Discover more from Personal Science
Personal Science Week - 230921 Cancer Diagnostics
Cancer and other diagnostic screening, plus links
Personal scientists love quantified data of every kind, and we jump at the chance to try new diagnostic tests. But how serious should we take the results?
This week we look at the effectiveness of cancer screening, plus some links worth your time.
Personal Science is delivered each Thursday to anyone who wants to use science for personal, rather than professional reasons.
Rabbits, Turtles, and Birds
Let’s say you have this nice garden that you want to keep safe from intruders. A nice fence will keep out rabbits. You don’t really need to worry about turtles, because you can stop them before they cause trouble. But birds are nearly impossible to stop; they’ll get in no matter what.
This analogy (via a discussion with Vinjay Prasad MD on the Econtalk podcast) is a good way to think about cancer and other diseases you might try to control with diagnostic testing — exactly the stuff we like to use as personal scientists. An effective test is like that fence against rabbits: works great and you’re glad you have it. A fence against turtles — i.e. a diagnostic for a disease that won’t harm you anyway — is a waste of time. But some conditions, unfortunately, are like those birds and will get you no matter what you do.
Before you get excited about a new diagnostic, ask yourself whether you’re dealing with rabbits, turtles, or birds. Unfortunately, most diagnostic situations are turtles or birds.
That basic metabolic panel (BMP) you get with a standard blood draw at your annual physical is a good example of how to stop a rabbit. Most of the anomalies you’ll catch are actionable if caught early. You’ll uncover treatable conditions that will benefit from early treatment: insulin resistance, kidney disease, pancreatic conditions, high blood pressure.
Unfortunately, many of other tests we think are important are more like turtles and birds. Last month, a Swedish research team published the results of an analysis of 1.8 million people enrolled in 10-year followup from randomized trial screening of six types of common cancers screening (mammography, colonoscopy, prostate):
The findings of this meta-analysis suggest that colorectal cancer screening with sigmoidoscopy may extend life by approximately 3 months; lifetime gain for other screening tests appears to be unlikely or uncertain.
Here’s the best evidence for mammography and breast cancer screening:
for every 2000 women invited for screening throughout 10 years, one will have her life prolonged and 10 healthy women, who would not have been diagnosed if there had not been screening, will be treated unnecessarily. Furthermore, more than 200 women will experience important psychological distress for many months because of false positive findings. It is thus not clear whether screening does more good than harm.
The (2022) NordICC randomized trial of 80,000 adults found no difference in mortality among those who underwent colonoscopy and those who didn’t.
Prostate cancer progresses very slowly over a lifetime, and you may be better off just waiting it out without treatment. In fact, the US government National Cancer Institute says don’t bother screening unless you’re in a high risk group. Many of the most famous cases (Ben Stiller, Warren Buffet, Rudy Gulliani) might have been just fine whether they had early screening or not.
For most of these tests, it’s impossible to know whether the cancer is a slow-moving and treatable turtle (prostate, skin), a fast-moving rabbit you can stop with early detection, or a bird that will get you no matter what. Unfortunately, the turtles make the most enthusiastic testimonials: nobody wants to admit, after enduring an expensive, life-altering treatment, that the early diagnosis did nothing.
So what’s the point? Should we just not bother with these tests?
Dr. Eric Topol admits that current screening tests are inadequate, but is optimistic that new technologies, including AI and microbiome testing, will improve the situation:
Our own thought is any diagnostic is only as good as the actions you’ll take when you get the results. It’s extremely difficult — even for a die-hard personal scientist — to ignore a positive or even inconclusive test result, no matter how much you understand the statistics. Before you sign up for a test, be sure you understand the difference between turtles, rabbits, and birds.
Links Worth Your Time
Supermoon Capital is an investment firm that specializes in sleep improvements. If you’re interested in the latest technology and interventions that can help, check out their Hypnogram Substack, especially The Market for Interventions for Sleep which gives a nice overview of all the trackers and diagnostics available in the space.
The Library of Digital Endpoints is a crowd-sourced list of all the various interventions under development that use digital technology. It has hundreds of entries in every category of interest to personal scientists, including glucose monitoring, blood pressure, and much more. Supermoon Capital, which pointed us to this list, has a subset table focused just on sleep.
If you have an $85 Polar H10 heart rate monitor and a little programming ability, every-breath-you-take is a Github repo that connects automatically to your device and lets you measure and train your heart rate variability (HRV) in real time.
About Personal Science
Personal Science is the rigorous application of scientific techniques toward personal questions. We’re skeptical but open-minded about all claims we read in the news, and prefer whenever possible to test things ourselves.
Our weekly newsletter is free to everyone. Paid subscribers have access to our “Unpopular Science” series, including our recent one on Lyme disease.
For interactive discussions with other Personal Science-minded people (including me), please join the Open Humans Weekly Self-Research Call, every Thursday at 10am Pacific Time. Open to everyone, and very friendly. (See Personal Science Week - 15 Sep 2022).
If you have additional comments or questions, especially about topics you’d like us to cover, please let us know.