Thoughts arising from the murder of a health insurance CEO

Restrictions in access to health care

Thirty days ago, the CEO of an American health insurance company named Brian Thompson was murdered on the streets of New York City.

There was an outpouring of glee rather than of outrage. Huge numbers of Americans took the position that the murder of a health insurance company executive fell in the category of being a good start. Many have said people like Thompson are evil creatures who put corporate profits before the needs of the sick, injured, or dying.

At the same time, in Canada, there is simmering anger about the state of our publicly-funded health care system. Fortunately, this has taken the form of punishing via the ballot box rather than with bullets, but the anger and resentment is real.

The reaction to Thompson’s murder set me to brooding.

Why are those who administer the health care system so unpopular?

In both Canada and the United States, improvements in medical treatment have contributed to an increase in life expectancies of almost two decades – in my lifetime. For myself, if I had been born a half century earlier, I’d be reaching the end of a normal life expectancy. I’d also be blind. As it is, I’m still chugging away and can see my computer screen as I type this posting.

Again, why are those who administer the health care system so unpopular?

I think there are four basic reasons.

1. Nobody likes to be in pain, sick, or injured. We are reluctant to die. As a result, want our afflictions removed or ameliorated.

2. Our capacity to treat many medical conditions has improved dramatically. We now have capacities that never before existed in the history of humanity. Since we know treatment for many things is now possible, we want our afflictions treated.

3. Many of these therapies are expensive. When the cost of these therapies is combined with our desire to have or afflictions ameliorated and our knowledge that this is often possible, spending on health care services is potentially almost infinite. At the very least, there are powerful cost escalation pressures.

4. Even though we are a rich society, we can’t expend all our resources on health care. Other priorities have to be met as well – and this means there must be constraints on the delivery of health care services. Put bluntly, somebody must use some method of saying no to some people who desperately want/need care.

There are three main ways to restrict access to health care services and thereby contain costs. These are using price, queueing, and clinical criteria. Because all involve saying “no” to some people, so all are unpopular.

In both Canada and the United States (and everywhere else) some mixture of each of these three methods is used to control costs – and thereby restrict access. In this article, I’ll explore some implications of each method. I’ll then wind up with looking at a common work-around people use to get the services they need or want – using “who you know” and “what you know.”

Using price to restrict access to health care

A simple way to allocate health care resources is through the market mechanism of pricing. If people are required to pay for health care costs, they will balance their health care needs and desires against their other requirement. People will have an incentive to seek the most cost-effective treatment alternative (although one’s ability to shop around are restricted if you have a broken leg or are bleeding).

There is a problem with using a price mechanism to allocate health care resources.

Health status is not tied to income. Indeed, there is usually a negative correlation. The poor tend to be in poorer health than the affluent. What’s more, many health care services are very expensive. Paying for something as important as ensuring the safe delivery of a child will be beyond the individual financial resources of most households.

If we reverted to a strict, point-of-delivery pricing mechanism, our health care expenditures would be reduced – but a lot of people would die prematurely.

In both Canada and the United States, both public and private health care financing schemes have been developed to pool the financial risk arising from injury or disease. We’ve made the moral judgement that income should not be the determining factor as to whether a person lives or dies – or continues to suffer from a treatable condition.

While limitations on using out-of-pocket payment to restrict access to health care are both just and necessary, the limitations on market forces produce some pathologies.

The first problem is that the elimination of price as a discouragement to the individual leads, for some individuals, to an excessive use of health care services. For most people, this is not a huge issue since going to a doctor’s office or getting some invasive procedure does not fall into the category of a pleasurable use of time. However, there are some people with very low toleration for discomfort or with vivid imaginations. When price is removed as a barrier, other mechanisms are needed to deny services to hypochondriacs.

The second problem is more difficult to grapple with. Simply put, neither public or private health insurance schemes cannot cover everything. Lines have to be drawn somewhere. Some things are paid for collectively while other things are not. This causes a migration to the “free” services even if they are not the most appropriate or effective. When Canada’s health care system was built up in the 1950s and 1960s, it became too hospital focused because care in hospitals was the first thing to be paid for by governments.

Removing the price mechanism from the point-of-service also removes the incentive for the health care consumer to seek the most cost-effective alternative. This, in turn, reduced the incentives for health care providers to operate efficiently.

Let’s look at two examples – hearing aids and prescription drugs.

