by Morley Leonard Evans © February 2003
|Struck by Lightning:
Risk, or probability, is a simple calculation: the numerator of a fraction is divided by the denominator of the fraction. Risk is then a single number which can be expressed as a percentage when it is multiplied by 100. The denominator is the total number of incidents in a sample and the numerator is the number of times the subject event occurred out of the total.
See: 200% of Nothing by prof. A.K. Dewdney
I have become aware that many people in the medical community are innumerate, just like many people in the general public. A lifelong sufferer of innumeracy myself, after having been terrified by a fraction in Grade Three, I am extremely careful when dealing with numbers and suspicious of what others do with them. Usually those casually throwing numbers around do not know what they are talking about.
Our friends at the Merck pharmaceutical company state in their product information that a percentage of people suffered from side effects when taking Zocor (1.5%). Pfizer indicates a similar figure for Lipitor (< 2%). For the sake of argument, let’s say it is only 1% for both. That would mean that 99% had no problems. (That isn’t exactly what that would mean, but let’s accept that it does for the sake of argument.) Both the pharmaceutical companies and the medical community seem to think that being close to 100% is the same as being 100%. Or to put it another way, the chances that a patient will suffer adverse effects are so remote they can be ignored. That is what I was told in those exact words at the “Lipid Clinic” at the Regina General Hospital.
One might think that anyone claiming to have suffered from statin drug side effects is mistaken simply because the probability of that occurring is so remote — as remote as, say, being struck by lightning. So how remote is it? What is the probability?
First of all how large is the sample population? Pfizer bases its number on a controlled clinical study of 2,502 people of whom fewer than 2% suffered problems. I personally have difficulty accepting that such a small number can represent millions of people, but let’s accept that it can. How many people in the study would be affected if the rate were 1%? The number of actual people adversely affected in the controlled clinical study would have been 25 (2502 x .01). Extrapolating, as the pharmaceutical companies have done, if two million people were taking a statin drug, we could expect to find 20,000 people suffering from side effects. In its advertisements, Merck has proudly proclaimed 140 million Zocor prescriptions have been written.
So what is the probability of suffering side effects from a statin drug like Zocor or Lipitor? If the rate were only 1%, one person in every one hundred would suffer. The rate would be .01 (1/100). What is the probability of being struck by lightning? Suppose 250 people a year are struck by lightning in a population of 250,000,000. The probability would be 250/250,000,000 which is .000001 (1e-06) and is also one in a million. If you buy a ticket in a simple lottery in which 14,000,000 people (including you) bought a ticket, your chance of winning the lottery would be 1/14,000,000 which is 0.00000007142857 which is also one in fourteen million. That can be rounded off to 0.0000001 or it can be expressed as 7.1e-08 or .00001%. (No wonder a lottery is a “sucker bet”.)
So how remote is one’s chance of suffering side effects when taking Zocor? Well let’s see: Zocor = 1.5%; lottery = .000001%. The probability of experiencing side effects from Zocor looks extremely high, not extremely low. We should keep in mind that risk is found in the range of small numbers, say, between 1 and zero. How high is statin risk, really? Let’s ask a doctor prescribing Zocor how risky it would be if he had to write a cheque for, say, a million dollars ($1,000,000) each to three patients in every two hundred or 1.5% of his Zocor patients. How risky would selling Zocor be if Merck had to pony up billions of dollars to do it? Risky.
Have these people in the medical industry been getting away with murder? Oh yeah. Let’s look around: In North America hundreds of millions of automobiles are on the road. Each of them is driven on average 10,000 miles a year. Yet out of several trillion annual passenger miles driven only a handful of incidents of exploding tires have recently cost Ford Motor Company and Firestone billions of dollars of lost business, plus billions of dollars in product replacement. This has also cost the top company executives their jobs. You may have heard about this. Both Ford and Firestone have settled lawsuits out of court for undisclosed huge sums (part of the deal is you keep-a you mouth shut). Yet the probability that you will be injured by exploding tires, or anything else while driving your car, is truly remote.
And what of the idea that if something is improbable it couldn’t happen? Although being struck by lightning is a one in a million chance, golfer Lee Trevino has been struck by lighting twice. It would be impossible to prove this didn’t happen to Trevino using the bogus statistical argument used by the medical community. And some people do win lotteries and on some cars tires do explode. Would a casino operator in Las Vegas try to turn the winners away by telling them they couldn’t have won because “the odds are against it?”
