Recommended Blood Tests by Age, and How to Read Your Own Results

The short answer: a small core panel repeated on a fixed cadence carries most of the information. What changes with age is the repeat interval and the interpretation. The test list is not what changes. In your twenties and thirties, draw a complete blood count with differential (CBC), a comprehensive metabolic panel (CMP), and a lipid panel with apolipoprotein B (apoB). Add a one-time lipoprotein(a), HbA1c, thyroid-stimulating hormone (TSH), ferritin with transferrin saturation, and 25-hydroxyvitamin D. Repeat the panel roughly every two to three years. From 40 to 59, move to annual or biennial and add measures of kidney and cardiometabolic strain: cystatin C-based estimated glomerular filtration rate (eGFR), urine albumin-to-creatinine ratio (UACR), fasting insulin, and high-sensitivity C-reactive protein (hs-CRP). From 60 on, keep the same core annually and add vitamin B12 and folate, with a clinician interpreting anything that moves. The reason is that the prior probability of a real finding rises, and the reference intervals you were compared against no longer describe your age group well. Everything below is measurement and interpretation. Deciding what to do about a result is a conversation with a physician.

What belongs in the core panel at any age

The core panel is small because most routine analytes are either highly stable within a person (so one good measurement lasts years) or so noisy that a single value tells you little. The core is CBC, CMP, lipids with apoB, HbA1c and TSH. Ferritin with transferrin saturation and 25-hydroxyvitamin D complete it. Together they cover hematopoiesis, renal and hepatic function, electrolytes and atherogenic particle burden. They also cover glycemia, thyroid axis, and the two deficiencies most likely to be silent and correctable.

Two analytes deserve special handling. Lipoprotein(a) concentration is largely genetically determined and stable through adult life, so a single accurate measurement in your twenties is usually sufficient, and it is the one lipid number most people have never seen. ApoB is preferable to LDL cholesterol as a repeated measure because it counts atherogenic particles directly rather than estimating their cholesterol content, and it is measured by immunoassay rather than by the Friedewald calculation, which breaks down at high triglycerides.

Frequency should follow the ratio of within-person biological variation to the size of change you care about. HbA1c has low intra-individual variation, on the order of a couple of percent, which is why a 0.3-point rise over two draws is worth attention. The opposite holds for hs-CRP, where within-person variation is large and commonly cited in the tens of percent. A single elevated value is therefore close to uninformative, and only a median of three draws weeks apart is interpretable. Ferritin behaves similarly during any inflammatory episode, which is why pairing it with transferrin saturation and hs-CRP prevents the common misreading of inflammation-driven ferritin as adequate iron stores.

Ages 20 to 39: build the baseline while nothing is wrong

The value of testing in this decade is almost entirely in establishing your personal reference interval. Population reference intervals are constructed as the central 95% of a reference population, which means roughly one in twenty healthy people falls outside any given interval and a panel of 20 analytes will flag something in most people by chance. Your own prior values are a far tighter comparator: if your creatinine has sat at 0.85 mg/dL across four draws, a value of 1.10 mg/dL is informative even though both are inside the population interval.

Concretely, we would draw the core panel at 25, 28, and 32, fasting, same laboratory, same time of morning, and keep the numeric results with units and the laboratory’s own reference intervals. Add lipoprotein(a) once. If there is a family history of premature cardiovascular disease, hemochromatosis, or thyroid disease, the case for starting earlier and repeating more often is straightforward, and family history is worth documenting formally before the relatives who remember it are gone.

Ages 40 to 59: where the signal changes

This is the decade where slow trajectories become visible and where kidney function measurement starts to matter. Creatinine-based eGFR is a function of muscle mass as much as filtration, so a lean, high-muscle, or high-protein-intake person can be systematically misclassified. Cystatin C-based eGFR is not muscle-dependent, and the combined creatinine-cystatin C equation is the one we would use when the two disagree. UACR on a first-morning urine adds information that no serum analyte provides, and it is inexpensive.

Add fasting insulin alongside fasting glucose, because the two together describe compensation rather than just the current glucose set point, and HbA1c alone will look unremarkable for years while insulin rises. Add hs-CRP as a median of three measurements rather than a single draw. Gamma-glutamyl transferase and alanine aminotransferase are already in most CMPs and are the cheapest available window on hepatic stress. For men, prostate-specific antigen enters the discussion in this decade as a shared decision with a clinician, not as an automatic addition, because the downstream biopsy and imaging cascade has real costs.

Ages 60 and over: interpretation does not get easier

After about 60, the test list barely changes but the reading of it does. Many analytes drift with age in ways that population reference intervals derived from younger adults do not capture, and comorbidity and polypharmacy make abnormal values common without being informative. The geriatric literature has described this problem for decades. Age-associated shifts in albumin, hemoglobin, and renal indices mean that applying adult reference intervals to an 80-year-old produces both false alarms and false reassurance.¹

Two additions are worth making. Vitamin B12 and folate become higher-yield with age, partly because absorption declines. And albumin, hemoglobin, and total lymphocyte count take on a second meaning as indicators of nutritional risk: a meta-analysis of blood biomarkers in older adults found albumin and hemoglobin to be consistently associated with malnutrition risk, though with limited discriminative power on their own.² Read them as one input among several. They are not a test result that settles anything.

Frequency: annual for the core panel is reasonable for most people over 60, with tighter intervals only when a clinician is following something specific. More frequent testing without a question in mind mostly generates incidental findings.

