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.

The only Apple Watch review you need

Lots of Apple watch reviews are out today.

(If you’re short on time, read Gruber’s review)

Taken together, the early reviews have been the tech equivalent of the multipurpose Southern maxim, “bless your heart.”

Apple Watch apparently needs work, even if no one wants to say it directly. From the NYTs Manjoo):

“The most exciting thing about the Apple Watch isn’t the device itself, but the new tech vistas that may be opened by the first mainstream wearable computer…For now, the dreams are hampered by the harsh realities of a new device.”

From Gruber:

“To me, the breakthrough in Apple Watch is the Taptic Engine and force touch. Technically, they’re two separate things. The Taptic Engine allows Apple Watch to tap you; force touch allows Apple Watch to recognize a stronger press from your finger.”

“And without taps, Apple Watch is rather dull.”

Call me contrary, but I think these reviews are pretty exciting. They sound precisely like those of the first iPhone.

ResearchKit is flawed [just like most research studies]

(Apologies for this week’s hyper-geeky posts. I’m about to double-down now. In fact, let’s call this part of an ongoing series on the promises and perils of ResearchKit for digital health science. How’s that?)

It seems like Research Kit is a winner out of the gates. Nevertheless, in the reaction to last Monday’s announcement, a number of reports have identified a common concern: limited generalizability.

The idea here is that people who own iPhones differ from the general population (particularly from those who pocket the Android). That’s true. TL; DR: iPhone users are more educated, higher income, and less likely to be males and racial/ethnic minorities, compared to Android users.

(Note: I’ve seen lots of blogs calling this selection bias. Selection bias is a potential issue with ResearchKit, and there are lots of potential selection biases depending on what kind of study you’re conducting. However, what seems to concern people most is limited generalizability).

This is a problem. From a research perspective, it means that what we learn from ResearchKit studies will only apply to iPhone users. But, here’s the thing: most studies have this problem.

Psychological research is based on studies of undergraduates — clear generalizability issues. Our most important research fundings about health risk factors come from big studies that are rife with generalizability issues. Nurse’s Health Study recruited nurses, in part because they are knowledgeable about health, and because they are extraordinarily conscientious about research participation. Framingham Heart Study recruited patients from a single city in the Northeast (an area that is far healthier than the rest of the country). We could go on all day like this.

That said (with widespread use), I think we can mitigate some of these concerns. Here’s how:

  • Recruit large sample sizes. This creates more variability (difference) in your study sample. With recruitment potential at a global scale, this is an area in which ResarchKit can really excel.
  • Expand ResearchKit to Android (and Windows). This would expand the pool of potential study participants to non-iPhone users. If Apple handles the open-sourcing appropriately (and there’s little reason to suspect otherwise), this shouldn’t be a major problem.This one is critical for another reason. Some historically disconnected populations are not only more likely to own smartphones, but they use their phone’s advanced data-related features (e.g., text messaging, watching videos, playing games, taking pictures) more than other groups. I think this might also extend to ResearchKit apps, when they’re properly designed.
  • Targeting specific populations. There is certainly merit in ResearchKit’s ability to recruit huge samples, but there’s also potential for micro-targeting specific groups. We’ll know more in a few weeks, but I suspect that many ResearchKit apps will be distributed using Apple’s enterprise features which allow one to [largely] bypass the App Store. This might allow us to identify folks who meet particular criteria (via social media, in clinic, Mturk) and screen in/out potential participants. I’m particularly bullish on the potential of ResearchKit to help us reach people with rare diseases — this has been a persistent challenge for the research world.

Look — I’m not minimizing the challenge here and the presence of biases in existing research studies is no excuse for introducing them into new studies. Indeed, we will have to be very careful about interpreting data from ResearchKit studies — at least in the short term.

Hyperbolic live blogging Apple’s ResearchKit

It’s always dangerous to post live comments during an Apple live event, particularly if you’re a rabid early adopter, fan-person admirer of Apple products, but oh well…

ResearchKit is an absolute gamechanger for health/medical research. It has potential to be the best thing to happen to behavioral research in a generation.

My real-time almost certain to be amended thoughts (in no order whatsoever):

  • ResearchKit will be open source. That’s great for all of the usual reasons. But it’s a savvy business move. This ensures less friction for integrating ResearchKit applications in National Institutes of Health grants. Counterintuitively [for those who don’t attend to these things], it will also help ease concerns about privacy.
  • We all struggle with patient recruitment, particularly when we don’t see them in clinic. Some of the biggest problems: finding people, recruiting, consenting, paying, and retaining them. Problem solved greatly mitigated.
  • ResearchKit might open a new market for study discovery and participant recruitment.
  • It its promotional materials, Apple is positioning ResearchKit for observational data collection. For this to work with intervention science, we’ll have to build ResearchKit hooks into health/medicine apps. It will be interesting to see what APIs Apple makes available. If Apple history is a guide, don’t expect this to happen right away.
  • If Apple allows ResearchKit to hook into non-resarch apps, watch out. Aside from cool new data, the commercial market for data aggregation will explode.
  • There is potential for changing the way that we run big cohort studies (e.g., Nurse’s Health Study, Jackson Heart Study, CARDIA, Framingham). Will it be cheaper to send every study participant an iPhone, versus the usual approach of creating, sending, scanning, and collating data from paper surveys? Probably. Incidentally, the National Institutes of Health has been funding fewer of these cohort studies, likely given resource constraints. Time to beef up on those epidemiology skills.
  • The ability to collect contextual data is going to be “great.” Beep. “We see that you’re inside Big Jo’s Burger Barn. How many minutes do you think it will take to burn that burger off?” Get ready for new science on in-vivo data collection.
  • We don’t yet know how ResearchKit will integrate with Apple Watch, but there is great potential for integrating new health metrics [particularly as Apple enhances Watch’s sensors].
  • Yes, some people will freak about the idea of a researcher collecting data from their Snapchatting device. There will be at least 1200 blog posts on the topic this week alone. I think that’ll be a short term problem though (how many cameras are pointed at you right now?).
  • That fancy new research data collection platform we’re creating [as I write]? History.

Prediction: apps are going the way of the dodo

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.

Humanities + Technology = Apple

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.