Guides

Decode Your Biomarkers

Do not read every lab flag as a diagnosis, and do not let one 'optimal range' chart set your treatment plan. First identify whether the number is a reference range, a diagnostic or risk threshold, or a personal treatment target. Then ask whether the result is persistent, whether context could distort it, and what decision would change next.

Hillary Lin, MD·Reviewed July 10, 2026·11 min read

Name the number

Reference range, diagnostic or risk threshold, and treatment target answer different questions.

Use your risk

LDL-C and ApoB goals get lower as cardiovascular risk rises. There is no single target for every adult.

Confirm before reacting

Check symptoms, medications, illness, and prior results before treating one mild abnormality as a diagnosis.

01

Start by naming the kind of number

Most lab confusion comes from treating three different numbers as if they mean the same thing. They do not.

A result can be statistically unusual without proving disease. A result can also sit inside the printed range and still be above a risk-based treatment goal. Before I interpret a value, I name the job that number is doing.

Protocol

The three numbers people confuse

Lab reference range

What it tells you
The range reported by that laboratory, often based on a comparison population and a specific assay.
What it does not tell you
It is not a guarantee of health and not an automatic treatment target.

Diagnostic or risk threshold

What it tells you
A cutoff used to classify a condition or a higher-risk state, such as the thresholds for diabetes or elevated Lp(a).
What it does not tell you
One result does not always establish a diagnosis; confirmation and clinical context may be required.

Personal treatment target

What it tells you
A goal chosen after accounting for baseline risk, known disease, treatment benefit, safety, and preference.
What it does not tell you
It is not a universal 'optimal range' that every adult needs to reach.

The same value can be normal by a lab reference range, meaningful as a risk marker, and still above a treatment target. Those are not contradictions.

02

Use a testing hierarchy

I start with a compact prevention baseline, then add tests when the result has a job. The practical question is not whether a biomarker is interesting. It is whether a high, low, or unchanged result would alter treatment, monitoring, or urgency.

If your immediate question is what to request at a visit, keep this interpretation guide open and use the separate companion guide, 'What labs should I ask my doctor for?', linked below.

Protocol

What belongs on each rung

Core prevention baseline

Typical tests
Standard lipid panel; A1C or fasting glucose; CBC; CMP; creatinine/eGFR; blood pressure, which is not a lab but changes the interpretation.
Why they are there
Classify common cardiovascular and metabolic risk and catch basic organ or medication-safety issues.

One-time or selective risk refinement

Typical tests
Lp(a) once in adulthood; ApoB when risk may be underestimated by LDL-C; uACR in people at kidney risk; hsCRP when inflammatory cardiovascular risk would change the discussion.
Why they are there
Refine a real decision rather than make the panel look comprehensive.

Story-driven testing

Typical tests
Ferritin and iron studies, B12, TSH, vitamin D, and other targeted tests when symptoms, diet, prior results, medications, or known disease make the answer actionable.
Why they are there
Investigate a plausible problem instead of screening broadly for every possible one.

Follow-up

Typical tests
The marker tied to the intervention, plus the safety labs that intervention requires.
Why they are there
Show whether the plan worked and whether it is safe on a timeline that matches the biology.

ApoB is selective and risk-based, not mandatory for every routine panel. Lp(a) is usually a once-or-rarely measurement because genetics drive most of the level.

Skip the 100-marker panel when nobody can say what each result would change.

Do not repeat stable markers monthly just because a portal can graph them.

Do not turn one mild abnormality into a diagnosis before checking whether it persists.

03

LDL-C and triglycerides: the target follows the risk

LDL-C is the clearest example of why a printed reference range is not enough. The 2026 ACC/AHA dyslipidemia guideline brought back explicit treatment goals, and the goal gets lower as cardiovascular risk rises.

I use the goal only after deciding which risk group fits. I also look at the starting value, the percent reduction, treatment tolerance, and the rest of the risk picture.

Protocol

Lipid numbers at a glance

LDL-C, borderline or intermediate primary prevention

Useful shorthand
<100 mg/dL
How I use it
A risk-based treatment goal when lipid-lowering therapy is chosen, not a diagnostic normal for everyone.

LDL-C, high risk

Useful shorthand
<70 mg/dL
How I use it
A lower goal for high-risk prevention and many people with established disease who are not in the very-high-risk tier.

LDL-C, very-high-risk ASCVD

Useful shorthand
<55 mg/dL
How I use it
A secondary-prevention goal for people at very high risk of another ASCVD event.

Triglycerides

Useful shorthand
<150 mg/dL
How I use it
The conventional normal threshold. Persistent elevation should trigger a look at alcohol, glycemia, diet, weight change, thyroid status, medications, and genetic risk.

