner than we might expect.

In this week’s addition of Another Day, Another Wearable, we present the latest offering from Pivotal Living.

c9464e_b76da95d89af4e8ea229a8b04a7edf6a.png_srz_384_390_75_22_0.50_1.20_0.00_png_srz

Wrist-worn? Check.
UI? Check.
Bluetooth? Check.
Hypoallergenic? Hopefully.
Price? Free (with a $12 annual subscription to their premium app … so that’s more like free-ish).

I’ve long been predicting that we’re going to see wearables (at least the non Apple variety) rapidly enter Commodity Land. It’s becoming more difficult to differentiate on features (although FitBit is trying), it’s hard to justify high prices for cheap innards, and all the action is in the software anyway. Now, for the dreaded self-quote:

I predict that we’ll see these things [on discount] lining the shelves at your local drugstore. Just like paper towels. — Me

Whoops, wrong quote. Try this one.

Commoditization … will redirect the market’s focus where it should be: on the software. After all, the challenge isn’t the [data] collecting, it’s the changing. Tracker makers have done an excellent job in developing cool tools for gathering data, but they’ve been much less creative in designing software that reliably improves people’s health. And it’s the software that offers tracker makers the greatest opportunity for differentiation. — Me

I haven’t yet used the Pivotal Living app and on first glance, I don’t see much that isn’t available in other apps. And remember, we don’t know if any of these apps or devices actually work. But, what’s important is that we’re beginning to see new business models for wearables that prize (and price) the software. If this business model catches on, I suspect that developers will need ways to better differentiate their apps. This might create opportunities for those of us who’ve been clamoring to get evidence-based approaches into the app market. Admittedly, that’s a lot of ifs, but I’m willing to play the long game because I’m convinced that focusing on the changing (vs. the collecting) will give us the best shot at using wearables to improve consumer health.

I’ve long been interested in Steve Jobs’ approach to design. True, I’m a bit of an Apple fan-person, but my appreciation for the Apple Way runs deeper than that. I constantly struggle with my approach to developing digital health technologies. Is it theoretically-grounded, using evidence-based approaches, technically sound, adopting optimal user experience practices, gathering the right amount of patient feedback and so on.

Jobs’ approach was different. He knew what he liked and he thought we should (vs would) too. And he was right.

I knew that Jobs’ liked pretty things, but I hadn’t tied it to his love of the humanities. This Brian Lehrer interview of Walter Isaacson is a must listen. They cover a range of topics but most interesting is Isaacson’s argument that Jobs’ love of the humanities pushed him to create mashups of great technology and humanistic design principles.

It’s a good reminder for all of us who do this work.