Skip to 0 minutes and 14 secondsSo now let's look at the grading evidence in CPGs and this is just an example each clinical practice guideline should list a table something like this and it to show you how it rated the level of evidence that it gave to different thing. So see on the left side there'll be the grade of the level and sometimes they'll these will be together or they'll be separate. I'll show you some examples of what you'll see in a clinical trial or in the guidelines. So you can see them they looked at the literature used to support it. Grade a is obviously or level one is the best evidence. It means there's well conducted clinical trials.
Skip to 0 minutes and 51 secondsthey clearly showed that the benefit was greater than the risk often language like is recommended is beneficial will be listed. Grade B level two, not so much information there may be post hoc analysis or subgroup analysis there may be some meta-analyses because of the individual trials may be didn't show something. For the most part it says the benefit is bet greater than the risk but it's not as clear so you'll have language is probably recommended could be beneficial and then grade C or D level three or four depending on how it's listed. This is more observational studies.
Skip to 1 minute and 31 secondsThere may be some clinical trials but they'll have some major study limitations so that's something that as a reviewer you will need to look at. This one says that the benefit is greater than or equal to the risk. Sometimes may not be might be reasonable effectiveness is not well established. Grade five level grade E level five. This is more excerpt expert consensus clinical experience. So this one you definitely need to make sure that you look at carefully because oftentimes the risk is not or equal to the benefit not usually recommended, might be beneficial, might be harmful. So that one takes a lot more clinical judgment. So use these level of evidence in your recommendations.
Skip to 2 minutes and 19 secondsBecause the majority of individuals should realize what these are but again they differ based on the different guidelines but you also need to be prepared to discuss what they mean in the context of the recommendation from that specific guideline. So you might want to take a look at this paper that was written recently. It's by a famous author that you all all hopefully known by now, Daniel Brown, and he wrote about think rethinking about clinical practice guidelines in pharmacy and had a couple of really good statements that he listed in there.
Skip to 2 minutes and 50 secondsIf you look at some of the reviews that were done and there were two of them that looked at quick clinical practice guideline one looked at 53 one looked at 41 and they found that only 11 percent and 14 percent respectively of the recommendations in those clinical practice guidelines came from level A evidence. And remember level A is the best level and 48 and 55 percent respectively were expert opinion level C. Now you would think that the best clinical trial should be in practice guidelines and you can see how poor quality these trials are even in the guidelines.
Skip to 3 minutes and 29 secondsThis should help you to see even more so why as a clinician you need to really look at these guidelines carefully and see what exactly where the evidence is coming from and how strong that evidence is. Author Berg listed in the journal citation below said on students failing to evaluate literature in CPGs that they will lose autonomy forced to follow poorly constructed guidelines subject their patients to substandard care without a strong scientific basis. And and summary Dr. Brown's trial rather than blindly memorizing parts of a CPG clinical learning and therefore patient care can be enhanced by analyzing critiquing the relevance and validity of the clinical practice guideline. So I'm not the only one.
Skip to 4 minutes and 16 secondsSo this was a very interesting both of these articles were very interesting highly recommend that if you get a chance that you could read them. They're both free to access on the internet. So no guidelines no the evidence. And somewhat clinical practice guidelines that evidence is the new natural. Natural doesn't always mean safe or better. So again always good to evaluate those guidelines and not just memorize what they say without knowing any of the context involved with them. So let me give you example that actually happened in class. I love it when this kind of stuff happens. We were doing a journal club based on an article and this new global gold guideline came out in 2014.
Skip to 4 minutes and 56 secondsAnd what it did was one of the things they changed in this guideline was they had new information that said a five-day course of steroids was better than a 14-day course of steroids for management of acute exacerbations of COPD. And it was all over the news at that time. It was showed shorter steroid treatment worked for COPD exacerbations Less is more. COPD best exacerbations 5 days better than the two weeks. But let's look at exactly what information there was. So if you pull up the 2014 gold guideline this was what was listed. A dose of it was changed from the previous guideline. A dose of 40 milligrams of prednisone per day for five days is recommended.
