Carlos Martins: Overuse of medical tests—a new health risk factor?

Carlos_MartinsA risk factor is, among other things, an aspect of personal behaviour or lifestyle that, on the basis of epidemiologic evidence, is known to be associated with health related conditions considered important to prevent (1).

A new behaviour may be observed in the general population of western countries; a behaviour related to the way patients use medical services with preventive intention and the frequency that they receive medical tests and other screenings. Evidence shows an alarming tendency to overuse preventive health services (2,3). In Portugal, for example, a recent population-based cross-sectional study showed that the majority of Portuguese adults think they should undergo general routine blood and urine tests on an annual basis. Other, even less-recommended screening tests, such as lung X-rays, breast ultrasounds, abdominal ultrasounds, and gynaecological ultrasounds, are deemed necessary by a great proportion of the population (3).

In part, this pattern of behaviour may be justified by the culturally rooted idea that routine general health checks are necessary. Obtaining general health checks is a practice that is not supported by the evidence because it does not reduce morbidity or mortality, neither overall nor for cardiovascular or cancer-related illnesses, although they increase the number of new diagnoses (4). Disease mongering campaigns, the greater variety and availability of medical tests, and the disseminated notion that more medicine is always better may be other factors that have contributed to the development of this behaviour. However, we should also consider the hypothesis that medical consultation has moved from a paternalistic model to that which we could call a consumeristic model. This is a kind of medical consultation where patients get what they want, and which is quite different from the ideal shared decision making model.

There is a growing amount of evidence that this behaviour is associated with harms, and those harms are considered important to prevent. False positives, incidentalomas, being overdiagnosed, and the resultant cascade of attitudes are some of those harms that may significantly impair the quality of life of healthy people (5).

Therefore, by definition, we may say that we are facing a new health risk factor: the excessive and inappropriate use of medical tests. Of course, there is a dose of irony in this statement, given that the excessive labelling of healthy people with “risk factors” has been one of the main driving forces that has converted so many healthy individuals into patients (6). However, it is quite probable that most people don’t know that they are at risk of suffering harm when they undergo medical tests. If we think about the current state of the pharmaceutical field, we all recognize that patients are aware of a potential drug’s adverse effects. But, most of our patients have never heard of a medical test’s possible adverse effects. Some patients don’t take a new medicine without first reading the Summary of Product Characteristics leaflet included with the drug. Is now the right time to add some regulations to medical testing and create a Summary of Test Characteristics? This may be a key communication tool in this area. An adequately-structured instrument should have language accessible to the population, but not necessarily as dense as the Summary of Product Characteristics. Sections such as “indications,” “possible harms,” “predictive value according to the prevalence,” “number needed to screen,” and “number needed to harm” could be included in this new Summary of Test Characteristics. This is a suggestion that deserves an opportunity to be tested. Would it have an impact on the way healthy patients view medical tests? This is an important question to be answered using proper research.

Nevertheless, if we want to prevent overdiagnosis and other potential harms related to excessive, not evidence-based, use of medical tests by healthy people, we should start designing strategies of communication to explain to our populations what is meant by potential harms, false positives, and overdiagnosis. Only by understanding these meanings and knowing the probabilities of being benefitted or harmed, will the patient be able to make a shared decision and to avoid a consumeristic model of medical tests utilization.

Competing interests: None declared.

Carlos Martins, Family Medicine Unit, Social Sciences and Health Department of the Faculty of
Medicine of Porto.


1) International Epidemiological Association. A dictionary of epidemiology 4th ed. Oxford University Press, 2001
2) Brotons C, Bulc M, Sammut MR, Sheehan M, Manuel da Silva Martins C, Björkelund C et al. Attitudes toward preventive services and lifestyle: the views of primary care patients in Europe. The EUROPREVIEW patient study. Fam Pract 2012; Suppl 1:i168-i176.
3) Martins C, Azevedo LF, Ribeiro O, Sa L, Santos P, Couto L, et al. A population-based nationwide cross sectional study on preventive health services utilization in Portugal— What services (and frequencies) are deemed necessary by patients? PLoS ONE 2013;8:e81256.
4) Krogsboll LT, Jorgensen KJ, Gronhoj Larsen C, Gotzsche PC. General health checks in adults for reducing morbidity and mortality from disease: Cochrane systematic review and metaanalysisBMJ 2012;345:e7191
5) Welch HG. Overdiagnosed: making people sick in the pursuit of health. Beacon Press, 2011
6) Gervas J, Starfield B, Heath I. Is clinical prevention better than cure? Lancet 2008;372:1997–9

  • Bill Cayley

    Great analysis. I think there is another angle to this as well. Not EVERY patient over-uses medical testing. Just as some patients want a pill for every ache and pain, while others just “gut it out” or muddle through rather severe pain; there are some patients who WANT a test for every possible “bad thing”, whether or not the evidence supports it, and there are others who decline to do tests and procedures, even when we think the evidence clearly indicates benefit.

    So, to go back to the “risk factor” paradigm, I do indeed think that we can think of “medical-procedure-seeking” as a behavioral risk factor: procedure avoidance may lead to missing things, but being overly “procedure seeking” would seem to predict worse self-ratings of health status, increased risk of iatrogenic harm, and increased medical expense.

  • Carlos Martins

    Thank you for your comment, Bill Cayley. It’s true: not every patient
    overuses medical testing… And in future initiatives, when trying to
    communicate the tests characteristics, we have to take care not to scare
    patients… Anyway, as health professionals we want the “right” tests
    for the “right” patients. There where the probability of benefiting the
    patient is the highest and the probability of harming the lowest. 🙂

  • Luís Filipe Gomes

    This is a very important issue, demanding adequate measures.

    If we really want our patients to achieve shared decisions on testing, we’ll have to help them into acquiring the competencies needed.
    We will probably need to spread the same level of information as professionals, associations, institutions and industry do to propose the same tests…

  • Dr Kalee

    Hello, Thank you for this very interesting article. I am a french Family Physician, and I was wondering if you would allow me to translate this article and post it on my own blog (link here:
    I would of course cite you and link to your original article.
    Thank you very much for an answer!

  • Carlos Martins

    OK! I thank you for your kind words and for your interest. I think you can make the translation…. I’d also appreciate if you could keep the bibliographic references… And congratulations for your blog! Merci beaucoup! 🙂

  • Dr Kalee

    Obrigada! 😉
    Here’s the link to the translated post, with the bibliography of course (I was going to keep it anyway…)

  • Carlos Martins

    Thank you very much, Kalee and thank you for “spreading the word”! 🙂 Kind regards, Carlos M