The American annual physical is a 20th-century artifact. It was designed when primary care had to triage tuberculosis, untreated hypertension, and undiagnosed type 2 diabetes from a single 30-minute visit. The standard lab orders, complete blood count, basic metabolic panel, lipid panel, reflect that triage logic. They are excellent at catching crisis. They are mediocre at catching the slow drift that defines aging.
If you are reading this site, you probably want a dashboard, not a smoke alarm. The 12 markers below are the cheapest, highest-yield additions to that dashboard. Each has at least one large prospective cohort or RCT behind it. None of them require a specialist or a concierge subscription.
What does a standard annual physical actually measure?
A typical primary-care annual visit in the US orders these labs:
- Complete Blood Count (CBC): WBC, RBC, hemoglobin, hematocrit, platelets, MCV, MCH, MCHC, RDW, plus a 5-part differential (neutrophils, lymphocytes, monocytes, eosinophils, basophils). About 14 reported values.
- Basic Metabolic Panel (BMP): glucose, BUN, creatinine, sodium, potassium, chloride, CO2, calcium. 8 values. Sometimes upgraded to Comprehensive Metabolic Panel which adds liver enzymes (ALT, AST, ALP) and albumin.
- Lipid panel: total cholesterol, LDL-C (calculated), HDL-C, triglycerides. 4 values, one of which (LDL-C) is an estimate from the others.
- HbA1c: ordered if you have risk factors for diabetes, or sometimes opportunistically.
- TSH: ordered if you have symptoms or are over 50.
That is roughly 30 distinct markers. Sounds like a lot. Now look at what is missing:
- No measure of atherogenic particle count (apoB) or genetic lipoprotein risk (Lp(a)).
- No measure of inflammation worth acting on (hs-CRP, not generic CRP).
- No measure of insulin resistance before glucose moves (fasting insulin, HOMA-IR).
- No measure of hormonal status beyond TSH (testosterone, estradiol, DHEA-S, SHBG).
- No measure of cardiovascular omega-3 status (omega-3 index).
- No measure of methylation/cardiovascular risk via homocysteine.
- No iron-storage signal (ferritin), even though iron overload is more common in men over 50 than iron deficiency.
The result: a panel that flags acute pathology and misses most of the trajectory information. The good news is the gap is mostly a markup problem, not a clinical-knowledge problem. Every marker below is in the catalogs of Quest, Labcorp, and every direct-to-consumer lab service. They just are not in the default order set.
Which 12 biomarkers should you add to your annual labs?
1. High-sensitivity C-reactive protein (hs-CRP)
What it measures: Low-grade systemic inflammation. The "high-sensitivity" assay detects values down to 0.1 mg/L, which is necessary because cardiovascular risk stratifies in the 0 to 3 mg/L band. Standard CRP is calibrated for acute infection and is useless for longevity work.
The gap: Standard panels do not include hs-CRP. Your doctor likely runs CRP only when they suspect active infection.
Optimal target: below 1.0 mg/L. The JUPITER trial (n=17,802) randomized adults with normal LDL but hs-CRP above 2.0 mg/L to rosuvastatin or placebo and cut MACE by 44 percent ( Ridker et al. (JUPITER) 2008, n=17802 ). The Emerging Risk Factors Collaboration (n=160,309 across 54 cohorts) found per-SD log hs-CRP increment associated with roughly 37 percent higher coronary heart disease risk after adjustment ( Emerging Risk Factors Collaboration 2010, n=160309 ).
Why it matters: hs-CRP is not specific to any one cause. Periodontal disease, visceral adiposity, undertrained aerobic fitness, broken sleep, and chronic infection all push it up. That non-specificity is a feature: a single value above 3.0 mg/L is a flag to investigate the lifestyle inputs before reflexively reaching for a statin.
2. Lipoprotein(a)
What it measures: Lp(a) is an LDL-like particle with an apolipoprotein(a) tail bolted on. The tail is genetically determined; concentration is largely fixed for life by your variant of the LPA gene.
The gap: Most primary-care doctors will not order it without a family history of premature cardiovascular events. About 20 percent of the population carries a clinically elevated Lp(a) above 50 mg/dL.
Optimal target: below 30 mg/dL. The Copenhagen Mendelian-randomization analysis (n=40,000) showed genetically elevated Lp(a) causally raises myocardial infarction risk independent of LDL-C ( Kamstrup, Tybjaerg-Hansen, Steffensen & Nordestgaard 2009, n=40000 ). Pelacarsen, an antisense Lp(a)-lowering agent, is being tested in the Lp(a)-HORIZON outcomes trial as the first targeted therapy ( Tsimikas et al. (Lp(a)-HORIZON) 2025 ).
