Screening Laboratory Tests for Runners, Part 1

Can having your blood tested–and making diet or lifestyle changes based on the results you receive–help make you a healthier or faster runner? This is the question we will try to answer in a two-part article series about biomarker testing in adult athletes.

This first article focuses on the basics of direct-to-consumer laboratory testing, iRunFar’s initial testing experience with a direct-to-consumer laboratory testing service for adult athletes called InsideTracker, and our feedback on the process. The second article covers follow-up testing and a summary of what we have learned or changed based on our multiple test results.

For this article series, InsideTracker provided free blood testing to iRunFar’s Bryon Powell and me.

Figure 1. Blood collected during a blood test. Photo: Tom Mallinson

Direct-to-Consumer Laboratory Testing Basics

Direct-to consumer laboratory testing allows adults to order their own blood testing without involving a physician or other health-care practitioner. The direct-to-consumer laboratory testing market is growing and reached a total of $208 million in 2018 and is expected to grow to $350 million by 2020 due to easing of regulations and increase in demand (1, 6). Laboratory-testing businesses such as LabCorp and Quest Diagnostics have partnered with Walgreens and Safeway respectively to reach consumers (1).

The general consumer is becoming better informed about what lab tests are available and, with the growing popularity of high-deductible insurance plans, consumers want to be in charge of what they are paying for. All this sounds exciting in theory, but it does come with challenges, such as consumers understanding their results and learning what to do if they receive abnormal test results.

With a cornucopia of blood tests available to you as an athlete, what, if anything, should you have checked? The short answer is that it depends on who you are, what you do, and how you feel. The more complicated–and important–answer is that more is not more. Specifically, more testing is generally not better. Let’s dive into this.

Laboratory Testing in Healthy Athletes

Here is one example of testing of healthy athletes which nicely illustrates the ‘more is not more’ principle: A recent research study (3) of healthy professional athletes of a wide range of ages found that 10.1% of initial screening lab results were abnormal. And 40.3% of these abnormal results led to additional testing. Out of all of the lab tests and abnormal results, only 0.35% of initial lab tests led to a change that resulted in a significant positive outcome. That was one out of every 300 tests. In this study, the positive outcomes were detecting non-vaccinated athletes who then became vaccinated against measles, mumps, and rubella as well as chickenpox. Tests of thyroid function, blood counts, screening for diabetes, and more led to no significant positive clinical outcome, only extra testing, money, and worries.

Granted, most of us are not professional athletes, but this provides some foundation for understanding why screening lab work among healthy athletes is generally not recommended. The fact is, current consensus regarding lab screening testing for otherwise healthy adult athletes is that annual screening lab testing should not be performed because it’s an unnecessary cost driver and creates unnecessary concern for a very low chance of beneficial outcome (4, 2, 7). This list of recommended health-screening tests shows that no blood lab screening tests are recommended in all adults at any age.

Why is this? In most (but not all) lab tests, there is a 5% chance a healthy person will end up with a result that is falsely flagged as abnormal. This is due to the pre-defined normal reference range which is based on the 95% inter-percentile range in a Gaussian distribution. In other words, 5% of most blood tests are expected to fall in the abnormal range but still be normal for that person. The healthier you are and the fewer symptoms you have of the specific disease being tested for, the more likely it is that an abnormal result is actually normal for you. We call this a false positive. And if you are a healthy, asymptomatic individual who has 20 labs tested, one is statistically expected to be falsely abnormal.

Pre-Test Probability/When Should You Test?

Was I not the one who wrote the iRunFar article on the high prevalence of iron deficiency among female runners? Yes! Could the testing of an iron panel in female runners thus be an exception to the above rule about not doing screening testing? Potentially. Estimates of the rate of iron deficiency among female runners vary, but I have seen rates as high as 50%. Can you thus see how this is different? If you have a 50% pre-test probability of abnormal results, then if/when you obtain an abnormal result, it is much more likely to be real.

Let’s take the opposite, male-gender scenario and say, despite not knowing your iron levels, you decide to take iron supplements. If you read that article, then you will understand that your iron level is much more likely to be too high. So if you have an iron level that is higher than the reference range, this again is much more likely to be truly abnormal.

