Tomorrow's Pharmacist October 1999 p67-68
Edited by Pamela Mason

Critical reading

The paper chase - journal papers

By Michelle Styles, BSc, MRPharmS

Reading - and being critical about what you read - is a vital skill and one of the most important ways of keeping yourself up to date. But how do you know whether what you are reading is sound or not? Read on to find out.

Many students believe that once they have left university and qualified as pharmacists they will never have to read another research paper or visit a library again. That may have been the case in the past, but nowadays, whatever the sphere of practice, you will find when you enter practice that to keep abreast of developments in medicine and science, it is simply not enough to skim through The Pharmaceutical Journal or read the edited highlights from the major medical journals.
Increasing numbers of pharmacists are working with nurse prescribers, GP practices and primary care groups on prescribing issues, and it is absolutely essential to keep your knowledge up-to-date in order to retain credibility. We all know that the information in textbooks is out-of-date almost as soon as the book is published because new research is being carried out all the time, and so it is essential to review the latest thinking, and hence, the latest research papers, before undertaking prescribing advice. You are likely to be asked to cite - and perhaps even defend - the evidence on which your recommendations are based.
In addition to developing prescribing guidelines, pharmacists are often asked specific clinical questions, such as "can drug A be used in condition B?" again, research papers are often the most useful source of up-to-date information.
How does a pharmacist decide which OTC medicines to recommend? Is the advice given based on evidence from the latest papers or does he or she recommend what they have always recommended or what their customers seem to like?
Health is news! The media knows that stories on health, particularly those that indicate danger of some sort, sell newspapers and increase viewing figures. As the most accessible healthcare professionals, pharmacists are frequently asked for their opinions on drugs and products mentioned in magazines and on television and so it is essential to be well informed and to have easy access to good information sources.
It has been estimated that a GP would need to read 19 articles every day just to keep abreast of changes in practice and the same can be said for pharmacists. It is easy to become overwhelmed with the sheer volume of information out there, but if a systematic approach is taken, the task becomes more manageable. As pharmacists become busier and the number of papers and journals increases, the ability to select those papers of greatest importance and read them quickly, yet critically, will become an essential skill, so where do you start?

Which journals?

Firstly, you need to decide which journals you are going to read regularly, taking into account subscription costs and the time needed for reading. Be sensible - you cannot possibly expect to read 15 journals a week and properly appraise and reflect on the information contained in them.
If you do not have access to or cannot afford to subscribe to lots of medical journals, many such as The New England Journal of Medicine, the Journal of the American Medical Association (JAMA) and The Lancet have websites which allow free access to abstracts from the main papers. You can even have the headlines from each week's current issue sent to you by e-mail as soon as the journal is published. Important and relevant papers can then be ordered from the Royal Pharmaceutical Society's library for a small fee.
We all read tabloid newspapers with a healthy degree of scepticism, yet when articles are published in learned medical journals, we assume that because they have been edited by experts and subjected to external peer review the information in them is "true". This is not always the case, as many studies, even those based on good ideas or presenting interesting findings, are scientifically unsound, particularly in their methodology and so the results cannot be relied on. With experience, you will quickly learn which journals publish good quality papers and which publish consistently poor ones.

Types of papers

Secondly, you need to be aware of exactly what sort of paper you are reading. Most of the studies published in medical journals are primary studies, for example, surveys or clinical trials of drug treatments. However, journals are publishing more and more secondary research, usually systematic reviews and meta-analyses, which summarise the results of several primary studies and formulate an overall conclusion. Indeed, there are now some dedicated journals available which publish only secondary research.
Most research journals publish studies in a standard format with a title and abstract, followed by the introduction, method, results and conclusion. A brief look at the title and abstract will tell you whether the study is of interest to your particular practice and you can quickly identify those articles you need to read more thoroughly. The introduction tells you why the authors decided to carry out the research, the method tells you how they carried out the study, results tell you what they found and the conclusion tells you what they think the results indicate.

How to read papers

It is often worth checking the authors of the paper. Well respected authors have a reputation to maintain and are less likely to be sloppy in their methodology. How many authors are mentioned (studies with a lot of assessors are more likely to have large variations in methodology)? Has a statistician been included? When you are short of time, it is very tempting to read only the conclusions of a paper, but to do so would be unwise. Before you let a paper influence your clinical decisions, you need to look carefully at how the study was designed and carried out (ie, at the methods section). It is always interesting to read the "Letters to the editor" pages of research journals as these often reveal methodological flaws in previously published papers.
The checklist below is intended to provide a general starting point for those new to assessing the methodology section of a paper. However, do not be put off by it - you may not find it necessary to apply all these questions to every paper you read. Often, common sense and experience will tell you whether a paper is likely to be worth reading.
When you have judged that a paper is scientifically sound, you need to decide whether it is one that should change practice. Obviously, this is the whole point of reviewing papers! However, it can be daunting to undertake such a decision alone, particularly if you are unfamiliar with the topic investigated. In such cases, you may find it helps to talk to colleagues and to discuss the matter with prescribers. And remember, confidence will come with experience and practice.

Further reading

Check List

• What type of study was carried out? (eg, is it a clinical trial, a meta-analysis, or a cohort study?)
• Who funded the research?
• How were patients recruited to the study? (random recruitment is best)
• Was the population studied relevant? (eg, was a drug to be marketed for use in the elderly instead studied in healthy young adults?)
• Which patients were included in the study and how representative are they of the patients you see? Remember that most of the patients seen by practising pharmacists will have more than one co-existing illness and may smoke, drink alcohol, take drugs and so on, yet most studies exclude such subjects.
• Were the treatments allocated randomly? How was the randomisation performed?
• Were the patient groups truly comparable? Were patients matched for age, gender, smoking status and so on?
• Was the sample size sufficient and does the paper give a sample size calculation?
• Who assessed the outcomes? If patients did this, then expect wide variation in results.
• How many assessors were used?
• What outcome was measured and was it relevant? For example, the blood levels of a drug may not necessarily equate with its efficacy in a certain disease state.
• Was the assessment of outcome blind? Were the subjects and their assessors ‘blind' to who was receiving which treatment?
• If a new treatment is being compared with an existing one, was the existing treatment used at the optimum dose?
• Are the statistical methods used described in the paper?
• Are the reasons for subject drop-out described?
• Do the numbers concur? Have all patients been accounted for at the end of the study?
• What references are cited at the end of the paper? Beware of "data on file" references as these are usually unpublished.

Mrs Styles is head of information services at the National Pharmaceutical Association

Tomorrow's Pharmacist is an annual publication produced within the editorial department of The Pharmaceutical Journal