John Hancock, a life insurer, is changing their business strategy per this Reuters article:
John Hancock, one of the oldest and largest North American life insurers, will stop underwriting traditional life insurance and instead sell only interactive policies that track fitness and health data through wearable devices and smartphones, the company said on Wednesday….
Policyholders score premium discounts for hitting exercise targets tracked on wearable devices such as a Fitbit or Apple Watch and get gift cards for retail stores and other perks by logging their workouts and healthy food purchases in an app.
data it has collected so far about customers’ activities suggest that it will, Tingle said, as Vitality policyholders worldwide live 13 to 21 years longer than the rest of the insured population.
From a social policy perspective, the question is whether or not the use of wearable exercise tracking devices are causative of improved mortality or merely correlated with underlying characteristics of a population that has longer lifespans than the rest of the population. From a business perspective, there is a market niche for being the life insurer that covers low risk, long lifespan individuals. This niche could be very profitable if the insurer has an expertise in being a superior hedge fund.
When I am thinking about this, I think of two things. The first is Centene; they are a Medicaid managed care insurer that is doing well to very well on the ACA exchanges. They have a deliberate business strategy to cover the lower risk cohort of a deliberately smaller market segment by compressing the Silver Spread. It is a profitable strategy as they can compress average state wide premium which allows them to under-compensate other insurers that cover the high cost risk. John Hancock may be doing the same thing.
The other thing I think about is this recent paper by Damon Jones, David Molitor and Julian Rief that looked at wellness programs. Aaron Carrol at the New York Times Upshot does a great job of digging into the niftiness of this paper.
Perhaps the greatest strength of the randomized controlled trial is in combating what’s known as selection bias. That occurs when groups being studied (intervention and control) are already significantly different after they are “selected” to be in the intervention or not. One of the most elegant examples of why we need such trials came recently in an examination of employer-sponsored wellness programs…
Almost all of those analyses are observational, though. They look at programs in a company and compare people who participate with those who don’t. When those who participate do better, we tend to think that wellness programs are associated with better outcomes. Some of us start to believe they’re causing better outcomes.The most common concern with such studies is that those who participate are different from those who don’t in ways unrelated to the program itself. …The results were disappointing. There seemed to be no causal effects.
Here’s the nerdy fun part, though. In addition to this analysis, the researchers also took the time to analyze the data as if it were an observational trial. In other words, they took the 3,300 who were offered the wellness program, then analyzed them the way a typical observational trial would, comparing those who participated with those who didn’t….
Wellness programs for health insurance purposes seem to be mainly a screening/cherry picking (choose your word based on your tolerable level of cynicism) exercise. They offer rewards to people who have underlying characteristics that predict low healthcare expenditures and they can punish either directly through surcharges or time sinks or indirectly by the non-receipt of rewards people with underlying characteristics that predict not low medical costs.
We have to ask ourselves if there are characteristics of people who use consumer wearables have fundamentally different characteristics that are relevant to mortality and morbidity compared to people who aren’t using or attracted to use wearables?
I would bet that someone with a Fitbit or an Apple Watch are likely to have higher incomes and higher net worth than people who don’t have one of those devices. I would bet that someone who has or is likely to buy a Fitbit already has some interest in physical activity and is (probably more importantly) in a social mileau that values physical activity or at least the body type that hints at a moderate level of physical activity. I would bet that people who can do a consistent 11,000 steps a day are likely to be younger than people who can’t do 11,000 steps a day. I would bet that people who can do a consistent 11,000 steps a day or at least see that as a reasonable goal, are in better physical condition than folks who look at a 5,000 step goal as a mountain to climb.
Maybe the presence of wearables and small incentives will meaningfully change behaviors and health outcomes. Or more likely, it is a trendy way for a life insurer to cherry pick their likely customer base while gracefully off-loading some of their bad risk to another insurer.
RobertB
My daughter’s high school had a Phys Ed class where you had to do X steps as recorded by Fitbit. Kids would sit there in class and tap their Fitbit in their hand to get the requisite number of steps. Just putting that out there.
Snarki, child of Loki
Are there insurers that cover BOTH life and health insurance?
Yes, I know that they are very different products, but having an economic incentive to provide health insurance so that their clients don’t DIE would seem to be a good thing.
Brachiator
I doubt that wearable data is meaningful. At best you have a group of younger, largrly healthy people recording the fact that they are healthy.
The wearables may correlate to a bunch of good activities, good diet, exercise, good sleep habits, which many people without smartphones can also do.