Why it matters: You only need to test once in your life. If yours is high, you do not change Lp(a); you compensate by driving apoB and blood pressure lower than population guidance to offset the genetic load.
3. Apolipoprotein B (apoB)
What it measures: The actual count of atherogenic particles in your bloodstream. Each LDL, IDL, VLDL, and Lp(a) particle carries one apoB molecule. apoB is what physically deposits in arterial walls; LDL-C is the cholesterol cargo inside those particles.
The gap: LDL-C is a calculation, not a measurement, in standard panels. In people with metabolic syndrome or insulin resistance, LDL-C systematically underestimates particle count.
Optimal target: below 80 mg/dL for healthy adults. Below 60 mg/dL for primary prevention if family history is heavy or if your CAC score is non-zero. The Cholesterol Treatment Trialists meta-analysis (n=186,854) showed each 1 mmol/L LDL-C reduction cut major vascular events by roughly 22 percent, with linear dose-response down to the lowest tested values ( Cholesterol Treatment Trialists Collaboration 2019, n=186854 ).
Why it matters: apoB resolves the disagreement between LDL-C and the actual cardiovascular risk in metabolically unhealthy people. If LDL-C looks fine but apoB is high, the lipid panel is lying to you.
4. Fasting insulin
What it measures: Your pancreas's compensation for early insulin resistance. Insulin starts climbing 5 to 10 years before fasting glucose moves out of normal range.
The gap: Standard panels measure glucose only. By the time fasting glucose is 100 mg/dL, the insulin signal has been deranged for years.
Optimal target: below 8 uIU/mL. HOMA-IR, calculated from fasting insulin and glucose, should be below 1.5. The American Diabetes Association sets fasting glucose 100 to 125 as prediabetes and 126 or above as diabetes ( American Diabetes Association 2023 ), but those thresholds catch the disease late. A 22-year follow-up of nondiabetic men found fasting glucose above 85 mg/dL was already associated with elevated cardiovascular mortality ( Bjornholt et al. 1999, n=1933 ).
Why it matters: Fasting insulin is the earliest practical signal that you are sliding toward metabolic dysfunction. It is also the single number that responds fastest to fixing diet, sleep, and movement, often within 8 to 12 weeks.
5. Free and total testosterone (with SHBG)
What it measures: Total testosterone is everything bound and unbound. Free testosterone is the fraction biologically active. SHBG (sex hormone-binding globulin) controls the ratio between the two and itself shifts with insulin resistance, alcohol intake, and thyroid status.
The gap: Total testosterone alone, the default order, can mislead. A man with high SHBG can have normal total testosterone and clinically low free testosterone. A man with insulin resistance can have low SHBG and look "normal" on free T while carrying metabolic dysfunction.
Optimal target: Total testosterone in the upper third of the age-adjusted reference range. Free testosterone above 9 ng/dL. SHBG between 20 and 60 nmol/L. These are not regulatory thresholds; they are longevity-cohort soft targets.
Why it matters: Low testosterone in midlife associates with cardiometabolic mortality, but the causal direction is contested. The pragmatic frame: test it because it is cheap and it changes with lifestyle interventions, not because every low value justifies replacement therapy.
6. Estradiol
What it measures: The dominant estrogen in premenopausal women. In men, estradiol is converted from testosterone via aromatase and matters for bone, brain, and cardiovascular health independently of testosterone.
The gap: Standard panels do not include estradiol for either sex outside fertility workups. Use the LC-MS/MS sensitive assay; the standard immunoassay is unreliable below 50 pg/mL.
Optimal target for men: 20 to 40 pg/mL. Men with very low estradiol have higher fracture risk and worse cognition than men with mid-normal values. Optimal target for women: depends on cycle phase and life stage; this is a marker for trend tracking, not single-value judgment.
Why it matters: Aromatase inhibitor abuse in men chasing "low estrogen" is a meaningful clinical harm that biologicalx readers commit at higher rates than the general population. Test before you blame estrogen for anything.
7. DHEA-sulfate (DHEA-S)
What it measures: The adrenal androgen reservoir. DHEA-S declines roughly 2 to 3 percent per year from age 30, reaching 10 to 20 percent of peak by age 70.
The gap: Not part of any standard panel. Often dismissed as "not actionable." It is at minimum a longevity trajectory marker.
