Over the course of a year two out of every 1,000 people will rack up $100,000 or more in medical bills. These are the “high cost patients” who make insurance premiums go up and bedevil policy makers. These are people who are really sick.
The conventional wisdom is that with better prevention, wellness, care-coordination, insurance reforms, and so forth, medical spending could be averted. But one thought I’ve had, in response, was that a lot of the really sick people I’ve known are suffering from maladies like cancer or dementia that are not so easily preventable or avoidable.
There’s a lot of hand-wringing about how expensive our medical system has become. But what if it’s because we take good care of people who are sick and need the help? (God forbid!)
I recently came across a report commissioned last year on this topic by Genentech. The biotechnology unit of Roche produces expensive but effective therapies for serious conditions like colon cancer and age-related macular degeneration. Genentech has a special interest in, and one might even argues contributes to, high-cost care.
Milliman, the actuarial firm, performed the analysis and made some interesting discoveries. For background: Milliman is based in Seattle, works with major payers, providers and employers on health care costs, and provided this report to me for free. (You can read it here.)
The ten events or conditions that are most commonly expensive are as follows. These are average costs, so many patients even with these conditions will not reach the $100,000 per year mark:
1. HIV $25,000
2. Cancer $49,000
3. Transplant $51,00
4. Stroke $61,000
5 Hemophilia $62,000
6. Heart Attack including Cardiac Revascularization (Angioplasty with or without Stent) $72,000
7. Coronary Artery Disease $75,000
8. Neonate (premature baby) with extreme problems $101,000
9. End-Stage Renal Disease $173,000
10. Respiratory Failure on Ventilator $314,000
The most expensive condition, respiratory failure on a ventilator, is another way of describing the intensive care that patients receive at the end of life for a variety of conditions. Kidney failure may be from diabetes or hypertension, both of which could have been preventable.
Hemophilia is obviously congenital. Cancer is not always preventable. Being born early can sometimes be the result of insufficient prenatal care, or just bad luck. Heart disease, as the number one killer in the country, is expensive to treat, though not the most expensive.
I was surprised how low cancer was on the list. The report provides a bit more detail. If you have cancer that’s not being intensively treated (no chemotherapy, no surgery) it costs only $14,000 a year. This would apply to 40% of total cancer patients. However, cancer patients who receive surgery or chemotherapy (15% of the total) cost $123,000 on average. It wasn’t clear how the other 35% are classified.
Some terminal cancer cases may end up as ICU patients. (So they may be double-represented on a list like this.) But most cancer patients are not “catastrophic” according to the report. Overall, cancer affects about 4% of the total population of a health insurance plan.
The Milliman paper had a few other interesting findings worth reporting. Costs total $4,000 per year per capita. Maternity care costs about $9,000 per year including when C-sections are necessary. Diabetes patients represent between 2% and 3% of a health insurance plan’s population and can cost quite a bit to treat. But on average, diabetics’ costs range around $10,000 a year.
Milliman warns that for high-cost patients it is risky to use conventional insurance plan design tactics to try to reduce expenses. A $30 copayment instead of a $10 one is not going to change the overall budget when the care costs $100,000. By upping coinsurance–where you make patients pay 20% of the total cost, for example–you could make the clinical outcome worse by reducing compliance. To use the classic example: a diabetic might take fewer medications and end up in the hospital.
Click HERE to read this article in its entirety.
Think about the impact of increasing coinsurance for someone with cancer or hemophilia. It wouldn’t make care cheaper or better. But you would be sticking vulnerable patients with thousands of dollars of out-of-pocket bills. Here care-coordination might lead to greater savings. Milliman recommends designing health plans that are generous for high-cost patients.
The verdict, as I read it: As long as there are sick people there will be high-cost patients. It’s not a problem that’s so easily solved.