Case #495

Comparison of CC exclusions in SWE versions

Added by Mats Fernström almost 5 years ago. Updated almost 4 years ago.

Status:AcceptedStart date:2016-12-20
Priority:MajorSpent time:-
Assignee:Mats Fernström
Target version:Expert Group 2017
Initiator:Sweden Target year:2018
MDC:CC Owner / responsible:National organisations
Target Grouper:SWE Old forum status:


For some years now we are writing the definition tables for the Swedish versions of NordDRG parallel to the process at the Nordcase Centre but we are always comparing our tables with those from the Centre. These parallel processes increase the quality of the definition tables because the comparison detects errors in both the Centre’s tables and our own tables. When we now compared the CC exclusion lists in the PR versions for 2017 we found thousands of differences so we decided to analyze the differences in order to suggest changes to make the CC exclusion lists identical. The CC exclusion lists are present in all national versions of NordDRG so it should be important to all members of the expert group that the lists are as accurate as possible.
NPK, Sweden – 2016-12-15
All the differences between the CC exclusion lists in the Swedish PR versions for 2017 are listed in the attached Excel file C660 Comparison of CC exclusions in SWE versions.xlsx but the file also includes the rows that are identical in both versions to make it easier to see the other diagnoses that are included in each exclusion list.
We have commented on all the differences (and a very few identical exclusions and a few exclusions that were missing in both versions) and suggest changes (IN or OUT). In some cases we are uncertain which is indicated by a question mark (?) after IN or OUT. This uncertainty is particularly true for cases in which the main diagnosis is congenital malformations, and we hope that Martti, who is a pediatrician, can contribute with the right decisions.
All our judgments are done from a medical perspective and they are based on our clinical experiences. An economic analysis is hardly meaningful because many of the diagnoses are rather rare.
An overriding principle in our assessment is that sub codes also must be on the exclusion list if the base code is on the list. We noticed that a lot of Swedish sub codes were missing in the Centre’s tables which explains most of the differences.
Note that the main purpose of this analysis is to discuss and decide only about the differences, although we happened to comment also on a few identical exclusions and a few exclusions that were missing in both versions. A complete judgement of all exclusions is a tremendous amount of work. It should be done in the future, however, but then the work must be divided between all the national organizations.
NPK, Sweden – 2016-12-15
We suggest changes according to the attached Excel file C660 Comparison of CC exclusions in SWE versions.xlsx but the final changes cannot be decided until the case has been discussed in the Expert network.

C660 Comparison of CC exclusions in SWE versions.xlsx (537 KB) Mats Fernström, 2016-12-20 16:38

Technical changes case #495.xlsx (7.67 MB) Martti Virtanen, 2017-03-06 11:30

Technical changes case #495 new.xlsx (7.67 MB) Martti Virtanen, 2017-05-29 18:15


#2 Updated by Martti Virtanen over 4 years ago

2017-03-06 Martti Virtanen
This is a very large case - The technical report from Mats has over 13'000 rows.
After deletion of the rows that Mats regards as identical and where we obviously agree, there are 2912 rows left.
The work was done in a phase where we already where aware of a number of differences that needed corrections.
Therefore I have done a comparison on those 2912 rows and found out that we seem to agree on 2371 rows.
There are 50 cases about code pairs where either of the combined codes has the demanded complex category. There is no reason the define the pair separately.
Two Swedish single codes are presented by a pair in ICD+ and one of the codes belongs to the demanded complex category. The Swedish code gets the property through the linkage.
In the Swedish CC-project cases with no or very limited data were not changed from the US model. The working group felt that there is now data to support a change and the americans might have had more data to support the decission. The table has 288 rows of this type (no data).

This all leaves 199 rows that may need actions.
Altogether 1196 cases have a definiton based on ICD+ mappping. Out of these 12 are cases where discussion might be needed.

This leaves 194 unique decissions. 80 of the rows are cases that were inluced in NordDRG 2016 (Swe). 94 of the rows are cases that were included in NordDRG 2016 but Mats wants the complex categories to be removed from the dx's.

#3 Updated by Kristiina Kahur over 4 years ago

Finnish National DRG-centre 2017-3-6

We agree to revise the CC-exclusion list and bring the clarity to Common and national versions. A tremendous work has been done by Mats and Martti which is a lot appreciated. Is there anything the national organizations (other than Sweden) can do or prepare to be ready for discussion in Expertnetwork meeting? Or the item will be discussed between Mats and Martti as Mats suggested in his email on 23 Feb in separate meeting in Stockholm?

#4 Updated by Anonymous over 4 years ago

Could you please send this mail to all of us in the expertgroup

Big hug
Susann Bengtsson-Berggren

Avdelningen för Statistik och jämförelser
Öppna jämförelse 1
106 30 Stockholm
Växel 075-247 30 00

Från: Berggren, Susann
Skickat: den 20 mars 2017 10:41
Till: ''
Ämne: SV: [Nordic Casemix Centre - Case #495] Comparison of CC exclusions in SWE versions

As we decided on the Expert group meeting in Riga, there will be a meeting in Stockholm to discuss the exclusion list. The meeting will be on Tuesday the 9th of May, 10:00-16:30, at Socialstyrelsen, Rålambsvägen 3.
Everybody that is interested is very welcome, but we need to know in advance who wants to attend to be able to book the right room.
Please answer to Susanns mail if you want to join.

#5 Updated by Martti Virtanen over 4 years ago

2017-03-13 Expert group
This item was postponed for a special meeting by interested members of the expert group.

#6 Updated by Martti Virtanen over 4 years ago

  • File Technical changes case #495 new.xlsx added

2017-05-09 Martti Virtanen
The small working group had its first meeing. The group could check a few hundred lines from the table. The list was also expanded during the work. We will continue the work next week.

The new technical changes document includes the codes from the previous Technical changes file but it has been expanded with rows that were added at the first meeting of the small Complex-woarking goup (marked at ncc ID column with 'Add'. In addtion the change so Komplex table from the case #377 have also been added. Theri NCC id is #377. The original rows are marked with #495.
At the first meeting we were aple to discuss less than half of the cses, so we will continue the work in a week.
The cases that have discussed are marked in the first column with dirty red colour. For these cases the first column 'IN/OUT Decicission' is the result of decission in the group. For others it is copied from the proposal from Sweden i.e. the next column.
The third column ('Current NCC') indicates the current NordDRG Com status. This is a result of a komparison with the current 'Komplex' table in the columsn N, O and P. The third column has value 'OUT' for cases where the property property is missing, 'IN' when the value is present. 'IN - *' when the manifestation code of a pair has the property . 'IN - # ' when the etiological code of a pair has the property
When one of the codes in a pair belongs to an exclusion group, the pair does not need to belong to the exclusion list. The exclusion is valid from either of the codes.

#7 Updated by Martti Virtanen over 4 years ago

2017-05-29 Martti Virtanen
At the second meeting we were able to check the rest of the cases. We did not check all cases that we agreed already.
In the attached technical changes all rows that were discussed are marked with dirty red color in the first 3 columns. The ones with white background were not individually checked, but as said we agreed in advance on those cases.
Int the checkup with Liselotte we noticed that there is one questionable decission. We decided that codes with dgcat 99M00 that cannot be used as principal dx, should not belong to any komplex-list. However, we included the code R5720. We might reconsider that case, but it is included now.

#8 Updated by Martti Virtanen over 4 years ago

  • File deleted (Technical changes case #495 new.xlsx)

#9 Updated by Martti Virtanen almost 4 years ago

  • Status changed from Active to Accepted

Also available in: Atom PDF