Having a health-care provider to help you navigate these situations is the best because they can help you determine–given your specific age, gender, symptoms, and by whether or not you take supplements, medications, and more–if you should have bloodwork drawn and, if so, which tests. This helps you avoid getting test results that read out as abnormal but are normal for you or do not require any treatment. As mentioned, unnecessary tests can get very expensive and can also be disconcerting if their value is off and you are not sure what, if anything, you should do.

InsideTracker

Shortly after I published the article on iron deficiency, InsideTracker contacted me to express interest in collaborating on research in iron deficiency and women’s health. As part of this, they offered me free comprehensive testing via their ‘Ultimate’ panel of tests. Knowing what I wrote above, I should have hesitated before saying yes, but of course I took them up on the offer.

I have a history of iron-deficiency anemia, so I suspected my iron level to be off. I was already taking iron supplements and wanted to know if I should continue with them. I recently had an iron panel and complete blood count (CBC) ordered by my doctor that would come at no charge to me due to my health care and that I had not yet fulfilled. I still went with InsideTracker because their testing center is a half mile from my house and I was curious what else I would learn as their ‘Ultimate’ panel of tests includes many more tests than an iron panel and CBC.

Figure 2. InsideTracker’s Ultimate panel tests. Image is a screenshot from the InsideTracker website.

InsideTracker has a unique platform compared with many other direct-to-consumer lab-testing companies in that they not only provide your lab results, but they also offer advice about how to improve your biomarkers. This advice, per InsideTracker, comes from “a team of nutrition and performance interns from Harvard, Tufts, and MIT.” They also have peer-reviewed published evidence (8) that people who have abnormal measurements tend to normalize many of their abnormal results by the time of the retest (at least 30 days later). They have deduced that this is because test takers follow InsideTracker’s advice.

Figure 3. A graph showing the normalization of InsideTracker’s tested biomarkers from baseline (first test) to follow-up testing. In the fourth column, if the p-value is less than 0.05, their data suggested that the lab value in question tended to normalize on repeat testing. Image courtesy of Westerman et al. 2018 (8).

In reality, this normalization may not be due to the recommendations they provide. Instead, another likely explanation is statistical regression to the mean. In other words, an abnormal value on initial testing tends to normalize with retesting due to sheer statistical likelihood.

For example, let’s say you design a study where you only include people whose blood pressure is very elevated on initial testing, then you tell them to wear blue shoes for two weeks and come back to be retested. When they are retested, their blood pressure has likely dropped simply because it was unlikely to stay as high as it was when it was first measured–not because of wearing blue shoes. This is why, for the InsideTracker study, having a control group of people who had abnormal biomarkers but did not receive any advice would have been useful in determining if the advice InsideTracker offers was causative in the change.

But each test should probably be looked at individually. If you find your vitamin D is low, that is easy to treat via supplements or sun exposure. For glucose normalizing, perhaps this is simply due to some test takers not realizing they should be fasting for their first blood draw. CPK normalization is likely regression to the mean. However, the normalization of LDL (one kind of cholesterol) is potentially more impressive, though again, without a control group, it is hard to establish what caused the improvement. Maybe their initial testing just happened to correspond with the start of a new exercise program, for example.

Figure 4. InsideTracker lab testing can be done a Quest Diagnostics. In this photo, Quiggles the Quest Diagnostics mascot gets their health checked. Photo: Quest Diagnostics Loxahatchee

The InsideTracker testing process is very smooth. They send you a lab slip which you print and take to your local Quest Diagnostics. They will even come test you in your home. When I checked in at our local Quest, the lady at the front counter said, “Wow, this is a lot of tests! How cool!” I thought so, too. After all, there were tests that I as a practicing physician had never ordered on a patient, such as sex hormone binding globulin (SHBG) and red blood cell (RBC) magnesium.

I had clearly shed my physician and statistician hats to have blood tests done while not knowing exactly what was being tested for and without any medical symptoms. This goes against both my medical and scientific educational background, but the opportunity to gain new knowledge is also very intriguing. Would these enticing esoterica be the key to my success as a runner?

