Whether you’re working on the front line or a member of the public with an interest in healthcare, reading academic journal articles helps you separate fact from fiction.
But don’t forget, there are good journal articles and also ones that have results we should take with a pinch of salt. Whilst we would hope all scientists are only providing the best data and only telling the absolute truth it is all our responsibility to try and educate ourselves and others.
As a doctor there is a good chance you were taught how to appraise these at medical school, but otherwise it all seems impossible, wordy and too complicated to even begin.
Fear not! Every new skill starts this way. Of course we will continue to summarise pertinent ones as best as we can but going to the source is always the most robust way to proceed.
1. Where do I find articles?
If you see a newspaper article talking about something interesting it is good practice for them to include the title of the article and the author somewhere in their article, some quick google detective work can often bring up the original source.
Most articles can be accessed via “pubmed” (https://www.ncbi.nlm.nih.gov/pubmed) a giant online repository of articles. Older articles you will be able to access for free and you will see all articles abstracts.
If you are asked to pay for articles but aren’t a member of an educational institution one trick is to directly email one of the authors of the article, they are often glad to help!
2. I’ve got the article, now what?!
We need to try and establish is this a good article or not and get a general feel for it. Where is this journal article published? Is it an internationally respected journal that is “peer reviewed” i.e. have other people read it and thought, yes, this is good research, or is it from an unreliable source.
Who wrote it? Having a quick look at the other works of the main authors is a good guide as you can see what other research they have done. You can find this out by putting their name into pubmed, or using a hyperlink over their name in pubmed.
Make sure you read about any “conflicts of interest” – is anyone being paid by a company that benefits from this paper showing positive results.
3. You’re ready to start reading.
The first thing to read is called “The abstract”. This is a bitesized summary of the paper which ideally gives all the relevant information contained within the article in a condensed format. If you are really pressed for time then you can get away with only reading this, but beware – you can get caught out!
4. The deep dive.
Now this is where the hard work begins, but where the rewards are to be had. You’ve found the paper, you’ve got an idea of what it is talking about, it’s from a reputable journal.
Introduction – this sets the scene for the article, gives us relevant background and reasoning behind study, can highlight points of further reading that are needed
Methods – often the hardest bit to unscramble, often there will be complicated talk of statistical analysis types but the thing to focus on is how the study was conducted. The best phrase you can see for a study is a randomized double blinded placebo control study (RDBPC). There are of course caveats to this but this is a good rule of thumb.
Results – As you read more and more this is the bit to really unpack and have a think for yourself. In this section start to have a look at the following things: who was selected and allowed to be in the study? (selection criteria) There will often be a table with selection criteria – are they all Caucasian men with no underlying health conditions? Will this drug only work for them?
Look at the randomized groups, are they split evenly with regards to the selection criteria or does one group (one “arm”) have more of one type of patient than another?
Next look at the actual results, does it have lots of patients or only a few patients. Bigger studies mean more reliable results. Are these results “significant”. This requires unfortunately a bit more reading but firstly try and see if the “P value” is less than 0.05. There is much more to discuss on this topic at a later date but for now that will do!
Be aware of the graphs they use. On a graph the numbers on the horizontal and vertical lines can be changed to make the graph more impressive. Smaller numbers on the lines (axis) make a bigger difference to the bars or lines on the graph!
The things they are measuring are the “Primary Endpoint” – Issue number one that they want to improve and then various “secondary endpoints” – things that can add to the study or some things that they may not have expected to be significant. Sometimes you will find the primary endpoint has not been met (the results showed no difference) and that in fact there were some useful secondary endpoints that had a positive (or negative) impact on a patients health.
Discussion – hopefully by this point you will have had some discussion points yourself to think of having analysed the data and you can agree with their findings or refute their findings and importantly say why.
Conclusions – this ties everything together and often raises any potential new avenues to explore in the future and any limitations of the study.
As you can see this is not an easy process and takes lifelong practice. To start with I would advise the following:
1. abstract
2. discussion
3. conclusion
4. results
5. methods
once comfortable read them in order and discuss them with friends! Journal clubs are a great way to force yourself to get into this method.
Learn Well!
@Medicine.Daily
Comments