CME Slides Forum - J. Galle

HOW TO SEARCH FOR THE BEST EVIDENCE IN CLINICAL PRACTICE

Jan Galle, Lüdenscheid, Germany

galle

Prof J. Galle
Director Department of Nephrology and Dialysis
Hospital Clinic Lüdenscheid
Lüdenscheid, Germany

Now, as I said when I introduced this session, what I will now try is to focus a bit more on the technical aspects after we got all the intellectual aspects. Just to be clear, first of all, I definitely have no conflict of interests, so if I advertise some approaches and also some software, I do not have any stocks in the market concerning these products and what I just want to do is to give you my very personal example of how I try to get organised because I think what we have learnt is that very easily we can completely be overwhelmed by the amount of knowledge we have and we have to find someway to discipline ourselves and to get a structure in how to extract data.

So, what I will try to do is share some personal experience and try to show you how to get organised and also show you some technical aspects.

My synopsis of my talk is not any longer the introduction. What is the best evidence has been nicely covered by Giovanni. But I will focus then on text books, electronic text books, guidelines, websites, societies, journals, review journals, lectures, databases and storage.

Now, text books are something which I will keep very short here during my talk and I’m sure everyone of us uses text books and we also need them if we want to have someone taking a picture of us with a textbook in the board behind us to make us look serious.

And we also need the textbooks because they are sort of a base but the problem with text books is that they are always relatively outdated. Every one of us who contributes to textbooks knows how long it takes from writing a chapter until you will have the book in your hands. That’s approximately 3 years. Then the textbook is out for let’s say 3, 4, 5 years. So if you have bad luck you buy a textbook and the information covered by the textbook is already 7 years old at the moment when you hold the textbook in your hands. Nevertheless, that does not imply that everything that is written there is wrong but it’s not capable to have the new actual data. Then textbooks are quite heavy, I do not like to travel with heavy suitcases and textbooks are usually not interactive. Frequently, now you get a cd with your textbook and you have some sort of interactivity and possibility to extract data but that’s very limited.

The question is, are there online alternatives for print media in nephrology? And that’s a rhetorical question, of course, there are alternatives.

The arguments which I have in mind are that it’s usually very comfortable to use it as a search tool, that it’s much more actual, I will show you how actual it can be, that it frequently has a multimedia content. You have easily the possibility for data extraction and to send it to colleagues. For example, you have a very specific question, as a team leader in hospital or anywhere a young assistant comes to you and asks you what about vasculitis and then you make a data extraction with vasculitis and send it to him. That’s also from a teaching point of view very impressive. It gives you usually the possibility to get access to secondary literature and some electronic textbooks have a so-called ‘what’s new’ section that means they emphasise what has been published on several topics very recently. You can also get CME credits by using these textbooks and it gives you the option to make slides which is for someone like me who gives frequently lectures very helpful but there is clearly a big disadvantage and that’s that they’re usually relatively expensive. But if people collect themselves together and try to get a group license, it breaks down to much lower numbers concerning the price.

