A trial reports a relative risk of 0.85 with a 95% confidence interval of 0.72 to 1.02. How should this be interpreted? A. The treatment significantly reduces risk B. The result is not statistically significant at the 5% level C. The treatment significantly increases risk D. The study proves the treatment has no effect ANSWER: B — not statistically significant at the 5% level. Why: this is a RATIO, so the no-effect value is 1. The interval runs from 0.72 to 1.02 and therefore CROSSES 1, so an effect of exactly zero remains compatible with the data. D is wrong and is the more important error: failing to demonstrate an effect is not the same as demonstrating there is none. The interval also includes a 28% reduction, which would matter clinically if real.

A drug reduces the risk of an outcome from 0.8% to 0.4%. Calculate the relative risk reduction, the absolute risk reduction and the number needed to treat. ANSWER: Relative risk reduction = (0.8 − 0.4) / 0.8 = 50% Absolute risk reduction = 0.8% − 0.4% = 0.4 percentage points NNT = 1 / 0.004 = 250 INTERPRETATION: the headline "halves the risk" is accurate, but 250 patients must be treated for one to benefit — and all 250 are exposed to the adverse effects and the cost. LEARNING POINT: benefits are advertised in relative terms and harms in absolute terms. Ask for the other number.

Which measure of central tendency best describes hospital length of stay, where most patients leave within days but a few stay for months? A. Mean B. Median C. Mode D. Standard deviation ANSWER: B — the median. Why: the distribution is SKEWED by a few very long stays. The mean is dragged upward by those outliers and ends up describing almost nobody. The median is the middle value and is barely affected by extremes, so it describes the typical patient. D is not a measure of central tendency at all — it describes spread.

Define the infant mortality rate and the maternal mortality ratio, with their denominators. ANSWER: INFANT MORTALITY RATE Deaths under 1 year of age per 1000 LIVE BIRTHS in the same year. MATERNAL MORTALITY RATIO Maternal deaths per 100,000 LIVE BIRTHS. TRAP: note the DIFFERENT MULTIPLIER. Infant deaths are per 1000; maternal deaths are per 100,000, because they are far rarer and a per-1000 figure would be an unhelpful decimal. WHY IMR IS USED SO WIDELY: it is sensitive to maternal health, nutrition, sanitation, immunisation and access to services all at once, so it indexes general development rather than medical care alone.

Why is an odds ratio used in case-control studies rather than a relative risk? ANSWER: because incidence cannot be measured in a case-control study. The investigator CHOOSES how many cases and how many controls to recruit, so the proportion of diseased people in the study is an artefact of the sampling, not a property of the population. Risk therefore cannot be calculated, and neither can a risk ratio. The odds ratio can be calculated from the same data, and when the disease is RARE it approximates the relative risk closely. When the disease is common, the odds ratio exaggerates the effect relative to the risk ratio.

A very large study finds a difference in mean blood pressure of 0.4 mmHg between two groups, with p = 0.001. How should this be interpreted? ANSWER: statistically significant, clinically meaningless. A large sample makes even a trivial difference detectable. The small p-value says the difference is probably not exactly zero; it says nothing about whether the difference matters. A 0.4 mmHg change in mean blood pressure has no bearing on any clinical decision. LEARNING POINT: statistical significance and clinical importance are separate questions. Always ask for the SIZE of the effect and its confidence interval, not the p-value alone.

In the demographic transition, why does a population grow rapidly during the middle phase? ANSWER: because death rates fall BEFORE birth rates do. 1. Sanitation, nutrition and infection control reduce mortality, especially in childhood. 2. Fertility stays high — family size responds slowly, and depends on child survival, women's education and access to family planning. 3. The GAP between a fallen death rate and a still-high birth rate is where rapid growth happens. 4. Birth rates fall later, and growth slows. LEARNING POINT: rapid growth in a developing country usually signals that MORTALITY HAS IMPROVED, not that fertility has risen.