According to the Wall Street Journal, in 1989 a guy appropriately named Sal Price started to lose his hearing. He was appalled at the cost of hearing aids. Price operated a chain of warehouse stores called Price Club. He set out to provide consumers with cheaper hearing aids. Price Club merged with COSTCO. The result – cheap hearing aids. COSTCO now sells three brands of hearing aids starting at $1,500. The same hearing aids cost between $3,500 and $7,000 from other venders. COSTCO now accounts for sixteen percent of hearing aid sales in America.

Here's the thing.

If Sal Price’s hearing aids had been covered by a public or private insurance plan, he would have never set out to sell hearing aids more efficiently.

Prescription drugs are another example of the impact of severing of a price mechanism.

Developing and getting regulatory approval for a new prescription drug is a lengthy and expensive process. To make it possible for new drugs to be developed, in both Canada and the United States patent protection is provided for twenty years. If a company develops a better drug, they have two decades to recoup their development costs (and make a lot of money). After the patent protection period expires, other companies can market the drug as a “generic” substitute. Competition then drives down the cost of both the generic and brand-name versions of the drug.

In most cases, the generic version is identical to the original brand-name drug in everything but price.

I’ve had the experience of going to a doctor. A prescription was deemed medically necessary. The doctor asked me if my drug costs were covered by an insurance plan. If the answer was “yes”, the doctor would prescribe the more expensive brand-name drug. If the answer was “no”, doctor prescribed the cheaper generic version. Pharmacists are pressured by insurance companies and government prescription drug plans to make the substitution to generics unless the prescribing doctor has specified “no substitution” (there is sometimes a valid medical reason for this). Back in the good old days, when doctors wrote out the prescriptions in their notoriously bad handwriting, pharmaceutical companies provided doctors with pads paper pre-marked with the words “no substitutions”.

If the patient does not pay out-of-pocket for the prescription – or only pays a small, flat-rate, co-payment – there is no incentive for the patient to want anything other than “the best.”

In the words of Dr. Randall Stafford of Stanford University, “in the treatment of high blood pressure, diabetes and some infections, there appears to be a tendency for new drugs to be used in a wider array of situations than supported by scientific studies."

Let’s look at the specific example of treatment for osteoarthritis and rheumatoid arthritis. These painful conditions are treated with non-steroidal anti-inflammatory drugs (NSAIDs). In the 1990s, the most common of these were over-the-counter drugs such as ibuprofen (Advil being the most popular brand name) and Aspirin. The most serious side effect of these drugs was that they caused stomach ulcers in about 5 percent of regular users.

Pharmaceutical companies saw an opportunity. In 1999, Pfizer introduced a new NSAID called Celebrex. Merck soon followed with Vioxx. These new drugs had the same clinical effectiveness as Ibuprofen or Aspirin but reduced the incidence of stomach ulcers. They also cost between 10 and 20 times as much. Pfizer and Merck launched massive advertising campaigns aimed at both patients and physicians. Industry observers estimated that Merck spent $135 million in 2001 to promote Vioxx – making it the most heavily advertised drug in North America.

Researchers from the University of Chicago and Stanford University developed clinical criteria published in the Archives of Internal Medicine. They suggested that arthritis sufferers with very low risk (31 percent of patients) or low risk (42 percent) should continue to receive the older, cheaper NSAIDs. The 25 percent of patients with moderate risk of stomach problems should be considered for Vioxx or Celebrex depending on the length of their treatment period. Only the two percent of arthritis sufferers deemed to be at high risk of stomach ulcers were recommended for the new, expensive medications on an ongoing basis.

What followed was a classic demonstration of the power of removing a price mechanism.

When people had to pay for the NSAID out of their own pocket, patients with low and very-low risk of stomach problems tended to stick with Ibuprofen or Aspirin. When the cost was paid for by insurance companies or government programs, the majority took the more expensive Celebrex or Vioxx. These drugs were introduced in 1999. By 2002, 66 percent of arthritis patients who had low risk of stomach problems were taking the new, expensive drugs. Almost all had these costs paid by public drug plans or private insurance.

But new and expensive is not necessarily better. In 2005, Vioxx was pulled from the market. While it reduced the risk of stomach ulcers, it significantly increased the risk of heart attacks.