Triglycerides

Useful shorthand
<100 mg/dL is often favorable
How I use it
A useful metabolic signal, but not a mandatory target for every adult and not a reason to chase medication in isolation.

These are decision thresholds and goals, not one universal definition of 'optimal.' Severe triglyceride elevation is a different clinical problem and deserves prompt evaluation.

04

ApoB and Lp(a): useful, but not interchangeable

ApoB estimates the number of atherogenic particles. I reach for it when LDL-C may understate particle burden, especially with higher triglycerides, diabetes, insulin resistance, very low treated LDL-C, or an otherwise discordant risk picture.

Lp(a) is a largely inherited risk marker. It is not another version of LDL-C, and a high result usually changes how seriously I manage the risk factors we can modify.

Protocol

Risk-based shorthand for ApoB and Lp(a)

ApoB, general primary prevention when measured

Level
<90 mg/dL
Clinical meaning
A reasonable shorthand goal for many borderline or intermediate-risk adults.

ApoB, high risk

Level
<70 mg/dL
Clinical meaning
A risk-based goal when a lower atherogenic-particle burden is warranted.

ApoB, very high risk

Level
Roughly <55–60 mg/dL
Clinical meaning
A context-dependent shorthand aligned with intensive prevention. ApoB targets are treatment goals, not diagnostic normal ranges.

Lp(a), lower risk

Level
<75 nmol/L or <30 mg/dL
Clinical meaning
Lower inherited Lp(a)-related risk.

Lp(a), intermediate

Level
75–124 nmol/L or 30–49 mg/dL
Clinical meaning
A gray zone that should be interpreted with the rest of the cardiovascular risk picture.

Lp(a), high

Level
≥125 nmol/L or ≥50 mg/dL
Clinical meaning
A risk-enhancing result that favors earlier, more intensive management of modifiable risk.

Use the unit your laboratory reports. Do not convert Lp(a) between nmol/L and mg/dL with a fixed mathematical factor; particle size makes that conversion unreliable.

05

Glucose: know the diagnostic thresholds

A1C, fasting glucose, and a 2-hour oral glucose tolerance test look at related but different parts of glucose physiology. A person can cross a threshold on one test and not another.

Unless hyperglycemia is unequivocal, a diabetes diagnosis requires confirmation with a repeat abnormal result or a second abnormal test. Results near a cutoff deserve follow-up, not a label delivered by portal notification.

Protocol

ADA 2026 criteria for nonpregnant adults

A1C

Below the prediabetes range
<5.7%
Prediabetes
5.7–6.4%
Diabetes threshold
≥6.5%

Fasting plasma glucose

Below the prediabetes range
70–99 mg/dL
Prediabetes
100–125 mg/dL
Diabetes threshold
≥126 mg/dL

2-hour 75-g OGTT

Below the prediabetes range
<140 mg/dL
Prediabetes
140–199 mg/dL
Diabetes threshold
≥200 mg/dL

A1C can be distorted by altered red blood cell turnover or assay interference, including iron or B12 deficiency, blood loss or transfusion, kidney disease, pregnancy, and some hemoglobin variants. When A1C and measured glucose do not agree, investigate the mismatch and use plasma glucose criteria when appropriate.

06

hsCRP: a risk signal, not a diagnosis

hsCRP can add cardiovascular risk information, but it does not tell you where inflammation comes from. I do not interpret it during an acute infection, just after an injury, or after unusually hard training.

Protocol

How to read hsCRP

<1 mg/L

Risk shorthand
Lower inflammatory cardiovascular risk
What changes next
Interpret with lipids, blood pressure, smoking, glycemia, kidney health, and known disease.

1–3 mg/L

Risk shorthand
Intermediate or average risk
What changes next
Use the full risk picture rather than treating hsCRP itself.

>3 mg/L

Risk shorthand
Higher risk if persistent
What changes next
Look for context and repeat when needed before calling it chronic low-grade inflammation.

≥2 mg/L

Risk shorthand
Risk-enhancing when persistent
What changes next
Can support a more intensive prevention discussion, especially when the lipid decision is uncertain.

>10 mg/L

Risk shorthand
Often an acute-phase signal
What changes next
Repeat after acute infection, injury, or another temporary cause has resolved.

The categories overlap: 2–3 mg/L sits in the traditional intermediate band and also meets the ≥2 mg/L risk-enhancing threshold. The result earns attention when it is measured in a stable state, persists, and changes cardiovascular management.

