[meteor_slideshow slideshow=”adssa” metadata=”height: 126, width: 630″]
• Each blog entry should be saved as a single Microsoft Word file (.docx) and uploaded to the appropriate repository on blackboard (below) and should include in addition to the body of the text:
• A title
• A reference in Harvard or Vancouver format.
• A hyperlink (web link) to the article upon which you have commented
• A word-count for the body of the text (ie excluding title, reference and hyperlink) which should be no more than 500 words in length.
Summative blog entries will be marked anonymously (i.e. with the marker blinded to the student’s identity) so it is important that you do not include your name anywhere in the files you submit.
General guidance on writing blog entries
Choice of Article
You need to comment on primary research articles. If you are in any doubt as to whether an article should be considered primary research, look and see whether it has a methods section. If it does not, it is almost certainly not primary research. Laboratory studies and clinical trials make excellent choices. You could also comment on systematic reviews and meta-analyses, however, the methods used in these types of publications are quite complex (including complex statistics) and may be difficult to summarise in a short blog entry. Studies which test an explicit hypothesis perhaps lent themselves more easily to critical analysis in comparison to observational studies which look at trends in diseases or prescribing.
The inspiration for your blog entries can come from anywhere: you might read about a study in the newspaper and decide to find the original paper in an academic journal. You might come across paper during your background reading for the module. At some point, you will have to use an academic database. I will recommend the names of some below, and I would encourage you to become familiar with them and to discover which ones you find most easy-to-use. They’re very powerful search tools, and allow you to filter the results by date, or by type of article (for instance you might want to restrict your search to clinical trials only).
Web of Knowledge
You may know of others, and are welcome to use them. in addition the University has its own search tool which is called ‘Discover’ and is particularly helpful, because it helps you find articles which are available in the universities electronic library collection.
Structure of your blog entry
You can structure your blog entry however you see fit, hoever you should include the following information:
You should briefly describe the purpose and aims of the study you are describing. Remember your audience is GPs and pharmacists. You shouldn’t need to give long explanations of common medical conditions. Remember that the marks are awarded for your description and critical analysis of the research.
Description of methods
Think carefully about including the most important details in the methods, because you will not be able to include all the details. You should include details of the the experimental protocol, but do not forgot to mention what was measured (and how) and what the primary end point of the study was (or which value was compared between groups.). As your critical analysis is largely dependent on the methods, it is important to make clear how the experiment was conducted.
Description of results
It is important to discuss the most important results quantitatively and to consider the most important information to include in a short summary. Don’t be tempted to write too much about statistical significance, without commenting on the size of the effect measured. Many papers will include lots of measurements, you need to consider which are the most important, as you won’t have room to discuss them all.
This is probably the most difficult section (and consequently, where the most marks are available). Essentially, you should aim to consider the work critically, rather than simply accepting the authors’ conclusion. You can approach this task by asking questions such as: Were the methods (and endpoints) appropriate? What do the results mean? Is the authors’ interpretation of the results supported by the data?
You should try to judge each paper on its own merits. If a paper set out to test the hypothesis that ‘dogs enjoy eating bones’ it’s not really fair to criticise it for not asking whether cats like eating bones. Proposing an extension to a study (however interesting) is not critical analysis. You may wish, briefly, to discuss the implications of the research which is again interesting but is not critical analysis.
Try to keep your critical analysis specific rather than general for example, rather than automatically saying ‘the experiment would have been better if the sample size had been bigger’ consider whether this is really the case. In very may experiments it is true, but it requires some justification. Experimenters don’t usually pick a sample size (n number) out of thin air, they will perform calculations to work out the sample size they need. A trial that is bigger than it needs to be costs more money and may have ethical implications relating to unnecessary experimentation on volunteers or animals. A comment along the lines of ‘the authors don’t state how they calculated their sample size’ or ‘the authors calculated their sample size but were not able to recruit enough volunteers’ is a much more useful indication that something is wrong.
It is important to comment on bias e.g. ‘there were more people with hypertension in the control group than the test group’ and to think carefully about critical analysis of the measurements and endpoints used in the trial. Many studies will claim that drug x reduces cardiovascular risk, when in fact they have only measured the effect of drug x on blood pressure, not on cardiovascular events. If a trial uses the ‘Penson depression score’ as its endpoint, you need to question what this score is? What does it mean? Has it been validated in other trials? What are its strengths and weaknesses?. Some experiments don’t seem to have a clear hypothesis and don’t state the primary endpoint in the methods. This is often the case with trials involving mental health where the patients will be assessed for severity of symptoms using 4 or 5 different scales before and after an intervention. You should be asking if it is necessary to use so many scales , or whether the authors were ‘hedging their bets’ and hoping they would see a significant difference in at least one of the measurements.
It is also interesting to comment on the way in which numerical data are treated. For example, Some trials of antihypertensives set out arbitrary categories for BP (ie normal<140/90<hypertensive) and then presented their results saying ‘at the end of the trial 20% of people in the control group were hypertensive and 10% of people in the treatment group were hypertensive’ When data is categorised like this, it is a good idea to ask why? Is there a good reason? Or would it have been better to present the mean BP in each group? You may find it helpful to commented on the statistics used in published work. Often these are very ropy!
These are by no means the only things you can comment on, but hopefully they give some food for thought.
[meteor_slideshow slideshow=”best” metadata=”height: 126, width: 630″]