Case #376

Outpatient visits with impossible/unspecified procedure

Added by Anonymous almost 5 years ago. Updated about 3 years ago.

Status:AcceptedStart date:2015-02-09
Priority:MinorSpent time:-
Assignee:Mats Fernström
Category:-
Target version:Expert Group 2015
Initiator:Sweden Target year:2016
Case type:Minor Owner / responsible:National organisations
MDC: Old forum status:
Target Grouper:SWE

Description

National ID: NPK ID C382
Initiated: 2011-08-31
Initiator: Mats Fernström, NPK, Sweden
Responsible at National organization: Mats Fernström
Sent to NordDRG Forum: 2015-01-27
Status: ACTITEM - Active item
----------------------------------------------------------------------------------------------------------------
Problem
The validation rule with ORD 000D000020 says that outpatient cases with the procedure property 00S99 (Unspecified or impossible procedure) but without the diagnosis property 00X99 (Intervention code valid for single use) shall be grouped to DRG Z74O (former DRG 470) but it is not working as intended. This depends on that all codes for type of visit and all codes for professional category have the diagnosis property 00X99. Thus a code for impossible/unspecified procedure can be registered as the only procedure code without leading to DRG Z74O.
Analysis
NPK, Sweden – 2014-12-16
The diagnosis property 00X99 is automatically generated in NDMS to all procedure codes that don’t have the procedure property 00S99. However, it is not possible to give 00S99 to the codes for type of visit or to the codes for professional category because then all cases without any further procedure code will be grouped to DRG Z74O. Besides, the codes for type of visit and the codes for professional category are neither “unspecified” nor “impossible”.
The validation rule can be made effective by the following steps:
• A new procedure property 00S90 (Type of visit) is added to all procedure codes for type of visit.
• A new procedure property 00S95 (Type of personal) is added to all procedure codes for type of personal.
• Diagnosis property 00X99 is withdrawn from all procedure codes with 00S90 or 00S95. (The automatic generation of 00X99 in NDMS is changed so that 00X99 is added to all codes that don’t have 00S90, 00S95 or 00S99.)
There could as well be one single new procedure property in common for the codes for type of visit and for type of personal but the suggested separate procedure properties make the system flexible if we in the future want the codes for type of visit and for type of personal to be handled differently in the grouping process.
Suggestion
NPK, Sweden – 2014-12-16
The validation rule with ORD 000D000020 is made effective in NordDRG 2016 by the steps described above. The suggestion is also recommended by the Swedish Expert Group for NordDRG (SWEX). The technical changes are described in detail in the file “Suggestion NPK ID C382.xls.
DRG change
Probably minor. Outpatient cases with only impossible or unspecified procedures will be grouped to DRG Z74O (former DRG 470) as wanted.

C382 to Forum.docx (17.2 KB) Anonymous, 2015-02-09 09:34

C382 Suggestion NPK ID C382.xls (50.5 KB) Anonymous, 2015-02-09 09:34

NCSPplus_ZZ_codes.jpg (114 KB) Martti Virtanen, 2015-03-24 21:56

NCSPplus_ZZ_codes.xlsx - Updated version (24.8 KB) Martti Virtanen, 2015-03-31 13:15

Updated Technical changes #376 SWE 20150408.xlsx (25.1 KB) Ralph Dahlgren, 2015-04-08 16:52

00S90&00S95interventions.jpg (240 KB) Martti Virtanen, 2015-04-14 14:23

Case #376 Technical changes.xlsx - Updated version (28.2 KB) Martti Virtanen, 2015-04-14 14:23

00S90_00S95interventions.jpg (225 KB) Martti Virtanen, 2015-04-14 14:28

Decision 2016 Case #376.xlsx (543 KB) Mats Fernström, 2016-01-28 18:10

Case #376 Technical changes for 2017.xlsx (1.38 MB) Mats Fernström, 2016-03-16 16:09

Technical changes #376.xlsx (129 KB) Martti Virtanen, 2016-04-16 00:46

History

#1 Updated by Anonymous almost 5 years ago

  • Case type set to Minor

#2 Updated by Martti Virtanen almost 5 years ago

2015-02-10 Martti Virtanen
I support this model.

