Biohacking Kompakt

Treatment & procedure

Genetic testing

Network

A genetic analysis reads from saliva or blood which variants of your genome you carry. In medical use this is routine, for example when a hereditary disease is suspected or before certain medicines. The lifestyle panels for nutrition, sport and disease risks promise more than the data have so far delivered.

In short

Genetic tests are strong where a single gene explains a lot: in hereditary diseases and in pharmacogenetics. In a large European study with 6,944 patients, clinically relevant adverse effects fell when medicines were selected according to a 12-gene panel. They are weak for everything that many genes and a lot of lifestyle determine together. Gene-based diet plans were no better than counselling without genes in two randomized trials. Statutory health insurance covers a genetic test only when there is a medical indication; you pay for lifestyle panels yourself.

What is behind it

Most offerings work with a genotyping chip. It checks hundreds of thousands of known single-base variants, so-called SNPs, at fixed positions in the genome. This is inexpensive and reliable for common variants. A chip, however, captures rare, disease-relevant changes poorly. That requires targeted sequencing in a clinical laboratory.

The variants give rise to three kinds of statements. First, monogenic findings: a variant in one gene explains a high risk, for example of certain hereditary forms of cancer. Second, pharmacogenetic findings: variants in enzyme genes change how quickly you break down a medicine. Third, polygenic risk scores and lifestyle profiles that bundle many small effects into one number, for example for heart risk, caffeine tolerance or the supposedly suitable type of training.

Where genetic tests really change something

Pharmacogenetics is the part that reaches everyday practice. The enzyme CYP2D6 converts tramadol and codeine, among others. According to the Association of Statutory Health Insurance Physicians of Baden-Württemberg, it is functionally absent in about 10 percent of people in Europe and works at a reduced level in about 30 percent. The painkiller then works less well than expected.

The case of chemotherapy with 5-fluorouracil and related agents is even clearer. The enzyme DPD breaks them down. Since 2020, BfArM and EMA have recommended testing for DPD deficiency before starting treatment, because those affected can develop severe, sometimes life-threatening side effects. An estimated 3 to 9 percent of the population of European descent have a partial deficiency. The genetic test for it has been covered by statutory health insurance since October 1, 2020.

What it costs and who pays

Statutory health insurance covers genetic diagnostics when there is a medical question: a suspected hereditary disease, a conspicuous family history, certain medicines. Many pharmacogenetic tests, however, are not yet in the benefits catalogue; reimbursement is then possible only in individual cases. Panels on nutrition, sport, skin ageing or general health risk are self-pay services.

How to recognize a reputable offering

In Germany, the Genetic Diagnostics Act (Gendiagnostikgesetz) regulates medical tests. A diagnostic test may only be ordered by a physician; a predictive one, that is, a prediction of future diseases, only by a specialist in human genetics or a suitably qualified physician. Genetic counselling is mandatory before and after a predictive test. An offering that delivers disease risks by post without medical information and counselling does not meet this framework.

Further points to check: Is it stated clearly which genes and variants are examined and with which method? Are abnormal findings confirmed in a clinical laboratory? Where are your data stored, and can you have them deleted? And for nutrition and sport recommendations, is it stated openly that the benefit is not proven?

What is well supported

The benefit of pharmacogenetics before certain medicines is supported. The PREPARE trial enrolled 6,944 patients in seven European countries who were receiving a new medicine with a known gene interaction. Depending on the study phase, therapy was guided by a panel of 12 genes or by the usual standard. Clinically relevant adverse effects occurred within 12 weeks in 21.5 instead of 28.6 percent, an odds ratio of 0.70.

Also well supported: personalized nutrition counselling works. But that is due to the counselling and not to the genes. In studies, the genetic information brought no measurable additional benefit.

What the studies show

PREPARE: medicines guided by a gene panel

Open-label, cluster-randomized crossover trial at hospitals, health centres and pharmacies in seven countries, 6,944 adults with a first prescription. 50 variants in 12 genes were tested, and therapy was adjusted according to the recommendations of a Dutch expert group. The primary endpoint, clinically relevant adverse effects within 12 weeks, was met: 21.5 versus 28.6 percent. Weakness: no blinding.

