Now Accepting FSA/HSA Payments
Last updated on

I Tried Function Health: Here is What I Learned from 100+ Tests

Article at a Glance

  • Function Health is a health tech startup that has democratized the world of advanced lab testing by allowing health conscious consumers to order the lab tests they want without a doctor as intermediary.
  • Annual memberships are $499 but add-ons like Autoimmunity, Lyme, and Advanced Cancer Screenings can rapidly increase the bill to well over $1,000.
  • While the product’s “digital bedside manner” could use some work, and the sheer volume of data may be overwhelming for some consumers, Function plays an important role in the market and will be worth the cost for many.
  • Function’s lab partner is Quest Diagnostics. If you have excellent health insurance, many of the labs, especially in the default panel, will be covered by insurance.

Genes Mentioned

Screenshot of Function Health website

Backed by Dr. Mark Hyman, Function Health is a relatively new personalized health testing platform that offers a panel of over 100 advanced lab tests with the slogan “100 healthy years.”

Gene Food’s health reporting recommends follow up lab testing our users can discuss with their doctors and my go-to has been Boston Heart Diagnostics, but that may have changed after using Function Health.

This is an excellent company and I had a positive experience. Anecdotally, the phlebotomist I used at the Quest Lab (which function uses) said they see a lot of Function patients, especially on weekends.

Let me dive into my results, including what I think could be improved, and what tests I believe most users will want to skip (because the service is pricey). I also have a recommendation for the Function brass about how they display their men’s health data and lipid reporting, plus some thoughts on “digital bedside manner.”

Get Started With Personalized Nutrition

Gene Food uses a proprietary algorithm to divide people into one of twenty diet types based on genetics. We score for cholesterol and sterol hyperabsorption, MTHFR status, histamine clearance, carbohydrate tolerance, and more. Where do you fit?

What I paid

I wanted to do a deep dive into the product so I ordered several add-on tests. Here is my full slate of tests and the price for each.

  • Annual membership ($499)
  • Lyme antibodies ($549)
  • Extended auto immunity ($249)
  • Celiac ($69)

As you can see, this is very expensive testing and it’s not covered by insurance. If you have good insurance, most of these labs will be available to you at no cost, but the focus here is on agency and the ability to order the labs you want, when you want them.

I ordered the Lyme panel because I am always curios to see those results due to the headspace hidden Lyme occupies in functional health circles.

For some reason, I think of Ross Douthat whenever the topic of Lyme comes up. Mr. Douthat, an opinion writer for The NY Times, contracted Lyme soon after moving to the Connecticut countryside, and shared his harrowing story in a tell all health memoir. Because the world of “hidden Lyme” is murky and the diagnostic tools can vary, I was curious to get my Lyme panel done at Function.

Same for advanced autoimmunity, I wanted to see what that test encompassed and the celiac panel I order from time to time to see where my tissue transglutaminase labs sit as a marker for wheat sensitivity.

Onboarding and setup

The first thing I noticed about Function after signup a similar questionnaire onboarding to the one Viome uses. Function wants to know whether you suffer from chronic illness, what supplements you take, how active you are, what type of exercise you do, your eating and sleep habits, and more.

The initial questionnaire doesn’t really tie back to the user’s results, which I believe is a missed opportunity, but it’s not that bad relative to some other apps I’ve tested.

The questionnaire is much shorter than Viome’s, less intrusive, and some of the questions do relate back to the process of getting blood drawn, so I give Function a pass here. It’s a light touch and time efficient, even if it does not tie back to the ultimate results of testing.

Function health membership labs overview

Report Type Biomarkers Comment
CardiovascularApoB, HDL, LDL-C, Non-HDL, Total, Cholesterol/HDL ratio, TG, Lp(a), HomocysteineWould like to see sterol panels. Non-HDL not necessary when testing ApoB
LiverALT, Albumin, Albumin/Globulin Ratio, ALP, AST, GGT, Globulin, Bilirubin, Total Protein Comprehensive
MetabolicGlucose, HbA1c, Insulin, LeptinComprehensive. Hard core biohackers may want post prandial blood glucose via a wearable
Nutrients Calcium, Ferritin, Iron, Iron % saturation, Iron Binding, Magnesium, MMA, Vitamin D, Zinc Opportunity to contextualize with genetic data
Stress & Aging DHEA Sulfate May shed light on adrenal function
Pancreas Amylase, Lipase Could reveal digestive issues related to enzyme insufficiency
Biological age Pulls from various lab tests Different from DNA based epigenetic clocks
Hormonal PSA Free, PSA Total, FSH, LH, Prolactin, PSA Total, SHBG, Testosterone free and totalPSA Free and % Free available to contextualize PSA risk when high
Thyroid TgAb, TPO, TSH, T4 free, T3 freeThese biomarkers are of interest to many people

The list above isn’t exhaustive, but designed to highlight some of the labs I found most interesting. Out of the box, you get 100+ lab tests, with a menu of expensive “add-ons” available.

Some of the add-ons are interesting, for example MTHFR and APOE gene analysis and IgE mediated allergy to foods like peanuts. Others, like cancer screenings are cutting edge, but have the potential to leave users “on their own” with what could be troubling and inconclusive results. I skipped the advanced cancer screening after digging around and finding that while only approximately 1% of users receive a positive result, of that user base, only 40% end up having cancer.

Function uses the Galleri Test for its advanced cancer screening add-on. The positive predictive value (PPV)—the likelihood that individuals with a positive test result truly have cancer—varies. In certain studies, approximately 38% to 44% of individuals with a “Cancer Signal Detected” result were confirmed to have cancer upon further diagnostic evaluation (R).