Despite the large amount of tests the InsideTracker Ultimate panel offers, I was surprised by certain tests not performed:

• Thyroid-stimulating hormone – This thyroid-function screening test is generally one of the first tests ordered if an athlete has low energy levels.
• Estradiol (Estrogen) – Why is testosterone tested and not estrogen? If SHBG is off, a woman will want to know what her estrogen level is.
• Creatinine or some measure of kidney function – If you are going to check CPK and electrolytes, then you should check kidney function.
• Hemoglobin A1c (average blood glucose over the last three months) – Hemoglobin A1c is a better screening test for diabetes since it does not have as many false positives and negatives as a plain glucose level test.

And I would have left these tests out:

• Glucose – Again, hemoglobin A1c is more useful.
• SHBG – SHBG seems to exist to protect people from elevated levels of sex hormones, testosterone and estradiol among others. It is high when a person has high levels of hormones to bind up the extra hormones and low when the sex hormones are low to free up the available hormones. But if it is too high or how, you don’t know what is off based on InsideTracker results alone because not all sex hormones are tested.
• RBC magnesium – This measures the magnesium levels inside your red blood cells and is supposedly more sensitive for detecting magnesium deficiency. A simple magnesium level should be sufficient.
• AST, ALT, and GGT (aspartate aminotransferase, alanine aminotransferase, and gamma-glutamyl transferase) – These are all measures of liver function. I would leave these to physicians to check when appropriate as a borderline abnormal result again is most likely normal in the population and an abnormal when it comes to the liver can be very worrisome and not something that is normally addressed with diet or lifestyle changes.

Six days later, I received my results by email. I felt giddy as I opened up the results after work. But as I looked through my sub-/supra-optimal or abnormal lab-number values, I found myself with a lot of questions (and frankly fears) about the meaning of what I was seeing:

1. My liver-function tests were all higher than optimal or abnormal. I felt fortunate that I had had these tested multiple times in the past and they had always been elevated with no cause ever identified. I thus knew this was nothing new for me. Mildly elevated liver-function tests also occur on my mother’s side of the family. Had I not known this, I likely would have freaked out. (Okay, I still freaked a little anyway.)
2. The most worrisome result was that my SHBG was suboptimal. It then said, “Your diet, weight, exercise level, and age all affect your SHBG level. Below-optimal levels of SHBG are associated with an increased risk of being overweight, having unstable blood sugar levels, and being less fertile.” I had previously attributed my lack of menstruation over the last year to my new intrauterine device (IUD), but I then wondered if my low SHBG was due to premature menopause. Basically, I did worry, and I didn’t have any other lab results that could give me a clue, other than my testosterone, which was low-normal. And honestly, I still do not know if my low SHBG is due to premature menopause. I have been too busy/lazy to drive the 30 minutes to get my anti-Mullerian hormone tested to see if I have gone through menopause and, to be honest, I prefer to go on believing it is my IUD causing my lack of menstrual periods. To make things even more complicated, hormone levels tend to drop throughout the day and I had my labs drawn in the afternoon, so maybe all of my hormone levels were normal for an afternoon test.
3. The most entertaining result was my elevated RBC magnesium. This result basically seemed to break the recommendation algorithm since the only recommendation I got was to stop magnesium supplementation, which I don’t take. It was reassuring, though, that my blood magnesium level was at the lower end of normal. I still do not know if my RBC magnesium was a false test result or perhaps there was something in my diet from the day before that had a high amount of magnesium, or maybe my RBC magnesium just runs high. This is where I could have used the help of another health-care professional!
4. My testosterone/cortisol ratio was borderline low as well, but my cortisone was on the low end of normal and my low testosterone level may have been related to having my testosterone tested at 2 p.m. Testosterone should be tested at 8 a.m. when it is around its highest level as it drops significantly throughout the day.
5. The most useful results for me were from the CBC and iron panel. My ferritin, hemoglobin, and transferrin saturation were all what InsideTracker calls suboptimal. While not frankly abnormal, this did reinforce my belief that I should continue iron supplementation, or at least that it was not dangerous.