What I now would like to do is to give you an example. And I hope that I am still online here. I go to up-to-date in medicine, let’s see whether that works. So, this is the website of up-to-date. As I said I have no conflict in interest, I’m not part of the company that’s producing up-to-date, it’s just one example which I find very helpful but there are certainly others on the market.
Now I’ll try to log-in and at this point I’m lost because I have to go to a different place to find my log-in data. It worked a few minutes ago because then it turned up automatically but nevertheless we still have the chance to get there. May you all please close your eyes now?! You see that’s typical for online demonstrations but nevertheless. You haven’t seen that number please, I will change it after the session. Ok it works. Now, now we are online and connected to the database of up-to-date and for example, we’re interested in vasculitis. Then there come a number of hits and now we should focus, for example, on ANCA positive vasculitis. It might be that this is what we are interested in. Then there are still many different topics and we have a patient where we have the idea that he may have got Wegener’s granulomatosis. Now we come more specifically to Wegener’s granulomatosis and then are here many different topics covered. You can look for diagnostic approaches, for treatment approaches and so on, also to experimental questions like animal models used to study this disease. Where do we want to go to? I also could have narrowed it by adding the term ‘treatment’ directly but I mean this is now not a systematic approach, it’s just an example of what the program might offer to you. What you see here is that there are links. If you read the text and then you come to another point which maybe of interest to you, you just click on it and you come to a different chapter. There are also citations, as you see here and if you click on this, you come to abstracts, you are directly linked to the Medline abstract, that’s what I mean by having access to secondary literature. But it’s not only text, it’s not only secondary literature which is covered by this program there are also, for example, radiology pictures and if I go to the pulmonary involvement somewhere here, I should hope to find an X-ray imaging chest radiograph. Do I have one here? No that’s still an abstract. Right we have one here, there is one. So this is something which is very helpful you get it online, you can print it, you can increase it in size and you can use it also for lectures by extracting it.
I think the very strong point about such a program is its actuality because it’s continuously renewed. If it is stored here on a database, it does not mean that you have to buy a new textbook, always if you go there and ask a question, you may trust that the relevant new information is already covered. I think the turn around time is about 3 months. At least, if you go to secondary literature covered by some topics, you will see that the very actual literature is already taken account of here in any clinical questions. As I said, you are able to extract data. You can use for example here a printer friendly format by pressing this button and then you could just print it here using your browser or your generated PDF file which can be sent by e-mail to a colleague or you press this button here which is also, of course, sort of an advertisement covered by the program where you enter someone and send to this person the information which you extracted here which makes it very simple to also teach your graduates, for example.
Any questions concerning up-to-date while we are online here? Otherwise I would go back to my presentation. So, the strong points are actuality, data extraction, radiographs, histological parameters are covered, laboratory values, you also get direct information concerning treatment, you get side effects of pharmaceuticals and so on. So to my opinion this is one of the most powerful programs using and getting new information and it’s the program which I use myself when I try to find something new or if I have a specific new question.

Fine, now coming back from up-to-date to guidelines which has already been covered by my colleague a little bit.

Here are some examples. Guidelines are published by, as Giovanni said, almost all societies very frequently and of course, guidelines are helpful but they may also have some disadvantages as it has been mentioned, if they are too many guidelines, many new relevant studies will never be undertaken because there is already a guideline.
We have the K-DOQI guidelines, Disease, Outcome, Quality, Initiative and k stands for kidney, so the K/DOQI means Kidney Disease, Outcome, Quality, Initiative and an important example would be for the bone disease guideline which was published in AJKD in 2003. Then there are smaller guidelines like one on lupus erythematosus published in Arthritis Rheumatism that was a very specific topic covering approximately ten pages. Then there is the big series of the European Best Practice Guidelines, the last issue was published just a few weeks ago in NDT on haemodialysis treatment. But we also should look at guidelines published by other societies. Here is an example with the American Heart Association because they also are interested in renal questions and publish their guidelines for their particular audience but it’s also relevant for us nephrologists.

How do you get there? It’s very simple for example, you go to Google and other search engines, you enter DOQI guidelines and then you will come with the first hit to the National Kidney Foundation and K/DOQI website.

And if you press this, you would come then to the website of the National Kidney Foundation where you can download the NKF-K/DOQI guideline.

You would just have to go clinical practice guidelines, press the button and then you could download the published K/DOQI guidelines in a PDF format, for example, diabetes, haemodialysis, peritoneal dialysis, vascular access, anaemia management.

If you press here haemodialysis adequacy, we would come to this page where you could get detailed information on your topic.

The other example, the bone disease guidelines published in 2003 is an example which I want to present to you because it also shows the limitations of these guidelines.

