Back in the mists of time, you may recall we described EBM as the combination of best evidence, clinical expertise and patient values. Which is pretty straightforward. But how to tell you have the best evidence might be a bit tricky – the RAMBo and FAST routes can appraise what you’ve got, but how to tell if you’ve actually got the right papers? As our guest blogger Sheeraz Khan asks – do all (search) roads lead to Rome?
Sometimes critically appraising search results and incorporating them into your practice might not be as challenging as finding the best evidence. Which could either be because there is no evidence, poor search techniques or unpublished data.
Searching the evidence: An example
Let us see how different search strategies come up with different results using just one search engine i.e. OVID Medline.
Clinical question: Is cycling CPAP a better strategy for the weaning of CPAP in preterm babies then reducing CPAP pressure?
The initial search done using OVID Medline 1996 to October week 3, 2015 using key words. CPAP, continuous positive pressure, cycling, intermittent, weaning, neonate, newborn, and preterm.
Only two articles were relevant from total of 35 results:
1. (A) CeasIng Cpap At standarD criteriA (CICADA): Impact on weight gain, time to full feeds and caffeine use.
2. (B) CPAP review.
The new search strategy included the key words of CPAP, Continuous positive airway pressure, positive expiratory pressure, infant, newborn neonate, preterm, premature, weaning, ceasing, cessation, decreasing, pressure, stop, stopping, time off, cycling and intermittent.
Only two articles appeared and both were relevant and different from the above results
1. (C) Methods of weaning preterm babies
2. (D) Withdrawal of neonatal continuous positive airway pressure: current practice in Australia.
I did a search on Google Scholar with my clinical question and picked up two systematic reviews to study their search strategies.
Using search strategy of the systematic review “Weaning of nasal CPAP in preterm infants: who, when and how? Amatya S. World J Pediatr, Vol 11 No 1. February 15, 2015” and got 82 articles however only seven were relevant. Four as above i.e. A, B, C and D with other three as below.
5. (E) Strategies to accelerate weaning from respiratory support. [Review]
6. (F) Gradual versus sudden weaning from nasal CPAP in preterm infants: a pilot randomized controlled trial.
7. (G) Strategies for the withdrawal of nasal continuous positive airway pressure in preterm infants.
Let us take a look at the strategies (key and text words) used by the Cochrane group in the systematic review.
“Strategies for the withdrawal of nasal continuous positive airway pressure (NCPAP) in preterm infants The Cochrane Library 2011.”
Seven articles were selected out of 554 results. The relevant articles were the same as in the previous search strategy 3 with the only difference of one missing (article E above) and one extra related article which is.
(H) Randomized controlled trial of discontinuation of nasal-CPAP in stable preterm infants breathing room air.
We can see the diversity of results with just a little change in the key and or the text words.
I searched Cochrane with the same clinical question however there were no results found after a search done with key words of “weaning of nasal CPAP” “withdrawal of CPAP” “withdrawal of CPAP” “weaning of CPAP in preterm babies”
But I do get the systematic review on a little change in the search words “weaning of NCPAP” “withdrawal of nasal continuous airway pressure” or “withdrawal of NCPAP” so just the letter “N” in this case can make your life easy!!
Anything else? – Unpublished data
After getting all the appropriate search results and applying it into the practice can we call it “Job done”?
However there is something still missing or hidden. We have the “Best Published Evidence” but still probably not the “Best Evidence”. There are thousands of clinical trials who have not reported their results. AllTrials campaign is a project advocating that clinical research adopt the principles of open research.
Since 2008 in the USA the FDA has required results of all trials to be posted within a year of completion of the trial. However an audit published in 2012 has shown that 80% of trials failed to comply with this law.
It is not very easy to say we are practicing evidence based medicine. For example, prescribing a new drug while more than half of the trial results were not published by the manufacturer.
Registering trials (tracking completed but unpublished studies) will help clinicians who prescribe new treatments and researchers publishing their results avoiding publication bias.
Obtaining the evidence to make decisions can often be complicated. Performing a comprehensive and well thought out search can lead to differing amounts of evidence obtained and often missing information. This can be because of variability in the availability of papers on different databases. Differing search algorithms leading to differing results or simply data that was never published in the first place. AllTrials will help to ensure that all data is available to be searched for. However, to ensure this can be found a more systematic, less variable and more comprehensive search formula should be found to obtain results.
1. Weaning of nasal CPAP in preterm infants: who, when and how? A systematic review of the literature. Amatya S., Rastogi D., Bhutada A., Rastogi S. World J Pediatr, Vol 11 No 1. February 15, 2015
2. Strategies for the withdrawal of nasal continuous positive airway pressure (NCPAP) in preterm infants (Review). Jardine LA, Inglis GDT, Davies MW. The Cochrane Library 2011.
3. Principles of evidence based medicine
A K Akobeng. Arch Dis Child 2005;90:837–840
4. AllTrials Compaign.
5. Compliance with mandatory reporting of clinical trial results on ClinicalTrials.gov: cross sectional study BMJ 2012; 344
6. “There is nothing like looking, if you want to find something” – asking questions and searching for answers – the evidence based approach. Munib Haroon Robert Philips. ADC ed prct ed 2010; 95; 34-39