Optimal target: age- and sex-adjusted upper third. Below the 25th percentile is associated with worse functional outcomes in older cohorts in observational data. Causality is unclear; replacement trials have been mostly null on hard outcomes.
Why it matters: DHEA-S is a useful adjunct to testosterone interpretation. A man with low total testosterone and adequate DHEA-S has different physiology, and different intervention options, than a man with both low.
8. Serum albumin
What it measures: Liver synthetic capacity, nutritional status, and inflammation simultaneously. It is one of the 9 components of Levine's PhenoAge clock ( Levine et al. 2018 ), the reference clock that powers our biological age calculator.
The gap: Albumin is included in the comprehensive metabolic panel (CMP) but rarely in the basic metabolic panel (BMP) that most physicals default to. Confirm yours is the comprehensive version.
Optimal target: above 4.4 g/dL. The Cabrerizo 2015 meta-analysis linked low serum albumin (under 35 to 38 g/L) to elevated all-cause mortality and frailty in older adults ( Cabrerizo et al. 2015 ). Goldwasser & Feldman 1997 found each 2.5 g/L decrement in serum albumin associated with 24 to 56 percent higher mortality risk across adult populations ( Goldwasser & Feldman 1997 ).
Why it matters: Albumin is downstream of nearly every chronic disease process. A trending decline over 5 years is a more reliable aging signal than any single value.
9. Red cell distribution width (RDW)
What it measures: Variation in size across your red blood cell population. Normally tight; widens under inflammation, oxidative stress, iron handling problems, and chronic disease.
The gap: RDW is in every CBC. Almost no primary-care doctor mentions it unless it triggers an explicit "high" flag.
Optimal target: below 13 percent. Each 1 percent rise in RDW associates with roughly 14 percent higher all-cause mortality in older adults across multiple cohorts ( Patel et al. 2010 ). In heart failure, RDW was the second-strongest mortality predictor after age in the CHARM cohort (n=2,679) ( Felker et al. 2007, n=2679 ).
Why it matters: RDW is free. You already paid for it. Read it.
10. Lymphocyte percent
What it measures: The fraction of your white blood cells that are lymphocytes (B cells, T cells, NK cells). Drops with age and chronic stress. Another PhenoAge component.
The gap: Reported in the CBC differential but rarely flagged unless extreme.
Optimal target: above 30 percent. Absolute lymphocyte count under 1.1 x 10^9/L doubled all-cause mortality and tripled infection mortality in a Danish cohort of 98,344 ( Warny et al. 2018, n=98344 ). Zidar 2019 (NHANES, n=31,178) found lymphopenia under 1.5 x 10^9/L associated with 1.6x all-cause mortality over 12 years ( Zidar et al. 2019, n=31178 ).
Why it matters: Lymphocyte percent integrates immune aging in a way that no single subset count does.
11. Homocysteine
What it measures: A sulfur-containing amino acid that accumulates when methylation pathways are inefficient (low B12, low folate, low B6, certain MTHFR variants).
The gap: Not standard. Sometimes ordered when family history is heavy or when MTHFR genotype is known.
Optimal target: below 9 umol/L. Above 15 is a clinical flag. Homocysteine is associated with vascular disease, cognitive decline, and bone fracture risk in cohort data, though intervention RCTs lowering homocysteine have shown smaller-than-expected effects on hard outcomes.
Why it matters: Cheap, fixable, often vitamin-mediated. If homocysteine is elevated, B12 and folate status are the first stops, not statins or aspirin.
12. Omega-3 index
What it measures: EPA + DHA as a percentage of total red-blood-cell membrane fatty acids. The blood-cell measurement is more stable than serum and reflects 3 to 4 months of intake, similar to HbA1c for glucose.
The gap: Not standard. Sold by Quest as OmegaCheck, by Labcorp as the Omega-3 Index, and by direct-to-consumer labs under various brand names.
Optimal target: 8 to 12 percent. Most Western adults sit between 4 and 6 percent. The REDUCE-IT trial (n=8,179) showed icosapent ethyl, a high-dose EPA-only formulation, cut major adverse cardiovascular events by 25 percent in patients with elevated triglycerides on statin therapy ( Bhatt et al. (REDUCE-IT) 2019, n=8179 ). The mixed-EPA-DHA STRENGTH trial was null, suggesting the omega-3 effect is dose- and isomer-specific.
Why it matters: Unlike most markers, this one tells you what to do: eat more fatty fish or supplement EPA/DHA until the index moves into the 8 to 12 percent band.
Where can you order these biomarker tests yourself?
You have four real options.
