Health Insurance is Good For You (?)

Kim Bellard, July 11, 2011

Finally, a study that indicates health insurance is good for you! 

Granted, the study looked at Medicaid, for a low-income population, in a specific state, but in a time when health insurers are commonly castigated as villains, perhaps they can take some comfort in the findings.  Or perhaps not.

A brief recap of the study.  In 2008 Oregon realized it had funds to expand their Medicaid rolls (having substantially dropped them several years earlier).  They knew demand would exceed the number of slots available, so they set up a lottery that allowed potentially eligible consumers to enter; almost 80,000 did.  Researchers realized they finally had a classic study design – populations randomly selected into those who got health insurance and those who remained uninsured.  This appears to be the first such ever done when it comes to health insurance. 

The results from the first year indicate that those with health insurance did appear to benefit.  I won’t recap all the results, but some examples include that the newly insured saw doctors 35% more often, went to the hospital 30% more often, got mammograms 60% more often, and were 25 percentage points more likely to say their health was good or excellent.

Not surprisingly, they also cost 25% more.

The increased utilization may be partly an artifact of the so-called “woodwork” effect, i.e., people tend to use benefits more when they first get them (or, conversely, are about to lose them).  The researchers are continuing to mine the data, looking in particular at longer term impacts on health results.  Those extra doctor or hospital visits may or may not actually be improving health; several of my previous blogs (e.g., here) have discussed some of the overuse and errors that appear to be prevalent in the health care system. 

Earlier studies have tried to quantify the impact of health insurance, or lack thereof, have on health outcomes, such as those by the IOM or Mathematica.  Numbers like 44,000 deaths annually attributable to lack of health insurance have been claimed.  It’s not even only a matter of the uninsured not receiving services; even when they obtain services they may have worse outcomes.  Another recently released study indicates that payor status impacts mortality and morbidity for patients undergoing cardiac value operations, with privately insured patients faring better than either the uninsured or those with Medicaid.  So, any way you cut it, having insurance seems like a good thing. 

Clive Riddle recently posted a blog that reminded us how disproportionate health expenditures are distributed, with 5% of the population accounting for about half of the spending.  The fifty percent of the population with the lowest spending accounted for only 3% of all spending.  And that, for insurance purposes, is good.  Historically, and still true in other sectors, insurance is supposed to protect against unpredictable and catastrophic losses.  Anyone can get hit by lightening, catch a rare virus, or develop cancer.  Protection for the high expenses caused by relatively rare events that are outside of the individual’s control is what insurance does best. 

Insurance doesn’t work well at all when the insureds either already are incurring expenses or know they soon will be; for example, persons with chronic conditions.  People in this situation are basically depending on the other insured people to subsidize their expenses.  The data that Clive cited in his blog also noted the persistence of high spending for persons with chronic conditions; high spenders are no longer primarily one-time catastrophic cases, but more often now involve ongoing situations.  At some point, low cost people may balk at continuing to subsidize higher cost persons who have ongoing expenses; many would argue that this is exactly what is happening already in the individual market.

Don’t get me wrong; I am all in favor of reducing financial barriers to care for low-income people.  I’m all in favor of ensuring that health expenses don’t bankrupt families.  Medicaid has many, many flaws, but in any health care system we’re likely to get there will need to be some kind of special consideration for populations on whom financial burdens of health care fall hardest.  Where I start to scratch my head is how far those special considerations need to apply.

A common perception seems to be that without essentially full or low cost coverage for services, people will not get them.  The antipathy towards high deductible or consumer-directed health plans reflect this, with critics fearful that consumers will not behave responsibly when they are accountable for the initial deductible of a few thousand dollars.  PPACA similarly added requirements for coverage without cost-sharing of various preventive services, on the premise that consumers are too irresponsible to obtain services if they bear any financial burden when obtaining them.

A little Insurance 101 might be helpful.  If 100 people are covered, and one of them is likely to incur a $100 charge, then charging an annual premium of $1 per person (plus an add-on for administrative overhead) suffices.  However, if every person was going to get, say, an annual preventive visit that cost $70, then each person will have to pay a $70 annual premium (plus the add-on for administrative overhead) to pay for that visit.   There is no insurance, simply dollar trading with a mark-up. 

For skeptics who say people are irresponsible and won’t obtain appropriate services if there is cost-sharing for them, then I would argue (and have, here) that we have a far bigger problem.  Rather than waiving any financial involvement from patients, we’d be better off figuring out how to motivate patients to take better care of themselves and how to motivate physicians to ensure their patients are doing so. 

I.e., it’s not the coverage; it’s the behavior.  Coverage is necessary but not sufficient.  Sadly, we’ve built our delivery system and much of our health behavior around what insurance pays for rather than around makes the most sense from a health outcomes standpoint.  Or perhaps we’ve built our coverage around what is easy to pay for rather than what we should pay for.   Either way, it’s a problem.