Genetic Health Testing: What It Can and Can’t Predict

Quick answer: A genetic health test reads parts of your DNA to flag inherited disease risk, carrier status for conditions you could pass to a child, and how your body may process certain drugs. It is good at confirming specific mutations like BRCA1, BRCA2, Lynch syndrome, and cystic fibrosis carrier status, but it cannot predict most common diseases on its own, and consumer kits like 23andMe screen only a handful of variants, not your whole genome. Treat the results as one input, not a verdict, and confirm anything serious with a clinical-grade test.

What a genetic health test actually measures

A genetic health test looks at specific spots in your DNA where a known change raises or lowers your risk for a condition. It does not scan everything. Most consumer kits use genotyping, which checks a fixed list of pre-selected variants (often a few hundred thousand spots out of more than three billion DNA letters). Clinical labs go deeper with sequencing, reading every letter in a targeted gene or panel, so they catch rare mutations a genotyping chip was never designed to see.

The distinction decides what your result means. Genotyping is like checking 50 specific addresses on a map to see if anyone is home. Sequencing reads the whole street. If a harmful mutation sits at an address the chip never checks, a consumer kit reports “no variants detected” and you feel reassured by a test that never looked. That single fact explains most of the false confidence people walk away with.

These tests do three jobs, and they are not interchangeable. People who lump them together end up disappointed, expecting a carrier screen to predict their heart attack or a pharmacogenomic panel to flag their cancer risk.

• Carrier screening. Tells you if you carry one copy of a recessive mutation (cystic fibrosis, sickle cell disease, Tay-Sachs). You are usually healthy yourself, but two carrier parents have a one in four chance per pregnancy of an affected child.
• Predictive and predisposition testing. Estimates your future risk for conditions like hereditary breast and ovarian cancer (BRCA1, BRCA2), Lynch syndrome, or late-onset Alzheimer’s (APOE e4). A “high risk” result is a probability, not a diagnosis.
• Pharmacogenomics. Predicts how you metabolize specific drugs, guiding dosing for some antidepressants, blood thinners like warfarin and clopidogrel, and chemotherapy agents.

Here is the insider detail most people miss: when 23andMe reports a “BRCA1/BRCA2 variant,” it tests only 3 of the more than 1,000 known harmful variants in those genes, and those 3 are most common in people of Ashkenazi Jewish descent. A normal result there does not clear you, because a clinical BRCA test sequences the full genes. Knowing this gap is the difference between false reassurance and real information.

Genotyping vs sequencing, in plain numbers

A typical consumer chip genotypes roughly 600,000 to 700,000 positions out of your genome’s 3.2 billion base pairs, so it examines well under one hundredth of one percent of your DNA, and only positions chosen in advance. Clinical panels flip the logic, reading a narrow set of medically important genes letter by letter to catch rare mutations the chip cannot represent. Neither is better in the abstract; they answer different questions. The mistake is using the wide-shallow tool for a question that needs the narrow-deep one.

Carrier screening, predictive testing, and pharmacogenomics: which one do you need?

Match the test to the question. Most people who feel let down by genetic testing bought the wrong category for what they wanted to know. The table maps the three jobs to who they serve and what a result changes.

Type	What it answers	Who it is for	What a result changes
Carrier screening	Could I pass a recessive condition to a child?	People planning or early in pregnancy	Partner testing, IVF with embryo screening, prenatal diagnosis
Predictive / predisposition	Am I at high inherited risk for a specific disease?	Strong family history of cancer or early disease	Earlier and more frequent screening, risk-reducing surgery or medication
Pharmacogenomics	How will my body handle certain drugs?	People with poor or unpredictable medication response	Drug choice and dose adjustments with a prescriber
Ancestry-style health (consumer)	General curiosity, a few flagged variants	Healthy people, no red flags	Usually nothing medical without confirmation

Notice the fourth row. The mass-market kit most people picture when they hear “genetic health test” is the one that changes the least, a mismatch between what it does and what buyers hope it does.

