How to read clinical paper with confidence?

Return to articles

Reading a clinical paper needs a critical mind with good research knowledge and basic interpretation of statistics test to start with. It is a skill that needs to be taught, trained and practiced regularly. Setting up a journal club is a good way to nurture the skill, but having a journal club without a trained moderator is like a blind leading a blind which happens unfortunately in some institutions.

Firstly, reading a paper starts with classifying the type of paper. The papers generally are categorised into case study, cohort, case control, systematic review, survey and so on. Each study has its advantages and limitations that need to be understood first. Generally, all the studies have a common sequence which starts with introduction or background, material and methods, result and discussion. The introduction will present the summary of literature review or what is known about the particular topic and highlights the gap in the knowledge and justifies the need for another study.

The method section will discuss the type of study design, the recruitment of subjects, the restriction criteria, matching, presence of bias, randomisation and highlights potential confounders. The experiment will be illustrated in a flow chart form and explained in detail.   The statistical tests that are applied in the study will be written with reasons behind the tests.  The ethical approval and other relevant concerns will be briefly mentioned in this section.  

The method section is followed by the results. The results will be depicted in a variety of charts/graphs which may demand some knowledge of interpretation. The discussion section thereafter will assess the ability of the researcher to analyse and interpret the results in the context of the study. There will be an extensive discussion in the study limitations and future track of the study at the end.

The front page of the research paper is known as abstract. It is a summary of the paper and more of sales pitch to capture reader’s attention to read the paper completely. In reality, the abstract section is unable to summarise effectively the entire paper and need to be read with a pinch of salt. The final section of a paper is conclusion and it summarises the result of the study in a few clear, concise sentences.

For a reader to critically appraise a paper, there are three aspects that he needs to answer. First, he has to comment on the research design, namely the type of design and reason for that, restriction criteria, matching or randomisation, and whether confounders are identified and accounted for. The second aspect would be the interpretation of the study results. He should be able to interpret the table or the graphs and build a simple 2 x 2 contingency table to identify CER (control event rate), EER (experimental event rate), number needed to treat (NNT) and other relevant parameters. The final aspect is to find out the generalisability or the applicability of the paper which is to see whether we can apply the evidence in the conscientious, explicit and judicious manner in making decisions about the care of our patients.  This is the gist on how to read a paper with confidence and outwit others. Good luck.