Canadian political debated tends to portray the use of price as an “all or nothing” proposition. Either the government pays the full cost of medical services or we move to “American-style private medicine”. This view is simplistic on several ways:

• It over-states the extent to which price is eliminated from Canadian health care. While physician services and hospital care are fully publicly funded, many other components of health care are funded out-of-pocket or through private insurance. Public funding accounts for about 70 percent of health care expenditures in Canada.

• It under-states the role of public health care funding in the United States. South of the 49^th parallel, almost half of health care expenditures are funded by state or federal governments.

But the most egregious simplification is to present a binary choice of “free” versus “full cost”. This overlooks the middle path of co-payment by the public/private insurance and the individual consumer of health care services.

Almost all prescription drug plans offered by private insurance companies have some level of co-payment. The person filling the prescription has to take out their wallet – even if the amount paid is nominal.

Why?

Because even a nominal payment reduced drug usage.

This raises a question that is impossible to answer definitively. Is this reduced usage a result of people not taking drugs that are medically necessary or is it the result of people not filling unnecessary prescriptions?

There has been a lot of research on this question. I’ve only tickled the surface of it for this article, but what it seems to show is that the answer lies in the amount of the out-of-pocket expenditure needed to purchase the drug in relation to the individual’s income. This makes intuitive sense. The studies I looked at suggest a nominal co-payment charge results in about a 10 percent increase in the number of prescriptions that are not filled. Interestingly enough, it also seems to result in an increase in the percentage of people who consume all of their prescribed medication. Patient co-payment seems to have the positive effect of reducing antibiotic resistance by both reducing people’s unnecessary consumption of antibiotics and increasing the likelihood people will take their full course of treatment.

Co-payment: the partial use of price

Canada and the United Kingdom are the only two OECD countries that do not have a patient co-payment for physician services and hospital care. In every other country – most of which have health care systems that deliver better health outcome – patients pay at least a nominal out-of-pocket charge.

In Canada, back in the early days of “Medicare”, we did have an experiment with co-payment within a system of public funding for physician services and hospital care.

In 1968, the Saskatchewan government imposed a $1.50 charge for each visit to a doctor’s office and a $2.00 charge for each out—patient visit to an emergency ward or home visit by a doctor (wow – that sounds quaint!). Patients in hospitals were charged $2.50 per day for thirty days and $1.50 per day after that.

The co-payment (or “user fee”) system was very unpopular. It contributed to the defeat of Ross Thatcher’s Liberal government in 1971. The NDP government led by Allan Blakeney quickly abolished the charges.

What was the effect of this co-payment system?

Two economists at the University of Saskatchewan – R.G. Beck and J.M. Horne – crunched the numbers. There were a lot of numbers to crunch. Dr. Beck told me the pair had to bring sleeping bags to the Health Department’s computer facility. The 1970s era mainframe computers would chug away for a long-weekend to produce results. Beck and Horne found that co-payment reduced the number of visits to the doctor by 5.6 percent but had no effect on the provision of hospital services.

Last month, I had eye surgery to remove a cataract. My ophthalmologist explained that the operation and the cost of a new lens was covered by the Saskatchewan government. The lens paid for by the government was, she said, excellent. However, one that did more to correct a stigmatism was available. It would cost me $1,500 out-of-pocket. After she explained the effect each lens would have on my vision, I chose the “free” lens. Quite happily. Three weeks after the operation, I now see a lot better.

But here is the thing.

If the other lens had been “free” (to me), I’d probably have picked that one even though the less expensive one works just fine.

Using Queuing to restrict access to health care services

To the extent that pricing and ability-to-pay are removed as ways to limit demand for health care services, queuing becomes a major feature of publicly-funded health care.

Queuing restricts the volume of health care services by limiting supply. Through budgetary and other resource allocation decisions, governments establish a specified volume of medical interventions that can or will be provided. People suffering ill health are then told to “wait your turn”.

This “waiting” can occur in many places in the delivery of health care services. The OECD breaks this down as follows:

• Timely access to effective care;

• Timely access to specialist care;

• Timely access to diagnostic tests;

• Timely access to hospital emergency departments;

• Timely access to primary care;

• Timely access to cancer care;

• Timely access to mental health services, and

• Timely access to cardiac care.

The logic of queuing is described in an OECD report as follows:

“Waiting times for health services in publicly-funded systems aim to create equality between patients: patients with similar needs are supposed to wait their turn irrespective of their ability to pay or other non clinical characteristics.”

