Darwinian medicine

The relatively new science of Darwinian medicine is based on the belief that diseases not only have proximate causes and mechanisms (e.g. viruses, bacteria, mutations) but also have evolutionary causes. Darwinian medicine focuses on the latter aspect and, while it may not yield cures for many serious diseases, it can help us to understand their current prevalence. Darwinian medicine is also rooted in the belief that natural selection favours reproductive success rather than health or life-span.

In Why we get sick: the new science of Darwinian medicine, Randolph Nesse, an evolutionary biologist, and George Williams, a psychiatrist, explain the application of evolutionary ideas to medicine with these examples:

Ageing and adaptation

One of the main features of ageing is progressive inability of the individual to deal with new or worsening environmental threats (Ch. 12). This is exemplified by the gradual impairment of immune responses, resulting in:

Disease predisposition as an adaptive advantage

Paradoxically, a disease or disease predisposition can have beneficial effects on the individual. A few diseases or susceptibilities to diseases, in addition to their deleterious effects, confer adaptive protection against specific environmental pathogens. This advantage may explain the high prevalence of a disease in areas where the specific pathogen for another disease is endemic.

The best examples are:

Probability of disease

The relationship between the quantity of causal agent and the probability that disease will result is not always simply linear (Fig. 2.2). For example, many infections occur only on exposure to a sufficient dose of micro-organisms; the body’s defence mechanisms have to be overcome before disease results. Some agents capable of causing disease, such as alcohol, are actually beneficial in small doses; those who abstain from alcohol have a slightly higher risk of premature death from ischaemic heart disease.

Fig. 2.2 Relationships between the amount of a causal agent and the probability of disease. Physical agents. For example, the risk of traumatic injury to a pedestrian increases in proportion to the kinetic energy of the motor vehicle. Infectious agents. Many infectious diseases result only if sufficient numbers of the micro-organism (e.g. bacterium, virus) are transmitted; smaller numbers are capable of being eliminated by the non-immune and immune defences. Allergens. In sensitised (i.e. allergic) individuals, minute amounts of an allergen will provoke a severe anaphylactic reaction. J-shaped curve. Best exemplified by alcohol, of which small doses (c. 1–2 units per day) reduce the risk of premature death from ischaemic heart disease, but larger doses progressively increase the risk of cirrhosis.

Host predisposition to disease

Many diseases are the predictable consequence of exposure to the initiating cause; host factors make relatively little contribution. This is particularly true of physical injury: the immediate results of mechanical trauma or radiation injury are dose-related; the outcome can be predicted from the strength of the injurious agent.

Other diseases are the probable consequence of exposure to causative factors, but they are not inevitable. This is exemplified by infections with potentially harmful bacteria: the outcome can be influenced by various host factors such as nutritional status, genetic influences and pre-existing immunity.

Some diseases occur more commonly in individuals with a congenital predisposition. For example, ankylosing spondylitis (Ch. 25), a disabling inflammatory disease of the spinal joints of unknown aetiology, occurs more commonly in individuals with the HLA-B27 allele.

Some diseases predispose patients to the risk of developing other diseases. Diseases associated with an increased risk of cancer are designated premalignant conditions; for example, hepatic cirrhosis predisposes to hepatocellular carcinoma, and ulcerative colitis predisposes to carcinoma of the large intestine. The histologically identifiable antecedent lesion from which the cancers directly develop is designated the premalignant lesion.

Some diseases predispose to others because they have a permissive effect, allowing environmental agents that are not normally pathogenic to cause disease. This is exemplified by opportunistic infections in patients with impaired defence mechanisms resulting in infection by organisms not normally harmful (i.e. non-pathogenic) to humans (Ch. 9). Patients with leukaemia or the acquired immune deficiency syndrome (AIDS), organ transplant recipients, or other patients treated with cytotoxic drugs or steroids, are susceptible to infections such as pneumonia due to Aspergillus fungi, cytomegalovirus or Pneumocystis jiroveci.

Causes and agents of disease

It is argued that a distinction should be made between the cause and the agent of a disease. For example, tuberculosis is caused, arguably, not by the tubercle bacillus (Mycobacterium tuberculosis) but by poverty, social deprivation and malnutrition—the tubercle bacillus is ‘merely’ the agent of the disease; the underlying cause is adverse socio-economic factors. There is, in fact, incontrovertible evidence that the decline in incidence of many serious infectious diseases is attributable substantially to improvements in hygiene, sanitation and general nutrition rather than to immunisation programmes or specific antimicrobial therapy. Such arguments are of relevance here only to emphasise that the socio-economic status of a country or individual may influence the prevalence of the environmental factor or the host susceptibility to it. In practice, causes and agents are conveniently embraced by the term aetiology.

Causal associations

A causal association is a marker for the risk of developing a disease, but it is not necessarily the actual cause of the disease. The stronger the causal association, the more likely it is to be the aetiology of the disease. Causal associations become more powerful if:

The utility of these statements is exemplified by reference to the association between lung cancer and cigarette smoking. Lung cancer is more common in smokers than in non-smokers; tobacco yields carcinogenic chemicals; the risk of lung cancer is proportional to cigarette consumption; population groups that have reduced their cigarette consumption (e.g. doctors) show a commensurate reduction in their risk of lung cancer.

Causal associations may be neither exclusive nor absolute. For example, because some heavy cigarette smokers never develop lung cancer, smoking cannot alone be regarded as a sufficient cause; other factors are required. Conversely, because some non-smokers develop lung cancer, smoking cannot be regarded as a necessary cause; other causative factors must exist.

Causal associations tend to be strongest with infections. For example syphilis, a venereal disease, is always due to infection by the spirochaete Treponema pallidum; there is no other possible cause for syphilis; syphilis is the only disease caused by Treponema pallidum.

Koch’s postulates

An infective (e.g. bacterial, viral) cause for a disease is not usually regarded as proven until it satisfies the criteria enunciated by Robert Koch (1843–1910), a German bacteriologist and Nobel prizewinner in 1905:

The last was added subsequently to Koch’s list. Although Koch’s postulates have lost their novelty, their relevance is undiminished. However, each postulate merits further comment because there are notable exceptions: