With an exciting new job this year comes–exciting new health insurance. I’m sure you all agree: picking a new insurance plan is a delightful experience. There’s a guided questionnaire, asking about my lifestyle and health choices, to offer up the best plan for me. I can click on scenarios and see immediately how much I’ll pay for what services, and for the whole year.
I love how, leaning into newly available technology, I this year have the choice to be rewarded for my healthy lifestyle by my insurance. Since 2017, I’ve been focused on adding healthy habits each year–eating whole foods and cooking at home, running 4x per week, doing weight workouts 2x per week, intermittent fasting, good sleep. I’ve tracked my progress with a Garmin smartwatch, an at-home blood pressure monitor, and either continuous glucose monitor (CGM) or Keto-Mojo, a ketone + glucose test kit.
I’ve agreed to share all this data with my insurance carrier. They track that I get in enough aerobic exercise, that I sustain or increase my muscle mass (via an annual dexa scan I’ve agreed to), and that keep my blood pressure and glucose in optimal ranges. My healthy choices are then rewarded with a “healthy living” discount. This month, that comes out to $250 off per month–plenty to pay for my gym membership and new running gear, all sent to me in an email titled “Congrats: you’re in the top 10% again–treat yourself!”
My health insurance is a dream! Then I wake up, and health insurance turns from a dream to the nightmare that all of us are all too familiar with.
In reality, I’m faced with insanely expensive insurance options that cover next-to-nothing. I have to navigate plan documents so confusing that even after building elaborate spreadsheets I can’t quite figure out what to budget for regular care for my family (let alone what a major surgery might cost us).
When I look beyond the crappy confines of health insurance in the US, I wonder why new tech and innovations from other markets haven’t made any inroads in healthcare–even as healthcare is a huge sector of the US economy.
Why is there a “safe driving” discount on car insurance, but not a “healthy living” discount for health insurance?
Encouraging and empowering people to live healthier should be a win-win proposal: People benefit from being well—not suffering from chronic diseases, not needing to pay for expensive medications, and having the energy and health to enjoy life to the fullest, maybe into their 80s and 90s. Employers benefit because healthier people can be more productive. Insurance companies benefit because healthier people cost less to insure.
Yet in the US, the best most insurance companies offer is a small subsidy for a fitness club membership (maybe). No discount for losing weight—although losing weight can help benefit the heart. No incentives for aerobic exercise—although VO2 max is one of the best predictors of lower all-cause mortality. No offer to pay for a supervised nutrition and exercise program (or fasting) instead of diabetes medication, even though companies such as VirtaHealth have shown that lifestyle interventions can cure diabetes (instead of it being a life-long chronic illness when treated with medication).
I haven’t done a systematic study of why there isn’t a healthy living discount in health insurance (someone should!). But I have two strong hypotheses:
Health insurance rules make it hard to create incentive programs.
My background is in business, so I naturally think about how business people might improve their bottom line. If I worked in health insurance and wanted to improve my bottom line, I might look at some expensive treatments that many of my insureds need. Medication for diabetes is one of those treatments. It costs an insurance company almost $10K to care for someone with diabetes; diabetics have medical expenditures about 2.3 times higher than non-diabetics. What if the insurer offered the patient a deal: if you reverse your diabetes (as measured by insulin and hemoglobin A1C or HbA1c test or via ongoing readings with a continuous glucose monitor or CGM), we’ll pay you $5,000 (or, if they pay privately for their insurance, we’ll give you a year-end cost refund of $$). The insurer would save $10,000, and pass $5,000 on to the insured—both win. (I’m simplifying here—of course there is a cost to the one-time treatment to reverse diabetes, whether that’s an intense nutrition and exercise program, or an on-site extended fasting program; more on that below.)*
Or, instead of paying for an outcome (healthy blood sugar), the insurer could incentivize behaviors they know lead to better health. For a diabetic person, this might involve eating a ketogenic diet or practicing intermittent or multi-day fasting, both of which can be monitored via a CGM. Or it could be committing to and tracking a certain level of exercise—say six sessions per week of 30 minutes at vigorous intensity (a defined heart rate, tracked via a smartwatch).
Of course, the details would need to be worked out and there are lots of questions (e.g., how does this work for people who are already metabolically healthy vs. those who aren’t). There are some non-insurance companies which use biofeedback and lifestyle incentives to improve health outcomes, such as VirtaHealth, Levels, Vitality, and HealthyWage.
So why aren’t the big insurers at least experimenting with these types of incentives, when the payoff could be so high?
