How to Read Science and Medicine Papers

A great deal of scientific material is now readily available for free but it can be difficult to read.  Scientific papers are important because they present evidence which we can weigh up when considering health and food claims.

For example, if someone claims that Goji berries improve the immune system, we can consult sites like PubMed  and read the scientific studies for ourselves.  There may be several such papers and we can sort through the evidence to decide whether or not the claim is credible.

Unfortunately, scientists have to adopt a certain format to these papers so that other scientists can quickly assess their relevance and content.  For the lay person, the non-scientist, this can seem rather intimidating and hard to follow but once we understand that the sections are about, we can make sense of a surprising amount of the research.

We do not need to understand all of the technical details to be able to appreciate the relevance of the work to our questions, and the significance of the results.

How scientific papers are structured
The paper will start with a brief introduction usually in fairly clear language which puts across the main purpose of the study and what will be investigated.  This is a good section for us to see if the paper is relevant to our question.

Typically it will say what was investigated and in general terms what was found.

The Methods section
The next section is usually one called Methods which explains in some detail how many people were involved in the study, how the groups were organised and what happened to each of them, what controls were put in place to ensure the results would be accurate and representative, and so on.

This section is very important, particularly in medical science because there are many sources of bias which could render the results useless.  For example, if the sample size was too low, then the results would not be statistically significant – they could have just been caused by chance.  If the groups were not rigorously controlled, then we could not be sure that the main cause under test was responsible for the effects.

So checking the details of the study is very important.  For example, if the paper describes writing down descriptions from patients about how they feel, we know this is subjective rather than objective and therefore cannot be taken as significant evidence, no matter what the authors might claim.  Anecdotes are not evidence.

The Results section
The plain results must be presented in a clear form for any scientific paper.  This is so that other researchers can use the methods to reproduce the work and compare their own results.

The interpretation of the results is left until later in the paper, so the Results section will typically contain tables of figures, graphs, and detailed factual text.  If you are not au fait with the science itself, this section may seem a little obscure but all the details are presented so that it is clear that the results have nothing to do with the opinions or beliefs of the scientists involved.

The results form the core evidence presented in the paper and anyone with whatever beliefs they might hold, can use the evidence to argue a particular interpretation, provided that the evidence supports them.  In any case, it demonstrates that science has nothing to do with belief.

The Discussion or Comment section
This is where the scientists will explain and argue what they think the results show, and explore possible reasons.  For example, if they find a consistent effect of a particular substance on some medical condition, they may try to offer an explanation, a possible mechanism of action.

They understand perfectly the difference between a possible mechanism, and an actual proven mechanism but the discussion will offer appropriate interpretations.  Often the scientists will be very cautious about what they claim from their results.  It is common for them to argue that much more work needs to be done before they can be sure.

Often, when science is reported in the press, these cautionary comments are ignored in the interests of producing a startling headline.  But we should read these comments carefully because they change the significance of the scientific report.

Statistics are important
In any scientific paper which presents a lot of numerical results, there will be some discussion of statistics.  In particular there is an assessment of whether the results are “statistically significant”.  This is because many rash conclusions can be wrongly drawn on the basis of wild intepretations.

For example, if the incidence of an illness has gone from 1 in a million to 2 in a million, we could say “risk of X has soared by 100%” but the result may be of negligible importance.  So the statistical treatment is very important.

Often you will see something like “P<0.05”.  What this means is that it is less than 5% likely that the results could have arisen by chance.  In other words, we are at least 95% sure that the results represent some real effect beyond chance.

Scientists do this to take into account chance events so they can realistically compare the likelihood of other circumstances affecting the results.

But scientists will also look at errors as well.  If they are measuring small quantities of chemicals in the blood for example, the equipment will not be 100% accurate so the figures will often include some variation, some plus or minus figure.

All this makes reading the figures somewhat tedious but it is important to understand how reliable the scientists themselves think the work is.

The Conclusions section
Nearly all scientific papers will have a useful summary at the end which contains a clear statement of the conclusions related to the questions posed in the introduction.  It will say whether or not the original idea, the hypothesis, was supported or disproved.  It will contain valuable information about the significance of the work and an indication of how much we can trust the results.

Scientists are remarkably honest and consistent in their work and have as much interest in disproving their ideas as in finding supporting evidence.  We should look at the conclusions and accept them in the context of the authors’ advice.

Science and opinion
In the popular press it is common to hear science talked about as if it is a collection of beliefs but this is quite the opposite of how science works.  Because of the objective controlled methods of working, the opinions and beliefs of the scientists are completely irrelevant to the work.

Science has to stand on the evidence regardless of who produced it.  Whether or not a scientists believes their own conclusions, they still have to present the data, the results for public open scrutiny.  If there is any bias, there are legions of other scientists only too keen to point it out.

But there are often vigorous disputes contesting the meaning of the results and advancing alternative explanations.  The disputes throw up new experiments, new tests, new theories which drive science onwards, disproving incorrect explanations and gathering new knowledge.

We should have no problem with the idea that there are disputes between scientists, but we should recognise that where the evidence is clear, such disputes are usually about the minor details of explanations, not about the main causes.  There is massive agreement about the vast majority of scientific knowledge.