Queuing can be a good way to promote efficiency as well as equality. When I had my eye operation last month, patients were lined up in a veritable assembly line. I don’t know how many operations were performed by one doctor that day, but it was more than could have been done if the procedures were not scheduled to maximize efficiency. This inevitably means some waiting until enough patients are available for the assembly line to be fired up.

But there is a limit to the effective use of queuing.

Treatment delayed can be treatment denied.

There is a steady trickle of news reports about people dying while waiting for care in emergency departments. Once diagnosed in the emergency room as needing to be admitted for further treatment, a patient can still wait a long time. According to a Canadian Medical Association report, half of patients wait longer than 14.7 hours and ten percent wait longer than two days. Remember, these are people deemed in need of treatment on an urgent basis. Writing in the Canadian Journal of Emergency Medicine in 2023, two emergency room physicians estimated that between 8,000 and 15,000 Canadians die in hospitals while waiting for treatment.

One reason Canadians wait so long in emergency departments is because they wait even longer to access physician care. According to an OECD study, 33 percent of Canadians report waiting longer than a day for a response from their family physician’s office when they call to get an appointment for an emergent condition. That’s the worst score in the 38 OECD countries. Oh. One more thing. According to Statistics Canada, 6.5 million Canadians did not have a family physician in 2023 – up from 4.6 million in 2019. People can wait a long time before finding a family physician willing and able to see them.

According to data compiled by the World Population Review, 61 percent of Canadians referred to a specialist had to wait longer than a month for their first appointment. This compares with 27 percent in the United States. Price is not the only barrier to care.

In 2023, waiting lists for cancer patients in British Columbia got so long that the provincial government began sending patient to Washington State for treatment. The American market-orientated system had capacity while the queuing-orientated Canadian system did not.

Canada has relied on queuing to ration health care services for a long time. It’s now threatening access. According to the OECD, Canada was one of six OECD countries having waiting times that threaten meaningful access to service in all eight areas listed above. We are joined by Latvia, Lithuania, Mexico, New Zealand, and Poland.

Using clinical criteria to restrict access to service

We’ve probably all had the experience of waiting for treatment in an emergency department for hours. An ambulance arrives. Some poor soul on a stretcher is rushed in. Doctors, nurses, and other professionals jump into action.

Nobody protests the “queue-jumping”. We all understand that some conditions require more urgent attention than do others. Clinical criteria determine priority of access.

Using clinical criteria to govern access to service often makes sense both from moral and common-sense perspectives but there are a few issues.

The first issue is a version of “an ounce of prevention is worth a pound of cure”. Some conditions get worse if left untreated. The worsened condition can require more intervention and use of resources than would have been required if treatment had been provided earlier. I think everyone understands this, so I’m not going to belabor this aspect.

The second issue is a little more subtle. The same clinical criteria can differentially affect the lives of different people. A blister on the finger of a baseball pitcher prevents them from working while it will be inconsequential for a sociology professor. A bad back has more impact on the life of a construction worker when they are working than after they are retired. People have different pain tolerances. And so on. Doctors and health care administrators have a complicated task managing priority access to service on clinical criteria alone. Asking them to consider the social and economic impacts of their diagnosis makes this task much more complicated – and exposes them to disputes about equity and value judgement.

There is a lot of contested terrain over the use of clinical criteria to determine access. I was personally involved in one such contest. During the 1990s, a drug called Betaseron came on the market. It was the first pharmaceutical product specifically designed to slow the progression of Multiple-Sclerosis. It was also hellishly expensive. This made coverage of the drug by provincial prescription drug plans a political issue.

Betaseron offered – for the first time – those suffering from MS some hope. People quite naturally wanted it – desperately. At the same time, even the manufacturer of the drug put very precise boundaries around when the drug would have some effectiveness. There was conflict between hope and what was scientifically proven. Doctors were caught in the middle between the hopes of their patients and the steely-eyed bureaucrats who – for good and valid resource allocation reasons – insisted on specific clinical criteria being met before coverage was authorized. It did not matter whether the bureaucracies were in governments or private insurance companies. “Following the science” on proven effectiveness is an essential aspect of cost containment for the health care system, but when this bumps against the denial of hope for a desperate person, hatred and anger results.

