Most of what we know in clinical medicine comes from treating individuals who are sick. The typical construct is “If a person consults a doctor with X (symptoms), and if the tests reveal Y (abnormal readings in a particular permutation/combination), then the person likely has disease/condition/syndrome Z, which responds to a certain course of treatment (or does not need any treatment). Most clinical medicine is therefore in the therapeutic domain.
What happens when we extrapolate the results of what we know (based on research) in the domain of individuals who are sick to those who are healthy? The answer, more often than not, is unnecessary testing, overdiagnosis and overtreatment (which not only does not help but can potentially harm).
The process of testing healthy individuals to “catch early disease” is called screening. For a screening test to be useful, several things must be known. The natural history of the disease (what happens if we do nothing) is the key first step. Only if we know the answer to that, will we know the benefits of doing something. For example, autopsy studies have often found abnormalities (such as tumors) in individuals who died of unrelated causes. Had we discovered these as part of screening, we would have unnecessarily subjected these individuals to multiple tests and treatments.
The second step is knowing whether the disease has an early phase in which detection is likely to make a difference. Tuberculosis screening programmes that had trimestral chest x-rays realized that most individuals would develop symptoms and go to a doctor between their scheduled x-rays, and therefore, waiting for symptoms was a more rational step to diagnosing the disease rather than frequent x-rays.
The third step is knowing whether early detection will result in treatment that will alter the course of the disease in a way that would be different from what would have happened had we waited for symptoms. We also need to know whether the ill-effects/cost of testing a huge number of people is worth the diagnosis of a condition that may be rare. This applies even more if testing is taxpayer funded.
Extremely few tests have been proven to meet these criteria. Even widely used tests, such as mammography screening, are still debated when applied in low-risk populations. Let us look at an example. Two individuals with similar risk profiles have different philosophies in life, but are destined to have identical life trajectories. The first gets screening tests performed frequently and discovers a tumor somewhere in her body despite having no symptoms. There is tremendous anxiety associated with the finding, leading to a whole gamut of blood tests and further imaging tests, consultations with experts and an eventual biopsy that tells her that the tumor is benign.
However, the knowledge that there is a tumor in the body leads her to undergo frequent tests and consultations periodically, costing her time, money and some mental peace. She eventually passes away of an unrelated cause. The second person has the same identical tumor in her body, but because of not undergoing screening, does not realize it, and has the same life course as the first person. Although this is an anecdote, research trials conducted to assess the usefulness of screening do exactly this – they randomly spilt individuals into two groups and assess whether screening makes any difference.
Most trials for a wide variety of screening tests have not proven conclusive benefits.
As a clinician, I see anxious individuals who get blood tests, chest x-rays/CT scans done with a threshold as low as a one-day cough or sneeze. “What’s the harm in testing and being safe?” is driving testing (and treatment) more often than “Do I get any meaningful information from this?”.
Since no diagnostic test is 100 per cent accurate (false positives and false negatives can occur), the more the number of tests done, the higher is the probability that there will be abnormal results just by chance (this can be mathematically proven). This gets further amplified if the pre-test probability is low, i.e. if the test is done in a population without a strong reason for doing so (such as in screening tests). The anxiety and fear associated with Covid lead to widespread unnecessary testing, including CT scans, and unfortunately may have normalized getting tested for a majority of individuals in India.
So what’s the harm? Imaging tests such as CT scans are associated with a dose of radiation that cannot be ignored, unless there is a strong reason to do so. Despite being a pulmonologist, not everyone who sees me is asked to get even a chest X-ray. Blood tests, when done in bulk without any specific reason, can lead to abnormal values by chance, leading to further unnecessary tests and treatment. The anxiety associated with abnormal results is not trivial, and this is completely unwarranted when the test does not meet the screening criteria mentioned above. On a bigger scale, we should not ignore the quantum of consumables that are utilized when it comes to collecting and disposing blood and tissue products. These simply cannot be justified unless they save lives.
We need to discuss unregulated and widespread testing that seems to be becoming the norm in India. As testing becomes more and more accessible, and since a majority of testing is paid for out-of-pocket, it is likely that individuals will be nudged and incentivised to test more frequently and with newer tests. We must let science guide such testing, and not allow for the overmedicalising of healthcare.
Dr Lancelot Pinto works with Hinduja Hospital in Mumbai
