Case #622

Access procedures for dialysis

Added by Mats Fernström 5 months ago. Updated about 1 month ago.

Status:AcceptedStart date:2019-01-29
Priority:MajorSpent time:-
Assignee:Mats Fernström
Category:-
Target version:NordDRG 2020
Initiator:Sweden Target year:2020
Case type:Major Owner / responsible:
MDC:MDC11 Old forum status:
Target Grouper:COMMON

Description

This case could can be regarded as a continuation of the closed case #503 but we think that it is best to have a new case instead of reactivating #503, otherwise it will be hard to see these new suggestions and decisions because there is already so much text and so many Excel files in #503.

Problem
The grouping of access procedures for dialysis is highly inconsequent in the Swedish version, especially if dialysis is performed during the same episode. Some of the problems, but not all, were solved in Common version by the split of DRG 315 (as decided in case #503) into DRG 315A ‘Kidney and urinary tract disorder, insertion of hemodialysis catheters’, 315D ‘Kidney and urinary tract disorder, construction or closure of arteriovenous fistula’ and 315E ‘Kidney and urinary tract disorder, insertion of peritoneal dialysis catheter. Sweden did not accept that split, mainly because we thought the groups would be too small, but we have now done a thorough analysis and found that we need a similar division but with minor adjustments. First a summary of the problems.
• A case with diabetes with renal insufficiency, e.g. E102 ‘Diabetes mellitus typ 1 med njurkomplikationer’ (E1020 ‘Insulin-dependent diabetes mellitus with renal complications) and a construction of an arteriovenous fistula, e.g. PBL30 ‘Anläggande av a-v fistel från a. radialis eller a. ulnaris’ (PBSL30 ‘Construction of arteriovenous fistula from radial or ulnar artery’) with or without dialysis during the same episode is grouped to DRG L39C/E ‘Andra operationer för endokrin, nutrit & metab sjukdomar …’ (292/293 ‘Other endocrine, nutritional or metabolic disease o. r. procedure …’).
A case with renal insufficiency of other causes, e.g. N185 ‘Kronisk njursvikt, stadium 5’ (N1880 ‘Chronic renal failure, stage 5’) and the same operation and without dialysis during the same episode is grouped to DRG M29A/C/E ‘Andra operationer vid njur- & urinvägsproblem …’ in the Swedish version and to DRG 315D ‘Kidney and urinary tract disorder, construction or closure of arteriovenous fistula’ in Common.
Thus, access surgery for hemodialysis is grouped differently depending on whether the patient has renal insufficiency due to diabetes or due to other reasons, but we think that it should be the same DRG.
• The second case mentioned above (renal failure of other causes than diabetes plus construction of an arteriovenous fistula) is grouped to the conservative DRG M31 ’Njursvikt, …’ (316 ‘Kidney failure’) in the Swedish version if dialysis (procpro 11S09) is performed during the same episode because the rules for DRG M29/315 don’t allow the presence of 11S09. It is true that hemodialysis isn’t done via a freshly made arteriovenous fistula but the patient might have an ongoing peritoneal dialysis and also the procedure codes for that have procpro 11S09. It is unreasonable that such an extensive operation as construction of an arteriovenous fistula is grouped to a conservative DRG. This problem is solved in Common by acceptance of 11S09 in the rules for DRG 315A, 315D and 315E.
• A case with diabetes with renal insufficiency and access surgery for peritoneal dialysis, e.g. JAK10 ‘Laparotomi och inläggande av peritoneal dialyskateter’ (JASK10 ‘Laparotomy and insertion of peritoneal dialysis catheter’) is grouped to DRG Z60 ’Annan sällsynt, eller felaktig, kombination av huvuddiagnos och åtgärd’ / 477 ‘Rare or incorrect combination of diagnosis and other procedure’ in the Swedish version if dialysis (procpro 11S09) is performed during the same episode. This is also unreasonable because it is fully possible to cautiously start dialysis via a freshly made PD catheter. This problem is also solved in Common by acceptance of 11S09 in the rules for DRG 315A, 315D and 315E.
• A case with diabetes with renal insufficiency and DJ004 ’Byte av peritoneal dialyskateter’ (JAXX34 ‘Laparoscopy and revision of peritoneal dialysis catheter’) is grouped to the conservative DRGs L40/294 ’Diabetes, >35 år’ or L45/295 ’Diabetes, 0-35 år’. This is because DJ004/JAXX34 hasn’t OR = 1, which it should have. It is normally not possible to use the same canal for the new PD catheter so “byte” (=exchange) is in practice the same as JAK10/JASK10. This is also reflected in the fact that the code text in NCSP+ has the word "laparoscopy". This problem is present also in Common.
• Despite the alternatives mentioned above, most of the cases with access procedures for dialysis are grouped to DRG M29A/C/E ‘Andra operationer vid njur- & urinvägsproblem …’ (315 ‘Other kidney & urinary tract o. r. procedures’) in the Swedish version. These are residual groups in MDC 11 and we have a standing recommendation from the Swedish Expert Network for NordDRG (SWEX) to try to break out significant procedures from such residual groups and instead let them go to more descriptive groups. The split of DRG 315 into DRG 315A, 315D and 315E (as decided in case #503) was from this aspect a step in the right direction.
• DRG 315A ‘Kidney and urinary tract disorder, insertion of hemodialysis catheters’ contains procedures with PROCPR 11S12. The selection of codes that have received this property is problematic.