DIETFITS: which diet suits my genes

Randomized trial with 609 adults with overweight, 12 months of a healthy low-fat or a healthy low-carbohydrate diet. Both groups lost a similar amount of weight, 5.3 and 6.0 kilograms. A gene pattern of three variants that was supposed to predict the suitable diet had no influence on success.

Food4Me: counselling with and without genes

Internet-based randomized trial in seven European countries; 1,269 participants completed it. After 6 months, personalized counselling improved diet compared with standard advice. When blood values and 5 nutrition-relevant gene variants were added, the counselling did not become more effective.

Hollands 2016: does a genetic risk change behaviour

Meta-analysis of 18 controlled studies. People who learned their DNA-based disease risk did not smoke less, did not exercise more and did not eat significantly differently. Anxiety or depression did not increase either. Study quality was mostly low.

Where the data stop

Polygenic risk scores sound precise but discriminate poorly for the individual. According to an analysis by the epidemiologist Aroon Hingorani, a typical score for a common disease detects 11 percent of later cases if 5 percent false alarms are allowed. That is useful for population research, but of little use as a personal health check.

A second problem is the measurement quality of consumer chips for rare variants. A clinical laboratory re-checked 49 samples in which raw data from a consumer test had reported a disease-relevant variant. 40 percent of them were false positives. The sample was small and selected; the direction is nevertheless clear. For sports genetics, that is, training plans based on a gene profile, no controlled study with a clinical endpoint was found.

Status, approval and legal

Medical genetic tests fall under the German Genetic Diagnostics Act: physician reservation, for predictive tests specialist reservation and mandatory counselling. Tests on the effect of medicines count as diagnostic. Statutory health insurance pays when there is a medical indication; individual pharmacogenetic tests such as DPD testing are in the benefits catalogue, many others are not. Insurers may neither require a genetic test nor use existing results. Exception: for life, occupational disability and long-term care annuity insurance with benefits above 300,000 euros or an annual annuity above 30,000 euros, known results must be disclosed.

Safety

Physically, a genetic analysis is harmless; a saliva sample is usually enough. The risks lie elsewhere: incorrect findings, false reassurance from an unremarkable profile, unnecessary worry from an abnormal one, and permanently sensitive data that also concern your relatives. Take a pharmacogenetic result to your physician instead of changing medicines yourself, because dose and choice of active substance depend on more than one gene.

Frequently asked questions about genetic testing

Does German statutory health insurance pay for a genetic test?

Yes, if there is a medical question, such as a suspected hereditary disease or certain medicines. The DPD test before chemotherapy with fluoropyrimidines is covered by statutory health insurance. You pay for many pharmacogenetic tests and all lifestyle panels yourself.

What does a genetic test do for nutrition?

In two large randomized trials, nothing additional. Neither DIETFITS nor Food4Me found that genetic information improves the choice of diet or the effect of counselling. Personalized counselling itself worked, even without genes.

What is pharmacogenetics?

The study of gene variants that determine how your body breaks down medicines. Slow or missing enzymes can increase side effects or weaken an effect. In the PREPARE trial, clinically relevant adverse effects fell when physicians adjusted therapy according to a gene panel.

How accurate are genetic tests from the internet?

Usually reliable for common variants, not for rare disease-relevant variants. When re-checked in a clinical laboratory, 40 percent of the variants reported from consumer raw data were false positives. Abnormal findings therefore always belong in a clinical laboratory.

Do I have to tell my insurer about a genetic test result?

As a rule, no. The German Genetic Diagnostics Act prohibits insurers from requiring tests or using results. Only for very high sums in life, occupational disability and long-term care annuity insurance must results that are already known be disclosed.

Does a genetic test predict my heart attack risk?

Only very roughly. According to one analysis, a polygenic risk score detects only a small share of later cases if few false alarms are allowed. Blood pressure, blood lipids, smoking and family history say much more for the individual.

Related

Sources

Open in the database – with search, filters and comparison (German app)

Information only, not medical advice and not a usage or dosage recommendation. Prescription-only and unapproved substances belong in the hands of a physician. Last updated: 2026-09-30.