Philadelphia Reflections

Economics of Chronic Disease and Catastrophic Illness

Until rather recently, most wounded soldiers could expect to die from their wounds. George Washington died of an infected throat, an unlikely outcome today. Infectious diseases are now much less likely to be fatal, while preventive measures are in the process of eliminating fatal heart attacks and strokes. It was mentioned in earlier sections that the two most medically expensive years of anyone's life are likely to become the first and last years of a considerably extended lifetime.

If we could be sure of that, or its timing, it would make modern health insurance design easier. Some people can be expected to outlive a formerly fatal condition, and it should be possible to raise annual premiums a little to account for it. If we were smart, we would try to lower premiums by taking advantage of the extended time for investments, which results. But a new and largely unexpected health cost is the treatment of chronic conditions. That might lead to fewer diseases but more expensive ones. In the past, it was possible to regard chronic illnesses as incompletely cured ones, just a step away from total elimination. However, the cost of diseases in the chronic category has risen so much it undermines attempts to fund it with insurance. People with chronic disease are also living longer. Take atrial fibrillation, from which your author happens to suffer, as an example.

Atrial fibrillation, or AF, is a pretty common disorder whose causes and mechanics do not concern us here. It has occasionally fatal or hospitalizable flare-ups, but for the most part, the patients (before 1990) would make one or two visits to the doctor a year, for checkups and renewal of inexpensive digitalis prescriptions. It was then realized that a small but definite proportion of the patients would have a serious stroke related to forming a clot within the chamber of the heart and throwing it into the brain. That is, although the patients were living a long time with their chronic condition, they were often later actually dying prematurely from preventable strokes. Accordingly, the patients were urged to take anti-clotting pills and monitor progress with weekly blood tests. Although it is true the maintenance costs of this disorder were abruptly raised, these costs would have to be offset by reducing the costs of hospital confinement and wheelchair life at the other end of life. To that would, unfortunately, have to be added the hospitalization and transfusion costs for the rare patient whose anti-clot medication gets out of control and provokes a massive bleeding episode. The net financial cost and gain from this change of treatment have not been completely worked out, but insured patients don't much care about that; they would rather have a weekly blood test than the gruesome experience of paralysis from a stroke. One thing is clear, however. A great many people paid for their blood tests with insurance that covered them when they were 40 or 50 years old and eventually saved money for Medicare because they didn't get a stroke when they were older. The tension between the two coverages held the potential for conflicts of interest to emerge between insurers.

Going forward, further changes occurred in the treatment of atrial fibrillation. By 2012, several medications emerged which did not require the cost and nuisance of weekly blood tests. Switching over to this improved medication, I discovered that the co-payment on my insurance was $68 a month, for the pills alone. By the formula which used to apply, this would imply another 80%, or $270, was paid by Medicare to the drug industry each month. There is no way the patient can be sure of these facts in a cross-subsidy world, but on the face of it, it becomes entirely possible for the drug costs to exceed my share of the cost risk of having a stroke. Since there are several of these drugs, it is possible that competition may drive down the drug cost. In any event, the drug cost will drop substantially when the patents expire, and we will finally be able to estimate the enduring cost/benefit of the new drug compared with its blood-test predecessor. But that's probably not the end of it. Sooner or later, someone will discover a way to prevent or cure atrial fibrillation, generating a new cost cycle, we hope a final and lower one. We go through this history of a single disease entity for the purpose of asking a simple question: How can you predict future health care costs in such a violently changing environment, well enough to construct an insurance design?

You can't. You cannot predict what mixture of diseases will lie in the future, but you can rouse yourself when they start to grow in frequency. You can't predict, so you monitor. And for that you need a professional monitoring organization, preferably acting in conjunction with other monitoring agencies throughout the world while remaining in competition with them for the glory of being right. It might also be well to isolate the monitors from the regulators, just enough to enable them to criticize each other's work because they didn't participate in it.

----------------------------------------------

The line of reasoning we have been following leads to an unexpected conclusion. When the insurance company is big enough and rich enough, it can ride out occasional additions to the chronic disease category, calculate its per-person cost, and run a responsible insurance company. Just how big it has to be, and how much it has to save in reserves, will, unfortunately, vary with the state of medical science. Eventually, we will get to that end-game of first and last years of life, with only accidents, epidemics, and self-inflicted disease during the interval. My first suggestion is we stop paying insurance for the first year after a new drug's introduction, giving the insurance company a chance to know about its new cost center that was unknown when they set the premiums. The patients could get the drug, but they would have to pay cash in full. While we are being fair, we should extend patent protection by a year to compensate. It would reduce premiums somewhat, because at present they have to set aside a reserve fund for such contingencies, and there might be other ripple effects, like giving competitors an extra year to catch up.

