Self & Other · A Glossary of Medical News

The Diseases and States

TNF inhibitor. A type of biologic that blocks tumor necrosis factor, a cytokine central to inflammation in rheumatoid arthritis, IBD, psoriasis, and several other diseases. Examples: adalimumab (Humira), infliximab (Remicade), etanercept (Enbrel).

JAK inhibitor. A small-molecule drug that blocks Janus kinase, an enzyme inside cells that many cytokines use to deliver their signals. Used in rheumatoid arthritis, certain inflammatory bowel diseases, severe atopic dermatitis. Examples: tofacitinib, baricitinib, upadacitinib.

Antiretroviral. A drug that blocks the replication of retroviruses (HIV being the famous one). Modern combination antiretroviral therapy can suppress HIV to undetectable levels and is one of the major medical achievements of the late twentieth century.

The Diseases and States

Terms used to describe immune-related conditions and the patterns they follow.

Autoimmune disease. A disease in which the immune system mistakenly attacks the body’s own tissue. Includes type 1 diabetes, multiple sclerosis, lupus, rheumatoid arthritis, Hashimoto’s, celiac, and dozens more. Roughly five to ten percent of people are affected.

Allergic / atopic. Allergic refers to immune reactions to harmless substances, mediated by IgE. Atopic is a related term used for the family of conditions that often run together: eczema, hay fever, asthma, food allergies. People with one are statistically more likely to have others.

Anaphylaxis. Severe systemic allergic reaction, with widespread mast cell activation, dropping blood pressure, swelling of airways. Treated immediately with epinephrine. Life-threatening if untreated.

Immunodeficiency. A state in which the immune system is impaired. Primary immunodeficiency is genetic (SCID, CVID, others). Secondary immunodeficiency is acquired (HIV, immunosuppressive drugs, chemotherapy, certain cancers).

Flare. A worsening of symptoms in a chronic autoimmune or inflammatory condition. Most autoimmune diseases have a relapsing-remitting pattern, with periods of more and less active disease. The terms “flare” and “relapse” are mostly interchangeable.

Remission. A state in which a disease is not actively producing symptoms or inflammation. May be partial or complete; may be drug-induced or spontaneous. In some cancers, “complete remission” means no detectable disease, but does not necessarily mean cure.

Seroconversion. The point at which antibodies against a particular antigen become detectable in blood. People are described as having “seroconverted” after a vaccine or after recovering from an infection. The opposite term, “seronegative,” means the antibodies cannot be detected.

Trained immunity. Recently described concept that some innate immune cells (macrophages, NK cells) can develop a kind of memory — not antigen-specific like adaptive immunity, but a more general state of readiness after a strong infection. An active research area.

The Evidence Vocabulary

Terms about how medical evidence is generated and reported. Understanding these makes most medical news much clearer.

Randomized controlled trial. The gold standard of medical evidence. Patients are randomly assigned to either a treatment group or a control group; the difference in outcomes between the groups is the evidence for the treatment’s effect. Often abbreviated RCT.

Placebo-controlled. A trial in which the control group receives a sham treatment (a sugar pill, a saline injection) that is indistinguishable from the active treatment. The placebo effect is real and substantial in many contexts; placebo-controlled trials are designed to measure the active treatment’s effect over and above placebo.

Double-blind. A trial in which neither the patient nor the researcher administering the treatment knows whether it is the active drug or the placebo. Reduces bias in both directions. Used in essentially all serious clinical trials.

Phase 1, 2, 3 trials. The standard sequence of clinical testing. Phase 1 tests safety in a small group of healthy volunteers. Phase 2 tests efficacy and dose in a moderate-sized group of patients. Phase 3 tests efficacy and safety in a large patient population, against the existing standard of care or placebo. Approval typically requires successful Phase 3 data.

Meta-analysis. A study that combines the results of multiple previous studies on the same question, providing a more statistically powerful answer than any single study. Generally a high-quality form of evidence, though only as good as the studies it includes.

Confidence interval. A range of values that, given the data, is statistically likely to contain the true value of whatever is being measured. A treatment that reduces risk by “20 percent (95 percent CI: 15 to 25 percent)” means: best estimate 20 percent, but the true effect is plausibly anywhere from 15 to 25 percent. Wide confidence intervals mean less certain results; narrow ones mean more certain.

Association vs. causation. A difference between groups (an association) does not, by itself, establish that one thing causes another. People who carry umbrellas get rained on more often than people who don’t — but the umbrellas are not causing the rain. Distinguishing association from causation is the entire job of the experimental design and statistical methods of medical research.

Statistical significance. A statistical statement that the result is unlikely to have arisen by chance alone. By itself, statistical significance does not say the result is large, important, or clinically meaningful. A statistically significant tiny effect can be irrelevant in practice. Always look at the size of the effect, not just whether it is “significant.”

Number needed to treat (NNT). How many patients have to receive a treatment for one of them to benefit. A drug with an NNT of 5 means: out of five patients treated, one will be helped by the treatment specifically (the others either would have done well without it or won’t respond). A useful concept for cutting through inflated relative-risk numbers.

A Few Big Ideas, Restated

Some of the recurring conceptual shorthand of the book, in case it is useful to find them in one place.

Self vs. non-self. The central question the immune system is constantly asking, in molecular terms. Most of immunology is the answer to this question, in different forms.

Innate vs. adaptive immunity. Two layers of the immune system. Innate is fast, ancient, recognizes broad categories of trouble, and does not learn. Adaptive is slow, vertebrate-only, recognizes specific molecular shapes, and remembers.

Regulated, not strong. The actual goal of immune health. The immune system you want is one that responds when it should, resolves cleanly when it can, and stays quiet otherwise. Anyone telling you to “boost” your immunity is selling you the wrong thing.

Population vs. individual. Lifestyle and treatment effects are mostly described as population averages. The same intervention that reduces a population’s risk by twenty percent does not protect any specific individual by twenty percent. Better odds, not guarantees.

Permission, not prescription. When something has gone wrong with the immune system — an autoimmune disease, an immunodeficiency, a cancer — nothing in the patient’s past behavior caused it, and nothing they could have done would reliably have prevented it. The relevant decisions were settled before the disease was diagnosable. Treatment matters; blame does not.

If you have read this far, this glossary is partly a service and partly a marker. The terms in it have been the building blocks of the entire book. They are also the building blocks of the medical conversations you will have, the news articles you will read, the doctor’s appointments you will sit through. The vocabulary is not a wall; it is a set of tools, and once you have them, the conversations get easier.

The next chapter is the last one. It is also the shortest. It returns, as promised, to the seven-year-old falling off a bicycle in the prologue. We owe that kid a return visit.

All 21 chapters of Self & Other · Library · All courses