In a casino, as in most aspects of life, few things are a sure bet. Yet, when it comes to matters of health, people want — and often expect — certainty. We'd like to think that if we see a doctor for a symptom, we'll leave the office knowing what we have, what we do not have, how to treat it, and confident that the treatment will work without side effects. In addition, we'd like to know how long it will take for the problem to get better and whether or not it will return.
In reality, leaving your doctor's office is often a bit more like entering the casino: even the best doctor can usually provide only possibilities, chances, or a likelihood that a condition is present, that other conditions are absent, and that a specific treatment will work.
In fact, you may hear the word "empiric" to describe the treatment approach. It means a therapy will be chosen based on your doctor's experience, rather than a treatment that is well studied and scientifically proven — a "best guess" about what might work well. If it doesn't work, another treatment will be recommended and so on, through a series of mini-experiments of trial and error to see which works best.
This is often the way things go when there is either no single "best" treatment for your diagnosis, or when the diagnosis is uncertain. Both are exceedingly common in medical practice because uncertainty is so pervasive.
Although most of my colleagues are not gamblers, doctors often deal with risk, chance and odds. For example, your doctor might evaluate your headache and think the chance of a brain tumor is very low, perhaps one in a million. While there is some small chance that there is a brain tumor, your doctor may elect to pursue other possibilities and treatments rather than ordering multiple tests that could detect a tumor.
You might wonder: Why not just order everything and see what turns up? In fact, some doctors do. But extensive testing for common conditions with no known cause (like most cases of headache) is time-consuming, expensive, potentially dangerous (see below), and produces little or no benefit. And because there are so many potential causes of a symptom, testing for every possible cause might literally be endless.
So, your doctor weighs the risk of missing an important or serious condition against the likelihood of that condition based on your symptoms, examination and available test results. This acceptance of risk, dealing with chance, and embracing uncertainty is business as usual for most doctors. If you've ever been admitted to a hospital or seen your doctor while you were sick, you may have experienced this uncertainty firsthand.
One of my classmates in medical school chose oncology, the specialty dealing with cancer, as a career path. When I asked him why, he said it was because he did not deal well with uncertainty and most oncologists only see patients who are known to have cancer. While all fields of medicine deal with uncertainty, some have more than others. In the practice of rheumatology (the specialty dealing with arthritis and related disorders), doctors may see patient after patient without making a diagnosis, a far different situation compared with oncology. There’s still plenty of uncertainty in all fields of medicine, and most good doctors get comfortable dealing with it.
One of the biggest challenges for patients is accepting the fact that their doctor may not have the answer, may not know a single best, highly effective treatment for their problems, and may not be able to tell them what to expect. So, while doctors may get used to the notion that uncertainty is common, patients may believe it is a sign of incompetence, inexperience or insecurity. The truth is that doctors are frequently unsure. Some of what your doctor may be saying to you is that he or she is uncertain, but in that uncertainty, there is still useful information.
One setting in which this comes up nearly every day is medication side effects. When I review the risks and benefits of a medication with my patients, I often hear appropriate concerns regarding side effects: "I don't want to take this if it's going to bother my stomach," for example. Because I mentioned stomach upset as a possible side effect, many people assume that it routinely happens; the fact is that there's a chance, typically low (for example, in the range of 5% or 10%), that a medication will cause significant stomach trouble. In addition, it's often impossible to predict who will or won't have that side effect. So, it's my job to communicate the chance that a side effect will occur while making it clear that most people never have that problem.
Another common scenario is an abnormal test result. Because "normal" is defined for many tests as the average result for 95% of the healthy population, 5% of the healthy population has abnormal results. Hearing that your abnormal result "may be nothing to worry about" is not completely reassuring, but that's often the best your doctor can say about it. No test is perfect, so even when abnormal, a single, clear interpretation of that abnormality may not be possible. Anyone with an abnormal mammogram that turned out to be insignificant or meaningless knows the anxiety and fear inspired by the uncertainty of an imperfect test. And there's an even darker side of imperfect tests: when a result is normal but the disease is present. Mammograms may be normal in a minority of patients with breast cancer. This sort of uncertainty is difficult to accept but lurks in the background of every discussion about test results.
The quest for a definite answer or an absolute declaration of "truth" may not only be fruitless, frustrating, and expensive — it can cause harm. Here's an example.