In the meantime, we must still make a guess as to the "actuarial sound" size of a viable subscriber group. In the environment which thinks about these things, the minimum actuarial size of the group is about 2000 subscribers, but only if coverage is limited to a single year. More conservative theorists would put the minimum actuarial size at 5000 members for a one-year premium. At either size or in fact at any size, there is a small but diminishing risk that some quirk will bankrupt the insurance. Therefore, we must return to the insurance, or reinsurance, described at the beginning of this section; that is, excess major medical insurance. The difference is that its proper role is to re-insure groups, not individuals. Although the risk becomes vanishingly small by increasing the size of the subscriber group, some risk always remains for reinsurance coverage. Although there are many twists and turns to be considered, it is worth mastering these complexities. Taken all together, they explain why the insurance industry favors large groups over small ones, one-year coverage instead of lifetime coverage, and either employer-based or government-based re-insurance with a deep pocket. These are entirely legitimate attitudes for people placed in charge of insurance solvency, and woe unto anyone who accepts this responsibility without providing for its inherent risks.

However, in recent years society has been acting as if the advice of insurance executives is to be taken without understanding its parochial limits. Yes, the biggest group we can imagine is the whole nation, and the biggest re-insurance carrier would be the Federal Government. But nothing in such stipulation suggests that it would be desirable for the government to run the insurance, or for national participation to be mandatory, or for a single size to fit everyone. To get to specific problem areas, it is much more desirable to design specific programs for the prison inmates, the mentally handicapped, and the non-citizens for a total of about thirty million people -- than to twist around the whole insurance system which is serving non-retarded, non-prisoner, non-illegal citizens moderately well. To say nothing of ending up with forty million who are still excluded, and an extra cost, estimated by the Congressional Budget Office to be a trillion dollars in ten years. Along with with the introduction of all the other curses of bigness, plus a certainty of political meddling. At the very least, don't make things worse.

Secondly, there is nothing in these ancient insurance cautions to preclude the gathering of investment income on the premiums, while at the same time despairing of predicting the costs of the medical care it is assigned to mitigate. There are risks to investing, but they are accepted every day by consenting adults. There is a great deal of bad investment advice in circulation and some outrageous fees. We can expect to hear plenty of it in discussions about sums this large. But surely our political system has the wit to consult with a variety of advisors, and ultimately let the public decide whether letting our elected leaders administer the matter, is cheaper than losing the opportunity entirely.

Third. It should be obvious that there are many important unknowns in the discussion of this subject and a deficit of timely monitoring information. The nation needs a monitoring system of great probity, immediate responses of experts to sudden developments, and the power to invade privacy occasionally to get the data. It is debatable whether a monitoring organization should have regulatory power, but it certainly should have close contact with those who have the power to act. There has to be a defined pathway between the complaint of any citizen, leading to access to those who have the power to do something about it. Both the legal and the medical professions have models to examine which are centuries old, functioning well in spite of having no official power to act. The Federal Reserve, on the other hand, has the power to command and the power to investigate within its field; it is its vaunted independence which might be questioned. It could even be suggested that its independence rests on the public perception that its viewpoints are more likely to be accurate, than are those of whoever seeks influence.

Fourth. Let us have incremental patchwork, by all means. That is the conservative approach to public affairs of great moment. But until the Supreme Court finds a way to remove the anti-trust blockade of Maricopa (1982) and medical malpractice's destruction of practitioner discretion, things are unlikely to make great leaps forward. The path to influencing the Legislative Branch is perhaps more direct, but its seventy-year continuation of income tax preferences for employer-purchased health insurance sends a clear warning. The Sustainable Growth Factor (the "Doc fix") sends a similarly unfortunate signal. Fifth. But who will watch these watchdogs? The careful reader will perceive how many of the proposed reforms in this book have little to do with the Affordable Care Act, except help ACA exploit public perception that something is wrong, and relies too much on special pleading. The electronic medical record is an example, reflecting reliance on advice from group practices. The dismal failure of the electronic insurance exchanges has indelibly stained the Administration's reputation as a computer authority, so this feature is now up for political sale. Nevertheless, it remains a good idea to permit interstate competition between health insurance companies, since that could create competition between fifty actuarily-sized monopolies while encouraging other insurance companies who do not enjoy similar hospital discount prices to seek equal treatment. It is almost a century past time to insist on equal prices for equal services, by imposing a "most favored nation" requirement, or fixed relationships between hospital costs and hospital prices, regardless of whether they apply to inpatients, emergency room, or outpatient satellite clinics. Strict segregation of patient-care revenues from research funding and totally non-medical activities in teaching hospitals within universities -- ought to require no mentioning. Oversight, perhaps regulation, of the whole accounting system of the medical industry definitely ought to be a function of the monitoring system mentioned in point Three.

Originally published: Wednesday, December 18, 2013; most-recently modified: Thursday, May 16, 2019