[marchenko]

At present, the claim made in the article is not as controversial within public health as it may seem: zip codes are a better predictor of many health outcomes, including life expectancy, than our present genome-based predictive models. The public tends to over-estimate the predictive value of polygenic risk scores, especially for complex conditions like metabolic syndrome. Zip codes capture a great deal of environmental variance, plus a bit of genetic clustering by shared ancestry. Most polygenic models fail to capture all of the relevant genetic variance, and are uninformative about environmental factors. Additionally, for some conditions, the magnitude of environmental variance and relative importance of environmental factors is greater than the variance and effect size of genetic factors.

[mcherm]

The original article did not provide any supporting evidence for this claim. Can you point us toward actual research to back up your statement?

[marchenko]

There are very few genetic risk score models that outperform traditional observational disease markers (here[1] is a non-paywalled discussion of cardiovascular GRS performance as an example). The best GRS results tend to be in relatively genetically-homogeneous populations that are similar to the population in which the GRS model was developed. In some cases, knowing the ethnicity or simple family history of a patient can buy you a good portion of the AUC of the relevant GRS.

So if you have a classifier like ZIP, that (1) epidemiologists have done a bit of legwork correlating to classical markers like obesity (or their correlates, such as income and dietary/smoking/prescription patterns) and (2) tends to follow familial/ethnicity clusters in (3) a heterogeneous population, you can amass a fair bit of predictive power on the cheap for complex disorders where environmental variance plays a role, as well as beating the spread on behaviourally-determined mortality/morbidity factors.

It is likely that the predictive power of GRS-based approaches will improve for many conditions in the future (they are of course already powerful for Mendelian disorders).

[1]https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4527979/

[mcherm]

I don't have the slightest doubt that a classifier like zip code (or traditional observational disease markers) delivers much greater value than a genetic risk score model because its accuracy is nearly as good while its cost is a couple of orders of magnitude less.

I do have some doubts that the zip code actually gives BETTER prediction than the genetic risk score. I have difficulty believing that if someone had done a genetic profile on a patient and was willing to tell me either the patient's zip code OR the genetic risk score, that I would be better off asking for the zip code because it had greater predictive value. It' certainly not impossible (because of environmental factors that correlate with zip code), but it is surprising and I haven't yet seen actual research supporting it.

[resu_nimda]

Why do you have those doubts? Do you have any "actual research" to support them?

As someone who doesn't really care either way, I will say that his arguments have been more convincing than you simply casting doubt and demanding proof based on what seems to be a gut feeling.