Let’s go to one topic. If serum calcium is higher than 10.2 mg/dL, then patients taking active vitamin D sterols the dose should be reduced or therapy discontinued until the serum levels of corrected total calcium return to the target range. That sounds reasonable but see here this was opinion based, of course, the people who wrote these guidelines were well known and recognised nephrologists but nevertheless they had to recognise, when they wrote these guidelines that there was no randomised controlled trial available to answer this question. I do adopt this guideline, I use it but I try to be aware and to distinguish where is it based on opinion and where is it based on randomised controlled trials. The K/DOQI bone disease guidelines cover approximately 120 pages and 80-90% of what is written in there is based on opinion, nevertheless it’s probably the best of what we have in hand, if we want to treat patients with secondary hyperparathyroidism and so on. So we have nothing better in hand but we should be aware that’s it’s opinion based and that it’s not derived from randomised controlled trials, so this nicely shows the conflict in which we are particularly in nephrology because so many clinical trials always excluded patients with renal disease and we are in bad luck that we really do have relatively few numbers of randomised controlled trials in nephrology for example, if you compare it to cardiology.

But nevertheless, we need the guidelines. Here’s the example for a guideline published by the American College of Rheumatology, the Ad Hoc Committee and they asked the question on systemic lupus erythematosus, how should this be treated?

Then I want to show you this example with the American Heart Association. Just remember that we should not look only into our own journals that we should also be aware that other societies answer questions related to nephrology in their guidelines and there was a very important paper published simultaneously in Circulation, as well as in Hypertension in December 2003 and it was a paper where kidney disease was for the first time defined as a risk factor for development of cardiovascular disease by cardiologists which has a big impact on their daily work.

Finally, the example for the very recently published guideline in NDT just a few weeks ago where the European Best Practice guidelines for haemodialysis were renewed and published. What is typical is that usually such a supplement in the NDT starts with an approach for guideline development where the authors very clearly should describe how they prepare these guidelines and where they also should mention to which extent the guidelines are based on opinion or on randomised controlled trials and how the data extraction was done.

Now, this brings me to websites of societies and of course, the websites national, as well as international societies, also link themselves to the guidelines like the K/DOQI guidelines or like the European Best Practice Guidelines.

They have usually sections like education and resources and if you press these buttons, you very frequently will get linked to important lectures, to the guidelines and to relevant literature.

This was the ISN here’s now the ASN Society guideline and of course, you should have in your browser all these addresses marked. I will show you how I do it that you easily get access if you need it.

Now, we come to the European Renal Association guidelines. This is the ERA-EDTA org website you see here again there’s a button education to press. And if you press here, you come to this site.

Which cover all the revised European Best Practice Guidelines. So if you press here, you will then come to the guidelines, the European ones, for dialysis, for anaemia, for diabetes and so forth. If you press here, you would have the possibility to download a PDF file.

Now, next step, journals. I would like to pick up the point raised by my co-chairman where he mentioned that searching in Medline is not sufficient. I gave you some examples that searching guidelines is important but also not sufficient. You certainly will have to cover at least some relevant journals for you to get a more complete follow up because you do not get all information from Medline. Of course, you need Medline search, if you have a new question, a new topic which you would like to cover but I would strongly give you the advice to regularly follow journals and again here’s my very personal example.

This is a word file which is a continuously followed word file by me. Here are the New England Journal, the Lancet, the Kidney International, the Nature Nephrology, the American Journal of Kidney Disease and JASN, Circulation, Hypertension, JAMA so some journals of which I personally think I should read them. Because I’m a lazy guy I have to get disciplined and I discipline myself by forcing myself to make a mark, if I have read issue number 21 in the New England Journal and issue number 75 in the LANCET. So, this is my actual stand this is what I performed last week and if I come home from this congress, I will have to make a mark here at number 22, number 23 and so on. I have been doing this already now for 10 years and it is a little bit helpful because it really disciplines me and it gives me also some security because I know, ok I have followed these issues, of course I cannot read every article but at least, I screen the actual issue and it forces me to concentrate and to extract something which is directly related to my daily work. So, it’s just a personal tip but I think it’s useful to get a more complete follow up of the literature. As I said, it’s not enough to look every now and then in Medline to see whether there is something new, you really have to follow your relevant important journals and in nephrology there are not so many. So this can be done.