Figure 5. The InsideTracker readout the author received regarding suboptimal sex hormone binding globulin (SHBG). To be clear, the author isn’t overweight. Her body mass index is around 18 and fat percent 11.1% the last time she was measured a few months ago. She also exercises daily, so she was left wondering how to ‘optimize’ this. Image is a screenshot of the author’s InsideTracker results.

InsideTracker’s Optimal Zone

The statistician in me is teeming with excitement to discuss the ‘optimal’ zone in the InsideTracker test results feedback. What InsideTracker has done is narrow the previously established ‘normal’ range and created a range of values that are technically normal but not optimal. As previously described, normal ranges for most lab tests are established based on the fact that 95% of all healthy tested subjects fall within normal. Normal ranges are established for each lab and on each patient population based on this.

InsideTracker’s optimal ranges may be based on one or more research study suggesting that one end of the normal reference range is less healthy or normal and thus labeled suboptimal. In theory, one could also have the opposite situation with a value higher or lower than normal which may be felt to be optimal based on one or more studies. InsideTracker appears to have only used the former rather than the latter for their optimal ranges. This creates a situation where a much higher percentage of the people tested are labeled as suboptimal.

The InsideTracker blog explains that the optimal ranges are based on the information they know about the test taker–age, ethnicity, medications, and more. But normally when you take these factors into account, the normal range just shifts and the area of non-normal results does not grow, at least not in general. InsideTracker suggests the currently used normal ranges are outdated, but again, it seems unlikely that all current research has suggested that the normal ranges of every test should be narrower.

This is an important distinction because if five out of every 100 normal people has a falsely abnormal test based on the classic reference ranges, imagine how many false positives you would find with optimal-range testing? Suddenly 10 or 20 out of every 100 people tested are told they are outside of the optimal zone. And one must ask whether or not this is beneficial overall and whether or not these results could or should be improved. Does it create more worry or does it result in positive health outcomes?

A New York Times article (5) describes the problem well: “Dr. Pieter Cohen, an assistant professor at Harvard Medical School and an internist at Cambridge Health Alliance, cautioned that the levels of vitamin D and other biomarkers that were optimal for one person might be very different from what is optimal for another person. He said InsideTracker’s lab reports, for example, classified vitamin D levels below 30 nanograms per milliliter as ‘low’—even though a level above 20 is perfectly normal and adequate for most people. Dr. Cohen said his major concern with direct-to-consumer blood tests was that they screened for so many biomarkers and created seemingly arbitrary ranges for what is considered normal. Then they give people advice that they already know they should be following.”

The other issue with the optimal zone is that it encourages repeat biomarker testing. One can easily see how attractive of a business model this is for InsideTracker. Customers may find it a fun challenge to try to optimize all of their values, but is this challenge realistic or just expensive? The statistical likelihood that all of the 43 results of InsideTracker’s Ultimate panel will be found to be optimal is only 1.1%. (Assuming 10% of test takers land outside of the optimal in each test. This is 0.9 ^ 43, if you want to follow along in my math.) So striving for perfection is highly unlikely and could get pretty expensive. And 1.1% is likely an overestimate of the likelihood given the optimal zones are often quite wide. As you can see above in Figure 5, the optimal zone for SHBG is almost as large as the normal zone.

Interview with Bryon Powell on His InsideTracker Testing

iRunFar’s Bryon Powell was also tested via InsideTracker’s Ultimate panel. Here we ask him questions about his experience.

iRunFar: Could you explain what kind of health care you have?

Bryon Powell: I have a high-deductible health plan. It’s bad enough that I don’t ever consider using it. It’s merely catastrophic insurance I picked up to avoid Affordable Care Act penalties following seven years of no health insurance. I now maintain it at the insistence of my dear wife. ;-)

iRunFar: What were you hoping to learn from InsideTracker?

Powell: For years, I’d wanted to have a series of InsideTracker tests with at least one in each of three periods: rest/easy base training, peak training, and after a focus race (following an appropriate acute recovery period). I’d hoped to see how various stress markers reacted across those time periods. As a bonus, I thought I might identify some nutrient deficiencies.

iRunFar: What was your experience with the testing itself?

