Some screenings detect anomalies which have no impact on a patient’s well-being. Such over-diagnosis can create significant economic and health consequences.
A healthy man is just a sick man who doesn’t know it,” wrote Jules Romains in his play Dr Knock, or the Triumph of Medicine (1923). Nowadays, “a person in good health is above all someone who has not yet undergone screening,” jokes Sauveur Boukris, general practitioner and author of the book La fabrique de malades: ces maladies qu’on nous invente (Disease Factory: The Diseases They Invent for Us). The reality is more mixed. With modern medical techniques, it is now possible to detect infinitesimal anomalies in the human body, such as lumps of cancerous cells a few millimetres in size. The benefits are sometimes clearly shown. For instance, screening for cervical cancer and colon cancer permits better management of these diseases.
“But certain screenings sometimes detect real anomalies that pose no health risk,” says Dr Arnaud Chiolero (Privat-Docent), epidemiologist at the University Institute for Social and Preventive Medicine at the Lausanne University Hospital (CHUV) and author of a literature review on the subject, published in the Swiss Medical Forum. “This is what is called over-diagnosis.”
Over-diagnosis is the consequence of preventive and highly technological medicine. Prevention through screening is based on the premise that the earlier a disease is detected, the better it can be treated. “It is clear that prevention has proved its case in many fields, such as cardiovascular disease (e.g. screening and treatment for severe arterial hypertension to prevent strokes) and cancer (e.g. cervical cancer screening),” continues Arnaud Chiolero. “Unfortunately, in some cases, the identification of an anomaly in a person leads to treatment when this anomaly would never have caused symptoms or led to death. It’s a major problem because over-diagnosis leads to over-treatment, which provides no benefit to the patient but can be a source of side effects and complications. Plus, the public-health costs can be quite high.”
Promoting awareness among the Swiss
The debate will heat up in Switzerland: this year the French-speaking Swiss Consumer Federation, together with its Italian- and German-speaking sister organisations, is launching an information campaign to raise public awareness about over-medicalisation. A number of corresponding initiatives will also be put in place. “It’s a complex problem that is tricky to publicise and that should not be lumped together with low-cost medicine. Rather, it should be clearly identified with medical acts whose benefits to health have not been scientifically demonstrated,” wrote the Federation on its website in January 2014.
Avoid invasive treatments
Screening for prostate cancer is a case in point. Tumours in the prostate are detected through blood analyses by measuring the level of PSA (Prostate Specific Antigen). This protein is produced naturally by prostate cells. But cancerous cells secrete ten times as much PSA as normal cells. By measuring the PSA concentration, the development of cancer in the patient can thus be suspected.
“Studies indicate that 30 to 70% of men over 60 years old have a prostate tumour,” says Arnaud Chiolero. “If it is detected during the person’s lifetime, the problem becomes understanding how it will evolve. Will it become an aggressive cancer, or will it grow very slowly without ever putting the patient at risk? The estimates are very uncertain, but it appears that 17 to 66% of the cases detected by PSA meet the definition of over-diagnosis. That is to say, the patient will not develop symptoms secondary to this tumour and will not die from it, whether treated or not.” In addition to the stress caused by such a diagnosis, the surgical treatment of prostate cancer is not without consequence for the patient, often causing impotence and incontinence.
Many anomalies are discovered by chance during radiological tests, which are another major source of over-diagnosis. American researchers analysed the images of full-body CT scans conducted on 1,192 people who presented no particular health problems. Their results, published in the journal Radiological Society of North America (RSNA), show that 86% of them presented anomalies, most of which were of no consequence to the patients’ health.
It is necessary to anticipate the consequences of discovering anomalies, and to aim for an informed decision shared with the patient.
“Apart from screening, a more subtle form of over-diagnosis is linked to the lowering of threshold values for risk factors,” continues Arnaud Chiolero. “For example, reducing the normal arterial blood pressure value automatically increases the number of people considered to be hypertensive. At certain thresholds, the benefit of treatment becomes minimal, even too low with regard to the costs and secondary risks of treatment. The difficulty for doctors is therefore to establish the correct threshold, avoiding over-diagnosis without going so far as to allow for under-diagnosis.”
The concern of doctors not to miss a diagnosis and the requests of patients for reassurance as to the absence of an anomaly can generate over-diagnosis. “Any screening test should be accompanied by information on the advantages and disadvantages of such a test, and on the risk of over-diagnosis. It is necessary to anticipate the consequences of discovering anomalies, and to aim for an informed decision shared with the patient,” says Arnaud Chiolero. “Over-diagnosis can be prevented by avoiding certain tests whose efficacy has not been proven and, in some cases, by carrying out targeted screening. Another approach, which is the subject of many research studies, is to better identify the nature of the anomalies discovered, so as to distinguish between those that will develop harmlessly and can be monitored without treatment (e.g. active monitoring of certain prostate cancers) and those that will develop unfavourably and require treatment.”
For some years now, a new acronym has been catching on in English-speaking medical circles: “VOMIT”, which stands for “Victim of Modern Imaging Technology”. It condemns futile treatments and operations to which certain patients fall victim after medical imaging tests. Most of these cases are due to errors of image interpretation.
Studies indicate that 30 to 70% of men over 60 years old have a prostate tumour. If it is detected during the person’s lifetime, the problem becomes understanding how it will evolve. Will it become an aggressive cancer, or will it grow very slowly without ever putting the patient at risk?