How should you do it? Save your good links to your favourites. So, I go back to Google and you see here --- is German and means favourites and here is one section for journals and now as you can see here, there are quite many journals. So this is for one time a little bit of work to get all of this done but I did this some years ago and if there’s a new journal, it will be added here but now it’s quite easy. I press JAMA and then I’m at JAMA and I do not have to look up in a browser for a long time to find it. So now we could do our JAMA search.

Going back to the presentation. If you go to a website now, for example, of JASN you look up the current issue and you screen the table of contents and if there’s something of interest for you, you could try to get the full paper but then you still have the difficulty, do you have access to it?

This is now my online journal slide. I put it off again, so that you cannot get all my personal online access data but you should use such a file, write it down somewhere because you are all travelling frequently and if you have it on hand, you can also while you are travelling use it for doing literature search. But what is probably more relevant is that you try to figure out whether you have access to a so-called virtual private network. This is exactly what I’m using while I’m standing here.

What is a virtual private network? Well, a virtual private network identifies you as member of, for example, a university. Also companies use virtual private networks but for us I think the main interest lies that we get contact to databases of a university. What it does actually is that it digs a hole beyond the tunnel, it creates a tunnel to get beyond the firewall of databases of your university or your hospital or whatsoever. This is a small software tool which is provided for me as a member of the University of Wurzburg by this university and if I activate this file as I’m doing now, now I get connected to the virtual private network of the University of Wuerzburg. Then I get a connection and now as I’m standing here identified as a member of the university, I get a very particular IP, identification protocol number which identifies myself within the internet as a member of the University of Wuerzburg and by doing so I go back to my favourites.

I get access to all the databases of the University of Wurzburg and there are very, very many journals covered and everywhere where there’s a green button I could now get access also to the full papers. I don’t know who of you is using such a virtual private network. Please hands up, a few hands, 1/3, ¼ approximately is going up, so the other 2/3 or ¾ should try to find out whether they have the possibilities by their companies or universities to get access because it allows you to work wherever you are and get full data access.

I will come back to journals to also demonstrate to you how to store data but this is the final point. Before I will go there I want to make the difference between journals and review journals. There are some on the market. Probably the most important one for us is the one which was introduced just 2 years ago coming from Kidney International when Kidney International moved to the Nature publishing group, the Nature Clinical Practice in Nephrology was initiated and the aim of Nature Clinical Practice in Nephrology is that it wants to deliver timely interpretations of key research developments translating the latest findings into clinical practice. What does this mean? This is a sort of – the information for you by some experts.

So there are people in the editorial boards of Nature Clinical Practice who follow the literature, try to find where are clinically relevant papers, trials, hopefully randomised controlled trials and then the editorial board will concentrate or extract the relevant information from these particular articles into something which is probably easier to understand than if you would read the full paper. So you see in the way as I pronounce it, it implies there’s some danger too because you do not make your own data extraction and interpretation, you follow the interpretation of someone else. This is someone helpful, if you do not have so much time and it is also helpful if you try to read an article of a topic of which you yourself are not specialised, then it maybe helpful to take an attempt because you get a quite good interpretation, it’s easily readable, much easier than it would be to go through the whole article yourself but you should be aware you read the interpretation of someone else. If I remember the interpretation of the 4D trial when it was commented here in this journal, I was not very happy with it. So, you should be cautious by using such a review journal as your main source for information but sometimes it’s helpful and it spares some time.

Here’s, for example, the actual table of contents of the June issue of this Nature Clinical Practice in Nephrology and you see it covers topics like --- haemoglobin, cigarette smoking and other relevant actual topics in clinical nephrology.