TPX10/PHXA20 ‘Implantation of vascular injection port’ is usually done for fluid therapy, nutrition, or cytostatic therapy but not for dialysis purpose.
PHXA30 and PHXA32 (Insertion of central venous catheter through …) are in NCSP+ mapped to Swedish codes for insertion of ordinary central venous catheters which are impossible to use for dialysis. Sweden has 17 codes for insertion of special central venous catheters for dialysis (they all have “dialys” in the code text) and only those should be included in a Swedish variant of DRG 315A. (They are mapped to PHXA50 in NCSP+ but there are also other codes that should not be mapped to PHXA50. Martti has to discuss this mapping with Ralph who is a specialist in nephrology.)

• DRG 315D ‘Kidney and urinary tract disorder, construction or closure of arteriovenous fistula’ is problematic because construction and closure of arteriovenous fistula are in the same group. Closure is a very simple procedure and we would like to limit the group to the access procedures for dialysis, i.e. constructions.
• DRG 315E ‘Kidney and urinary tract disorder, insertion of peritoneal dialysis catheter’ should contain more codes than JAK10/JASK10.

Analysis

NPK, Sweden – 2019-01-29
The complete analysis is reported here for interested. If you don’t want to follow all details of the analysis, go directly to the summary at the end of this section.
For this analysis we invented eight temporary DRGs and regrouped the national cost database (KPP database) for 2017 that initially had been grouped with the Swedish PR version of NordDRG for 2019. The new temporary DRGs were:
• M23N ‘Access surgery for hemodialysis, inpatients’ and
• M23O ‘Access surgery for hemodialysis, outpatients’ for cases with constructions of arteriovenous fistulas.
• M24N ‘Catheter for hemodialysis, inpatients’ and
• M24O ‘Catheter for hemodialysis, outpatients’ for cases with insertion of central venous catheters for hemodialysis.
• M25N ‘Access surgery for peritoneal dialysis, inpatients’ and
• M25O ‘Access surgery for peritoneal dialysis, outpatients’ for cases with insertion of catheter for peritoneal dialysis (PD catheter) via laparotomy or laparoscopy
• M26N ‘Catheter for peritoneal dialysis, inpatients’ and
• M26O ‘Catheter for peritoneal dialysis, outpatients’ for cases with percutaneous insertion or exchange of PD catheter.
The technical changes for implementation of the temporary DRGs, and the procedures involved, are specified in the file Appendix_C772.xlsx, in the sheets beginning with “TC”.
Abbreviations and calculations in Appendix_C772.xlsx:
AC_1 and AC_2 = average cost (SEK) for the group before and after regrouping.
N_1 and N_2 = number of cases in the group before and after regrouping.
N_DIFF_ABS = N_2 minus N_1 in absolute numbers, i.e. without any possible minus signs.
CV_1 and CV_2 = coefficient of variation (calculated as standard deviation divided with the average) before and after regrouping.
CV_DIFF = CV_2 minus CV_1. A negative value thus means that the group became less heterogeneous after regrouping.
W_CV_DIFF (weighted CV difference) = N_DIFF_ABS multiplied with CV_DIFF. Thus, a minor change in the number of patients per group does not weigh as heavily as a larger one when summing the changes in the CV for several groups.
SUM W_CV_DIFF = the sum of the weighted CV differences for the groups involved. This value shows the average change of CV for the groups and a negative value indicates that the groups on average became less heterogeneous after regrouping.
Y/I_1 = the original cost outlier categorizing in the KPP database where Y = outlier ("ytterfall" in Swedish) and I = inlier ("innerfall" in Swedish).
Y/I_2 = cost outlier categorizing according to the quartile method (cases more expensive than 1,5 interquartile range above the 3rd quartile = outliers). Still Y = outlier and I = inlier. This method was used for the temporary DRGs because the original method used in the KPP database demands recalculation of the entire database even if the content of a single DRG is changed, which we considered to be too cumbersome and time consuming. Thus Y/I_1 and Y/I_2 are not quite comparable.
Sheet “AFFECTED_DRGs” shows all groups where the content was changed by the regrouping. Note that the cost data is untrimmed which explains the relative high variation (CV). All together 42 DRGs were changed but for 32 of them the changes in numbers and CV were very small or even negligible. The number of patients in the DRGs M29/315 ‘Other kidney & urinary tract o. r. procedures’, M31/316 ‘Kidney failure’ and M32/317 ‘Admit for renal dialysis’ were significantly reduced, as expected, since several of them were regrouped to the new temporary DRGs. The CVs were both increased and decreased but the overall effect (= the sum of the weighted CV difference) was that the CV was decreased (= better homogeneity) if DRG M29C/ 315C ‘Other kidney & urinary tract o. r. procedures w cc’ was excluded from the calculations. We think that we have to accept the increase in CV for DRG M29/315 because it is a residual group and it is almost impossible to get such groups homogeneous.
