Dec 10 2018

To Screen Or Not To Screen

Should you regularly see a physician for preventive medicine and to screen to potential health problems? Of course, and this has been consistent messaging to the public for decades. However, specific decisions about whether or not to perform a specific screening test can be complicated, and this muddies the messaging.

The problem is that there is a disconnect between how optimal medical decisions are made, and how individuals approach their own medical care (or that of their loved-ones). Optimal medical decision-making, which results in the best possible outcomes, are based on careful analysis of the best evidence available. Specifically, it considers risk vs benefit – what is the net effect of doing, or not doing, any medical intervention compared to the alternatives? This is necessarily a statistical determination, because we cannot literally see the future.

But people don’t like making cold, hard statistical decisions, especially about something as personal and important as health care. They prefer to prioritize hope. Also, people tend to be risk-averse, but also wish to avoid missing out on a potential benefit. Therefore, psychologically we will tend to go for the option that offers the most hope, not the option that has the statistical best outcome. This is part of the role of the physician – to advise their patients with the hard analysis.

All this is just considering individual decisions, but increasingly we are making societal decisions. These often include cost-effectiveness. This is because we are resource-limited, and decisions about what health care to provide and how to provide it has a dramatic impact on, again, statistical outcomes. If you are on the board of health of a state deciding how to spend your Medicare dollars, then you have to decide, for example, to pay for one liver transplant to save one life, or more basic medical care that might save hundreds of lives for the same money.

Another source of the disconnect between how medical decisions are made and public perception comes from a poor understanding of the statistical nature of medicine and the risk vs benefit analysis. Screening tests are perhaps the most dramatic example. At first it seems that screening tests, looking for the presence of a disease or risk factor before it obviously manifests, is a no-brainer. If you detect disease early, you can treat it more effectively (and at lower cost), with better outcomes. Screening tests are a critical component of effective preventive medicine.

This, certainly, has been the public messaging, and it is true, as far as it goes. However, it is also more complicated. In order to decide if a screening test is worth it you have to consider several variables: what is the sensitivity (how many people with the disease will be detected) and specificity (how many people with a positive test actually have the disease) of the test, what is the natural course of the disease, what treatments are available and how do they effect outcome, what are the risks of the screening test, and the demographics of individual people. All of these factors are put together into an analysis of the net risks vs benefits of screening vs not screening.

So, for example, if you have a low-risk population with a moderately specific test, you may end up with many more false-positives than true-positives. What are the results of having a false positive test? This may result in further testing (usually more invasive), or even treatment, each of which can have its own risks. You may, therefore, cause more harm than good by screening a low risk population (not to mention wasting a lot of limited resources).

We know these complexities produce confusion in the public about the value of screening tests, and a recent study published in JAMA gives us more evidence of this. The study specifically looked at attitudes among VA patients regarding stopping colonoscopies to screen for colorectal cancer when the benefit was too low. The results show that 29% of subjects were not at all comfortable stopping screening, 49% thought age should not be ever used to determine who gets screened, 32% said life expectancy should not be used, and 25% said it was not reasonable to use risk calculators.

This is understandable, but it shows the difference sometimes between how patients think and how doctors think. If a screening test is only beneficial if your life expectancy at the time you take it is >5 years, and yours is <5 years, then there is no benefit to the test. You are therefore getting some risk with no benefit, so why do it? Of course it can be emotionally difficult to face the reality of your own life expectancy. This decision is easily interpreted as a doctor saying that you are not worth it, because you’re too old, or too sick. Really it’s about risk vs benefit, and not wanting to cause unnecessary harm. What if you do a statistically unnecessary colonoscopy on a patient, and they have a rare complication which actually causes their death?

The study also showed that a major predictor of whether or not subjects thought that taking risk into consideration was reasonable was their trust in doctors and the medical system. This is why trust is so important – medical professionals need to adhere to strict ethical guidelines, and also be certified to have had proper training, in part because the system only really works if there is a certain amount of trust. If a doctor is going to look you in the face and give you very uncomfortable information, combined with an objective analysis, this only benefits you if you trust their professionalism. This does not mean blind trust – we are still operating in a context of informed consent, and this is why second-opinions are a good idea for any important decision.

But without any trust, the system cannot really function. This is yet another reason why so-called alternative medicine is so pernicious. It erodes trust in doctors, science, and the system. Also, its practitioners often do not adhere to the scientific model of medicine, meaning they don’t practice according to a careful analysis of risk vs benefit. Rather, they practice emotion-based medicine, giving their patients any hope, even false hope. In so doing they position themselves as the good guys, and science-based practitioners as uncaring, or in the pocket of bigger interests.

This narrative has been frustratingly successful, because it plays directly to patient psychology, and exploits the gap between the public understanding and professional understanding of clinical decision-making. This is also why I always take the time to carefully explain my decision-making to my patients. However, this is increasingly challenging in a world of rising health-care costs, where there is tremendous pressure to practice in a more cost-effective way. Things that are perceived as having a soft-benefit, like spending time explaining decision-making to patients, tend to be squeezed out. (I practice in an academic setting, so these pressures are less.)

This is yet another reason why weakening the scientific basis of medicine, and softening professional ethics, is so dangerous. Alternative medicine is having a corrupting influence on our medical system, and it threatens to destroy it if we let it. In its place we will have a system rife with waste and exploitation.

No responses yet