Which of the following data sources do you think would provide the best data to evaluate how well a national primary prevention programme of behavioural change
To decrease the mortality rate from a cancer is working?
A)A cohort study
B)A prevalence survey in the community to measure behaviour change
C)A large‐scale randomised trial
D)Routinely collected mortality data
E)Incidence rates from a cancer registry

Question

Quizplus · Accepted Answer

These data directly address the aim of the programme (have mortality rates
fallen?), but if other control measures (e.g.a new treatment) have also been introduced
it can be difficult to determine the specific contribution of the prevention programme
per se.E gives a direct measure of the effect of primary prevention on the occurrence
of new disease, so a better judgement can be made by looking at time trends in both
mortality and incidence; if the former parallels the latter after an appropriate lag time
(the average survival time post‐diagnosis) then the mortality decrease may well be
largely due to the prevention strategy.B can be a useful process measure of whether
there is good uptake of the behavioural change - if there is no or minimal change then
there is little basis for continuing; but showing behaviours have changed (if that is what
is found) is not enough to declare the programme a success until disease and mortality
rates drop.In the context of a national programme, it would not be ethical to conduct a
randomised trial (C) and a cohort study (A) would almost certainly be subject to
confounding because those who changed their behaviour would almost certainly differ
in other ways from those who did not change (i.e.the groups would not be
exchangeable).
Note - in practice this decision might not be straightforward.For a more complex
situation see e.g.Knox, K.L., Pflanz, S., Talcott, G.W., Campise, R.L., Lavigne, J.E., Bajorska,
A., Tu X., Caine, E.D.(2010).The US air force suicide prevention program: Implications for
public health policy.American Journal of Public Health, 100: 2457‐2463.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2978162/

Which of the Following Data Sources Do You Think Would