#3 Updated by Martti Virtanen over 4 years ago

2015-03-13 Expert group/MV
The proposal was accepted.

#4 Updated by Martti Virtanen over 4 years ago

  • File NCSPplus_ZZ_codes.jpg added
  • File NCSPplus_ZZ_codes.xlsx added
  • File Case #376 Technical changes.xlsx added

2015-03-24 Martti Virtanen
When I checked the case for all versions I realized that there are some problems. The list of codes that should be considered are the NCSP+ codes starting with ZZ

The attached Excel-file includes the linked national codes. Most of these are on the Swedish list. Most of these codes get currently automatically 00X99 propety and may cause the error presented in the Swedish proposal.

The division of the group to codes for personnel and contact type is both difficult and theoretical. For the grouping it is unnecessary.
I propose that all ZZ-codes of NCSP+ (and all linked national codes) are given a new property 00S98 'Contact attribute'.

The Foxpro NDMS I have already modifies so that all '00S9x' properties block the automatic addition of Dgprop 00X99 to the intervention code at issue.

#5 Updated by Ralph Dahlgren over 4 years ago

  • File NCSPplus_ZZ_codes_Revised_by_SWE.xlsx added

We have solved the problem with the help of Fredrik Hansen. We do not think that it is correct to put all of these additional codes together. These codes are of different types and should have different properties.
Our suggestions are shown within the enclosed Excel file "NCSPplus_ZZ_codes_Revised_by_SWE".

#6 Updated by Martti Virtanen over 4 years ago

2015-03-31 Martti Virtanen
It is not technically any problem to divide the codes to two groups. I have done it following the Swedish proposal. However, this does not affect the DRG assignment in any way.
All codes in the list have now either 00S90 or 00S95. It does not matter that some of them did already before get the property 00X99. I prefer not to put any of these to 00S99, although that does neither affect the DRG assignment.

The updated version of NCSPplus_ZZ_codes includes the Swedish comments and the Technical changes is updated accordingly.

#7 Updated by Martti Virtanen over 4 years ago

  • File deleted (NCSPplus_ZZ_codes.xlsx)

#8 Updated by Martti Virtanen over 4 years ago

  • File deleted (Case #376 Technical changes.xlsx)

#9 Updated by Martti Virtanen over 4 years ago

  • File deleted (NCSPplus_ZZ_codes_Revised_by_SWE.xlsx)

#10 Updated by Ralph Dahlgren over 4 years ago

2015-04-08 NPK Sweden
We have compared "Case #376 Technical Changes.xlsx" with our original proposal.
In the enclosed Excelfile "Updated Technical Changes #376 SWE 20150408" on the first sheet we have noticed some problems.

We do not see the reason for having procpro 00S98 Contact attribute. We can not find this procedure property in "Case #376 Technical Changes.xlsx".

#11 Updated by Martti Virtanen over 4 years ago

2015-04-14 Martti Virtanen
I have missed the codes that do not belong to the ZZ-group. It is obvious that they should behave in similar manner.
Have of the cases on the lates list from Sweden are cases that are linked to codes, that are on the list. They will chang automatically. In principle each NCSP+ code should appear on the list only once, since the other changes will follow from the change of the NCSP+ code. Althoug this is not an error, I have not been quite systematic in this respect.

I have now added the missing codes on the Technical changes list.