Imagine that you have been having headaches for years and lately they have been worse. Maybe it's the stress at work, or the way you sit for hours in front of the new computer, or one of many other possible explanations. It’s entirely reasonable to see your doctor, review your symptoms in detail, and have a physical examination. Tests or referrals to other health care professionals may be recommended, or they may not, depending on the symptoms and examination findings. At the end of the evaluation, you may get a diagnosis of "tension headache," but you wonder, "Could it be something else? I've always had stress, why are my headaches worse now? And my boss has more stress than I do; why doesn't she have headaches?"
For the vast majority of people with headaches, there are no definitive answers to these questions. Even if you were sent for extensive testing and consultation, there's an excellent chance you'd be no further along in your knowledge than when you started. And if you do have a CT scan or MRI for your headache, there's a chance that it will be abnormal in a way that will not account for your headache but could lead to more tests, worry, anxiety, and, yes, more stress. It is not rare for tests to show abnormalities that turn out to be unimportant;, however, these abnormalities are not known to be meaningless when first detected, so additional tests, including surgical biopsy or other "invasive testing" and weeks or months of worry, may follow. While the effort to eliminate uncertainty is understandable, that effort can do more harm than good.
Medical textbooks often describe a condition (or even define it) based on the average experience of many patients thought definitely to have had that disease; "expert consensus" is often used to establish the diagnosis in those patients, with the experts being nationally recognized clinicians who regularly take care of such patients. In order to establish a standard definition, these patients tend to have similar features. In "real life," however, many people do not fit the classic description — we often joke that "the patient did not read the textbook."
In addition, over a career, physicians often encounter situations that they have never seen or heard of before. Even after seeing specialists in reputable academic hospitals, it is not rare to have an undiagnosed problem. These cases are the ultimate in uncertainty but are by no means rare.
The "gold standard" is a concept commonly embraced by doctors — it's the best evidence there is, the ultimate defining features or crucial characteristic for a particular condition or disease. Some situations are black and white. A biopsy, for example, can show cancer (or not), and most of the time the reading is clearly one or the other. The biopsy is the gold standard, and there is generally little uncertainty about the diagnosis. However, there are times when the reading of the biopsy is not clear-cut, when doctors disagree about the presence of cancer and suggest additional specialized tests on the tissue or even another biopsy. Here, the gold standard is not good enough to be sure of what is going on.
Another common scenario is that many conditions or diseases have gold standards, but they are not truly "gold" — they are imperfect, prone to error. For example, for the syndrome called fibromyalgia, the gold standard is a list of criteria designed for research studies; however, these criteria were not intended to be applied to an individual patient in the office, because they do not perform particularly well in that setting. Yet, it’s the best we have — it's the gold standard though not a particularly accurate one. In practice, the diagnosis of fibromyalgia relies on the judgment of an individual physician. Because judgment can vary widely, and doctors may disagree, uncertainty regarding a diagnosis is common in conditions without a highly accurate gold standard.
When you see your doctor, don’t expect certainty. The notion that an authoritative, all-knowing doctor will announce your diagnosis and treatment with supreme confidence — and that he or she will be correct — is an unrealistic expectation promoted by movies and television, one that does not reflect real life in real time at the doctor’s office.
For much of medical care, there simply is no gold standard; fortunately, there is a lot your health care professional can provide for you even when he or she is uncertain. And if a condition does not respond as quickly as expected or if symptoms change, your doctor will re-evaluate you. As long as you and your doctor keep an open mind about the situation, uncertainty can be manageable.
The next time you see your doctor, don't be surprised if he or she uses terms such as "possible," "potential," "likely," or "maybe." While it may sound like your doctor is hedging his or her bets, what your doctor is probably saying is that each patient is different, there is much that remains unknown, and your situation is still unclear. Not knowing is certainly unsettling, but uncertainty is probably the only sure thing in medicine.
Robert H. Shmerling, M.D. is associate physician at Beth Israel Deaconess Medical Center and associate professor at Harvard Medical School. He has been a practicing rheumatologist for over 20 years at Beth Israel Deaconess Medical Center. He is an active teacher in the Internal Medicine Residency Program, serving as the Robinson Firm Chief. He is also a teacher in the Rheumatology Fellowship Program.