I asked some people who work in the D.C. policy world, and one of them pointed me to this write-up on HIPAA and the Affordable Care Act Wellness Program Requirements. It is dense reading—but the basic gist, as far as I can tell, is that insurers cannot differentiate rates based on health factors. There is an exception for wellness programs—however, the wellness benefit needs to be made available to all participants, whether or not they actually achieve any health-relevant improvement. So, for example, a reward for participating in a smoking cessation program would need to be provided “without regard to whether the employee quits smoking”. The requirements are complex and I imagine that insurance plan administrators might be worried about experimenting lest they be taken to court for failing one of the five elaborate non-discrimination standards, especially as there is also a strict limit on how much of an incentive can be offered.
To profit from a program, an insurer would need to have some confidence that a large number of the people they reward for a certain behavior or outcome would actually end up healthier and needing less costly care. If they need to provide incentives to both those who become healthier and those who don’t the program is much less likely to pay.
Employer-linked health insurance misaligns incentives by making insurance companies short-term focused.
While I think that the complexity of regulation (and fear of lawsuits) makes it hard for these programs to work, I think the fundamental problem is probably much more deeply embedded in our health insurance system.
In the US, most people get health insurance through their employer. The employer decides, anew each year, what insurance plans to offer, and from which insurer. Plus, employees change jobs frequently: the median tenure in a job in 2022 was 4.3 years for men and 3.8 years for women. As a result, the insurer typically has only a four-year relationship with any given insured person. Their financial time horizon is thus short: US insurers don’t optimize for the lowest possible healthcare cost over a person’s lifetime; they optimize for four years.
Going back to our diabetes example above, if it cost $20,000 to reverse someone’s diabetes and the insurer passed on half of the $10,000 savings to the insured, the insurer wouldn’t break even on until somewhere in year 5—by which time the insured is likely covered by another insurance company which then reaps the financial benefits.
If insurance was portable—if it was a contract between the insured and the insurer that stayed consistent beyond the 4-year-term of employment—incentives would be quite different. Germany, my country-of-origin, has a very different system, explained in detail in this NBER working paper, German Long-Term Health Insurance: Theory Meets Evidence. In Germany, individuals enrolled in private health insurance sign long-term contracts that are theirs to keep as long as they want to keep them, independent of employment status:
When individuals apply for a GLTHI [German Long-Term Health Insurance] contract, insurers have the right to deny applicants with bad risks or impose pre-existing condition clauses. However, once contracts are signed, insurers cannot terminate them. GLTHI contracts are not annual contracts, but permanent lifetime contracts without an end date. In other words, GLTHI contracts are guaranteed renewable over the lifecycle. However, policyholders can terminate these permanent contracts and switch insurers. Thus, the GLTHI is a market with one-sided commitment. It is very common that policyholders keep their GLHTI contract until they die: Medicare does not exist in Germany.
The paper goes into 80 pages of detail on this system and its many differences to the US system. What’s relevant here is that Germany’s private insurance system is an example of a life-long relationship between an insurer and an insured. This makes it possible to align incentives: in a system like this, at least theoretically, an insurer would profit from paying $20,000 once to treat diabetes, as they could recoup the annual savings potentially for decades. (I don’t know if insurers actually experiment with this type of thing in Germany—that would be an interesting thing for someone to research)**.
Interestingly, the tie between health insurance and employment in the US wasn’t something strategically planned. Instead, it grew out of a labor shortage in World War II, where employers offered health insurance to compete for workers—and an IRS decision to allow employers to pay this insurance cost with pre-tax dollars, while individuals had to keep paying for insurance in after-tax dollars, a pattern that still is the case today. As the New York Times puts it, “There are other countries with private insurance systems, but none that rely so heavily on employer-sponsored insurance. There are almost no economists I can think of who wouldn’t favor decoupling insurance from employment.”
Employer-based insurance isn’t a purposefully developed system. Instead, it’s what I call a “red tape fossil”—an artifact of historical developments that exists not because it makes sense (it so doesn’t!), but only because it came into existence and is now so deeply entrenched that it’s hard to imagine a world where it didn’t exist.
