Benefits of Epidemiological Studies

Q1, Risk data

1.1 Exposure X: 1 in 100,000 × 4.0 = 4 in 100,000, that is 0.004% [1]. Exposure Y: 1 in 500 × 1.3 = 1.3 in 500, that is 0.26% or about 2.6 in 1,000 [1].

1.2 Relative risk describes how much an individual's risk multiplies, not how many cases occur [1]. Exposure X multiplies a very small baseline risk and affects only 0.5% of the population, so it generates few cases in total. Exposure Y multiplies a much larger baseline risk and affects 55% of a population of two million, so even a modest 1.3-fold increase applied to more than a million people produces far more cases overall [1]. Because physical inactivity is also modifiable at population scale, removing it would prevent a much larger number of cases, which is what a prevention budget is trying to achieve [1].

1.3 The measure is attributable risk [1]. It estimates the share of cases in the population that would be prevented if the exposure were removed [1], which is why a small relative risk on a common exposure can outrank a large relative risk on a rare one [1].

Q2, Melanoma incidence graph

2.1 In the under-40 group, incidence rises from about 45 per 100,000 in 1985 to a peak of about 55 per 100,000 around 1997, then falls steadily to about 33 per 100,000 by 2020 [1]. In the 65-and-over group, incidence rises continuously across the whole period, from about 40 per 100,000 in 1985 to about 117 per 100,000 in 2020 [1]. The two trends diverge from the late 1990s onward [1]. Accept values read reasonably from the graph.

2.2 Melanoma develops decades after the ultraviolet exposure that causes it, because UV damage to DNA in skin cells accumulates and the disease appears much later in life [1]. Australians who were already 65 or over during the study period had their formative sun exposure before the campaign began in 1981, so a campaign launched in 1981 could not have altered the exposure that is now producing their disease [1]. The rising older-age curve therefore reflects historical exposure rather than campaign failure, and judging the campaign on it would be reading the data before the expected lag has passed [1].

2.3 The pattern is strong evidence precisely because the two age groups move in opposite directions in the way the biology predicts [1]. If the fall in the under-40s were caused by something general, such as changed diagnostic practice or reporting, it should have affected both age groups similarly [1]. A decline confined to the cohort young enough to have been reached by the campaign, alongside a continued rise in the cohort that was not, matches the predicted exposure-to-disease lag and is difficult to explain by an alternative cause [1].

Q3, Folate cause-and-effect chain

Step 2. Folate supplies the one-carbon groups used to build nucleotides, so a shortage limits the rapid cell division required to close the neural tube, which happens within the first four weeks of development [1 to 2].

Step 3. The neural tube closes before most people know they are pregnant, so advice directed at pregnant women arrives too late, and it cannot work at all for an unplanned pregnancy [1 to 2].

Step 4. Mandatory folic acid fortification of bread-making flour, introduced in 2009, which removes the need for any individual action [1].

Step 5. Neural tube defect rates fell after fortification, and the largest falls occurred in the groups with the highest rates beforehand, including teenage mothers and Aboriginal and Torres Strait Islander women, so the benefit and the equity test passed together [1 to 2].

Maximum 6 marks.

Q4, Bowel cancer screening

4.1 Any three, 1 mark each: the disease is common and serious enough that finding cases matters at population scale; it has a detectable early stage, developing slowly from polyps, so there is a window in which the test can find it before symptoms; early treatment gives a better outcome than late treatment, since polyps can simply be removed; and the test is cheap and acceptable enough that people will actually complete it, which is a precondition for any benefit at all.

4.2 Participation near 40 percent means roughly six in ten people invited do not return a test, so the program can only deliver a fraction of the benefit it would achieve at full uptake [1]. A screening program's benefit depends on the whole chain holding, and implementation is the link failing here [1]. Because participation is lower again in remote areas and in the lowest income group, the benefit is distributed unequally, and those groups often carry a higher disease burden or face longer delays to diagnosis [1]. A national average of 40 percent conceals this: an equity analysis reporting participation separately by region and income is required to show whether the program is narrowing or widening the existing health gap [1].

Q5, Comparison table

What it counts. Incidence: the number of new cases arising in a defined period [1]. Prevalence: the number of people living with the disease at one point in time, including both new and long-standing cases [1].

Planning question. Incidence: how fast is fresh demand arriving, and is prevention working [1]. Prevalence: how many people need ongoing care, staffing and services right now, for example how many dialysis chairs must be funded [1].

What it cannot tell you alone. Incidence: it says nothing about the accumulated burden of people already living with the disease, so a planner using it alone could cut services while demand grows [1]. Prevalence: it cannot show whether prevention is succeeding, because it rises when survival improves as well as when prevention fails [1].

Falling incidence with rising prevalence. This usually means treatment improved so people live longer with the disease, while prevention is simultaneously reducing new cases. It is a sign of success on both fronts, not of prevention failing, and it means prevention funding and ongoing-care funding must be judged on different measures [2].