07

Kidney function needs both filtration and urine context

An eGFR of 60 or higher is often reassuring, but it is not enough to rule out kidney disease in someone at risk. Urine albumin can rise before filtration falls, which is why eGFR and urine albumin-to-creatinine ratio, or uACR, answer different questions.

Protocol

eGFR and uACR

eGFR

Category
≥60 mL/min/1.73 m²
How I interpret it
Often reassuring, but not sufficient by itself in people with diabetes, hypertension, cardiovascular disease, prior kidney injury, or other kidney risk.

uACR

Category
<30 mg/g
How I interpret it
The KDIGO normal-to-mildly increased category. It does not cancel an abnormal eGFR or other evidence of kidney disease.

uACR

Category
30–299 mg/g
How I interpret it
Moderately increased albuminuria. Confirm persistence and address blood pressure, glycemia, medications, and cardiovascular risk.

uACR

Category
≥300 mg/g
How I interpret it
Severely increased albuminuria and a higher-risk finding that warrants timely clinical follow-up.

Do not diagnose chronic kidney disease from one incidental result. Repeat an unexpected abnormality under stable conditions and interpret it with the urine sample, trend, medications, and clinical history.

08

Context can change the number

Some of the most over-optimized biomarkers are also the most context-dependent. This is where I slow down and ask what could move the assay, what symptoms fit, and whether treating the number improves anything that matters.

Protocol

Iron, B12, vitamin D, thyroid, and reproductive hormones

Ferritin

Practical interpretation
<30 ng/mL supports depleted iron stores in the right context. Inflammation can raise ferritin and hide deficiency, so CBC, transferrin saturation, symptoms, and the reason for iron loss matter.
What I would not do
I would not prescribe a universal ferritin target of 100 ng/mL.

Vitamin B12

Practical interpretation
<200–250 pg/mL is generally low. If B12 is 150–399 pg/mL and symptoms or risk raise concern, check methylmalonic acid, or MMA, while accounting for kidney function.
What I would not do
I would not chase a high B12 level after deficiency is corrected.

25-hydroxyvitamin D

Practical interpretation
There is no validated universal optimal target for disease prevention in healthy adults. Test for a reason, such as a bone, calcium, absorption, or treated-deficiency question.
What I would not do
I would not use routine serial testing to tune every healthy adult to the same number.

TSH

Practical interpretation
Use the laboratory and assay-specific range, roughly 0.4–4.0 mIU/L for many adults. Age, medications, and life stage can change interpretation.
What I would not do
I would not mandate a TSH of 1–2 mIU/L for everyone.

Estradiol and FSH

Practical interpretation
There is no single optimal range during the menopausal transition. Levels fluctuate and should answer a specific diagnostic question.
What I would not do
I would not use one generic hormone panel to diagnose symptoms or titrate routine treatment.

Low or borderline values should be interpreted with symptoms, diet, absorption, medications, organ function, and prior results. Treat a deficiency or disease, not a wellness leaderboard.

AgeTSH, kidney estimates, and nutrient markers can shift with age, and treatment benefit depends on baseline risk.

Medications and supplementssteroids can raise glucose; biotin can interfere with some thyroid assays; acid suppression, metformin, diuretics, and many other drugs can change what should be measured.

Kidney functionreduced filtration can raise MMA and changes how several biomarkers and medications should be interpreted.

Ancestry and hemoglobin variantssome variants can distort A1C depending on the assay. Ancestry can raise the chance of a variant, but it should prompt the right question, not an assumption.

Training and illnesshard exercise, dehydration, infection, injury, and poor recovery can temporarily move inflammatory, liver, kidney, glucose, and muscle markers.

Menstruation and pregnancy historybleeding patterns can change iron stores; prior gestational diabetes or preeclampsia changes future cardiometabolic risk; current pregnancy uses different glucose and thyroid criteria.

Perimenopause and menopausethere is no single optimal estradiol or FSH range. In adults over 45 with a typical symptom and cycle history, that history usually leads. Test when symptoms begin younger, the pattern is atypical, or a competing diagnosis needs to be ruled out.

09

What changes next

A useful interpretation ends with a decision. The next move may be no action, a repeat, a second test, a risk calculation, a treatment change, or prompt evaluation. It should not automatically be a supplement and another panel.

Protocol

From result to action

One mild result outside the reference range

Next step
Check preparation, recent illness or training, medications, and prior values; repeat if the result could be transient.
Why
A flag is not the same as a diagnosis.

A diagnostic threshold is crossed

Next step
Confirm as the relevant guideline requires and assess symptoms and competing explanations.
Why
The diagnostic label changes follow-up and treatment.

A risk threshold is crossed

Next step
Recalculate the full risk picture and discuss whether prevention intensity should change.
Why
Risk markers matter through the decisions they alter.

