Narinder Gosall
It is important that as an MSL that you develop a trusted relationship with your KOLs so that you can work in partnership to deliver patient centred evidence based healthcare. Transparent, objective and evidence based discussions are essential to ensure KOLs receive the right information for their patients. This is why critical appraisal skills are fundamental for MSL’s to allow these evidence based discussions. Here are some key tips to consider when having these discussions:
Why do I need to critically appraise an article that is already published in a prestigious journal?
All academic journals have different requirements from authors and publish in many different therapy areas. As an MSL, it is important to be knowledgeable about your study rationale and how the new data adds value. It is vital to be credible in the way you present and interpret the information. Critical appraisal allows you to objectively assess the validity of the study and assess its usefulness and applicability.
What is peer reviewing and isn’t that sufficient?
The peer review process is widely accepted as a method to validate the research being published by ensuring quality and integrity. Critically appraising the published article allows you to assess the finer detail so that you can be better informed for decision making.
Is critical appraisal an attempt at dismissing the study findings?
No. Critical appraisal skills should be a balanced assessment of the research process and results, as well as highlighting strengths and limitations to assess the validity of the research. Critical appraisal is not designed to be dismissive of research, it focuses on assessment of evidence to ensure the data is reliable and valid.
How can critical appraisal of an article help my KOL?
This would allow your KOL to distinguish evidence from opinion and consider the clinical applicability of the research to their own patients. Different study designs have their own challenges and appraising a study can allow your KOL to see how some of those challenges were met, why the study design was used and what the clinical and statistical significance of the research indicates for their patients. As a result, KOLs can keep abreast of the evidence being published and be better informed for their own decision making.
How much detail should I know?
Ensure you are aware of strengths, limitations and any relevant information for your study. This may or may not be in your study slides so it is important that you have assessed the actual clinical paper so that you have your own understanding of the study. If there is an appendix or supplementary document with relevant information then it is important that you have also assessed that resource. Some studies may be published online and have a link to the study protocol; for some healthcare professionals this may be important to mention.
What about the all the statistical analyses?
Statistical techniques that have been applied should be specified in the publication and it is important to have an understanding of why these analyses were used. However, it isn’t necessary for you to be an expert in how these were performed. You are not seeing your KOL in the capacity as the statistician who conducted the analysis. Therefore, it is perfectly fine to say that you can provide further information from your organisation, if necessary.
How do I present a non-significant study?
Transparency is key. It is vital that the study is presented in an unbiased and objective manner and if there were challenges in the study design that led to the non-significant results, these can be discussed. Nobody designs a study to fail but dealing with real patients in real situations means that bias and confounders are inevitable. Have a good understanding of the additional literature for your treatment and ensure you can present the treatment in the context of all the evidence you have.
What should my conclusion be for a statistically significant study?
The term ‘statistical significance’ is often misinterpreted as a ‘clinically important’ result even when it isn’t. It is important to consider the external generalisability of the findings and consider the clinical relevance of the findings and not just focus on the P value. Ensure that the P-values do not neglect the estimation of effects.