Blood-based Alzheimer’s biomarkers: what the guidelines say

Plasma phosphorylated tau (p-tau217) and amyloid-beta 42/40 ratio assays are the most consequential new blood biomarkers of the last decade, and they are also the ones most often misused in consumer testing. The Alzheimer’s Association clinical practice guideline restricts their use to the diagnostic workup of patients with cognitive impairment in specialized care settings, with defined accuracy thresholds, and does not endorse testing in people without symptoms.³ Commentary in the BMJ has gone further, arguing that the evidence base for population-level or asymptomatic use remains thin.⁴ Work from the EU/US CTAD Task Force on translating these assays into practice makes the same point about the gap between research performance and clinical deployment.⁵

If you are tested anyway, pre-analytical handling dominates the result. Published guidance for Alzheimer blood biomarkers specifies plasma collection in EDTA tubes, low-binding polypropylene aliquots, minimal freeze-thaw cycles, and tight control of time-to-centrifugation, because deviations shift measured concentrations enough to change interpretation.⁶ A p-tau217 value from a sample that sat at room temperature for six hours in a serum separator tube is not a measurement of anything. The same principle applies less dramatically to routine analytes: potassium rises with hemolysis and delayed separation, and glucose falls in unseparated tubes without glycolysis inhibitor.

Keeping your own record

Ask the laboratory for results with numeric values, units, and the laboratory’s reference interval for each analyte, and keep the LOINC code where available so that results from different laboratories align. Most portals expose a FHIR API or at minimum a downloadable CDA or CSV export. We would normalize into a long-format table with columns for date, LOINC, analyte name, value and unit. Further columns hold reference low, reference high, laboratory, and fasting status. We would then fit a simple per-analyte linear trend and flag any point more than two within-person standard deviations from your own trend. Population reference intervals become a secondary check. Large-scale cohorts that collect venous blood biomarkers document exactly this discipline around collection protocol, assay platform, and laboratory identity, because switching platforms mid-study introduces step changes that look like biology.⁷

The single most common failure we see is comparing values across laboratories that use different assay platforms, most often for ferritin, TSH, testosterone, and vitamin D, where between-method differences can exceed the biological change you are trying to detect. Pick one laboratory and stay with it.

Questions people also ask

How often should seniors get bloodwork done? For most people over 60 without an active problem being followed, the core panel annually is a reasonable default, and more frequent testing should be driven by a specific clinical question rather than a schedule. A clinician following a known condition will set their own interval.

Which tests should be questioned after 70? The ones whose downstream cascade is likely to cost more than the finding is worth.

• prostate-specific antigen screening in men with limited life expectancy
• routine tumor markers such as CA-125 in asymptomatic people
• broad panels ordered annually without a question
• plasma amyloid or p-tau testing in the absence of cognitive symptoms
• repeat vitamin D testing after a normal result with no change in circumstances

Each of these is a shared decision with a physician. The list is not a set of rules.

What blood tests should a 70-year-old have? The same core panel described above, with vitamin B12 and folate added. Cystatin C-based eGFR is preferred when muscle mass is low, and results are read against that person’s own history rather than a young-adult reference interval.¹

How often will Medicare pay for routine blood work? Medicare covers diagnostic laboratory tests that are medically necessary and ordered by a clinician, plus specific preventive services such as lipid screening as part of the cardiovascular disease risk reduction visit, generally at defined intervals. Coverage for panels ordered without an indication is limited, and the current details are worth confirming with the provider’s billing office before the draw.

Does a normal panel mean nothing is wrong? No. A normal panel means the measured analytes sat within a population interval on that day. Its main value is as the next point in your own series.

Footnotes

1. Kalman Kafetz. Blood tests in elderly people and their interpretation. Reviews in Clinical Gerontology, 1998. https://doi.org/10.1017/s0959259898008442 ↩ ↩²

2. Zhiying Zhang, Suzette Pereira, Menghua Luo, et al. Evaluation of Blood Biomarkers Associated with Risk of Malnutrition in Older Adults: A Systematic Review and Meta-Analysis. Nutrients, 2017. https://doi.org/10.3390/nu9080829 ↩

3. Sebastian Palmqvist, Heather E. Whitson, Laura A. Allen, et al. Alzheimer’s Association Clinical Practice Guideline on the use of blood‐based biomarkers in the diagnostic workup of suspected Alzheimer’s disease within specialized care settings. Alzheimer’s & Dementia, 2025. https://doi.org/10.1002/alz.70535 ↩

4. Willem A Van Gool, J A Floria Siebrand, Carol Brayne, et al. Evidence gap in blood biomarkers for Alzheimer’s disease. BMJ, 2025. https://doi.org/10.1136/bmj-2025-084781 ↩

5. Davide Angioni, J. Delrieu, O. Hansson, et al. Blood Biomarkers from Research Use to Clinical Practice: What Must Be Done? A Report from the EU/US CTAD Task Force. The Journal of Prevention of Alzheimer’s Disease, 2022. https://doi.org/10.14283/jpad.2022.85 ↩

6. Xuemei Zeng, Yijun Chen, Anuradha Sehrawat, et al. Alzheimer blood biomarkers: practical guidelines for study design, sample collection, processing, biobanking, measurement and result reporting. Molecular Neurodegeneration, 2024. https://doi.org/10.1186/s13024-024-00711-1 ↩

7. Xinxin Chen, Eileen Crimmins, Peifeng (Perry) Hu, et al. Venous Blood-Based Biomarkers in the China Health and Retirement Longitudinal Study: Rationale, Design, and Results From the 2015 Wave. American Journal of Epidemiology, 2019. https://doi.org/10.1093/aje/kwz170 ↩