MomSense
Will they provide these devices or will we be forced to purchase them.
There’s actually one I’ve had my eye on, but it is not in my budget. Well I’d have to give up my gym membership for a year and then I could buy it.
p.a.
There are studies of fitbit etc accuracy and effectivenes. NIH for one:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4683756/
different-church-lady
Nice way to ruin the only segment of insurance that isn’t designed to be a game you can’t win.
different-church-lady
We’ve had sophisticated nutritional awareness for decades now, and yet people weigh more than ever. So I’m thinking you were just being charitable with that first scenario.
p.a.
Ugh… effectiveness. Also this compilation of studies of ‘effectivenes’ isn’t about whether these units enable positive behavioral changes. TL/DR: never mind.
evap
My workplace instituted a program where employees could earn money towards health insurance deductibles by using a fitness device and logging enough steps. There was also a team challenge with prizes. Many, many people participated and to encourage people to sign up, they sold Fitbits with a steep discount. I think it *did* get people to walk more.
Gin & Tonic
Does anyone know of one of these units for which you own and control your data, and not the vendor?
Steve in the ATL
I got a FitBit for a company challenge (the Labor team finished second out of fifty!). I move around a lot anyway, so I didn’t need that reminder, but it was helpful to see how much I sleep. Or, more accurately, how little I sleep.
After a few weeks of this, I left it on an airplane (hello Cleveland!). I feel so free now, without the tyranny of the tracker….
Cheryl from Maryland
Ugh. Thank God my husband got life insurance and long term heath care insurance before this skimming took place. Also that his insurers are NOT John Hancock. He lost the genetic lottery with multiple health issues which have nothing to do with lifestyle. His cardiac issues are due to calcium buildup in his arteries; no amount of Fitbitting will change anything. I feel sorrow for those younger who will get screwed.
Steve in the ATL
@MomSense: at my company, we provided low end ones for free to anyone who asked, and subsidized some of the nicer ones. My iPhone also tracks steps, so I don’t even need the FitBit anymore.
beth
I was enrolled in a “wellness” program run by our company’s insurance company – it consisted of a nurse calling me once a month and discussing nutrition and exercise. Didn’t change a thing I was doing or my health but I did get a discount on my premiums. My doctor suggested I join the hospital system’s program which is overseen by a cardiologist and consists of monthly individual meetings with a nutritionist and monthly nutrition classes, twice a week hour long workouts with a personal trainer in a small gym in groups of three or four, and regular weigh ins and physical assessments. This costs about $900 for six months. Since participating, I’m eating much better, I’m losing 1-3 pounds a week, my cholesterol, blood sugar and blood pressure are all down and now that I’m used to regular exercise I’m much more active on days I don’t have a workout session. I am truly lucky to be able to afford to do this. It’s a great program to get you on track to eating right and exercising regularly, not meant for you to be on it forever (I think they cap you at two years). This is the kind of program that would probably work for most people and would save the insurance companies a bunch of money in the long run. Of course it isn’t covered by insurance…..
OGLiberal
@Snarki, child of Loki: When I worked for Prudential years ago they provided both health and life insurance. I believe the health coverage was strictly group stuff, no individual, while the life insurance was both. (I worked in the individual life group – they were all very separate divisions) That was when Prudential did everything – life, health, P & C, brokerage, real estate, asset management, banking, etc. They have since gotten rid of health, P & C, banking and brokerage and are doing better as a result.
Brachiator
@p.a.:
Of course, even if the data is accurate, it is not necessarily meaningful or significant for an otherwise healthy person.
I think a couple of companies recently released devices that monitor the heart and oxygen levels. But for healthy people, this stuff is not data, it’s just an anecdote.
ProfDamatu
@Brachiator:
Well, sort of. I’d say you can get individually useful information from a Fitbit. For example, I have one of the models that tracks heart rate and, among other things, estimates resting heart rate each day. I can look at how my resting heart rate changes and, if it seems to be creeping up, I know that I’m not fully recovered from my workouts and need to back off for a couple of days. The actual number isn’t important, but the trend is useful. :-)
All of that is, of course, separate from the issue of whether it’s appropriate to use fitness tracker data to help set insurance rates. I vote no on that one.
stinger
Wow. I come from long-lived women on both my mother’s and father’s sides, I jog 2-3 miles most days, have healthy low blood pressure, am at a healthy weight, eat low fat and 6-8 vegs/fruit daily, get 7+ hours of sleep, am on no medications — but John Hancock won’t cover me unless I agree to wearing a “device” to “prove” I’m behaving in ways they approve of? This is like the opposite of what insurance is supposed to be. Sorry, Warden, I’m leaving the house without your damn ankle bracelet. I’ll pay my premiums to somebody else, thanks.