And that’s the big issue with screening like this – there is no guarantee you won’t receive information that is stressful. Using a service like this is a balance between empowerment and anxiety, as it gives you access to knowledge that can help you take control of your health, but also opens the door to uncertainty and potential worry about findings you may not fully understand.

There are several Reddit threads full of people very concerned about what they learned from their Function results.

One area I think Function could improve is with their content and risk tiers. In many cases, visual reports break down into either risk (out of range) or no risk (in range), but I think use of a third tier that classifies some users as borderline would help ease the concerns of some users.

For example, in the screenshot below, you see my LDL-C number, which was 97 mg/dL on the day I tested. In the graphic there is either “In Range” or “Above Range,” but someone with LDL-C at 165 mg/dL (which is very high) will see the same content as someone with LDL-C at 101 mg/dL (which is a reasonable number). Later in the report, they offer more information on risk in text, but this content should be built into the visuals the user sees because many won’t take the time to read the written content further down the page.

More visual stratification in the UX would allow users who have only a borderline elevation the peace of mind to know that their results aren’t wildly out of range, and that their risk remains low. As someone who has designed products like this which have to place people into buckets based on numerical values, I am not saying its easy, but an overhaul of the content in this way would, I think, put many people’s minds at ease who use the product.

The advanced lipid panels (the area I am most interested in) contained the following biomarkers:

  • ApoB
  • hs-CRP
  • Lp(a)
  • Homocysteine (included in the Nutrients report)
  • Non-HDL (which isn’t necessary since they have ApoB)
  • Total cholesterol
  • Triglycerides
  • Total Cholesterol / HDL ratio

While this is a strong start, I would love to see Function add sterol panels for users. Sterol panels are a key missing link here because biomarkers like sitosterol and desmosterol are the guideposts that tell us why we are dyslipidemic, not just what the lab values say.

Boston Heart offers sterol panels through their Cholesterol Balance Test.

Predictably, my Lp(a) value came back moderately elevated, which is consistent with my genetic profile.

Gerald Dropped His LDL-C by 100 Points With Diet Alone

Gerald was an ultramarathon runner, but despite his dedication to fitness, he was struggling with rising cholesterol levels, increasing blood pressure, and low energy.

Methylation and homocysteine

I am sharing my homocysteine results, which were slightly out of range, to illustrate two points:

  • Methylation goes beyond the MTHFR gene
  • Function should add a third “borderline” or “moderately elevated” tier to their visual content (not just in text)

According to the visual lab values, my homocysteine is out of range, but barely so. Further down the report, the homocysteine values listed don’t correspond with the visual plot that maps the user’s results.

According to the visual, my results are just out of range, but further down the report, “optimal” homocysteine for men is listed as 6.0-8.0 umol/L and “slightly above” is listed as 8.0-12,o umol/L. This is all confusing because the text report and visual plot don’t align. The inconsistency between the visual plot and the written report should be addressed by the Function content team.

Although my homocysteine wasn’t terribly out of range, my results did prompt me to increase my methylated B vitamin intake. Although, my MTHFR function is normal (with only one SNP for MTHFR A1298C), my methylation score flags as elevated risk in our scoring model, and I do carry a homozygous MTRR “mutation.” The Function Health reporting was a good reminder to be more mindful of B vitamin supplementation and my knowledge of genetics clued me in on why homocysteine had fallen out of range.

MTHFR results Gene Food

Digital bedside manner

In the age of AI and Telehealth, digital bedside manner will become that much more important. Offering laypeople access to advanced tests is important, but with all this data comes great responsibility to contextualize and properly frame these tests as part of a broader picture of health. The various tests don’t “speak to each other,” and in my view, greater integration would make a great product that much better.

To illustrate the point, the PSA and PSA Free % reporting in the Mens Health reporting dashboard could use some work.

Since the PSA Free % (the portion of PSA in the blood that is not bound to proteins) can be elevated due to recent ejaculation, or prostatitis, so reporting on this number when the PSA value is very low seems like it could cause undue stress in some men who take the test. The mental health ramifications of this testing can not be overlooked because you are dropping an enormous amount of information on laypeople, so Function is tasked with building “digital bedside manner” and I think they could improve in this area.

Overall, I think the Function team would be wise to audit all of the content in the app with an eye towards:

  • Offering greater personalization based on a user’s intake questionnaire and results of other labs (i.e. – “you flag as high risk for [LAB VALUE], however based on [RELATED LAB VALUES] your risk remains moderate. We always recommend speaking with a physician in these cases, but your overall health appears good…”)
  • Digital bedside manner which could include a directory of physicians nearby who are equipped to answer questions on a given issue.

Final thoughts

Function Health is an excellent company. I only had to drive 15 minutes to a Quest Diagnostics lab to get blood drawn, the results came back much faster than had I been waiting for my doctor, and the presentation is professional. The best value is with the $499 annual membership. Next time, I would skip the add-ons and focus on the biomarkers that are part of the core product, which are extensive.

As I mentioned above, I do think there are several ways the content and visual presentation of the reports could be improved, but overall Function earned an A- from me as a user. Congratulations to the Function team on building a fantastic business.

Gerald Dropped His LDL-C by 100 Points With Diet Alone

Gerald was an ultramarathon runner, but despite his dedication to fitness, he was struggling with rising cholesterol levels, increasing blood pressure, and low energy.

John O'Connor

John O'Connor is the founder of Gene Food, a nutrigenomic startup helping people all over the world personalize nutrition. John is the host of the Gene Food Podcast and a health coach trained at Duke's Integrative Medicine Program. Read his full bio here.

The very latest on genetics, nutrition and supplements delivered to your inbox!

Facebook icon Twitter icon Instagram icon Pinterest icon Google+ icon YouTube icon LinkedIn icon Contact icon Info icon Email icon Phone icon Pin icon Back to top