Health optimisation: evidence vs marketing in the UK

Health optimisation is worth doing when it targets risks that can be measured reliably and changed in ways that improve outcomes. The marketing version often sells more tests, scans, supplements and dashboards than the evidence can support. The useful question is simple: will this result change a decision that is likely to make me healthier, or will it only create a report?

The evidence test

Health optimisation has a good version and a weak version. The good version identifies your biggest modifiable risks, measures them accurately enough, and changes something that matters. The weak version gives you numbers, colour-coded dashboards and "biological age" scores without proving that acting on them improves health.

The difference is not whether a test is private or NHS. A private test can be useful if it answers a real question. An NHS test can be low value if it is repeated without a reason. The difference is clinical utility: does the result change the next decision, and is that decision supported by evidence?

NICE's shared decision making guideline is useful here because it emphasises risks, benefits, consequences and people's values.⁵ A good optimisation plan should be able to answer the same questions: what are we looking for, what are the false positive and false negative risks, what happens after an abnormal result, and what would we do differently if the result is normal?

Screening is not the same as diagnosis

Screening means testing apparently healthy people to find increased risk before symptoms. Diagnosis means investigating a symptom, sign or known risk. Confusing the two is where much health optimisation marketing becomes slippery. A scan for unexplained weight loss is not the same as a scan sold to a healthy person because "early detection saves lives".

The UK National Screening Committee says screening identifies apparently healthy people who may have an increased chance of a condition and can then be offered further information, tests or treatment.² That sounds simple, but population screening is only justified when the whole pathway works: a suitable test, a meaningful target condition, an accepted treatment, quality assurance, fair access and more benefit than harm.

Government guidance on private screening warns that private providers may offer screening not recommended by the NHS, and that people should consider whether there is clear evidence of benefit, possible harms, and what happens after a positive result.³ The harms are not theoretical. False positives can trigger anxiety, repeat imaging, biopsies and incidental findings. False negatives can falsely reassure. Overdiagnosis can label slow or harmless changes as disease.

The Cochrane health-check review is a useful reality check. It included 17 randomised trials, with outcome data from 15 trials and more than 250,000 participants. General health checks found more conditions, but had little or no effect on total mortality, cancer mortality or fatal and non-fatal heart disease.¹ That does not mean all checks are useless. It means broad, untargeted "check everything" packages should not be assumed to improve outcomes.

The NHS Health Check is narrower. It is offered in England to adults aged 40 to 74 who do not already have certain conditions, and it looks for risk of heart disease, stroke, kidney disease, diabetes and some dementia risk factors.⁴ The logic is targeted risk reduction, not total certainty.

Where marketing gets ahead of evidence

Full-body MRI is a good example. MRI is powerful when there is a clinical question. Whole-body MRI for a healthy, average-risk adult is different. A systematic review of whole-body MRI for preventive health screening found that critical and indeterminate incidental findings were common and concluded that, based on current evidence, providers should not offer it to asymptomatic people outside research settings.⁹ This may change for specific high-risk groups or future protocols, but that is not the same as a general consumer promise.

Large private blood panels have a similar problem. A one-off panel can reveal iron deficiency, diabetes risk, kidney disease or thyroid disease if chosen for the right person. But running dozens of markers because "more data" feels scientific creates predictable noise. The more low-pretest-probability tests you run, the more borderline abnormalities you find. Some are useful. Some are random biological variation. Some trigger months of repeat testing without changing health.

Direct-to-consumer genomic tests can also be over-sold. NHS Genomics Education describes direct-to-consumer genomic testing as testing arranged directly by an individual without being ordered by a clinician or genetic counsellor, and notes that this can bypass NHS processes for clinical appropriateness and informed consent.¹⁰ A negative consumer result should not reassure someone with a strong family history, and a positive result often needs clinical confirmation.