Miss Bianca
Best suggestion I’ve heard so far (via Slashdot): Strapping your Fitbit to your dog. In my case, a choice of two hyperactive mountain doggies who take themselves for walks that are *miles* longer than mine, these days. ; )
H’mm…wonder if “rolling on the ground”, with or without the presence of cow flop, or some other noisomely stenchous thing, shows up on the monitor…?
ProfDamatu
The other thing that jumps out at me here is the reference to wearable devices *and smartphones.* That could just refer to the fact that most smartphones have some sort of pedometer function built in, or it could indicate that they’re wanting people to use an app to collect data about things like diet. And that, of course, is the definition of “garbage in, garbage out” – people have no incentive to tell the truth if doing so would cost them more in premiums. Not to mention that, even if you’re OK with your insurer collecting that information, it’s still one more daily task that I imagine lots of people don’t want to do (even healthy/active ones – I’ve been wearing a Fitbit and using the app for years, but I still struggle with recording food intake; on a few months, off a few months, rinse and repeat).
ProfDamatu
@RobertB: Yup! Heck, even I sometimes sit at my desk and wave my arm around to get those last few “steps” for the hour. It reminds me of seeing students doing homework in the very noisy foyer of the gym at the university where I teach. I always used to wonder why they’d choose such a loud place to work…until one of my students told me that part of the required “health 100” class was going to the gym and doing some type of workout a couple of times a week. To use gym facilities here, you swipe in and out with your ID card, and the class used that information to monitor students’ completion of that requirement…so, students would swipe in, do homework in the foyer for an hour or so, and then swipe out, their “workout” complete. :-P
Martin
Thanks for covering this, David.
@Gin & Tonic: The Apple Watch. All the data is yours, stored in your account. They cannot use it, and they do not sell user data ever. If you want other apps/services to tap into that data, that’s done through a local app that you provide explicit authorization for, and which you can delete whenever you want.
I feel this is the best approach to user privacy – local rather than cloud based access means you can add/remove services in an obvious way. The interface for managing this on the Watch is through the connected iPhone. There are tabs for Settings->Privacy->Health and Settings->Privacy->Motion & Fitness that list all apps that you’ve approved to access that data, and you can remove them from that tab.
I asked the question of David yesterday, because there are some other cases here other than the individual market. I have an employer based plan, so there’s really no cherry picking. But I have a minor heart condition (paroxysmal supraventricular tachycardia) which means that during exercise my heart rate can go very high, so for me having a monitor when I exercise is something of a necessity. My daughter has a different heart condition (postural orthostatic tachycardia) which causes her heart rate to sometimes spike when she stands up suddenly, which then drop suddenly upwards of a minute later, which for a while caused her to pass out in the bathroom every morning when she was getting ready for school.
For years I thought I was healthy and that feeling I got after strenuous exercise was normal, but it turns out it wasn’t – it was basically my heart trying to not explode. It took a fitness class in college to spot the problem and then get it tested. My daughter feeling lightheaded in the morning was normal for her and it wasn’t until she started fainting that her’s got diagnosed. There are lots of seemingly perfectly healthy people walking around with AFib and other conditions that are undiagnosed. And it’s not that these things necessarily need medical treatment, mine and my daughter’s mostly just need a feedback system. I need a little alarm to tell me to back off my exercise. My daughter needs a warning that she shouldn’t get up at that moment, or she should lay down if she’s about to pass out. These are a form of prevention.
Martin
@stinger:
No, insurance is just risk pooling for low likelihood/high cost events. If you can decrease either the likelihood or cost of those events, that’s (in an ideal system) equivalent to a lower premium. That’s why you get a lower car insurance rate when you have more airbags, or active braking systems, etc. The former reduces the cost of the event, the latter the likelihood of it.
ProfDamatu
@Martin:
Right. But again, for me, the sticking point with the kind of data John Hancock wants to use in the health insurance context is that many expensive conditions aren’t amenable to the kind of reduction in likelihood that’s supposed to reduce costs. So we’re now talking about instituting fairly intrusive monitoring and/or imposing bookkeeping tasks (e.g., dietary monitoring) on people for little benefit. And, as mentioned in the OP, there’s also the fact that when employers have tried to reduce insurance costs through wellness programs, the research seems to indicate that selection bias is responsible for most, if not all, of the positive results; the people who hit the Fitbit targets and such are the people who would be exercising and eating right regardless, and the tracker use isn’t motivating most people, at least not over the long term.