A personal treatment target is not reached

Next step
Check adherence, response, side effects, secondary causes, and whether the target still fits before intensifying.
Why
Targets are chosen for a reason and should be revisited with the same logic.

The result is extreme or comes with concerning symptoms

Next step
Contact a clinician promptly or seek urgent care based on the symptom and result.
Why
Routine optimization rules no longer apply.

Repeat the marker you are trying to change on a timeline that matches the intervention. Do not keep expanding the panel unless a new question appears.

Clinical lens

How I’d decide

Use this section as a second pass after the main answer, not as homework before you know what the page is saying.

Who it’s for

Adults trying to understand common prevention labs, decide which results deserve a repeat or workup, and have a more useful conversation about cardiovascular, metabolic, kidney, thyroid, or nutrient risk.

Who should skip it

Do not use a general guide to manage a severe result or new symptoms on your own. Very high glucose or triglycerides, major electrolyte abnormalities, rapidly worsening kidney function, severe anemia, chest pain, fainting, or neurologic symptoms need timely clinical care.

Measure before / after

Start with the smallest panel that answers the question: a standard lipid panel, A1C or fasting glucose, CBC, CMP, creatinine/eGFR, and blood pressure. Add Lp(a), ApoB, uACR, hsCRP, thyroid, iron, B12, vitamin D, or other tests only when risk, symptoms, medications, or prior results make them actionable.

What I’d do first

I would classify the number before trying to improve it. Confirm a possible diagnosis, calculate risk before choosing a lipid target, repeat results that may be transient, and follow the marker that is tied to the treatment decision.

What would change my mind

Persistence, symptoms, family history, known disease, medication exposure, and a result that crosses a diagnostic or treatment threshold all raise the stakes. A single mild flag without a coherent story often lowers them.

Frequently Asked Questions

Is the reference range the same as a healthy or optimal range?

No. A reference range describes how a laboratory reports results, often from a comparison population and a specific assay. Diagnostic thresholds classify disease or risk. Treatment targets are chosen from your risk and clinical context. They may overlap, but they are not interchangeable.

What should I do with one result just outside the range?

Check whether you were ill, dehydrated, fasting, training hard, or taking a medication or supplement that affects the test. Compare prior values and repeat when the result could be transient. Act faster when the abnormality is severe, fits symptoms, or crosses a validated diagnostic threshold.

Should everyone get ApoB?

Not necessarily. ApoB is most useful when it could change cardiovascular risk assessment or treatment. I prioritize it when triglycerides are elevated, diabetes or insulin resistance is present, LDL-C is already low on treatment, or LDL-C seems discordant with the rest of the risk picture.

How often should Lp(a) be checked?

Usually once in adulthood because genetics drive most of the level. Repeat testing can make sense in selected clinical situations or as Lp(a)-directed treatments evolve, but routine frequent checks rarely add useful information.

How often should I repeat prevention labs?

Repeat on the timeline of the decision. A medication change may call for weeks-to-months follow-up, a treated deficiency should be checked after enough time to respond, and stable prevention markers often need much less frequent testing. Monthly broad panels usually add noise, not control.

References & citations

  1. 1.2026 ACC/AHA Guideline on the Management of Dyslipidemia. PMID 41824590; DOI 10.1161/CIR.0000000000001423
  2. 2.American Diabetes Association. Diagnosis and Classification of Diabetes: Standards of Care in Diabetes—2026
  3. 3.National Lipid Association focused update on lipoprotein(a), 2024. PMID 38565461
  4. 4.National Lipid Association Expert Clinical Consensus on ApoB, 2024. PMID 39256087
  5. 5.Inflammation and Cardiovascular Disease: 2025 ACC Scientific Statement. PMID 41020749; DOI 10.1016/j.jacc.2025.08.047
  6. 6.KDIGO 2024 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease
  7. 7.Recommendations for diagnosis, treatment, and prevention of iron deficiency and iron deficiency anemia. HemaSphere, 2024
  8. 8.NIH Office of Dietary Supplements: Vitamin B12 Fact Sheet for Health Professionals
  9. 9.Endocrine Society Vitamin D for the Prevention of Disease guideline, 2024. PMID 38828931
  10. 10.American Thyroid Association-commissioned review of TSH and thyroid hormone testing. PMID 37655789
  11. 11.ACOG 2025 guidance on hormone testing during the menopausal transition

Related Guides

Next step

Turn the guide into the right next decision.

If this page raised a real clinical question, start with the practice details. If you are still learning, get the weekly letter. If you are comparing tests, use the testing hub before buying another panel.