Correct Answer: C. Green tea reduces the risk of diabetes
A risk ratio (RR) of 0.84 in a cohort study directly indicates the magnitude and direction of association between the exposure (green tea consumption) and the outcome (diabetes mellitus). The RR is calculated as the ratio of incidence in the exposed group to the incidence in the unexposed group: RR = Risk(exposed) / Risk(unexposed). When RR < 1, the exposed group has lower risk than the unexposed group, indicating a protective effect. Specifically, an RR of 0.84 means drinkers of green tea have 84% of the risk compared to non-drinkers, or equivalently, a 16% risk reduction (1 − 0.84 = 0.16). This is a statistically meaningful reduction in relative risk. In the Indian context, where diabetes prevalence is rising (IDF 2021 data: ~77 million diabetics), identifying modifiable protective factors like green tea consumption is clinically relevant. The RR value alone, when <1, establishes the direction of association; whether it is statistically significant depends on the 95% confidence interval (CI), but the question asks about the interpretation of the point estimate itself. An RR of 0.84 with a CI not crossing 1.0 would establish causality at the epidemiological level, though biological plausibility and dose-response relationships strengthen causal inference per Bradford Hill criteria.
Why the other options are wrong
A. Data insufficient to establish causal association — This is incorrect because a cohort study with a calculated RR of 0.84 provides sufficient data to establish an association (not causation, which requires additional criteria). The RR value itself is derived from the study data and directly quantifies the relationship. While causation requires Bradford Hill criteria (dose-response, temporal sequence, biological plausibility, consistency), the question asks about establishing association, which the RR clearly does. Confusing association with causation and then claiming data insufficiency is an NBE trap. B. The value (0.85) tends to be close to 41, hence there is no effect — This option contains a mathematical error and logical fallacy. First, the RR is 0.84, not 0.85 (minor typo in the option). Second, the statement 'close to 41' is nonsensical—0.84 is nowhere near 41. This appears to be a distractor testing whether students blindly accept illogical statements. The correct interpretation is that RR = 0.84 is significantly different from 1.0 (null value), indicating a real effect. This is a classic NBE trap using mathematical confusion. D. Green tea increases the risk of diabetes — This is incorrect because an RR < 1 indicates a protective effect, not an increased risk. An RR > 1 would indicate increased risk. Students who misinterpret the RR scale or confuse the direction of association may select this. In Indian epidemiological studies on traditional beverages (e.g., ICMR studies on tea consumption), RR < 1 consistently indicates protective associations. This option reverses the correct interpretation and is a common distractor for students unfamiliar with RR interpretation.
High-Yield Facts
- Risk Ratio (RR) < 1 indicates a protective effect; RR > 1 indicates increased risk; RR = 1 indicates no association (null value).
- RR of 0.84 means the exposed group has 84% of the risk of the unexposed group, equivalent to a 16% relative risk reduction.
- Cohort studies measure incidence directly and calculate RR; case-control studies calculate odds ratio (OR) as a proxy for RR.
- Bradford Hill criteria (temporal sequence, dose-response, biological plausibility, consistency) are needed to infer causation; RR alone establishes association.
- Confidence interval crossing 1.0 indicates statistical non-significance; if CI does not cross 1.0, the RR is statistically significant.
Mnemonics
RR Interpretation Rule RR < 1 = Protective (Risk Reduced); RR > 1 = Harmful (Risk increased); RR = 1 = Null (No effect). Use this when interpreting any relative risk value in seconds. Cohort Study Memory Hook 'Cohort Counts Incidence' — Cohort studies follow people forward in time and directly measure incidence rates, allowing calculation of RR. Case-control studies work backward and use OR.
NBE Trap
NBE pairs a mathematically nonsensical distractor (option B: "close to 41") with a plausible-sounding but incorrect option (option D: increased risk) to test whether students understand RR scale and direction. Students unfamiliar with RR interpretation may be lured into option A (claiming insufficient data) by conflating association with causation.
Clinical Pearl
In Indian clinical practice, green tea's protective effect against diabetes (RR 0.84) aligns with traditional Ayurvedic and modern epidemiological evidence. This finding is particularly relevant for Indian populations with high diabetes prevalence, making green tea a low-cost, culturally acceptable preventive intervention alongside lifestyle modification per ICMR guidelines.
_Reference: Park's Textbook of Preventive and Social Medicine, Ch. 8 (Epidemiology); Harrison's Principles of Internal Medicine, Ch. 3 (Epidemiology and Evidence-Based Medicine)_