Direct to consumer testing- liberating or a liability?

Mary is a 72-year-old lady presenting with abdominal pain. She is very apologetic for taking up an appointment: ‘I know you’re very busy doctor, so I didn’t want to trouble you’. On further exploration, it transpires that Mary has been suffering from a change in bowel habit and intermittent abdominal pain for around a year. She’s finally come in because her weight is plummeting and she’s exhausted. Clearly a concerning picture.

So why has it taken so long for Mary to come in? 

‘It’s not a very nice thing to talk about, is it doctor? I did one of those bowel health tests from the chemist and that was fine. So I thought it can’t be anything too serious, I just need to get on with it!’. 

As discussed in a recent BMJ editorial, direct to consumer testing has exploded onto the shelves in recent years, offering patients a new way to access information outside of the traditional medical model. It is not an entirely new phenomenon- pregnancy tests are well established (despite the initial scepticism from the medical profession that ‘women could not be trusted to use them’). More recently, covid–19 testing required everyone to become familiar with at home diagnostics, and the NHS endorses HIV testing for populations at high risk. Purported benefits of self-tests include enabling patients to take charge of their own health and reducing the barriers to testing.

So, what are the concerns? 

Firstly, do the kits provide the information patients need to make informed decisions about testing? A cross sectional review published in the BMJ (BMJ 2025;390:e085546) looked at information on self-test boxes and patient information leaflets. They obtained self-testing kits from supermarkets, community pharmacies, and health and wellbeing shops in a 10-mile radius of Edgbaston. Overall, 30 testing kits were reviewed, assessing 20 biomarkers for 19 different conditions, including hypothyroidism, iron deficiency anaemia, menopause, fertility, kidney health, bowel health, prostate health, infections (HIV, chlamydia, Group A strep, flu, covid-19), vitamin D, cholesterol and gluten sensitivity.

Information on the box to guide purchase decision was often sparse. Who should use the test and when was only present on 27% of boxes, the recommended action after the test result was on 23%, and the numerical test performance on 33%. Only 50% explicitly stated whether the test was intended for screening, diagnosis or monitoring purposes. Overall, the authors concluded that ‘information provided by the tests in our sample is unlikely to be sufficient to support appropriate use’.

Further, the use of tests for 11 out the 19 conditions was judged to be in contrary to current evidence-based guidance (in terms of intended population, frequency of testing, test threshold or approach). There were also concerns about the usability of some tests and challenges for lay people in reading and interpreting the results outside of a laboratory environment. A second team raised concerns over the accuracy claims (BMJ 2025;390:e085547), encountering significant challenges as the manufactures declined access to the necessary data for over half of the tests studied.

What do patients do with the result? 90% recommended follow up with a health professional if the test was positive (and 47% recommended follow up with a health professional if the tests were normal/negative), meaning it is highly likely that patients will be seeking our help following a test result. This could potentially put us in the tricky position of being asked for advice on an inappropriate test we haven’t requested, with unclear accuracy.

As the editorial states, false positives have the potential for patient distress, triggering over-investigation, overdiagnosis and possibly overtreatment. Conversely false negatives could have even more concerning consequences. Take the example of Mary, how many people may decide not to trouble the GP with their ‘embarrassing’ bowel symptoms if an over-the-counter bowel health test says it’s all ok? Was it clear to Mary what the test was actually measuring? Was it set at a level for screening or symptomatic patients? Can Mary reasonably be expected to understand that distinction? If seen within a clinical setting then tests would be interpreted in the broader context of the patients’ symptoms and background, with other factors guiding joint decisions on further investigation.

This is tricky area to navigate, and greater regulation is called for. In the interim, we need to be wary of being drawn into interpreting private tests, or feeling pressured to offer non clinically indicated tests as a consequence. In the words of the editorial ‘the NHS should not be expected to provide a “free” follow-up service for companies offering inappropriate, oversold, and low value tests’. For Mary, however, there is a clear and urgent need to investigate her symptoms further.