A system like Germany’s, where insurance is bought by a person and stays with them for life would provide many benefits (several of which are discussed in the New York Times and NBER pieces). I think one of the biggest indirect benefits might be aligning incentives so we can reap the benefits for human health and flourishing, especially now that we have so much personal health technology available to track both behaviors and outcomes. Healthcare spending in the US was around $3.5 trillion per year in 2017, almost 18% of GDP—and much of it is spent on the treatment of chronic illness, exactly the type of medical cost that could be avoided if only we used technology and aligned incentives to enable people to improve their health before they get really sick. More importantly, as Eli Dourado points out, extending life by improving health could have huge pay-offs:
The monetary value of a quality-adjusted life-year is often estimated to be on the order of $100,000. With more than 300 million people in the US, every QALY we add to average US lifespan adds around $30 trillion in value.
I wonder what progress we could unleash if we found a way to fix our broken health insurance system? What if insurers were encouraged to experiment with incentives? What if there was a way to even on a small scale experiment with portable health insurance that stays with you for years, instead of being tied to employment?
What innovations might we see if it was possible and profitable to offer “healthy living” incentives on health insurance, just like we have “safe driving” discounts for car insurance?
I’ll keep writing on this topic as I think it’s one of the big needles to drive progress in the US. I hope people reading this will take action, too: Write to your representative and ask them to think about ways to enable more experimentation and innovation in preventative care. Be mindful of the red tape fossils like employer-based care and how they mess up incentives. And, if you work in policy or health insurance I’d love to hear any thoughts you have on how to work towards a world where preventative care gets the attention it needs to help Americans live longer, healthier lives.
* I recently read of a rare case of insurance paying for extended fasting to treat diabetes. In 2001, a California union made a fasting program available as a covered benefit to their four hundred thousand members. The fasting program reduced the costs of medical and drug treatment per member by an average of $2,700 in the first year alone, more than paying back the cost of treatment. Interestingly, this self-insured union had long-term relationships with its employees, which means it stood to save for years to come from every member who participates in this program.
** Many private insurance plans in Germany have at least one feature that aligns incentives. When I was enrolled in private insurance in the 1990s at the start of my career, if I didn’t submit any medical expenses for reimbursement I received a year-end refund (“Beitragsrückerstattung”) on part of my health insurance premiums. The refund amount increased over time: the more years I didn’t need to tap into my insurance, the higher the annual refund–up to 50% of all the premiums paid! Apparently, this is still the case today: this article (in German) shares that people who only use preventative care may receive up to 30% off the annual premium back–in the example, Euro 1,800. In this scenario, someone with diabetes who reversed their disease and can cut or eliminate their reliance on expensive medication would be able to capture some of the financial benefits. (The article interestingly also mentions that insurance premiums can be deducted from individual tax returns–and that refunds need to be netted out.)
Why is there a “safe driving” discount on car insurance, but not a “healthy living” discount for health insurance?
Link post
With an exciting new job this year comes–exciting new health insurance. I’m sure you all agree: picking a new insurance plan is a delightful experience. There’s a guided questionnaire, asking about my lifestyle and health choices, to offer up the best plan for me. I can click on scenarios and see immediately how much I’ll pay for what services, and for the whole year.
I love how, leaning into newly available technology, I this year have the choice to be rewarded for my healthy lifestyle by my insurance. Since 2017, I’ve been focused on adding healthy habits each year–eating whole foods and cooking at home, running 4x per week, doing weight workouts 2x per week, intermittent fasting, good sleep. I’ve tracked my progress with a Garmin smartwatch, an at-home blood pressure monitor, and either continuous glucose monitor (CGM) or Keto-Mojo, a ketone + glucose test kit.
I’ve agreed to share all this data with my insurance carrier. They track that I get in enough aerobic exercise, that I sustain or increase my muscle mass (via an annual dexa scan I’ve agreed to), and that keep my blood pressure and glucose in optimal ranges. My healthy choices are then rewarded with a “healthy living” discount. This month, that comes out to $250 off per month–plenty to pay for my gym membership and new running gear, all sent to me in an email titled “Congrats: you’re in the top 10% again–treat yourself!”
My health insurance is a dream! Then I wake up, and health insurance turns from a dream to the nightmare that all of us are all too familiar with.
In reality, I’m faced with insanely expensive insurance options that cover next-to-nothing. I have to navigate plan documents so confusing that even after building elaborate spreadsheets I can’t quite figure out what to budget for regular care for my family (let alone what a major surgery might cost us).
When I look beyond the crappy confines of health insurance in the US, I wonder why new tech and innovations from other markets haven’t made any inroads in healthcare–even as healthcare is a huge sector of the US economy.
Take other insurance, for example. Car insurance companies offer safe driver discounts—and now, with technology, even higher discounts if you agree to share your driving data with the insurance company. Life insurance companies similarly offer fitness discounts for healthy individuals. Home insurance offers discounts for such things having sprinklers, having a monitored security system, or creating fire-safe zones around your home.
Why is there a “safe driving” discount on car insurance, but not a “healthy living” discount for health insurance?
Encouraging and empowering people to live healthier should be a win-win proposal: People benefit from being well—not suffering from chronic diseases, not needing to pay for expensive medications, and having the energy and health to enjoy life to the fullest, maybe into their 80s and 90s. Employers benefit because healthier people can be more productive. Insurance companies benefit because healthier people cost less to insure.
Yet in the US, the best most insurance companies offer is a small subsidy for a fitness club membership (maybe). No discount for losing weight—although losing weight can help benefit the heart. No incentives for aerobic exercise—although VO2 max is one of the best predictors of lower all-cause mortality. No offer to pay for a supervised nutrition and exercise program (or fasting) instead of diabetes medication, even though companies such as VirtaHealth have shown that lifestyle interventions can cure diabetes (instead of it being a life-long chronic illness when treated with medication).
I haven’t done a systematic study of why there isn’t a healthy living discount in health insurance (someone should!). But I have two strong hypotheses:
Health insurance rules make it hard to create incentive programs.
My background is in business, so I naturally think about how business people might improve their bottom line. If I worked in health insurance and wanted to improve my bottom line, I might look at some expensive treatments that many of my insureds need. Medication for diabetes is one of those treatments. It costs an insurance company almost $10K to care for someone with diabetes; diabetics have medical expenditures about 2.3 times higher than non-diabetics. What if the insurer offered the patient a deal: if you reverse your diabetes (as measured by insulin and hemoglobin A1C or HbA1c test or via ongoing readings with a continuous glucose monitor or CGM), we’ll pay you $5,000 (or, if they pay privately for their insurance, we’ll give you a year-end cost refund of $$). The insurer would save $10,000, and pass $5,000 on to the insured—both win. (I’m simplifying here—of course there is a cost to the one-time treatment to reverse diabetes, whether that’s an intense nutrition and exercise program, or an on-site extended fasting program; more on that below.)*
Or, instead of paying for an outcome (healthy blood sugar), the insurer could incentivize behaviors they know lead to better health. For a diabetic person, this might involve eating a ketogenic diet or practicing intermittent or multi-day fasting, both of which can be monitored via a CGM. Or it could be committing to and tracking a certain level of exercise—say six sessions per week of 30 minutes at vigorous intensity (a defined heart rate, tracked via a smartwatch).
Of course, the details would need to be worked out and there are lots of questions (e.g., how does this work for people who are already metabolically healthy vs. those who aren’t). There are some non-insurance companies which use biofeedback and lifestyle incentives to improve health outcomes, such as VirtaHealth, Levels, Vitality, and HealthyWage.
So why aren’t the big insurers at least experimenting with these types of incentives, when the payoff could be so high?
I asked some people who work in the D.C. policy world, and one of them pointed me to this write-up on HIPAA and the Affordable Care Act Wellness Program Requirements. It is dense reading—but the basic gist, as far as I can tell, is that insurers cannot differentiate rates based on health factors. There is an exception for wellness programs—however, the wellness benefit needs to be made available to all participants, whether or not they actually achieve any health-relevant improvement. So, for example, a reward for participating in a smoking cessation program would need to be provided “without regard to whether the employee quits smoking”. The requirements are complex and I imagine that insurance plan administrators might be worried about experimenting lest they be taken to court for failing one of the five elaborate non-discrimination standards, especially as there is also a strict limit on how much of an incentive can be offered.
To profit from a program, an insurer would need to have some confidence that a large number of the people they reward for a certain behavior or outcome would actually end up healthier and needing less costly care. If they need to provide incentives to both those who become healthier and those who don’t the program is much less likely to pay.
Employer-linked health insurance misaligns incentives by making insurance companies short-term focused.
While I think that the complexity of regulation (and fear of lawsuits) makes it hard for these programs to work, I think the fundamental problem is probably much more deeply embedded in our health insurance system.
In the US, most people get health insurance through their employer. The employer decides, anew each year, what insurance plans to offer, and from which insurer. Plus, employees change jobs frequently: the median tenure in a job in 2022 was 4.3 years for men and 3.8 years for women. As a result, the insurer typically has only a four-year relationship with any given insured person. Their financial time horizon is thus short: US insurers don’t optimize for the lowest possible healthcare cost over a person’s lifetime; they optimize for four years.