Supplements sit in an awkward middle ground. Some are useful when there is a deficiency, pregnancy need, restricted diet, proven indication or medication interaction. But "supports immunity", "balances hormones" or "detoxes the liver" often means the claim has been shaped to sound medical while avoiding a disease claim. GOV.UK says food businesses cannot claim or imply that food can treat, prevent or cure a disease or medical condition.⁷ CAP rules also require health and nutrition claims in ads for food and supplements to meet the conditions of the relevant register.⁶

Medicines are regulated differently from foods. MHRA guidance on advertising medicines sits in a separate regime, because medicinal claims need appropriate authorisation and rules.⁸ If a supplement advert sounds like it is treating anxiety, arthritis, insulin resistance, menopause, ADHD, dementia or cancer, treat that as a red flag rather than a sign of bold science.

What is actually worth optimising

The strongest optimisation targets are boring because they sit close to disease mechanisms and have clear action pathways. Blood pressure is useful because treatment lowers stroke and heart risk. Lipids and cardiovascular risk are useful because they guide lifestyle and medication decisions. HbA1c or fasting glucose is useful when it identifies diabetes or prediabetes risk. Kidney, liver, thyroid, ferritin, B12 or vitamin D tests can be useful when symptoms, diet, medication, age, ethnicity or disease risk make them relevant.

Physical activity is a good example of evidence that does not need a luxury wrapper. The UK Chief Medical Officers' report sets out activity guidance because movement, strength and reduced sedentary time are linked to broad health benefits.¹¹ The practical version is not just "exercise more". It is: preserve muscle, raise aerobic capacity, reduce long sitting blocks, protect sleep, and make activity repeatable enough to survive real life.

Sleep is another target where basics beat novelty. If the problem is snoring, witnessed apnoeas and morning headaches, the optimisation route may be obstructive sleep apnoea assessment, not magnesium. If the problem is insomnia, the useful path may be sleep schedule, light timing, alcohol reduction, CBT-I principles and checking medication effects. A wearable can help spot patterns, but it should not become the authority over how you feel and function.

The same logic applies to nutrition. A high-value nutrition plan improves the pattern: fibre, protein adequacy, unsaturated fats, fruit and vegetables, oily fish where suitable, fewer ultra-processed defaults, less excess alcohol, and a calorie pattern that matches the goal. A low-value plan sells a single villain, a secret protocol or a supplement that claims to fix everything. Use the health library to understand conditions, insights to challenge claims, Start Here to organise your timeline, and the stack builder to track what you are actually taking.

How to decide what to pay for

Before paying for any optimisation service, ask five questions. First, what decision will this change? Second, what is the chance of a false positive, false negative or incidental finding? Third, who interprets the result and with what qualifications? Fourth, what happens if it is abnormal? Fifth, would my GP or specialist accept the result, or would it need repeating?

Be especially cautious when the sales page uses certainty language: "root cause of all disease", "doctors do not test this", "clinical grade detox", "reverse ageing", "balance hormones naturally", "personalised to your DNA", or "catch cancer before symptoms" without saying which cancer, which test performance, which population and which evidence of better outcomes. Precision language should make claims narrower, not grander.

There is a sensible middle ground. You can be proactive without buying every test. Know your blood pressure, waist, smoking status, alcohol intake, family history, medicines, vaccination gaps, sleep pattern, activity level and core metabolic risk. Investigate symptoms properly. Use private care when it answers a real access or expertise problem. Ignore dashboards that cannot explain how a result changes what you do on Monday morning.

This article is educational and does not constitute medical advice, diagnosis, or a treatment recommendation. Medication uses described as “off-label” are not licensed for that purpose in the UK and should only be considered under qualified clinical supervision. Always speak to your GP, pharmacist, or a registered specialist before starting, stopping, or changing any treatment. If you have severe or alarm symptoms - unintentional weight loss, blood in your stool, difficulty swallowing, persistent vomiting, a fever, or severe pain - seek urgent medical care.