Option A: ask your doctor. Some primary care physicians will sign a requisition for any of the 12 above without much pushback, especially if you frame the request as "I would like to add hs-CRP, Lp(a), and apoB to my next blood draw." Insurance often covers most of the panel under "preventive care" if your physician documents intent. The variance is enormous: some doctors will sign anything; some refuse anything off the standard pathway.
Option B: Quest Health or Labcorp OnDemand (DTC). Both major US reference labs sell direct-to-consumer panels. Comprehensive add-on bundles run $150 to $300 cash. You order online, walk into the same lab your doctor would have sent you to, and get results in your patient portal. No physician interpretation included.
Option C: aggregator services (Function Health, InsideTracker, Marek). $499 to $1,500 per year. They bundle 50 to 100 markers with a software dashboard and, in some cases, a clinician review. Function Health runs around $499 per year for roughly 100 markers including all 12 above. Useful if you value the dashboard and the clinician review more than the cash savings of going direct.
Option D: research-grade panels. Companies like SiPhox and Trubeswell sell smaller panels via at-home finger-stick kits. Convenient. Less reliable than venous draw for some markers (Lp(a) and hs-CRP in particular run noisier on finger-stick than on venous serum).
The honest comparison: Option A is cheapest if your doctor will play. Option B is the best value for self-directed people who do not need hand-holding. Option C is for people who would rather pay a markup than read a marker article. Option D is for people who refuse to schedule a phlebotomy appointment.
What to do with the numbers
A blood marker that you do not act on is a vanity number. The 12-marker panel above is most useful when you feed it into one of three workflows:
- PhenoAge calculation. Albumin, RDW, lymphocyte percent, plus 6 other commonly-available markers compute a single biological-age estimate using the Levine 2018 algorithm ( Levine et al. 2018 ). Run yours at our biological age calculator.
- Bloodwork tracker. Your panel becomes more useful with each annual repeat. Trend matters more than any single value. Track yours at /tools/bloodwork-tracker/.
- Per-marker deep dives. We have detailed articles on each of the most impactful markers: albumin, creatinine, RDW, MCV, lymphocyte percent, hs-CRP, fasting glucose, white blood cell count, and alkaline phosphatase.
Two more workflow notes. First, repeat the panel annually if anything is out of the optimal band, every 2 to 3 years if everything is dialed in and your lifestyle has not changed. Second, for the markers whose optimal range is age- or sex-stratified (testosterone, DHEA-S, estradiol), do not anchor on the laboratory's reference range. Lab reference ranges are statistical descriptions of "the population that gets tested," which skews toward the unwell. Use the longevity-cohort optimal ranges referenced above, not the lab's "normal" arrows.
Counter-view: when not to bother
The case against this panel is real and worth hearing.
You are 25, healthy, no family history. The marginal information yield is genuinely low. Lp(a) once is sufficient because it is genetic. Skip the rest until 35 unless you have an active reason. Testing too often in low-risk people generates incidental findings that lead to follow-up tests, biopsies, and anxiety, with no improvement in outcomes.
You are 80, have multiple active conditions, and your clinical priorities are functional. A 12-marker longevity panel adds noise to a treatment plan that is already balancing tradeoffs. In this population, your geriatrician's prioritization is more useful than your apoB.
You are pregnant or recently postpartum. Many of these markers (hs-CRP, ferritin, SHBG, free testosterone) shift dramatically in ways that do not reflect underlying health. Defer the longevity panel to 6 months postpartum.
You will not act on the numbers. This is the most common version of "do not bother." If you are going to test apoB, see 110 mg/dL, and do nothing, you have learned nothing useful. Lab numbers are not magic. They become information only when paired with a behavior change you would not otherwise have made.
How to talk to your doctor
Most resistance you will get is procedural, not clinical. Doctors order from a default order set; deviations require typing. Make typing easy.
A working script:
"I would like to add a few markers to my next blood draw beyond the standard panel. Specifically: hs-CRP, lipoprotein(a), apolipoprotein B, fasting insulin, and the omega-3 index. They are all available at Quest. I have been reading the longevity literature and these are the markers that show up consistently in mortality cohorts. I am happy to pay out of pocket for any my insurance does not cover."
Most physicians will say yes to this. Some will push back on Lp(a) ("we already have your LDL"). The counter is that Lp(a) is genetically determined, not captured by LDL, and you only need to test it once in your life. Most physicians will then sign.
If your physician refuses without a clinical reason, you have learned something useful about the practice and can route through Option B (Quest direct-to-consumer) without explanation. Cost is $150 to $300. No physician signature required.
