Define epidemiology and describe its aims, uses and scope.

Definition, Components and Aims

• Every phrase of that definition carries meaning, and unpacking it is what a good answer does:
• Distribution — the descriptive component, concerned with time, place and person: WHO gets the disease, where and when
• Determinants — the analytical component, concerned with why and how: the factors that influence the occurrence of disease
• Health-related states or events — deliberately broad; epidemiology is not confined to infectious disease, and now encompasses chronic disease, injuries, behaviour, health services and positive health
• Specified populations — the defining feature of the discipline. The unit of study is the population or group, not the individual patient, and a denominator is always required
• Application to control — epidemiology is an applied science with a purpose, not merely an observational one; the point of measuring is to intervene
• The three components to be remembered — distribution, determinants and application (or, as often taught, frequency, distribution and determinants)
• The aims of epidemiology (International Epidemiological Association):
• 1. To describe the distribution and magnitude of health problems in human populations
• 2. To identify the aetiological factors in the pathogenesis of disease
• 3. To provide the data essential for the planning, implementation and evaluation of services for prevention, control and treatment, and for setting priorities
• The uses of epidemiology, listed by MORRIS, are examined directly and should be known as a set:
• To study the historical trends of disease in a community, which reveals the epidemiological transition
• Community diagnosis — identifying the health problems and needs of a defined community, which is the basis of all planning
• To plan and evaluate health services
• To evaluate individual risk and chances, which is what allows a clinician to give a prognosis
• To identify syndromes, by defining the distribution and association of clinical phenomena
• To complete the clinical picture, by revealing the full spectrum of disease including the mild and subclinical forms the clinician never sees
• To search for causes and determinants of health and disease
• To work the health services, meaning to operate them on the basis of evidence

Approach, Tools and Scope

• The epidemiological approach rests on two simple questions, and stating them is the clearest way to convey what the discipline does:
• 1. Ask questions — what is the condition; WHO is affected; where; when; why and how; and what can be done about it
• 2. Make comparisons — comparison is the fundamental tool of epidemiology. A rate has no meaning in isolation; it acquires meaning only when compared with another group, another place or another time
• The three broad types of epidemiological study:
• Descriptive — describing the distribution by time, place and person, and generating hypotheses
• Analytical — case-control and cohort studies, which test hypotheses
• Experimental (interventional) — randomised controlled trials and field and community trials, which confirm causal relationships and evaluate interventions
• The logical sequence is descriptive → analytical → experimental, with each stage narrowing the question
• Observational versus experimental is the fundamental division — in observational studies the investigator observes what happens; in experimental studies the investigator allocates the exposure. Only experimental studies control for unknown confounders, through randomisation
• The tools of epidemiology — rates, ratios and proportions, which are the basic measurements; and a denominator, without which numbers are uninterpretable
• The importance OF the denominator cannot be overstated — "200 deaths from a disease" means nothing until we know 200 out of how many, over what period; and much misleading reporting of health statistics consists precisely of numerators without denominators
• The sources of epidemiological data — census; registration of vital events; sample registration system; notification of disease; hospital and clinic records; disease registries, such as cancer registries; record linkage; epidemiological surveys such as NFHS; surveillance systems (IDSP); environmental and occupational data; and demographic and health service statistics
• The scope of modern epidemiology extends far beyond infection — chronic disease epidemiology; clinical epidemiology, which applies these methods to the care of individual patients and underlies evidence-based medicine; genetic and molecular epidemiology; nutritional, occupational and environmental epidemiology; pharmaco-epidemiology; social epidemiology; health services research; and epidemiology of injuries and of mental health
• The limitations must be acknowledged in a good answer — epidemiology demonstrates association, not causation, and inference requires judgement; observational studies are vulnerable to bias and confounding; the ecological fallacy arises when conclusions about individuals are drawn from group data; data quality limits every conclusion; and statistical significance is not the same as clinical or public health importance
• The ecological fallacy deserves an example — a country with high average fat consumption may have high heart disease mortality, but that does not establish that the individuals eating the fat are the ones dying
• The ultimate purpose remains prevention and control — and an epidemiological finding that does not eventually change practice or policy has not completed its work
• The population is the unit of study, which is the single feature separating epidemiology from clinical medicine
• Comparison is the fundamental tool; a rate in isolation means nothing until set beside another group, place or time
• Never accept a numerator without a denominator, which is the form most misleading health statistics take
• Distribution, determinants and application are the three components, and the definition should be unpacked rather than recited
• Descriptive then analytical then experimental is the logical sequence, each stage narrowing the question
• Only experimental studies control unknown confounders, through randomisation, which no analysis can achieve
• Epidemiology completes the clinical picture by revealing the mild and subclinical disease the clinician never sees
• Community diagnosis is the basis of planning, identifying the health problems and needs of a defined population
• The ecological fallacy draws individual conclusions from group data, and is a standard trap in international comparisons
• Association is not causation, and epidemiology demonstrates the first while inference to the second requires judgement
• Statistical significance is not importance, clinical or public health, and conflating them misleads policy
• Clinical epidemiology underlies evidence-based medicine, applying population methods to the care of individual patients
• Epidemiology is not confined to infection, and now covers chronic disease, injuries, behaviour and health services
• The Sample Registration System is a key Indian source, alongside the census, NFHS and the surveillance programme
• Cancer registries provide incidence data that no hospital return can, since they are population-based
• Record linkage joins data across sources, and multiplies the value of each individual dataset
• Data quality limits every conclusion, however sophisticated the analysis applied to it
• The purpose is prevention and control, and a finding that never changes practice has not completed its work
• Ask what, who, where, when, why and how, which is the epidemiological approach reduced to its simplest form
• Epidemiology identifies syndromes by defining the distribution and association of clinical phenomena
• It evaluates individual risk, which is what allows a clinician to give a prognosis at all
• Historical trends reveal the epidemiological transition, which is one of Morris uses and shapes national planning