Going back to our diabetes example above, if it cost $20,000 to reverse someone’s diabetes and the insurer passed on half of the $10,000 savings to the insured, the insurer wouldn’t break even on until somewhere in year 5—by which time the insured is likely covered by another insurance company which then reaps the financial benefits.
If insurance was portable—if it was a contract between the insured and the insurer that stayed consistent beyond the 4-year-term of employment—incentives would be quite different. Germany, my country-of-origin, has a very different system, explained in detail in this NBER working paper, German Long-Term Health Insurance: Theory Meets Evidence. In Germany, individuals enrolled in private health insurance sign long-term contracts that are theirs to keep as long as they want to keep them, independent of employment status:
The paper goes into 80 pages of detail on this system and its many differences to the US system. What’s relevant here is that Germany’s private insurance system is an example of a life-long relationship between an insurer and an insured. This makes it possible to align incentives: in a system like this, at least theoretically, an insurer would profit from paying $20,000 once to treat diabetes, as they could recoup the annual savings potentially for decades. (I don’t know if insurers actually experiment with this type of thing in Germany—that would be an interesting thing for someone to research)**.
Interestingly, the tie between health insurance and employment in the US wasn’t something strategically planned. Instead, it grew out of a labor shortage in World War II, where employers offered health insurance to compete for workers—and an IRS decision to allow employers to pay this insurance cost with pre-tax dollars, while individuals had to keep paying for insurance in after-tax dollars, a pattern that still is the case today. As the New York Times puts it, “There are other countries with private insurance systems, but none that rely so heavily on employer-sponsored insurance. There are almost no economists I can think of who wouldn’t favor decoupling insurance from employment.”
Employer-based insurance isn’t a purposefully developed system. Instead, it’s what I call a “red tape fossil”—an artifact of historical developments that exists not because it makes sense (it so doesn’t!), but only because it came into existence and is now so deeply entrenched that it’s hard to imagine a world where it didn’t exist.
A system like Germany’s, where insurance is bought by a person and stays with them for life would provide many benefits (several of which are discussed in the New York Times and NBER pieces). I think one of the biggest indirect benefits might be aligning incentives so we can reap the benefits for human health and flourishing, especially now that we have so much personal health technology available to track both behaviors and outcomes. Healthcare spending in the US was around $3.5 trillion per year in 2017, almost 18% of GDP—and much of it is spent on the treatment of chronic illness, exactly the type of medical cost that could be avoided if only we used technology and aligned incentives to enable people to improve their health before they get really sick. More importantly, as Eli Dourado points out, extending life by improving health could have huge pay-offs:
I wonder what progress we could unleash if we found a way to fix our broken health insurance system? What if insurers were encouraged to experiment with incentives? What if there was a way to even on a small scale experiment with portable health insurance that stays with you for years, instead of being tied to employment?
What innovations might we see if it was possible and profitable to offer “healthy living” incentives on health insurance, just like we have “safe driving” discounts for car insurance?
I’ll keep writing on this topic as I think it’s one of the big needles to drive progress in the US. I hope people reading this will take action, too: Write to your representative and ask them to think about ways to enable more experimentation and innovation in preventative care. Be mindful of the red tape fossils like employer-based care and how they mess up incentives. And, if you work in policy or health insurance I’d love to hear any thoughts you have on how to work towards a world where preventative care gets the attention it needs to help Americans live longer, healthier lives.
* I recently read of a rare case of insurance paying for extended fasting to treat diabetes. In 2001, a California union made a fasting program available as a covered benefit to their four hundred thousand members. The fasting program reduced the costs of medical and drug treatment per member by an average of $2,700 in the first year alone, more than paying back the cost of treatment. Interestingly, this self-insured union had long-term relationships with its employees, which means it stood to save for years to come from every member who participates in this program.
** Many private insurance plans in Germany have at least one feature that aligns incentives. When I was enrolled in private insurance in the 1990s at the start of my career, if I didn’t submit any medical expenses for reimbursement I received a year-end refund (“Beitragsrückerstattung”) on part of my health insurance premiums. The refund amount increased over time: the more years I didn’t need to tap into my insurance, the higher the annual refund–up to 50% of all the premiums paid! Apparently, this is still the case today: this article (in German) shares that people who only use preventative care may receive up to 30% off the annual premium back–in the example, Euro 1,800. In this scenario, someone with diabetes who reversed their disease and can cut or eliminate their reliance on expensive medication would be able to capture some of the financial benefits. (The article interestingly also mentions that insurance premiums can be deducted from individual tax returns–and that refunds need to be netted out.)