Etiology is the study of the causes and origins of disease. In medicine, the term also denotes the cause or set of causes of a particular condition: for example, an infectious etiology means that an infection causes the condition. Etiological explanations may involve biological agents, inherited variation, environmental exposures, nutritional deficiencies, physical injury, or interacting social and behavioral circumstances. Many diseases result from several contributing factors rather than a single identifiable cause.
Scope and related concepts
Etiology addresses why a disease occurs. It is related to, but distinct from, several other medical concepts:
- Pathogenesis describes the sequence of mechanisms through which a disease develops after its initiating causes act.
- Pathology concerns disease processes and the structural, biochemical, and functional changes associated with them.
- Diagnosis identifies a condition from clinical and investigative evidence. A diagnosis may establish the disease without establishing its cause.
- Epidemiology studies the distribution and determinants of health-related events in populations; investigating etiology is one of its central purposes.
These distinctions are not absolute boundaries. A molecular change may be both part of a pathogenic mechanism and a causal contributor, depending on the question being asked. Likewise, an etiological explanation can operate at several levels: the organism responsible for an infection, the route of exposure, and the circumstances enabling transmission can all be relevant causes.
A syndrome is a recognizable grouping of findings that may have several different etiologies. Conversely, a single cause can produce different clinical presentations.
Causes, risk factors, and causal models
A cause contributes to the occurrence of an outcome. In modern causal research, an important question is whether changing an exposure would change the occurrence of disease under specified conditions. A risk factor, by contrast, is a characteristic or exposure associated with an increased probability of disease. Some risk factors are causal; others are markers of an underlying cause or consequences of an already developing disease.
Necessary and sufficient causes
A necessary cause must be present for a particular disease to occur, but its presence need not produce disease in every exposed person. A sufficient cause is a complete set of conditions capable of producing the outcome.
In the sufficient-component cause model, disease can arise through different combinations of contributing factors. Each combination constitutes a sufficient causal mechanism, and individual factors are components of that mechanism. This explains why:
- The same disease may develop through different causal pathways.
- An exposure may cause disease in some people but not others.
- Removing one causal component may prevent some cases without preventing all cases.
An organism can therefore be necessary for a specifically defined infection while still requiring suitable exposure conditions and host susceptibility to produce clinical illness.
Multifactorial causation
Many common diseases have multifactorial etiologies. Type 2 diabetes, for example, involves interactions among inherited susceptibility, insulin resistance, pancreatic beta-cell dysfunction, and environmental and behavioral factors. No single exposure explains all cases.
Causal factors may also act at different distances from disease onset. A toxic exposure can directly damage tissue, while housing conditions or workplace practices can determine who experiences that exposure. Social determinants of health are therefore relevant to etiological explanation rather than merely background characteristics.
Interaction means that the effect of one factor depends on another. Its presence and magnitude can differ according to the statistical scale used, such as absolute risk differences or relative risks.
Major categories of etiological factors
Etiological categories overlap, and a disease may involve several of them.
| Category | Description and examples |
|---|---|
| Infectious | Disease caused by infectious agents, including bacteria, viruses, fungi, and parasites. Clinical expression also depends on host and environmental conditions. |
| Genetic and chromosomal | Conditions involving inherited or newly arising genetic changes. [[sickle-cell-disease |
| Immune-mediated | Disease involving inappropriate or misdirected immune activity. [[autoimmune-disease |
| Nutritional and metabolic | Disease caused or promoted by nutrient deficiency, excess, or disturbed metabolism. [[scurvy |
| Physical, chemical, and occupational | Injury or disease associated with mechanical forces, radiation, heat, toxins, or workplace exposures. [[silicosis |
| Iatrogenic | Disease or injury resulting from medical investigation or treatment. The term does not, by itself, imply negligence or preventability. |
| Multifactorial | Disease arising through interacting genetic, biological, environmental, and social contributions. |
Congenital means present at birth; it does not necessarily mean inherited. Congenital conditions can result from genetic changes, prenatal exposures, infections, or other developmental disturbances.
Idiopathic indicates that the cause is unknown or has not been established. It does not mean that the condition has no cause. Cryptogenic also denotes an obscure or unidentified origin, although its precise use varies by specialty.
Establishing an etiology
Etiological conclusions depend on converging evidence rather than association alone. Relevant evidence can come from population studies, laboratory experiments, clinical observations, genetic investigations, and interventions.
Observational and experimental studies
A cohort study can examine whether an exposure precedes disease and compare disease occurrence between exposure groups. A case–control study compares earlier exposures in people with and without an outcome and is particularly useful for studying uncommon diseases.
A randomized controlled trial can provide strong evidence about the effects of an intervention because random allocation reduces systematic differences between comparison groups. However, many suspected causes cannot ethically or practically be assigned experimentally. Evidence about harmful exposures therefore often depends heavily on observational research.
Important threats to causal interpretation include:
- Confounding: an exposure–outcome association is distorted by other causal factors.
- Selection bias: the way participants enter or remain in a study produces a misleading association.
- Measurement error: exposures, outcomes, or other relevant variables are measured inaccurately.
- Reverse causation: early or unrecognized disease changes the apparent exposure.
- Chance: sampling variation produces an apparent pattern.
Statistical adjustment can address some of these problems, but it does not automatically establish causation. The validity of an analysis depends on its design, measurements, and assumptions.
Interpreting converging evidence
The Bradford Hill viewpoints, introduced in 1965, provide considerations for interpreting associations. They include temporality, strength and consistency of association, biological gradient, plausibility, coherence, experimental evidence, specificity, and analogy.
They are not a checklist that proves or disproves causation. A cause must precede its effect, but a causal association need not be strong, specific to one disease, or accompanied by a simple dose–response pattern. Mechanisms may remain incompletely understood even when the causal evidence is persuasive.
A biomarker can indicate exposure, biological response, susceptibility, or disease activity. Its association with disease does not by itself show that the marker causes the disease.
Historical development
Medical explanations of disease have included environmental, humoral, and supernatural accounts. In European medical traditions, humoral theory attributed illness to disturbances in bodily humors. During the nineteenth century, experimental physiology, pathology, and microbiology increasingly supported explanations based on specific physical and biological processes.
John Snow investigated cholera transmission through comparisons of water supplies, demonstrating the value of population evidence for identifying causes before the responsible organism was firmly established. The development of the germ theory of disease subsequently connected particular microorganisms with particular infections.
Koch's postulates provided an influential framework for linking microorganisms to disease. Their classical form cannot be applied universally: some organisms cannot be grown using standard culture methods, some infections occur without symptoms, and some diseases depend on complex host conditions. Molecular methods and broader epidemiological evidence have extended the investigation of infectious causes beyond those original criteria.
During the twentieth century, research on nutritional deficiencies, occupational hazards, smoking, and chronic disease expanded etiological investigation beyond single-agent infection models. Genetics and molecular biology further clarified inherited susceptibility and biological pathways without eliminating the importance of environmental and social causes.
Clinical and population significance
Etiological knowledge can distinguish conditions that appear similar but require different forms of treatment or prevention. Peptic ulcer disease, for example, can be associated with Helicobacter pylori infection or exposure to nonsteroidal anti-inflammatory drugs, among other causes.
In public health, identifying causes supports interventions directed at transmission, harmful exposures, nutrition, and living conditions. The importance of a cause at population level depends not only on the size of its effect but also on how common the exposure is.
Population-level causation does not always establish why a particular individual became ill. A well-established causal exposure can increase disease risk without allowing any single case to be assigned uniquely to that exposure. Conversely, an unknown etiology does not invalidate a clinical diagnosis or imply that effective treatment is unavailable.