Powell: I live pretty remotely, so I had an in-home blood draw… twice. InsideTracker sent me a kit and I scheduled an in-house appointment with a phlebotomist. Unfortunately, I was only able to schedule a draw on a Friday which meant my blood wasn’t tested quickly enough for various tests to be completed, including the complete blood count (CBC). That said, InsideTracker quickly resolved this by scheduling a complete redraw, so I got a bonus set of results for a majority of the tests as well as CBC results. And they were kind to facilitate my rescheduling of the redraw when I learned of my grandmother’s passing two days before the test and left to be with family. As for the actual draws, the same woman came out both times. I hate needles and blood draws but she was compassionate and understanding.

iRunFar: Nice! What did you learn from the results?

Powell: Uh, I learned that I was poisoning myself [with iron supplements]. Seriously! My iron measures were through the roof and could have serious health consequences as well as result in various negative symptoms in life and running, including low energy. Guys [men], don’t take iron supplements and possibly even multi-vitamins with iron unless you’ve had bloodwork that confirms you have low iron and a medical professional recommends iron supplementation!

iRunFar: Wow! Great example of how supplementation without known deficiency can be dangerous. Thanks so much for sharing this. I am so glad it was caught!

Powell: I also learned that I’ve got really low free testosterone, that is biologically available/active. Surprisingly, this could be tied to my extremely elevated iron levels. This could be part of why I’ve been low energy/low motivation for so long.

iRunFar: Thanks for sharing all of this, Bryon. Absolutely, elevated iron can lower testosterone levels. What do you plan to do with the results?

Powell: Well, I immediately stopped taking both a 65-milligram iron supplement and a multi-vitamin with iron. Honestly, while I glanced at the rest of the results, this was my key takeaway at this point. I’m omnivorous these days, but eat very little meat, especially red meat. I’m crossing my fingers that my retest in May shows big improvement here. Even in the few weeks between my two tests, my iron measures went from catastrophically bad to merely very, very bad.

More long term, I’ll have to see if my free testosterone levels recover as my iron levels decrease. If not, I’ll chat with some medical professionals about this and, then, weigh the health and ethical issues that surround [treating this with supplementation].

iRunFar: It’s worth it to explain the ethics Bryon mentions. Testosterone supplementation is prohibited by the World Anti-Doping Agency without a therapeutic use exemption, which is only granted when it’s medically shown that you have one of a very few genetic diseases with low testosterone. Bryon, what are your thoughts in general on being tested? Do you think it was overall beneficial either for your health or performance? If yes, how? If not, why?

Powell: Oh, gosh, yes! I never would have even thought that my iron levels could be too high. Here’s hoping that I start feeling better in both running and life by stopping taking two little pills! Also, I’d prefer to avoid organ damage where possible!

iRunFar: Makes sense! Do you recommend all runners have this testing? Or are there certain things that should be included or left out?

Powell: I’m not sure that all runners should have this testing. If you’ve got good health insurance and can get a good blood panel and medical analysis and advice through the insurance, that might be a good route. Similarly, if you’re a casual runner and feel good, there’s not much reason to get such testing.

I think two groups could benefit most from InsideTracker testing. The first are athletes looking to maximize performance who have some disposable income, as I think InsideTracker provides some really good analysis and advice based on an individual’s test results. Second, if you’re an athlete who’s been feeling like crap for a while (if that athlete has a long-enough athletic history to sense that something’s off), especially if you don’t have good health insurance that might facilitate testing with no/low out-of-pocket costs, it’d be great to identify if there is something wrong with you, whether that’s too high or too low a level of one or more macronutrients, or one or more hormones that are out of whack.

I think it’s worth considering whether or not the sex-hormone-related results should be included. Yes, they are an important marker of health but, in the context of athletic performance, it’d be very easy for a completely well-meaning athlete to see low testosterone results, consult his or her physician, and begin medication without thinking about the ethics. On the other hand, such results might simply indicate that an athlete is overtraining and should back off for a while, or the test could reveal a significant health issue. Maybe I’d just like to see more information presented more prominently here.

Does Blood Testing Encourage Doping?

Bryon Powell brings up the great point that testing hormone levels could unintentionally encourage doping, specifically with testosterone. When an athlete, male or female, finds out their testosterone–or another hormone such as dehydroepiandrosterone (DHEA)–is low or suboptimal, they may wonder if they should supplement it. Especially if that person thinks, Maybe eating yardlong beans as InsideTracker recommends to raise my testosterone is not going to do it.