Skip to 5 minutes and 41 secondsEvidence B and you can see when you pull up that reference 567 it was based on this Journal article and this was actually the one we and that we were doing in journal club and after we got done reviewing it this guideline came up. So you could see that level a grade B level two from what I had shown you showed that it should come from post hoc or subgroup analysis of randomized control trials or meta-analysis of control trials benefit is greater than risk is probably recommended could be beneficial. You could see what they had listed in there remember you need to pull up there what they said.
Skip to 6 minutes and 19 secondsThey said randomized controlled trials limited body of data, evidence is from endpoints of intervention studies that include only a limited number of patients postdoc or subgroup analysis of randomized control trials or meta-analysis control trials in general category B pertain pertains when few randomized trials exist, they are small in size. They were undertaken in a population that differs from the target population of the recommendation or the results are somewhat inconsistent. I don't know that doesn't sound too positive to me. So let's look back at that clinical trial. because that recommendation that they changed was based on this one randomized control trial.
Skip to 7 minutes and 0 secondsNow I know you don't have time for us to do a full journal club because that would take longer than this session is allowed but let me just tell you a few major inconsistencies that we found. When I do journal club with students I make them rip that thing apart and find every single little thing wrong with it. but then we we step back and come up okay. Let's talk about what the major inconsistencies were that make may make us doubt the results and that's what I'm on a list here. There first thing was the primary outcome was not the preferred measure and this was according to the gold guidelines.
Skip to 7 minutes and 32 secondsIt says that the rate of exacerbations not the time to the next exacerbation is the preferred primary outcome measure and they use time and not rate. They didn't include diet and exercise that they monitored that was hugely important in COPD management. They didn't have an adequate and inadequate study duration and again according to the guidelines that it says that it should be 12 months in this study was only six months. The study sample was not well-defined and narrow so again extrapolation to the general population is going to be difficult from this study and that was what they had said in there that it may not be similar to the patient population which was true.
Skip to 8 minutes and 12 secondsThey didn't mention several ancillary types of medications that could significantly interfere with the results. The Comrbidities that the patients Tad's were not listed and we know that patients with COPD are gonna have a variety of comorbid states that again would make it difficult for us to know who to extrapolate this information to. Patient inhaler technique was not addressed and if you know anything about asthma or COPD literature that patient inhaler technique is is horrible. That can often lead to inappropriate results at the end just because it's not the drug that's the problem is the patient's technique with using the inhaler. And then adherence was not monitored which goes back to patient inhaler technique hugely important.
Skip to 8 minutes and 57 secondsSo to me that it says in there that it was a few major inconsistencies. I would call this more than a few. So even the student so in my pharmacy class or in the second year of their pharmacy education clearly thought that there were too many limitations with this trial to be able to provide some a clear recommendation from this study.
Grading in Clinical Practice Guidelines
Prof. Mary Ferrill clarifies grading in CPGs and major inconsistencies in this video.
To begin with, there are 5 grades or levels in total. Firstly, the studies in GRADE A (Level 1) are well-conducted randomized controlled clinical trials (RCTs). Their benefits are much more than risk.
Secondly, the papers in grade B (Level 2) are probably recommended, and they could be beneficial.
The lowest one goes to grade E (Level 5), and it may be some experts’ consensus or clinical practice experiences, which is not usually recommended because the risk is higher than the benefit.
Additionally, we are given several examples to understand Grade B category. In GOLD, it means when few RCTs exist, they are small in size, or they were undertaken in a population that differs from the target population of the recommendation.
Ultimately, she gives us several major inconsistencies in Leuppi, et al Study, including primary outcome not preferred measure, inadequate study duration, and no mention of ancillary medications.
Have you found any inconsistency in the study you are working on? Please share it below.