Skip to main content

How John Oliver misrepresents medical studies

In a recent episode of Last Week Tonight With John Oliver the eponymous host presented the case that there's bias in medicine with regards to how doctors deal with people based on their sex and race.

In particular, he claimed that there are many extensive studies that show such bias. One such paper showed the result that doctors are less likely to prescribe knee replacement surgery to women than men. Another paper showed that doctors are less likely to give opioid analgesics to women than to men. The presentation of these results was sandwiched between clips of people claiming that doctors regularly believe that women are just "faking it" and making stuff up for attention. Thus, the clear implication was that doctors are biased against women, and thus eg. prescribing less operations and less analgesics to them because they are prejudiced against them.

As you might guess, John Oliver was very deceitful in this, as he was simply cherry-picking tiny tidbits from those studies, deliberately hiding other details. The studies are real alright, but they show more than what he presented.

For example, the same study that showed that doctors are less likely to administer opioid-based analgesics to women than to men pointed out that for some reason there was no such difference when it comes to non-opioid-based analgesics. Curiously, John Oliver left that part out of his presentation.

If it were true that doctors are biased and prejudiced against women, and tend to think that women are just exaggerating and "faking" it, why wouldn't the difference in administering analgesics be the same for all types of analgesic? The fact that there's no difference when it comes to non-opioid based analgesics erodes what John Oliver was trying to convey, which is why he conveniently left that part out.

(What's most probably happening is that doctors are wary of prescribing opioid-based analgesics because those run the danger of causing strong addiction. So what's probably happening is that doctors are actually more protective of women than of men, and thus more careful about giving them things that might cause them long-lasting addiction problems. Therefore if there's any bias here, it's against men, not against women. Doctors may be less careful with men than with women.)

The paper that talked about knee replacement surgeries actually presented several reasons why that may be happening, most of them not related to bias. One of the strongest widely reported reasons was that of misdiagnosis or underdiagnosis due to the differences in how men and women describe their symptoms. Men tend to be very dry and to-the-point, describing their symptoms succinctly and in technical detail, with no unrelated chitchat. Women, on the other hand, tend to make describing their symptoms into a life story, where they present anecdotes and describe things that happened to them with unnecessary detail, which may not help the doctor at all. In other words, many women tend to talk to the doctor as if they were talking to a friend, rather than going straight to the point without meandering all over the place. The paper hypothesizes that this makes it more difficult for doctors to extract the relevant information from the patient's description, and thus may easily misdiagnose or underdiagnose the problem.

These become clear in the publications in question, but of course none of this was conveyed by John Oliver. I'm sure on purpose.

Another example of misinformation is when he states that (in the United States) black people are less likely to receive proper medical treatment than white people, and thus mortality rates are higher among the former than the latter.

There is some truth to this (in the United States), although that gap has been narrowing down considerably (the study he sites is like 15 years old). However, while the claim is true to an extent, what's wrong is what he is implying with it. He is quite obviously implying that black people get poorer medical treatment because doctors are prejudiced against black people, and less eager to offer them the same quality of care as white people.

However, that's not the main reason (if it's a reason at all). The main reason is that, still to this day, the majority of black people in the United States on average live in poorer areas than the majority of white people, and the quality of life of the society is reflected in the quality of medical care in the hospitals in those areas. Poorer cities, and poorer neighborhoods, tend to have lower-quality hospitals than richer areas, which is reflected in the quality of medical treatment.

This in itself is a problem that would be good to solve. However, it's not a problem of racial prejudice by doctors and medical personnel against black people, as John Oliver insinuates. What he's doing is taking a real problem and injecting his own agenda-driven false interpretation into it.

Comments