There’s No Such Thing as a Disease

Written by Geoff Kane, MD, MPH. Posted on
Print Friendly, PDF & Email

Healthcare providers are charged with helping individuals who come to them with physical, emotional, and behavioral problems. As they prepare to help, providers usually follow a routine—they get to know the person and their problem(s), examine the person, and, frequently, obtain additional information such as blood tests or x-rays. Prior to recommending specific treatment, providers “make a diagnosis,” which then guides providers and patients to treatment options relevant to the problem at hand.

Diagnoses are commonly expressed in terms of the manifestations of a problem (hives, for example) or the cause of a problem (for example, penicillin allergy). Clinicians sometimes make diagnoses quickly and confidently or, at other times, slowly and tentatively. They may entertain several candidate diagnoses, “the differential diagnosis,” before settling on a provisional, or working, diagnosis.

A biology professor periodically reminded his students, “Variation is the law of life!” Clinicians can testify to this. No two patients—or the problems they present—are identical, which means that when clinicians make the same diagnosis in two individuals, they are not saying the two people have exactly the same problem. They are saying that the problems of the two patients have important characteristics in common, often at a cellular level, and that both patients are likely to respond to similar treatments.

For example, the thick, scratched blotches on the neck of the woman under stress do not look precisely like the raised, itchy patches that appeared on the arm of the young man after a dose of penicillin. But if clinicians diagnose hives in both cases, both patients will likely obtain relief if they accept treatment with an antihistamine.

The two people and two skin eruptions are not identical, but the underlying cellular processes and responses to medication are so much alike that both individuals, with different but similar problems, benefit from receiving the same diagnosis—and ensuing treatment guided by that diagnosis.

When a clinician makes a diagnosis, she or he judges the problem of this particular patient to be biologically similar to the problems of a previously known or described group of patients. Technology (lab tests, x-rays, and such) may detect or confirm the similarities. Once a current patient becomes identified with a like group of previous patients, the current patient benefits from an existing body of knowledge regarding causes of the problem and treatments that helped others.

The diagnoses that name patient problems are terms used to identify or label the many separate groups of individuals afflicted with biologically similar problems. These terms are called diseases. References such as the International Classification of Diseases (World Health Organization) and the Diagnostic and Statistical Manual of Mental Disorders (American Psychiatric Association) classify and assign numbers to the many disease names in use at a given time. These volumes have to be regularly revised because diseases are practical categories that guide treatment and, to remain useful, the categories must change to keep pace with advances in the science and technology that describe and treat health problems.

The point of this article is to explain that diseases are categories of people with biologically similar health problems. Diseases are concepts, not things. Diseases do not exist on their own, apart from the groups of people they designate.

There is confusion about this—inside healthcare professions as well as outside. Some people talk as if diseases attack individuals from the outside, as a monster would. This mistake, thinking of a concept as if it has independent existence, has been called the fallacy of reification.

Classifying addiction as disease is controversial. Participants in that debate sometimes reify diseases. The next Addiction Medicine Update will examine whether addiction can be thought of as disease. And whether it should.