Figure 6. Yardlong beans. Photo: Wikimedia Commons

I would be curious to know what percentage of people have found out through this sort of testing that their testosterone was lower than optimal and started supplementing without understanding or considering the ethics of doing so. I do not believe it is InsideTracker’s intention to encourage doping behavior, however, I would not be surprised if unintentional doping in our sport actually increases with testing like this becoming more common.

Interpretation of Your Results

Consumers correctly interpreting their results is a big issue with direct-to-consumer lab testing. I personally have been contacted on numerous occasions by athletes who obtain abnormal results and don’t know how to interpret them. InsideTracker offers helpful advice, but, even with it, I feel that many results could be alarming to consumers if they do not have easy access to a health-care provider to interpret the results and order more testing if appropriate. Readers who are considering a comprehensive type of screening lab panel should consider beforehand that they may end up with worrisome results that they are not able to interpret without the assistance of a health-care provider and/or may end up needing to have multiple extra tests because of initial abnormalities found. This, of course, can get expensive and time-consuming.

In Summary

• Screening lab tests are not routinely recommended for healthy athletes.
• Women athletes may consider being tested for iron deficiency and anemia with an iron panel and complete blood count because the prevalence of iron-deficiency anemia is high in this group.
• If you are not feeling well, discussing your specific symptoms with your health-care provider is the ideal approach. However, rather than neglecting your health if you have a high insurance deductible, direct-to-consumer lab testing may help you look into the most common causes of your symptoms. The more specific you can be with the tests you want, the better. Involving a health-care practitioner in this process can only benefit you.
• Vitamin and mineral supplements beyond the daily recommended allowance should not be taken without a known deficiency or doctor’s recommendation.
• Unnecessary testing can yield confusing and worrying results, and is expensive.
• Consider carefully exactly which tests you want to have taken and what you will do if one of the levels is abnormal.
• Striving for perfection in your bloodwork or in any aspect of life can become a tiring pursuit of the impossible. Sometimes a healthy dose of acceptance can be the best medicine if you feel healthy.

Call for Comments (from Meghan)

• Have you had direct-to-consumer lab testing done?
• If so, can you explain what the process was like for you and what you learned from it or how you modified your diet and lifestyle as a result?

Acknowledgements

Thank you to InsideTracker for providing testing. I hope that my honest critique was appreciated. I tried to remain as unbiased as possible in writing this article. The InsideTracker team is a pleasure to work with and sales manager Jonathan Levitt deserves special thanks.

References

1. Carlson, B. 2019. Direct-to-Consumer Diagnostic Testing Market Exceeds $200M: Report. https://www.prweb.com/releases/direct_to_consumer_diagnostic_testing_market_exceeds_200m_report/prweb16036400.htm.
2. Conley KM et al. National Athletic Trainers’ Association position statement: pre-participation physical examinations and disqualifying conditions. J Athl Train. 2014; 49: 102-120.
3. Darche, JP, et al. 2019. Assessing the utility of yearly pre-season laboratory screening for athletes on a major professional sports team. J Sci Med Sport. 22 (4) 484-487.
4. Fallon KE. The clinical utility of screening biochemical parameters in elite athletes: analysis of 100 cases. Br J Sport Med. 2008; 42.
5. O’Connor A. Direct-to-Consumer Lab Tests, No Doctor Visit Required. New York Times. Well. June 6^th, 2016.
6. O’Reilly KB. What your patients must know about direct-to-consumer lab tests. American Medical Association. Precision Medicine. 11/20/2018. https://www.ama-assn.org/delivering-care/precision-medicine/what-your-patients-must-know-about-direct-consumer-lab-tests.
7. Sanders B, Blackburn TA and Boucher B. Pre-participation screening – the sports physical therapy perspective. Int J Sports Phys Ther. 2013; 8: 180-193.
8. Westerman, K et al. 2018. Longitudinal analysis of biomarker data from a personalized nutrition platform in healthy subjects. Sci Rep. 8. 14685