Appendix - Devising a methodology for open rationing

            B. Prioritizing health services - What Oregon did

The key to the Universal Basic Health Plan (Tier 2 in our proposed scheme for open rationing) is to provide to all Americans all health care services that have sufficiently favorable cost-effectiveness scores.  In other words, the key is to figure out how to prioritize health care services.

The notion of having to actually prioritize services to determine which services will be offered and which will be withheld is an odious one.  Indeed, wanting to avoid such a prioritization is probably the main reason people have been unwilling to consider open rationing in the first place.  Given the visceral repulsion we all feel toward this task, how can we best accomplish it in a way that is fair, equitable, and ultimately acceptable?

Perhaps the first thing we should do in trying to answer this question is to take a brief look at how they did it in Oregon.  Examining the brave efforts of the Oregonians should help us to place our priorities in order as we establish our own methodology.

What Oregon did

In 1989, the state of Oregon passed legislation designed to provide a package of basic health care to all state residents who were on Medicaid.  This package required that health services be prioritized in order to determine what would be covered and what would not be covered.  In other words, it required a rationing plan.

To accomplish this task, the legislature created a Health Services Commission that consisted of five primary care physicians, a public health nurse, a social worker, and four “consumers.”  The Commission was charged with the task of producing a list of health care services ranked in priority “according to comparative benefits of each service to the entire population being served.”

The Commission approached its work as follows.  First, it divided the world of health care into “condition-treatment pairs.”  Each pair consisted of one medical treatment, paired with the condition it is intended to treat.  So, for instance, “appendectomy for acute appendicitis” was one of the condition-treatment pairs.

Next, the Commission attempted to measure the level of clinical effectiveness for each pair (i.e., how well the treatment worked for the condition with which it was paired).  To do this, they heard testimony of numerous panels of physicians from every specialty (physicians in Oregon donated over 7000 hours of their time to the effort).  They also assessed how well each treatment affected quality of life.  From these data, they computed a cost-effectiveness value for each condition-treatment pair.

The Commission, in accordance with its charter, took pains to gain broad public support for their efforts.  They conducted numerous public meetings and public hearings, and specifically sought out opinions from a full range of health care advocacy groups, including groups advocating for the poor, the uninsured, and general consumers. These public discussions generally centered on ethical issues pertinent to rationing health care.  The discussions were summarized in an extensive report for the Commission, which then attempted to integrate the social values expressed by Oregonians into the priority list of health care services.

Based on such public testimony, the condition-treatment pairs were divided into 17 major categories, and these categories themselves were prioritized in order, from 1 to 17.  For instance, the category of “therapies that treated acute, fatal conditions and restored normal health” was given the top priority.  The categories of “preventive care for children,” and of “maternal health,” also were ranked high.  Then, within each category, the condition-treatment pairs included under that category were ordered according to their numerical cost-effectiveness score.

The final product was a priority list of 709 condition-treatment pairs in ranked order.  The list was then given to a private actuarial firm that estimated the cost of paying for each condition-treatment pair per year in the state of Oregon.  Next, the state legislature determined (through an open budgetary process) how much money there was to spend on health care.  Based on that dollar figure and the cumulative costs of providing the services on the priority list, a line was drawn on the list after the 587^th  condition-treatment pair.  Any treatment above the line was covered; any treatment below the line was not covered.

How well did it work?

Probably the most encouraging thing about Oregon’s experience in devising a rationing scheme is that it proved that it is possible to do so without producing a general uprising.  Oregon even demonstrated that under the right circumstances significant public support for the rationing process, from both the professional community and from the community at large, is possible.  In fact, from all appearances, something approaching a general public consensus in support of a system of rationing health care was achieved.  For these reasons alone, the Oregon state officials and all Oregon citizens are owed a huge debt by us all.  

Other features of the Oregon rationing system proved to be more problematic.  For instance, the Commission’s original intent to rank condition-treatment pairs in strict numerical order depending on their cost-effectiveness analysis had to be abandoned early on.  This proved necessary because strict cost-effectiveness calculations produced rankings that were counter-intuitive and seemingly absurd.  In the original list, for instance, the insertion of dental caps was ranked higher than surgery for ectopic pregnancy (a condition which is fatal without surgery).

It was because of such counter-intuitive results that the Commission created the 17 categories under which the condition-treatment pairs were prioritized.  With this new methodology, all life-saving therapies were ranked high. (Treatments that prevented death and restored normal health were in the top-ranked category; treatments that prevented death but left the patient in a reduced state of health were in the third ranked category.)  Unfortunately, even this modification left seemingly illogical rankings.  Because all life-saving therapy was ranked high, for instance (i.e., in either the first or third category), expensive treatments that merely postponed death in terminally ill patients received higher overall priorities than most clearly effective non-life-saving therapies.

Another general problem with the Oregon system was its use of quality measures.  To measure cost-effectiveness of therapies, Oregon used a Quality of Well-Being (QWB) scale.  This scale defines 24 distinct states of health, examples of which are “generalized tiredness, weakness, or weight loss,” and “upset stomach, vomiting or loose bowel movements.”  One thousand patients were then surveyed and asked to assign a numerical score to each of these 24 states of health, from a scale of zero to 100 (with 100 being perfect health and zero being as good as dead).  Each of the 24 states of health was thus assigned an overall numerical weight based on these surveys.  These QWB scores were used to calculate the cost-effectiveness, and therefore the priority ranking, of the condition-treatment pairs.

One general problem that resulted from this methodology is illustrated by the fact that the QWB state of “nausea and vomiting” received a score that was one-third as high as the state of perfect health.  Thus, in the cost-effectiveness calculations, preventing vomiting in three patients was numerically equivalent to saving a life.

The more significant problem with this methodology is that it is clearly discriminatory (as was recognized by the federal government, which prevented implementation of Oregon’s plan for this reason.)  Since any patient with a disability (such as walking with a cane) would automatically receive a lower QWB score than a patient without such a disability, patients with disabilities were systematically ranked lower on the priority list. (This is not strictly true since it was the condition-treatment pairs, and not the patients, that were ranked.  But the cost-effectiveness of a treatment – which determines its ranking – depends on the quality of the outcome, and the quality of outcome is directly affected by patients’ concomitant diseases and disabilities, as we will shortly see.) The Oregon system thus overtly sacrifices distributive justice for maximization of beneficence. Of course, this was precisely in accordance with the Commission’s charter, which charged them with ranking priorities “according to comparative benefits of each service to the entire population being served,” in other words, to maximize the public good.

Next: The Quality problem

Return to YourDoctorintheFamily.com home page