Providing care based on clinical criteria is the heart of the need for “prior authorization” that has driven much of the anger directed at health insurance companies in the United States following the murder of Brian Thompson. To contain costs (and thus keep premiums somewhat affordable), insurance companies will insist on reviewing and authorizing treatment before committing to pay for it. At least in principle, the insurance companies are attempting to “follow the science” by only paying for treatment that conforms to established clinical criteria. People in pain or facing death often have different criteria – sometimes valid, sometimes based only on desperate hope.

In the debate following Thompson’s murder, many Americans have blamed the for-profit nature of insurance companies as the source of the perceived evils of prior authorization. But don’t kid yourself. Publicly funded programs do the same thing. Just watch in the media for the regular stories of folk who have headed to the United States for treatment without prior authorization and who are now attempting to mobilize public and political support for the government to pick up the bill. Fortunately, no Health Ministers have been murdered.

Yet.

Using who you know to access service

Price, queuing, and clinical criteria all restrict access to health care services. When these restrictions become too problematic for the sick or injured, people search for ways to work around the barriers. One’s interest in the ideal of equity is lessened with personal pain and suffering – particularly if there is suspicion that others are working the system to get their needs looked after before yours.

Since people are intrinsically social creatures, we turn to our social networks to get the care we need or want. When the barriers are high, calling on who you know becomes the most common workaround.

When we talk of using personal connections to get service, it conjures up the image of powerful politicians or the rich jumping the barriers of price, waitlists, or clinical criteria. The wealthy can fly to the Mayo Clinic. The Minister of Health gets priority service. The star professional athlete gets an MRI the day after stubbing their toe.

This sort of thing happens, but the reality of personal connections to access service is usually more prosaic.

When my youngest son was about 13, he played on the same hockey team as the son of an orthopedic surgeon. In the second period of a game on a Saturday night, a player broke his arm. The Orthopedic surgeon bundled the kid into his car. They returned before the game ended with the kid’s arm in a cast. Everyone at the rink was happy the kid received such excellent and prompt treatment, but I could not help wondering what the experience of another kid playing at a different rink would have been.

At scale, the most important “who you know” determinate of access to care is access to a family physician. In our complex and specialized health care system, referrals are the portal to care. Without an ongoing relationship to a family physician, access to a lot of care becomes – well – inaccessible. About 6.5 million Canadians do not have access to a family physician. This means they have less access to specialist services as well. In Quebec, the lack of access to family physicians has reached the point where an estimated 500,000 people with complex or chronic health conditions such as cancer, mental health issues, cardiovascular disease or diabetes do not have a family physician.

The Quebec government’s proposed solution?

Restricting access to family physicians for those without major diagnosed health issues.

This is a proposal to allocate care on the basis of clinical criteria. The problems, of course, are that emergent conditions are much more likely to remain undiagnosed – and that access to required specialized for those categorized as “healthy enough” will be diminished.

In addition to “who you know” as a means to access health care services, “what you know” is also important. The health care system is complex, opaque, and mysterious. Complexity punishes the unsophisticated. People with the ability to understand rules and procedures, articulate their needs, and call on relationships are in a better position to obtain the best possible care than those without these skills.

The reality is that the poor are, on average, in poorer health than the affluent. As a result, they consume more “drop-in” health care services such as those provided by walk-in clinics and emergency departments. But the affluent consume more specialist services, even though they are healthier.

Who you know and what you know matters.

A few final observations (at last)

This article has been way too long. If you are still reading, it means you really care about health care.

The point these 4,500+ words is that we have a societal need to sometimes deny people the care they want. Sometimes we have to deny care they need. It’s a dirty job, but it has to be done because we need more than health care services to survive and prosper.

Saying “no” is intrinsically unpopular – particularly when you are saying “no” to someone sick, injured, in pain, or facing death. Price, queuing, and clinical criteria are the mechanisms we’ve developed to say “no” in a cold and impersonal manner. Each method brings with it different negative effects. Each method involves trade-offs.

Two final thoughts:

• Simple solutions and scapegoating do not improve our health care system. There’s a lot of complexity and a lot of trade-offs. In the end, while we want lots of health care services, we also want low taxes and insurance premiums.

• In the end, someone has to administer a system that says “no” as well as saying “yes.” We can (and should) hold these folk accountable for their decisions and insist they perform their duties in a caring and thoughtful manner but gunning them down in the streets is not going to do the trick.