The NCSP+ codes that get 00S9x -propery and loose 00X90 property are following:

#12 Updated by Martti Virtanen over 4 years ago

  • File deleted (Case #376 Technical changes.xlsx)

#14 Updated by Martti Virtanen over 4 years ago

  • File deleted (00S90&00S95interventions.xlsx)

#15 Updated by Mats Fernström almost 4 years ago

Mats Fernström, NPK, Sweden 2016-01-28
We have now turned 180 degrees and want to abolish DRG Z74 (Impossible or too unspecified procedure). There are mainly two reasons for this turnaround.
Firstly, we see it as a waste of time to continue discussions about which procedure codes that shall have the property 00S90 (Type of visit) etcetera. In relation to the total number of cases in the patient registry, only a few cases are grouped into DRG Z74 and there are more important things to investigate (or to quote Shakespeare, it is “Much Ado About Nothing”).
Secondly, DRG Z74 gives no medical information. With the changes described below the cases will instead be grouped to conservative DRGs (based on diagnoses) which provides significantly better medical information.
Technical changes:
• In the table Drglogic, the rules with ORD 000D000020 and 000D000021 leading to DRG Z74 are deleted.
• In the table Proc1, all codes with PROCPR 00S99 are transferred to the table Proc0 and the property 00S99 is removed.
• In the table Proc1, all rows with DGPROP 00X99, PROCPR 00S90, 00S91 or 00S95 are deleted. (This will make the table more than 10 700 rows shorter!)
• In the table Drgnames, DRG Z74N and Z74O are deleted.
• In the table Dgprop, the row with 00X99 is deleted.
• In the table Procpro, the rows with 00S90, 00S91, 00S95 and 00S99 are deleted.
• In the table Rtc, the row with B is deleted.
The changes are also described in Decision 2016 Case #376.xlsx. Making that file we noticed that the rule for DRG Z70N with ORD 000D000030 has wrong Rtc. It should be changed to 1.
We want these changes to be introduced in the Swedish planning version for 2017.

#16 Updated by Anonymous almost 4 years ago

  • Target version changed from Minor-Proposals-for-2016 to Expert Group 2015

#17 Updated by Martti Virtanen over 3 years ago

2016-03-10 Nordcase - Martti Virtanen
I agree with the two points and remind you of the third one: This is a case where we want to teach hospitals coding through DRG. Our record here is extremely poor!

The change cannot be done for Sweden only or it would at least be very complicated. As the idea originally came from Sweden, I expect that the other countries have great problems in removing it.

Technically, I think we will retain the actively defined properties, but I will probably remove the automatical creation of 00X99.

#18 Updated by Mats Fernström over 3 years ago

Mats Fernström, NPK, Sweden, 2016-03-16
This case was discussed at the Expert Network meeting 2016-03-14 - 15 and it was decided to implement the changes according to the Swedish proposal. However, the file Decision 2016 Case #376.xlsx has some errors. The attached file Case #376 Technical changes for 2017.xlsx contains the correct changes to create NordDRG SWE 2017. Note that the instructions for the tables Proc0 and Proc 1 are separated for the NCC and the SOS versions. Which version to change is specified in the first column.

#19 Updated by Martti Virtanen over 3 years ago

2016-03-15 Expert group (MV)

Expert group accepted the Swedish proposal that means that the original proposal of case #241 will rejected.
All changes related to that proposal will also be rejected. This includes at least cases #101, #241, #278, #350, #376 and #389.

The two rules using property 00S99 (000D000020 and 000D000021) will be inactivated and consequently the property 00X99 is neither no more needed. The property 00X99 is automatically created to all codes that do not have a property from the group 00S9x. This rule will be inactivated. The properties in the group 00S9x will also become inactive and will be invisible in the definition tables.

The technical changes produced by Sweden are unnecessary complicated. We do not report 00X99 property changes in the technical changes tables because they are secondary and automatic. We do neither report proc0 table changes since proc0 table is only for easing the analysis of the tables and not part of the actual definition of grouping. However, there are also some obvious errors in the tables delivered from Sweden.

The two changes that need to be done are:
1) Removal of the 2 lines
2) Programme change to inactivate the production of the 00X99 properties.

The 00S9x properties will be retained inside NDMS.

#20 Updated by Anonymous about 3 years ago

  • Status changed from Active to Accepted

Also available in: Atom PDF