Now, my second last point covers lectures and those of us who are giving lectures themselves are always glad when they get good lectures by colleagues, for example, my colleague Giovanni Strippoli saved his lecture, his power point presentation on my notebook, thanks very much nice slides, I will use them.

But there are other chances to get presentations from other people. Very official ones and that is again if you go to websites of societies, there is the point CME slides forum. So if you click here, you have the chance to get selected lectures, selected by the editorial board where you can download those lectures as PowerPoint and can use them as educational slides in your own presentations. There are many websites who offer this actually all topics, big issues in internal medicine are covered, so I showed you just this one example.

Now, I would like to end up with databases and storage and we’ll go back again to some for my opinion useful software and it was introduced already by Giovanni.

So, what you can do, if you look up for journals is that you use the Pubmed website but I think this is not very comfortable. You could enter a search term here but I do not use this website because as an example was also given by you in your talk, it’s quite complicated to really come to a good extraction. What to my opinion is more helpful if you have access to software such as Reference Manager, EndNote or others? I personally have been working with Reference Manager for many years, again no conflict of interest and that is why I use this as an example but I think end note and probably some others are as powerful too.

What you can do is, if you use Reference Manager, you can make an internet search. That’s what we’re doing now online. You see it takes us to the Pubmed and now, we look up what Giovanni published in 2006. Let’s see whether he was successful. 13, well you mentioned a number like 20 yes, of course not everything is covered in Medline, that’s the point. So, now we have captured 13 references of Giovanni and let’s say I myself am interested in this one. Now, I copy between databases and extract the highlighted reference to my own big database and this reference is now being transferred to my big database. Now, it’s inside there. Why did I put inside there? Not only to have it stored in my database but now I could use this citation in the text I’m going to write. I go now to my own database and Strippoli, here we are this was the one, I already have some of yours. This was the one which I just extracted and let’s say that I’m writing an article where I would like to insert this reference. I mean, now it’s completely out of context but nevertheless. I choose to insert it exactly here. So what I’m doing now is going back to my Reference Manager, reference, insert this marked reference into my text file and now, look what happens, we’re back in word and here is the De Nicola paper. Let’s see what happens. It was just 2-3 and now it’s 2-4 and it’s already incorporated into the file. At the end a citation list has been created and somewhere here it must be now number 3. This paper has been inserted into the citation list. So, you see it’s extremely comfortable. I want to submit this to Circulation, then it will be rejected and then I just press another button so that the citation list will have the format of let’s say Kidney International, JASN or whatsoever.

So, to summarise this and say it in other words what this program offers is that you use it as a searching tool for citations in the Internet mainly going to the Pubmed or to other sources, that you use it as a database where you collect all your citations but you use it also as a tool to generate papers and citational lists within the papers and for this purpose it’s really extremely powerful and elegant and easy to use. But then, of course, it’s my own database and I’ve been collecting data for many years electronically, my database consists now of some 18,000-19,000 citations, so it’s for my topics in which I’m interested in quite complete. So if I do a literature search, I go first of all into my own database because it’s easy to find things because I can link them to certain key words and for example, I was always interested in let’s use the key word ‘oxidised LDL’ and I used it as a shortcut key word in that way and then I come to those citations which have been published during the last years relevant for this point. But then, of course, it could be linked with particular authors. So, let’s say I want to see my own papers related to this because I need them for a review article. Then with this term it’s not so many. Well, it doesn’t matter. You see that there are many possibilities of how you can perform a search within this Reference Manager and I think it’s much more comfortable than the website which is offered by Pubmed.
So to summarise it again such a tool offers you data collection, data storage, it gives you the chance to find the papers again and it helps you if you create a manuscript, if you want to write a paper, to generate citation lists.