In a tort, the question is what happened and who was responsible, not what are the chances that it happened. The entire issue of probability is a stinky red herring dragged across the trail to conceal the tracks of the guilty. It’s bunk.
Probability of Disease:
The medical community here in Regina (the birthplace of our beloved Canadian Medicare system) seems to believe that the incidence of side effects from Zocor or Lipitor (or any other prescription drug) is very rare: so rare it can be ignored.
I have written my thoughts on this above; here are a few more.
According to Pfizer the incidence of side effects caused by Lipitor is < 2% out of a clinical study of 2,502 people. According to Merck, the incidence of side effects caused by Zocor is 1.5%. So how safe is that? Not very safe actually.
Above, I compared that to the chances of being struck by lightning and the chances of winning a lottery and the chances of having the tires explode on your vehicle. These are truly remote occurrences. Drug side effects are not.
Here I suggest that the incidence of side effects would be considered an epidemic if we were talking about anything other than pharmaceuticals that are earning billions of dollars.
Let’s compare this to the AIDS epidemic:
3,000,000 estimated deaths world-wide in 1999
5,000,000,000 estimated world population in 1999
0.0006 x 100 = .06% probability of death from AIDS
30,000,000 estimated living with AIDS in 1999
5,000,000,000 estimated world population in 1999
0.006 x 100 = .6% probability of acquiring AIDS
Note: the incidence of AIDS is much higher on the continent of Africa by itself.
Now, let’s compare this to the Influenza Pandemic of 1918:
• between 20 and 40 million were killed world-wide
30,000,000 estimated killed world-wide in 1918
1,500,000,000 estimated world population in 1918
0.02 x 100 = 2% probability of death from the pandemic
So the incidence that the pharmaceutical companies proudly stipulate on their own product information warnings is about the same as the Influenza Pandemic of 1918 which was one of the most devastating public health disasters in history. (And statin drug risk is much higher than the AIDS epidemic.) The Flu killed more people world-wide than were killed in the First World War. Far from being risk free, the risk from taking Lipitor or Zocor is getting right up there into the same ball park as the Black Death in the Middle Ages!
Medicine is risky business. (But only if you are a patient.) If you are a health care provider (a doctor or a pharmaceutical company) it’s a different story. Then — here in Canada at least — you can expect to enjoy a monopoly with a lock on the Public Treasury while you operate Above the Law. Such a situation can produce and is producing gross abuse. It is criminal. Good doctors should do something about it.
Both outrage and sensationalism are relative concepts. The pharmaceutical industry, unlike the automobile industry, currently enjoys the extraordinary tolerance once enjoyed by the tobacco industry. People will overlook much.
How much? Let’s see:
Because it is an extraordinarily sensational story, as well as an unbelievable outrage, the CBC (Canadian Broadcasting Corporation) aired an exposé on the anti-malarial drug, Lariam (mefloquine hydrochloride from Roche Pharmaceutical).
It seems that approximately one in three people taking Lariam experiences “side effects”. That in itself is quite extraordinary. But it gets better. The side effects they experience are “psychotic events”. They go nuts. Does that matter? Well, yes it does if you become a victim because some people who take Lariam turn into homocidal maniacs who kill family members, like their wives and children, and others.
Remember the disgraced Canadian Airborne Regiment which was disbanded after members were convicted of torture and murder while on duty in Somalia? Remember the Fort Bragg family murders committed by members of the 101st Airborne Regiment who had returned from duty in Afghanistan? It seems the soldiers were all given Lariam to guard against malaria while they were overseas.
Both the U.S. and Canadian governments deny any connection while Roche continues to manufacture and sell Lariam. It’s business as usual here. In Europe, however, other anti-malarials are issued to avoid these problems.
It seems that no pharmaceutical creation is ever too dangerous to remove from the market. Remember Thalidamide? (Bayer deserves honourable mention for removing Baycol (cerivastatin) from the market. Then Merck and Pfizer rushed in to grab Bayer's share of the market with their own statins.)
But things eventually do change. Pharmaceutical executives (and doctors) can ask the people at Johns Manville (asbestos) and Phillip Morris (tobacco) if anything happened to their businesses. Did liability matter at Union Carbide? They can inquire into the whereabouts of recent top executives at Ford and Firestone. Where are they now?