Carrier screening: the one with the clearest payoff

Carrier screening has the cleanest cost-to-benefit case because it leads to a concrete decision. If both you and your partner carry a cystic fibrosis mutation, that is a one in four chance per pregnancy of an affected child, and you can act on it: preimplantation genetic testing during IVF, prenatal diagnosis, or going in with eyes open. Expanded carrier panels now screen 100 to 500 conditions at once. The American College of Obstetricians and Gynecologists supports offering carrier screening to anyone considering pregnancy, which is why an OB often orders it without hesitation.

Predictive testing: BRCA and Lynch syndrome

This is where genetic testing genuinely saves lives, and where consumer kits most badly underdeliver. BRCA1 and BRCA2 mutations raise lifetime breast cancer risk to roughly 45 to 72 percent and ovarian cancer risk to about 17 to 44 percent, against a general-population breast cancer risk near 13 percent. A confirmed clinical BRCA result can justify earlier MRI screening, risk-reducing mastectomy, or removal of the ovaries and tubes, interventions that change survival odds. Lynch syndrome, caused by mismatch-repair gene mutations (MLH1, MSH2, MSH6, PMS2), sharply raises lifetime risk of colorectal and endometrial cancer and shifts colonoscopy from every ten years to every one to two, often starting in your twenties. These are the cases where a single result rewrites a screening calendar.

The catch ties back to method. A consumer kit checking 3 BRCA variants misses the overwhelming majority of harmful mutations in non-Ashkenazi families, and does not test the Lynch genes at all. If cancer runs in your family, a consumer “no variants detected” is close to meaningless. You need a doctor-ordered hereditary cancer panel.

Pharmacogenomics: useful, narrow, often misunderstood

Pharmacogenomic testing reads genes like CYP2D6, CYP2C19, and CYP2C9 that govern how fast you break drugs down. A “poor metabolizer” of CYP2C19 may get little benefit from clopidogrel after a stent. It is real and clinically used, but narrow: it does not tell you which antidepressant will work, only how your body processes a subset of them. Treat it as input for a prescriber, never a self-service reason to change a dose. The pairs below come up most often in real prescribing, each with clinical dosing guidance behind it from groups like the Clinical Pharmacogenetics Implementation Consortium.

Gene	Drugs affected	Why it matters
CYP2C19	Clopidogrel, some antidepressants, proton pump inhibitors	Poor metabolizers get weak clopidogrel effect, higher clot risk after a stent
CYP2D6	Codeine, tamoxifen, many antidepressants and antipsychotics	Ultra-rapid metabolizers turn codeine into dangerous morphine levels; poor metabolizers get little pain relief
VKORC1 and CYP2C9	Warfarin	Sets a safer starting dose and reduces bleeding risk while levels stabilize
TPMT and NUDT15	Azathioprine, mercaptopurine (chemo and immune drugs)	Low activity raises the risk of severe, life-threatening bone-marrow toxicity
DPYD	5-fluorouracil, capecitabine (chemo)	Deficiency causes severe, sometimes fatal toxicity at standard chemo doses

Notice the pattern. The strongest cases are not “which mood medication suits me” but “will a standard dose of this drug harm me,” which is why oncology and cardiology order these tests far more than primary care, and why a consumer trait report is no substitute for a clinical panel before a high-stakes prescription.

Is genetic health testing worth it for most people?

For most healthy adults with no strong family history, a consumer genetic health test is interesting more than medically useful. It rarely changes what a doctor would already tell you: eat well, move, screen on schedule. It earns its price when you have a specific reason, a relative with early breast or ovarian cancer, a known family mutation, a planned pregnancy, or recurrent reactions to medication.

The honest tradeoff: these tests can surface findings you cannot un-know, including risk markers for conditions with no prevention or cure. The clearest example is APOE e4, which raises Alzheimer’s risk but has no proven preventive treatment behind it. Some find that knowledge empowering, others find it anxiety with no action attached. Decide which camp you are in before you swab.

When testing clearly makes sense

• A first-degree relative had cancer before 50, or multiple relatives had the same cancer.
• You and a partner are planning a pregnancy and want carrier screening, especially if your ancestry carries higher rates of specific recessive conditions.
• You have had unpredictable responses to common medications, or a close relative already tested positive for a specific mutation, so you know exactly what to look for.

When to skip it, or at least wait

• You are healthy, with no notable family history, and want a number to feel proactive. Current bloodwork will teach you more.
• You would spiral over an unactionable risk result, or you are about to apply for life or long-term care insurance (more on why below).

