Bio-markers and Bio-plausible not bio-promoting

Sometimes when my patients and I discuss a treatment option, I might get a little rant going about how researchers and health professionals and also patients are confusing bio-markers with actual health outcomes. This is especially significant when trying to evaluate whether a treatment is (1) evidence-based medicine and (2) whether the evidence is actually useful evidence.

There is a long but very interesting article co-published by ProPublica and The Atlantic, which everyone interested in health should read:
When Evidence Says No, But Doctors Say Yes, by David Epstein, published 2/22/2017. This article is primarily about how bio-plausible (it just makes sense!) treatments keep getting recommended to patients despite in-depth quality evidence showing that they aren’t useful.

I am very interested in this in general, but considering my particular pet peeve of biomarkers being confused for outcomes, this paragraph stuck out:

In 1997, a Swedish hospital began a trial of more than 9,000 patients with high blood pressure who were randomly assigned to take either atenolol or a competitor drug that was designed to lower blood pressure for at least four years. The competitor-drug group had fewer deaths (204) than the atenolol group (234) and fewer strokes (232 compared with 309). But the study also found that both drugs lowered blood pressure by the exact same amount, so why wasn’t the vaunted atenolol saving more people? That odd result prompted a subsequent study, which compared atenolol with sugar pills. It found that atenolol didn’t prevent heart attacks or extend life at all; it just lowered blood pressure. A 2004 analysis of clinical trials — including eight randomized controlled trials comprising more than 24,000 patients — concluded that atenolol did not reduce heart attacks or deaths compared with using no treatment whatsoever; patients on atenolol just had better blood-pressure numbers when they died.

I’m diverging from the topic of the article to point this out, because it is very important when weighing your options.

A colleague asked the Facebook hivemind what questions she, as an accompanying advocate, should ask the physician/surgeon of a friend who had debilitating back pain about a proposed surgery. Many people chimed in that in addition to her already proposed “risks, alternatives, and physician experience” and etc., she should absolutely ask about the expected result of the surgery. This goes for medications and nutritional changes and physical therapies as well. What are you trying to achieve? What can you expect to achieve?

A lot of interventions (medications, nutritional changes, exercises and physical therapy… anything you change in order to make a difference) have been tested against whether they make a difference in bio-markers because it is too costly and risky to see if they make a difference in actually important things like quality of life, mobility and individual agency, and duration of life. They’ll therefore conflate a biomarker known to be involved in the outcome, such as LDL cholesterol or blood pressure, with the outcome of Not Dying Suddenly From A Heart Attack or Permanent Disability After A Stroke. Again, the reason is that it’s often too costly or too difficult/unethical to get permission to run a study long enough to find out; but that means that there’s lots and lots and lots of treatments out there that make your numbers look pretty without addressing the cause for those numbers to be “off.” They also confuse people as to whether the biomarker is the cause of an outcome, or just related to the outcome. Things like LDL cholesterol can be a predictor of cardiovascular disease but don’t necessarily cause CVD!

This doesn’t even get into the difficulty of studying one particular outcome, like cardiovascular disease or skin cancer, while ignoring all other outcomes, like autoimmune disease, diabetes, other types of cancers, infection, and all sorts of other things that can make life difficult or make life end. Or being so focused on reducing the biomarkers of one disease (blood pressure as an indicator of CVD) that you take risks with other diseases (blood pressure medications making diabetes mellitus type II more common) which can increase mortality or reduce quality of life.

It’s a very tricky balance. It’s not something that you can just figure out by finding the one right study.

That’s why it’s important to find clinicians who will work with you holistically –as a whole complex person– to figure out which risks are the real risks, which biomarkers are worth chasing down into pretty, non-threatening ranges, and which interventions are actually helpful versus which ones are just to cover everyone’s behind.