What is the ratio of the incidence of the disease in the exposed to that of the non-exposed?

Explanation

## Correct Answer: A. Relative risk **Relative Risk (RR)** is the ratio of the incidence of disease in the exposed group to the incidence in the non-exposed group. Mathematically: RR = Incidence in exposed / Incidence in non-exposed. This is the fundamental epidemiological measure used in **cohort studies** to quantify the strength of association between an exposure and an outcome. The question explicitly asks for "ratio of incidence in exposed to non-exposed," which is the textbook definition of RR. An RR of 1 means no association; RR > 1 indicates increased risk with exposure; RR < 1 indicates protective effect. For example, if the incidence of tuberculosis in healthcare workers exposed to TB patients is 50 per 1000 person-years and in non-exposed is 5 per 1000 person-years, RR = 50/5 = 10, meaning exposed workers are 10 times more likely to develop TB. This is the standard measure reported in Indian epidemiological surveys (e.g., RNTCP surveillance data) and is the primary outcome measure in prospective cohort studies used to establish causation in occupational and environmental health research in India. ## Why the other options are wrong **B. Odds ratio** — Odds ratio (OR) is the ratio of odds of disease in exposed to odds in non-exposed, NOT the ratio of incidences. OR is used in **case-control studies** where incidence cannot be directly measured because sampling is based on disease status. While OR approximates RR when disease is rare, the question specifically asks for incidence ratio, which is RR by definition. OR is calculated from a 2×2 table as (a×d)/(b×c), fundamentally different from the incidence ratio requested. **C. Population attributable risk** — Population attributable risk (PAR) is the difference in incidence between the total population and the non-exposed group, expressed as the proportion of disease in the population attributable to the exposure. It answers 'what proportion of disease in the population is due to this exposure?' — not the ratio of incidences. PAR depends on both the strength of association AND the prevalence of exposure in the population, making it a population-level measure, not an individual-level ratio. **D. Attributable risk** — Attributable risk (AR), also called risk difference, is the **difference** between incidence in exposed and non-exposed (Incidence exposed − Incidence non-exposed), not their ratio. AR answers 'how much additional risk does exposure confer?' and is expressed in absolute terms (e.g., 45 per 1000 person-years). The question explicitly asks for a ratio, not a difference, eliminating this option. ## High-Yield Facts - **Relative Risk (RR)** = Incidence in exposed / Incidence in non-exposed; used in **cohort studies** to measure association strength. - RR = 1 (no association), RR > 1 (increased risk), RR < 1 (protective effect); RR of 10 means 10-fold higher incidence in exposed group. - **Odds Ratio (OR)** is used in **case-control studies** where incidence cannot be measured; OR ≈ RR when disease is rare (< 10% prevalence). - **Attributable Risk (AR)** is the absolute difference in incidence (exposed − non-exposed), measured in cases per person-years; answers 'excess cases due to exposure.' - **Population Attributable Risk (PAR)** = proportion of disease in total population attributable to exposure; depends on both RR and prevalence of exposure in population. ## Mnemonics **RR vs OR vs AR** **R**elative Risk = **R**atio (cohort); **O**dds **R**atio = case-control; **A**ttributable **R**isk = difference. Use RR when you can measure incidence directly in exposed and non-exposed groups prospectively. **When to use each measure** **Cohort study** → RR (follow exposed & non-exposed forward); **Case-control study** → OR (identify cases & controls, look backward); **Public health planning** → PAR (what % of disease can we prevent by removing exposure). ## NBE Trap NBE pairs "odds ratio" with "ratio of disease" to trap students who confuse OR with RR. The key discriminator is that the question explicitly asks for "ratio of **incidence**" — incidence is measured prospectively in cohort studies (RR), while odds are calculated from case-control designs (OR). ## Clinical Pearl In Indian occupational health surveillance (e.g., silicosis in miners, asbestos-related disease in construction workers), RR is reported to quantify excess disease burden in exposed workers compared to unexposed controls. An RR > 1 triggers regulatory action under the Occupational Safety, Health and Working Conditions Code, 2020. _Reference: Park's Textbook of Preventive and Social Medicine, Ch. 9 (Epidemiology); Harrison's Principles of Internal Medicine, Ch. 3 (Epidemiology)_