Consumer kits vs clinical genetic testing

The biggest mistake is treating a $99 mail-in kit as equivalent to a doctor-ordered test. They differ in depth, accuracy standards, and what you can do with the result. Clinical tests are CLIA-certified and go in your medical record; most consumer results are not meant to be acted on without confirmation. The FDA authorizes consumer health reports as screening information, not diagnosis, and the fine print on every reputable kit says exactly that.

Feature	Consumer kit (e.g. 23andMe, AncestryHealth)	Clinical genetic test
Method	Genotyping (fixed variant list)	Sequencing (reads full genes)
BRCA coverage	3 variants (Ashkenazi-focused)	Full BRCA1/BRCA2 genes
Ordered by	You, online	Doctor or genetic counselor
Typical cost	$99 to $229 cash	$250 to $2,000+, often insurance-covered
Counseling included	Rarely	Usually
Goes in medical record	No	Yes
Use for medical decisions	Not without confirmation	Yes

Does Ancestry health test for the BRCA gene? AncestryHealth offered limited BRCA screening on the same narrow-variant logic and has since wound down its health product. The takeaway holds regardless of brand: a consumer result is a screen, not a diagnosis.

A worked example: the same finding, two very different paths

Picture two women, both with a mother who had breast cancer at 44. The first orders a $129 consumer kit, gets “no BRCA variants detected,” and relaxes. The kit checked 3 of more than 1,000 harmful variants, none of them the one in her non-Ashkenazi family, so her real risk is untouched and now hidden behind false comfort. The second asks for a referral. A counselor orders a full hereditary cancer panel, insurance covers it because her family history meets criteria, and it finds a pathogenic BRCA2 variant. She starts annual breast MRI at 30. Same history, same question, opposite outcomes, driven by which test was used.

How a genetic health test actually works, step by step

Knowing the workflow helps you spot where quality varies.

1. Sample collection. Consumer kits use a saliva tube or cheek swab you mail back. Clinical tests usually use a blood draw or saliva ordered through a provider.
2. DNA extraction and prep. The lab isolates DNA from the sample. Quality here affects whether a result comes back at all, which is why some people get a “sample failed, re-swab” email.
3. Analysis. Genotyping runs your DNA across a chip that reads the preset variants. Sequencing reads the targeted genes base by base, then compares your sequence to reference databases.
4. Interpretation. Variants get classified: pathogenic, likely pathogenic, variant of uncertain significance (VUS), likely benign, or benign. That middle category, VUS, confuses everyone, because it means “we found a change but do not yet know if it matters.”
5. Reporting. Consumer kits return a web dashboard in two to six weeks. Clinical labs return a formal report to your provider, usually with a counselor to walk you through it.

The step that matters most is the human one at the end. A “variant of uncertain significance” with no one to interpret it causes far more anxiety than the same finding explained by a counselor who can tell you it currently warrants no action.

Does insurance cover a BRCA test or genetic counseling?

Yes, when it is medically indicated. Under the Affordable Care Act, BRCA risk assessment and counseling are covered preventive services for women who meet family-history criteria, often with no out-of-pocket cost. If your history flags risk, your insurer frequently covers both the counseling and the clinical test. What insurance generally does not cover is a curiosity-driven consumer kit you bought yourself.

One protection worth knowing: the Genetic Information Nondiscrimination Act (GINA) bars health insurers and employers from using your genetic results against you. It does not extend to life, disability, or long-term care insurance, a real reason to settle any such applications before testing if a high-risk result is plausible. If results matter to you medically, ask your primary care doctor for a referral to a genetic counselor first. They order the right test, and insurance is far more likely to pay.

Costs swing widely. A clinical hereditary cancer panel might list at $1,500 to $2,500, but with documented family history and insurance many people pay a small fraction or nothing, and self-pay programs have pushed some full-panel cash prices to the $250 range. The expensive-looking clinical test is frequently the cheaper one once insurance applies, since the consumer kit is rarely covered.

Privacy: who else can see your DNA?

