Slide 1

Thank you, good morning Chairman, Ladies and Gentlemen. A short presentation to bring you up to date with the work of the ERA-EDTA coding group entitled ‘How can we improve our ERA-EDTA coding systems and definitions to increase comparability between renal registry data?

Slide 2

Slide 3

In order to fulfil these functions it's essential that we are able to compare our data.

Slide 4

At the moment our data are something like this, they’re beautiful, perfectly formed but they’re all a bit different. This is Cezanne’s ‘Apples and oranges’ and actually if you look carefully, it’s quite difficult to tell in some of the bowls which are apples and which are oranges. That’s our problem in the registries. All of us think that our data are perfect, we all have our own in-house definitions but none of us really agree and we’re not sure which are apples and which are oranges.

Slide 5

So the question is, do we need new codes for the renal registries?  I think the answer to this has to be yes, if we want to continue doing the things we’ve been doing but do them properly, certainly if we want to pursue new topics and this morning we’ve heard about some fascinating new topics. If we want to participate in international comparisons, we have to have international definitions. If we want to move from simply counting our patients and reporting them to improving the quality of their treatment using audit and other tools which is actually the purpose of the QUEST initiative, then we have to. If we want to use our data for service planning and many of us do, we will need to improve our definitions. I've put in grey print three things that our data might be used fro in the future. The answer to the question therefore is “no”, if we are happy to continue doing exactly what we have done in the past. I don’t think anybody is really happy with that idea.

Slide 6

In order to see whether there was an appetite for changing and improving our coding system we asked the representatives of the national registries and we received a number of very helpful comments but also some very helpful cautions.
The national registries told us that the current system, although it has its flaws, it is very well liked, it’s well known and it’s simple. They acknowledged the problems particularly that it wasn’t comprehensive. It isn’t possible to split groups or even sometimes to combine groups because we haven't defined what we meant by them. There's insufficient detail. It's very difficult to cross-map our data to other mapping systems. There's no quality assurance or validation included and the last two points are primarily because we actually don't have any definitions for many of our terms and so for one nephrologist what is diabetic nephropathy would be chronic renal failure cause not known. Similarly another nephrologist might code a patient with proteinuria as having glomerulonephritis even if they hadn’t had a biopsy, whereas many would again say the cause is not known. They commented that ICD had its flaws too, they said it was not clinical enough, it was really meant for coding specialists but not for clinicians and also they commented that it was not available in all languages.

Slide 7

Their recommendations to us were that we should indeed review the codes, we should expand them and we should improve them but we should keep it simple, they should be comprehensive, it should enable international comparisons and we should retain the existing structure. Well, that’s quite a challenge really because we’ve been asked to do everything but do nothing. So the correct solution for this is form a committee. That’s what we’ve done. The committee recognises that we have to achieve technical perfection but in a sort of pragmatic way.

Slide 8

This group has been formed as a spin-off of the ERA-EDTA registry committee and specifically of the QUEST initiative and of the data acquisition group of the QUEST initiative and we will report back to them. We’re a short term group that is we will try and do the task that has been set us and then we’ll expand I hope leaving the ERA-EDTA registry with a system that it can implement in the future. We accept that some of the definitions will have to be pragmatic.

Slide 9

We will consult with national and regional registries affiliated with the ERA-EDTA and if possible we’ll consult with our colleagues in other international registries. We’ll also consult with other international standards organisations but we’ve been given a clear instruction by our affiliated registries that we mustn’t be dictated to by them.

Slide 10

Our aim is pretty simple, it is to improve and standardise the coding definitions used for renal registries in Europe.

Slide 11

The topic we will tackle first is the list of primary renal diagnosis. I think that is certainly the most important thing we have to do but we will then look at the definition and description of the modes of renal replacement therapy and how you define a change in mode, for example, when a patient on home haemodialysis is transferred to hospital outpatient HD because the person who helps them with their home haemodialysis has gone on holiday, should this be classified as a failure of home renal dialysis?  I think we would agree probably not but for how long can a home HD patient  dialyse in centre before you say that your home dialysis has failed?

We’ll look at vascular access; I think everybody recognises this is our Achilles’ heel but we’re not actually sure what we mean by all the different terms, the different types of fistulas, different types of catheters.

Slide 12

We will stick to the principle that the codes we produce should be usable by ordinary clinicians for their routine work and the hope is that the data will then flow through electronic records to registries. We’re trying not to set up a parallel system which will be useless in clinical care. We acknowledge that there will still be uncertainties even with a good coding system and good definitions. We’ve been told very clearly that we must start with our current coding system and expand it incorporating all that is currently there. So we’ve been told not simply to rub it out and start again. We would like to be consistent with other emerging medical standards, for example, the Snomed classification and we hope that it will be possible to map all our codes into the Snomed terminology. It should be rigorous enough to support primary and epidemiology research, as I’ve said comparable with other international registries and capable of transmission in the scheme recommended by the QUEST data capture group which will probably be based on a well formed XML schema using HL7 version 3. Obviously it has to be acceptable to our parent body, the registry committee and affiliated registries.

Slide 13

The output will be in the form of guidelines and definitions for use in registries and in renal centres  by clinicians and we hope that they will be understandable by non-clinicians who undertake quite a lot of the coding work in many hospitals. These coding staff may not have sufficient background knowledge to determine the diagnosis from first principals and they will not be able to do their job unless we give them good definitions. Our work will be published on the registry website and if it’s of high enough quality, it maybe submitted as a manuscript for publication.

Slide 14

In summary I think most people now agree that we need new codes and we need to take the new step of adding definitions to these codes. They must be reliable and consistent so that comparable data will support research, quality improvement, service planning, teaching and funding. They have to be based on an expansion of our existing codes. We’ve been instructed to use routine data if possible and to try and keep it simple. Compatibility with other coding systems will be kept in mind and the short-term working group has now been established, has met once so far and will meet immediately after this presentation for its second meeting and we’ll report back the other chair of the QUEST data capture group. Thank you very much.