As we progress through this part and the next, you will be introduced to the different ways in which epidemiologists go about analysing the factors that are associated with people becoming ill or getting better. Each of these has a role to play in building up our knowledge about what influences human health. Our objective here is to provide an overview of the range of techniques that are available and to develop your understanding of which of these might be more appropriate in any given situation. One way to think about these techniques is as a set of tools for tackling a range of problems, much as a carpenter has a box full of tools for tackling different aspects of building a house. No one tool is useful in every situation, and some are more useful at certain stages of the construction process than others. Some even have features that make them useful in a variety of situations. Of course, context is everything, so even when a tool might not look like it’s the ‘right’ one in a particular situation, if the results are robust and reliable then that might be all that matters.

So, here we are at the final chapter, and at this point you might be minded to ask ‘So what?’ Although some of you may have found this book to be so compelling that you have decided to become an epidemiologist, it is likely that most of you will looking for other ways for this epidemiology stuff to value add to your health science learning and ongoing professional or academic lives. In modern life, we are deluged with health information that is provided in multiple formats, including social media, news websites, online videos, televised news bulletins and chat shows, and even academic texts and other published literature. How are we to find something approaching the truth in this plethora of often contradictory information? In its focus on epidemiology, this book has aimed to provide you with the tools for evaluating scientific information using critical thinking – a way of identifying and evaluating evidence that has wide applicability to just about every area of human endeavour.

Epidemiology is the study of patterns and determinants of disease and other health states in populations. It primarily uses quantitative methods (those methods dealing with counting, measuring and comparing things) that definitely use statistics and include statistical methods, but in this book we will not be talking about performing any statistical acrobatics more complicated than completing a sudoku puzzle.

Arriving at evidence-based solutions requires strong evidence. Usually, this evidence will be derived from quality research, such as is often published in reputable scientific journals. But how do we know whether even these studies are good through and through? There is always the potential that pesky flaws, such as bias and confounding, might can beset even the most (otherwise) perfect of studies. This is why the methods taken to avoid bias and confounding are always well-described in all good published studies, as is the potential for remaining sources of error for which the design is (inevitably) unable to account, but which might still influence findings. There is always a bit of uncertainty about any evidence provided by studies and, to add to this, the very real possibility that we are not getting the full story at all times. In a phenomenon known as ‘publication bias’, even really high quality studies may not get published if they report non-significant results.

In this chapter, we are moving across to the experimental branch of the epidemiological research tree to focus specifically on randomised controlled trials (RCTs). These are often referred to as the ‘gold standard’ of health research designs, and are ranked pretty much at the top of the hierarchy of scientific evidence. Representations of the hierarchy of evidence in the form of pyramid are more or less ubiquitous on the internet and in textbooks.

They were great for settling questions of logical truth, validity, equivalence, and so on, but became unwieldy in an exponential hurry as the number of relevant atomic sentences increased. They also foundered on the rocks of $\mathcal{Q}\mathcal{L}$’s non-truth-functional constructions.