Separating Hyperplanes

LA Times partnered up with Kaiser to conduct a survey of employer-sponsored insurance. We've known that High Deductible Health Plans (HDHP) have been on the rise since Congress granted them special tax privileges beginning in 2005, and a key finding of the survey is that we're approaching the point where the average employer health plan is HDHP:

The average deductible for a single worker with a job-based insurance plan in 2006 was just $379, adjusted for inflation, according to an annual employer survey that KFF has conducted for more than two decades. By 2018, that figure had more than tripled to $1,350. Four in 10 U.S. workers have at least a $1,500 deductible — the threshold the poll used for high-deductible coverage for individuals.

The qualification to be a HDHP when they first came on the market in 2005 was a deductible at least $1,000, and that has risen to $1,350 today.

Now that HDHPs have taken over, we need to be honest that the empirical results show they aren't working out as we intended. These plans are great for employers because they shift expenses onto workers and cause them to indiscriminately cut all types of healthcare spending. But that isn't what was supposed to happen. Yes, HDHPs were suppose to restrain spending, but they were only supposed to do so through two mechanisms:

  1. encouraging patients to shop around for the lowest priced providers, and
  2. encouraging patients to prioritize high-value spending while cutting low-value services

The evidence shows that's not happening. Brot-Goldberg et al. (2017) looked at employers who switched to HDHP and found that patients slashed spending indiscriminately, without prioritizing high-value care nor shoping around for the lowest prices

The switch caused a spending reduction between 11.8% and 13.8% of total firm-wide health spending. We decompose this spending reduction into the components of (i) consumer price shopping, (ii) quantity reductions, and (iii) quantity substitutions and find that spending reductions are entirely due to outright reductions in quantity. We find no evidence of consumers learning to price shop after two years in high-deductible coverage. Consumers reduce quantities across the spectrum of health care services, including potentially valuable care (e.g., preventive services) and potentially wasteful care (e.g., imaging services).[emphasis added]

In other words, we've reduced healthcare spending by cutting valuable care rather than just waste, and harmed patients' finances in the process.

Employers, insurers, and governments have attempted on several occasions to rectify this by building price-transparency tools. Here's a study of two employers who built price-comparison tools that showed patients how much they'd owe out of pocket at each of the in network providers for various services. They didn't work:

After adjusting for demographic and health characteristics, being offered the tool was associated with a mean $59 (95% CI, $25-$93) increase in outpatient spending. Mean outpatient out-of-pocket spending among those offered the tool was $507 in the year before introduction of the tool and $555 in the year after. Among the comparison group, mean outpatient out-of-pocket spending changed from $490 to $520. Being offered the price transparency tool was associated with a mean $18 (95% CI, $12-$25) increase in out-of-pocket spending after adjusting for relevant factors. In the first 12 months, 10% of employees who were offered the tool used it at least once.[emphasis added]

Other attempts have met similar fates.

New Hampshire had a bit more luck with a multi-pronged effort that mandated all insurers disclose all claims to the state, paired with an app called HealthCost that would let patients enter their insurance plan and see projections of their out-of-pocket costs at each of the providers in the database. The results of the HealthCost experiment show a modest reduction in spending associated with shopping on the app.

Unlike other price transparency tools, the website could be used by all privately insured individuals in the state, potentially generating both demand- and supply-side effects. Exploiting variation across procedures available on the website as well as the timing of the introduction, estimates imply a 3 percent reduction in spending for visits with information available on the website. This is due in part to a shift to lower cost providers, especially for patients paying the highest proportion of costs. Furthermore, supply-side effects play a significant role—there are lower negotiated prices in the long-run, benefiting all insured individuals even if they do not use the website. Supply-side effects reduce price dispersion and are especially relevant when medical providers operate in concentrated markets.

With apps like this, an important part of the research question is always whether the app itself drives the result, or the context that accompanied the app's introduction. As the study makes clear, the context—mandatory disclosure of claims to the state—contributed a significant piece of the program without which supply-side effects driving prices down could not happen. One wonders, then, whether making these disclosures public and available to insurers directly would obviate the need for the consumer app.