NCSteve
No different than Vegas throwing you out of the casino for winning too much at the skill games against the house. The question is whether regulators will put up with a scheme intended to render vast swaths of the general population incapable of getting life insurance.
J R in WV
My Medicare Suppemental insurance, acquired via my retirement from state Insurance, is via Humana. I get phone calls from them asking stupid questions about my last doctor’s appointment. How long did I have to wait less than 15 minutes, less than 30 minutes, etc. When I visit my doctor it’s usually an all day thing for annual physical, just a blood draw, weight, BP quarterly, but still a long visit.
Humana also wanted HARD to have an in-home RN visit, always “optional, my choice” and I always turn it down. They send me literature about my diabetes, which I do not have, and my cardiac condition, which I do not have. I assume they get my weight/height/age from my doctor, and leap to conclusions from that tiny slice of data.
I stopped answering their phone quizes and have turned down the home health care visits, hope they leave me alone. My family (members who don’t smoke, as I do not) tend to live a very long time w/o much medical intervention. Hope they leave me alone!
J R in WV
@NCSteve:
I worked with a very smart Burmese guy with degrees from UT El Paso, who went to Las Vegas and made money at Black Jack often, and after being asked to leave one casino, found that all the other casinos wouldn’t allw him to “gamble” any more.
Personally, if they aren’t going to allow “winners” in, they shouldn’t be allowed to call in gambling, if they throw out the people who are competent at the “games” they offer. It should be called STEALING Houses, not gambling casinos. I don’t get that brain buzz from gambling, have spent maybe $40 in my whole life on bets.
Brachiator
There’s an annoying puff piece/review of a new Samsung smartwatch over on Gizmodo that includes this little nugget:
It costs $330 and the review says nothing about privacy of data. It has settings that supposedly tracks an oddly specific set of exercises, and this only emphasizes the stupidity of an insurance company that would focus on people who use these devices. They would be essentially saying, “we will insure you only if you go rowing as part of your exercise regime.”
ProfDamatu
@Brachiator:
Yeah, that’s quite the puff piece. I would make one small defense here, though – exercises that are automatically tracked aren’t really *oddly* specific to my mind; they’re all going to be things that result in characteristic arm movement patterns in most people. For anything else, you have to tell it what you’re doing; the value there would be, I suppose, finding out just how taxing doing planks is for you in terms of heart rate, lol. The automatic exercise tracking is just a bonus to make the device more user friendly.
This does really point up another pitfall in terms of insurers trying to use this sort of data to set rates, though. It’s very possible that users won’t bother with the non-autodetected workout modes, and so if they don’t do any of the exercises the device auto-detects, it’s going to look to the company like they don’t exercise (I don’t trust that the insurance company would look at, say, HR data and realize that yeah, that’s an unmarked workout there).
EthylEster
I recall reading an article recently, probably in the NYT, that reported that the gains from these activity programs did not really make much difference in the long run. That is, good things happen at the beginning but the positive trends disappear after not very long. Like a lot of things.
ProfDamatu
@EthylEster: I remember reading something like that too. No different from most diet and exercise programs – many if not most people can stick to the program for awhile, and see benefits, but after a few months most people have stopped following the program and the gains disappear. And of course that’s on top of the tendency for the “workplace wellness” versions of these programs to (however inadvertently) cherry-pick the people who were going to exercise/eat right regardless of any program.
I have to say, even the fairly benign version offered by my employer irritated me a bit in terms of its structure. Basically, the program was set up to reward both becoming more active as recorded by a pedometer, and weight loss, with the best rewards being determined by percentage of body weight lost over a certain period of time. I had to keep reminding myself that the purpose of the program was to encourage the adoption of healthy habits; therefore by definition it was not meant for me, so I needed to chill the heck out about the fact that I couldn’t get the cool rewards, lol. And I also have reservations about encouraging people to compete in terms of weight loss generally. Percentage of body weight lost is at least better than basing it on total amount lost (both in terms of fairness – if it’s just number of pounds, it’s going to be much, much more difficult for a woman to win – and in terms of possibly encouraging obsession with the number on the scale), but when the program is essentially unmonitored…I’m not sure how healthy it is to get people focused on that goal.