So this has mainly been what I wanted to present to you as technical issues but we are still on line and of course, we are here having to answer questions, if you have any concerning these technical issues or the more intellectual issues. But I would like to summarise with a sentence. I think before you get lost in this overwhelming fluid of data, you really should try to get organised because that makes work much easier and you should keep in mind what Giovanni has told you that you always have to ask the right questions, if you want a good answer. So thanks very much for listening and now we’re both available for your questions.

Questions: I will slightly deviate from what we have spoken about. That’s mainly for the researchers but as a clinician, I’m in a private practice, I have a patient and I find some difficulties in analysing that patient. Then how do I interconnect with the specialists and get their opinion about certain questions about a particular patient?

Prof. Galle: Well, if you were German, you could contact the website of the German Society of Nephrology and then you would be connected to the press related officer that would be me in person and you could ask your question and either I could answer your question or I could make a link to another specialist in the field. This is how it could work in Germany. I don’t know how other international societies would deal with such a problem. How would it be in Italy, do you have such a tool like a press related officer?

Dr. Strippoli: Yes, I think actually most nations have these tools at the moment, we do in Italy as well.

Question: Is your personal reference manager on your personal list where’s that kept and how do you get the citations in there? Is that your personal reference manager?

Prof. Galle: The database is a big file which is stored on my very individual notebook. So this is not anywhere internationally this is my notebook and there’s a big file, the database which I named Jan Galle which I have continued to increase over the years. If I use the program here like I did and perform let’s say an internet search, then I can, as I did download all these topics, all these citations directly into my database. So, this is how my database increases but I also, of course, could enter references newly by typing them by hand into the database.

Prof. Galle: No it’s a Reference manager database. It’s a particular database which is typical of this program. It’s part of the program, if you want.

Prof. Galle: Within Reference Manager. Yes a file of Reference Manager. Ok, so I do not see any open questions. You have one. Ok go ahead.

Question: What would you suggest I think for both, of course we have a lot of questions in clinical practice and we don’t know the answer with interventional studies. So what do you suggest in that case?

Dr. Strippoli: Well, the first thing that I suggest as I said is that you should not make a decision yourself that there are no intervention studies. So as I said previously, the best thing to do when you have a clinical question is to look in the Renal Health Library and see whether or not there are reviews and trials. You will be surprised that many times in fact there are. That’s the first thing to do. The second thing to do is then look at those guidelines as he suggested but being able to critically appraise what they tell you and never forget that there are lots of powers on the background of these guidelines. Now, recently this scenario of how you write a guideline has been changed. I don’t know if you have heard about this so-called GRADE system because we do have a problem that we want guidelines, many times we don’t have the trials, many times we never will have the trials. So what do you do in these settings? In the past you just got those statements from the guidelines saying there is insufficient evidence to do this. But I’ll give you an easy example. Prevention of peritonitis in a patient who has a PD catheter. Ok? I think everybody does the disinfection of that catheter with for example, soap and water or povidone-iodine. However, there are no trials that look at whether soap and water are effective. In the past, the guidelines would have said this is an intervention of unproven efficacy or if soap and water were produced by a pharmaceutical company the guidelines might have said soap and water is very good but this is an opinion. Today with the GRADE system, so in short look at guidelines that use this system they say soap and water is an intervention of unproven efficacy the likelihood that there will be more trials is virtually inexistent. The ratio between potential benefit and potential harms is in favour of the benefits because we don’t see how soap and water may harm you. So because of that we cannot highly recommend it but we can suggest that you use it. So look in the Renal Health Library, if there are no indications, look at guidelines but good guidelines which have been based upon these principals and the KDIGO guidelines which are now coming out have been based on these principles, so probably this is a good way to start.

Question: But of course these principals may not prevent you or keep you off from acting. There’s this well known example to all of us an infected appendicitis has to be removed but there’s no trial which ever proved versus placebo that removal of an infected appendicitis is appropriate nevertheless the surgeon has to do it.

Prof. Galle: Ok. So, I thank you very much for you interest and in the name of my co-Chair leader and I we close the session thank you.