Coronavirus #7: The Desperation of Hydroxychloroquine

I recently came across an Internet story about a woman who had brought her COVID positive father home from the hospital to hospice.* The hospital physicians who discharged him (rightly, I believe) refused to prescribe hydroxychloroquine when the family requested it. The daughter asked plaintively, why not try it? What has he got to lose? Her father will likely die soon, anyway. Even a million to one shot is still a chance.

There are, however, good reasons.

Not that I don’t sympathize with the argument. If there is little to lose, taking a long shot can seem like a good idea. As Nobel Prize-winning psychologist Daniel Kahneman notes in his book Thinking, Fast and Slow , humans have a tendency to be conservative with risk-taking when they have a lot to lose, but very aggressive when they have nothing. For example, millionaires buy very few lottery tickets, while poor people buy many. The rich already have money, and would rather place their bets on a more certain thing such as a financial investment that pays 7% interest, or more practically, a dinner at a restaurant they are 100% certain of enjoying. Meanwhile, the poor, who have no such options, will wager a dollar on ten-million-to-one shot at wealth, even if they can’t pay the rent. People who are desperate will risk everything because they reason, as the dying man’s daughter did, that there is nothing to lose.

But this is flawed logic. The odds of a winning lottery ticket are the same, no matter what your financial straits are. That is, you are no more likely to win the lottery if you have $2 to your name than if you own Apple Computer, Inc. Poverty does not improve your odds. But it does increase the harm in loss: If a millionaire loses $10 in the lottery his life will be no different, but a poor person may have to skip lunch. For the poor man, the money is better left in his pocket, spent in a way that guarantees benefit. On a good sandwich, for example.

The same goes for COVID. Hydroxychloroquine isn’t going to work for someone just because he is desperate. The odds don’t improve with desperation, and so logically, there is no better reason to give it to the dying than to people who healthier. In fact, with the dying, the outcome could be worse. If hydroxychloroquine has a 1% risk of side effects but only a one in a thousand chance of saving a dying person, that means the chance of harm is a ten times higher than of benefit (1% = 1/100, and 1/100 is ten times greater than 1/1000). When dealing with long odds, because the compared risks tend to be small, people often fail to notice this. Another example: for someone hauling rocks, it is much easier to tell the difference between a 10 pound stone and a 100 pound stone than a one ounce pebble and a ten ounce pebble. But nonetheless, the proportion is the same, a factor of 10. 

But there is more. For a dying person, while the risk of side effects may be the same as for anyone else, the outcome could be more devastating. Getting a little sicker may be tolerable for a healthy person. For a dying patient who lacks the ability to recover from a minor insult, “a little sicker” could mean death. When you have less to gamble with, the chances of losing everything are much higher.

Another problem with longshot medications is something called availability bias. Availability bias is the tendency to believe a statement familiar to you is more likely to be correct than an idea you have not about heard about. You have heard of hydroxychloroquine. Because you have heard of it, you automatically assume it must be better than the billions of compounds you have never heard of. It must be better…because you’ve heard of it!

This is a logical fallacy. Why does the most well-known product on the market have to be the best one? In fact, fame is not proof of quality. There was once a time when bleeding and leeches were the most famous treatments for disease, but that didn’t make these treatments good. Medical history is littered with treatments that not only didn’t work but were outright harmful. Doctors prescribed them not because they were good but because they were treatments they knew. Hydroxychloroquine, judging by several studies that are out, could be a harmful treatment. The fact that it is well-known is not evidence that it is good.

But if the error of playing desperate odds and the error of availability bias do not convince you, there is a third even more compelling reason to reject hydroxychloroquine. Even more significant than the physical risk of the drug, which is admittedly not always serious, there is a risk of psychological harm.

My experience with COVID and end-of-life decisions is growing daily but remains fairly limited. But I have a wealth of experience with cancer, and I have seen first hand the horror a “do whatever it takes” attitude can wreak. 

Cancer patients who are in the terminal stages of their disease often clamor for a last chance at life. I have had patients who had less than a week to live tell me they wanted another round of chemotherapy. These patients spend the last days of their lives in the hospital battling failing kidneys, intestinal obstruction, and severe pain instead of choosing to go home with hospice and die in peace. They think they have nothing to lose by fighting to the last breath, but they are wrong. What they lose in their relentless fight is a precious thing: the time to say goodbye to their families. How do you say goodbye when you are still playing to win? So instead of dying at home, they die in the hospital — a place no one says they want to die in — sacrificing the precious opportunity to be at home in their own beds, with their spouses, friends, children, and grandchildren gathered around them. 

I have seen patients do this more times than I can count. It used to frustrate me, but now it just makes me sad to watch people I know will die cling to false hope, only to have it dashed. People say they want to go down fighting, because society (and sometimes, their own families) have told them the lie that fighting to the end is the only way to go. 

Dylan Thomas was a great poet. But his famous poem about death,

Do not go gentle into the dark night

Old age should burn and rave at close of day;

Rage, rage against the dying of the light

is probably the worst advice about dying ever written.

I don’t want to go down fighting. I want to go down in peace, a prayer on my lips, my family and friends at the bedside. This is much better than dying at 2am, drowning in your own body fluids while a doctor tries to shove a breathing tube down your throat. That is not a heroic death. It is a miserable one that would be shameful to inflict on a dog.

No, taking a pill that doesn’t work isn’t exactly the same as dying of cancer in the hospital. But it plays into the same flawed logic. It cultivates hope that will disappoint. It encourages the patient to fight against dying, when dying is inevitable and a necessary part of life. Fighting doesn’t make life longer. In fact, there is substantial medical evidence that people who resist death do not live as long as people who agree to hospice care.

Fighting death only increases the fear of the dying process, and makes the end, when it comes, even more painful. We all have to die sometime. The stunning thing is how little effort we put into preparing for it. Even when we have been given the great gift of being warned ahead of time.

I believe in hope as much as the next person. I believe in God, too, the ultimate hope. But God doesn’t prevent anyone from physically dying. Even Lazarus, raised from the dead, died later on. While I support seeking to prolong life when the treatment is realistic, I cannot in good conscience give a medication that is highly unlikely to work, but will almost certainly enable patient and family to pretend death isn’t coming. Hope is valuable. Self-deception is not.

One of the four principal ethics of modern medicine is the value of nonmalfeasance.** Sometimes equated with the phrase “First, do no harm,” nonmalfeasance means that it is not ethical to provide medical treatment that is more likely to hurt patients than benefit them. In the case of hydroxychlorquine, we have a drug that has real toxicities, little or no evidence of efficacy, and the psychological effect of giving patients false hope. Does this make it harmful? I think so.

______

* Hospice is end-of-life comfort care. This implies that the doctors at the hospital felt the patient would die fairly soon.

** The others, according to philopshers James Childress and Tom Beauchamp, are beneficence, autonomy, and justice.