This is the question people ask last and should ask first. When you send saliva to a consumer company, you are handing over the most permanent identifier you have, one you cannot reset like a password. Read three things before you swab. First, whether the company shares or sells de-identified data to research or pharmaceutical partners, and whether you can opt out. Second, what happens to your sample if the company is sold or goes bankrupt, since a DNA database transfers with the company as an asset. Third, the law-enforcement policy, because investigators have used consumer genealogy databases to identify relatives of people who never tested.

GINA covers health-insurance and employment discrimination, but it does not govern what a private company does with data you consented to share, nor does it cover life or disability insurers. Clinical testing lives under HIPAA and inside your medical record, a tighter regime than a consumer terms-of-service agreement. If privacy is a priority, that difference alone can point you to the clinical path.

What genetic testing cannot tell you

A genetic health test cannot predict most of your future. Common conditions like type 2 diabetes, heart disease, and depression are shaped by hundreds of genes plus diet, sleep, stress, and environment. A single variant nudges the odds; it does not set them. This is where consumer reports oversell. A “slightly elevated” heart-disease risk score tells you far less than your actual blood pressure, ApoB, and A1C measured today.

Polygenic risk scores, the newer approach that sums many small-effect variants into one number, improve on single-variant reports but still explain only a slice of risk for most conventional diseases, and they perform worse in people of non-European ancestry because the reference data skews European. A reassuring polygenic score is not permission to ignore a bad lipid panel, and a worrying one is not a diagnosis.

Consider the ancestry-bias problem concretely. Two people get the same coronary artery disease polygenic score from the same report. One is of Northern European descent, the population the score was trained on, so the number carries real, if modest, weight. The other is of West African descent, sparsely represented in the training data, so the same printed score is far less reliable and may even point the wrong direction. The report looks identical for both. The validity underneath is not, which is why a clinician weights your measured biomarkers and family history above any polygenic number.

It also says nothing outside DNA. Epigenetics, infections, your microbiome, lifestyle, and plain chance all shape disease, which is why identical twins with the same DNA routinely diverge in health outcomes. The genome is the hand you were dealt. It is not the game.

That gap is why genetics and current biomarkers do different jobs. DNA shows a fixed predisposition; bloodwork shows what your body is doing right now and what you can change. If you are mapping your overall health, current labs usually move the needle more than a static DNA report, so anyone getting one test done is better served by a single full baseline draw. Here is the Superpower blood test review for how a full-body panel compares. For how DNA categories differ, see DNA Tests Explained, and if you are mixing up health screening with lineage questions, Paternity Testing covers that path.

Common mistakes people make with genetic health tests

• Reading “no variants detected” as “no risk.” It means the specific variants checked were absent, nothing more. With a 3-variant BRCA screen, that is a tiny slice of possible mutations.
• Acting on a consumer result without confirmation. Reputable clinical labs re-test a positive consumer finding first, because consumer reports carry a real false-positive rate. Never schedule surgery off a mail-in kit.
• Testing the worried-well instead of the symptomatic relative. When a family has cancer, the most informative person to test first is the affected relative. If they carry a known mutation, the rest of the family knows exactly what to check for.
• Testing right before an insurance application. GINA does not cover life, disability, or long-term care insurance. Order matters.

Edge cases: minors, the uninsured, employer programs, and Medicare

Standard guidance bends in specific situations worth knowing before you assume the default applies to you.

• Children and predictive testing. The consensus is to avoid predictive adult-onset testing in healthy minors. A child cannot consent to learning their APOE or BRCA status, and there is usually no childhood action to take, so most guidelines say wait until they can decide as an adult. Exceptions exist for conditions that present in childhood where early screening helps.
• Uninsured or self-pay. Clinical labs run patient-assistance and flat-rate self-pay programs, sometimes a few hundred dollars for a full hereditary cancer panel, which can undercut a consumer kit while delivering a far more complete result. Ask the lab directly about cash pricing before assuming clinical means unaffordable.
• Employer or wellness-program testing. Be cautious. GINA limits what employers can require, but an employer-sponsored genetic perk still raises the question of where the data lands. Voluntary does not always mean private.
• Medicare. Medicare covers genetic testing in narrower circumstances, generally when you already have a relevant personal cancer diagnosis or clear medical indication, not for healthy screening. Confirm coverage before testing.

Who should pick what: a quick decision guide

Strip away the marketing and the choice comes down to your reason for testing.

• Planning a pregnancy? Expanded carrier screening through an OB or counselor, with your partner tested if you screen positive.
• Strong cancer family history? A clinical hereditary cancer panel through a genetic counselor, not a consumer kit. This is the highest-stakes mismatch to get right.
• Recurrent medication problems? Pharmacogenomic testing arranged through your prescriber so the results actually change a prescription.
• Generally healthy and curious? A consumer kit is fine as entertainment and ancestry, but treat any health flag as a prompt to see a clinician, not a result. Your current bloodwork tells you more about your real, changeable risk today.

The throughline: a genetic health test is a starting point, not an endpoint. The people who get value from it pair it with a clinician who can act on the result.

Why a genetic counselor changes the result you get

The single biggest predictor of whether genetic testing helps you is not the lab or the panel. It is whether a genetic counselor is in the loop. A board-certified counselor does three things a swab-and-app product cannot. Before testing, they take a family history and decide which test answers your question, often ordering a narrow, cheaper, more accurate panel instead of a broad expensive one. They make sure the right family member is tested first, usually the affected relative, so a negative result in you actually means something. After testing, they translate “pathogenic BRCA2 variant” or “VUS” into a concrete plan: what to screen, when, and what to tell your relatives.

That last point, telling relatives, is the part people forget. A pathogenic finding is not just your information. Your siblings, parents, and children each have a meaningful chance of carrying the same variant, and they cannot act on what they do not know. Counselors coach you through that conversation and provide a family letter relatives can hand to their own doctors. Many consult by phone or video, so you do not need a specialist in your town. If a consumer kit ever returns a serious flag, paying out of pocket for one counseling session before you act is the highest-value step you can take.

FAQ

What genetic health risks does 23andMe test for?

23andMe’s FDA-authorized health reports cover selected risk variants for conditions including late-onset Alzheimer’s (APOE), Parkinson’s, certain hereditary thrombophilia, and a limited set of BRCA1/BRCA2 variants, plus carrier status for conditions like cystic fibrosis and sickle cell. It screens specific variants, not every cause, so a normal report does not rule risk out.

Does the 23andMe genetic health test show predisposition to addiction?

No. 23andMe does not provide a clinical addiction-risk report. Addiction involves many genes and heavy environmental influence, so no consumer DNA test reliably predicts it. Any wellness-style trait estimate is informational, not diagnostic.

What is the most accurate DNA test for genetic health screening?

For health decisions, a clinical sequencing test ordered through a doctor or counselor is the most accurate, because it reads full genes rather than a short variant list. Among consumer options, accuracy is similar across major brands since they use comparable genotyping; the real difference is how many variants they check and whether counseling is included.

What is the difference between carrier screening and predictive testing?

Carrier screening tells you whether you carry a recessive mutation you could pass to a child, even though you are healthy yourself. Predictive testing estimates your own future risk of developing a condition, such as hereditary breast cancer. One is about your children’s risk, the other about yours, and a kit built for one does not answer the other.

Should I get tested for Lynch syndrome or BRCA at home?

No, not for a real medical decision. Consumer kits do not run a proper Lynch syndrome panel and check only a few BRCA variants, missing most harmful ones in non-Ashkenazi families. If colorectal, breast, or ovarian cancer runs in your family, ask for a clinical hereditary cancer panel through a counselor.

Will a genetic test affect my insurance?

For health insurance and employment, the Genetic Information Nondiscrimination Act (GINA) prohibits using your results against you. GINA does not cover life, disability, or long-term care insurance, so a high-risk result could in theory affect those policies. If that worries you, finalize those applications before testing.

What is a variant of uncertain significance (VUS)?

A VUS is a genetic change the lab found but cannot yet classify as harmful or harmless, because the evidence is incomplete. The standard guidance is not to act on a VUS medically. A counselor notes it and watches for reclassification, since many are eventually downgraded to benign as more data arrives.

Is genetic health testing a good idea?

It is a good idea when you have a clear reason: a strong family history, pregnancy planning, or a medication question. For a healthy person with no red flags, it is optional and most useful paired with a clinician who can interpret anything concerning. Talk to a counselor before you make any medical decision based on results.