There's evidence this trend towards HDHP has harmed patient care. Wharam et al. (2019) find that switching to high deductible plans caused women to skip breast cancer screenings and delay treatment:

We examined time to first breast cancer diagnostic testing, diagnosis, and chemotherapy among a group of women whose employers switched their insurance coverage from health plans with low deductibles ($500 or less) to plans with high deductibles ($1,000 or more) between 2004 and 2014. Primary subgroups of interest comprised 54,403 low-income and 76,776 high-income women continuously enrolled in low-deductible plans for a year and then up to four years in HDHPs. Matched controls had contemporaneous low-deductible enrollment. Low-income women in HDHPs experienced relative delays of 1.6 months to first breast imaging, 2.7 months to first biopsy, 6.6 months to incident early-stage breast cancer diagnosis, and 8.7 months to first chemotherapy. High-income HDHP members had shorter delays that did not differ significantly from those of their low-income counterparts. HDHP members living in metropolitan, nonmetropolitan, predominantly white, and predominantly nonwhite areas also experienced delayed breast cancer care.

The harms were mostly concentrated on the poorest most disadvantaged groups.

The HSA side of the HDHP has not performed any better. Brot-Goldberg et al. (2017) were not able to look at HSA balances but the Kaiser survey did:

Most of these people [who have HDHP] report making relatively modest contributions and having relatively modest savings in their HSA, particularly if they have lower incomes. For example, about half (47 percent) of people with an HSA plan say they have contributed less than $1,000 to their account in the past 12 months, and just 11 percent say they’ve contributed $5,000 or more. Among those with household incomes under $75,000 a year, 72 percent say they contributed less than $1,000 and just 3 percent put in at least $5,000.

HSA contribution limits are $7,000 for a family and $3,500 for individuals. Because of the tax treatment, there is a strong incentive to max these out regardless of your health needs, and yet most people enrolled in a HDHP do not even save enough to cover their deductible.

Personally I have a HDHP and I love my HSA. I max it out every year, and instead of reimbursing medical expenses from the HSA I keep the receipt so that I can withdraw that amount as totally tax-free income when I retire. It turns out there's no time limit on how long after you pay a medical bill you have to reimburse that expense to HSA. That is, you're substantially better off not using your HSA for healthcare at all.

You're better off not using your HSA for healthcare at all.

Your HSA is a second IRA, a super-IRA that's even better than the normal one, for people rich enough to be able to contribute to it and savvy enough to know the gimmicks. This is not good healthcare policy and it's time we admit that.

5/02/2019 12:30:00 PM

Rep. Pramila Jayapal has introduced a new Medicare for All bill. Like most of the other "Medicare for All" bills, this bill would not just extend Medicare to everyone, but also radically reform it by eliminating all-cost sharing and expanding what's covered. RAND estimates the program would increase healthcare spending by 1.8 percent, which is not very much, but they also imply that waiting times would increase:

We also include a supply constraint that results in unmet demand equal to 50 percent of the new demand. If there were no supply constraint, we estimate that total health expenditures would increase by 9.8 percent to $4.20 trillion.

This all makes plenty of sense. But as I always say, the most important part of these proposals is how they answer the question: "How will prices be determined?"

Although Jayapal's bill calls for reform in how CMS sets prices—something not in previous versions of Medicare for All—RAND did not believe the bill offered enough details to score the pricing reforms, and instead assume CMS initially continues to use the current price-setting process. Like Urban Institute's analysis of a previous Medicare for All bill (but unlike Mercatus), RAND does not believe that under the current process Medicare prices would follow current trajectory. Instead, they believe CMS would raise reimbursements to hospitals by 24 percent and to physicians by 7 percent. This is not enough to fully offset the loss of higher reimbursements from private insurers, but it does mean that on averagte hospitals are not looking at a huge cut.

4/15/2019 01:16:00 PM
(This meme generator was inspired by this)
2/28/2019 11:34:00 AM