Sheet “DIFF_IP” shows all 1 609 inpatients where the regrouping resulted in a different DRG compared to the initial grouping. Eight of the 1 609 inpatients were regrouped to DRG Z60N ’Annan sällsynt, eller felaktig, kombination av huvuddiagnos och åtgärd’ (477 ‘Rare or incorrect combination of diagnosis and other procedure’) because we had forgotten to construct rules for the new temporary groups in the post-MDC area. With such rules, these eight patients would have been correctly grouped into the new temporary groups that are written in the same field as Z60N.
Seven of the 1 609 inpatients were regrouped from a conservative DRG to an existing surgical DRG due to the OR change from 2 to 1 for the procedure code DJ004 ’Byte av peritoneal dialyskateter’ (JAXX34 ‘Laparoscopy and revision of peritoneal dialysis catheter’) which actually means a better description. The cost perspective is less interesting when it, as here, only are a few cases per DRG.
The absolute majority, 1 594 of the 1 609 inpatients, were regrouped to one of the new temporary DRGs which are discussed further down.
Sheet “DIFF_OP” shows all 280 outpatients where the regrouping resulted in a different DRG compared to the initial grouping. One single case was transferred from the Swedish DRG F77O ‘Punktion/biopsi andra bukorgan, öppenvård’ (Puncture or biopsy of other abdominal organs, outpatients) to DRG U19O/443O 'Other o. r. procedures for injuries, short therapy' due to the OR change from 2 to 1 for the procedure code DJ004/JAXX34 which is OK. All other outpatients were regrouped to one of the new temporary DRGs (see sheets M23O, M24O, M25O and M26O) but the groups were too small to become separate DRGs.
Sheet “M23N” shows the temporary DRG 'Access surgery for hemodialysis, inpatients' which has 732 cases including those that by mistake were grouped to DRG Z60N. After trimming according to the quartile method (selection “I” in the column “Y/I_2”) the number is 695 cases with average cost = 56 178 SEK and CV = 34%. We believe that these are good values and that the group is medically meaningful and therefore it should be implemented in NordDRG Swe.
Sheet “M24N” shows the temporary DRG 'Catheter for hemodialysis, inpatients' which has 571 cases including those that by mistake were grouped to DRG Z60N. After trimming according to the quartile method (selection “I” in the column “Y/I_2”) the number is 531 cases with average cost = 138 611 SEK and CV = 86%.
As an attempt to get more homogeneous groups and fewer outliers, we tried to divide M24N into M24A (very complicated), M24C (complicated) and M24E (not complicated) – see those sheets – but M24E became too small (less than 100) for a separate DRG. We then tried to keep M24A but merge M24C and M24E (see sheet M24CE). The improvement in CV and reduction of outliers was marginal, however, and we should not increase the number of groups in the system unnecessarily so we will not suggest a split of the group M24N.
A somewhat unexpected finding was that the cases with insertion of catheter for hemodialysis (M24N) were so much more expensive than the cases with construction of arteriovenous fistulas (M23N). We believe that a possible explanation may be that these patients are more ill and have a more severe kidney failure so the need for dialysis is more urgent. They cannot wait for the a-v fistula to be done and the postoperative healing.
Sheet “M25N” shows the temporary DRG ‘Access surgery for peritoneal dialysis, inpatients’ which has 249 cases. Trimmed values for average cost and CV are 41 771 SEK and 37% which is OK.
Sheet “M26N” shows the temporary DRG 'Catheter for peritoneal dialysis, inpatients' which has only 50 cases which is far too few to become a separate DRG. We therefor suggest merging of M25N and M26N.
Sheet “M25N_M26N” shows the merging of M25N and M26N. There are 299 cases. Trimmed values for average cost and CV are 44 983 SEK and 43% which is OK.
Sheet “Access for PD” compares the different procedures for insertion of the PD catheter. The number of cases with exchange of the PD catheter (DJ004/JAXX34) is rather low but the cost data don’t contradict our clinical experience that exchange is as resource demanding as a primary insertion. Thus we think that the change from OR=2 to OR=1 is OK.
An unexpected finding was that the cases with TJA33/JAXX33 ‘Perkutan inläggning av peritoneal dialyskateter’ (Percutaneous introduction of peritoneal dialysis catheter) were that expensive but the explanation is that most of them had another procedure with OR = 1 and the others had general anesthesia. This study cannot answer the question whether TJA33/JAXX33 should have OR=1 or not.

Analysis summary
Cases with access procedures for dialysis (construction of arteriovenous fistulas, insertion of central venous catheters for hemodialysis and insertion of catheters for peritoneal dialysis) are scattered on a lot of different DRGs in the Swedish version (see Appendix_C772.xlsx, sheet “DIFF_IP”, column C or D), especially when dialysis is performed during the same episode. Without dialysis most cases are grouped to DRG M29/315 ‘Other kidney & urinary tract o. r. procedures’ which are residual groups that give a rather poor medical description. We have studied the possibility to introduce more descriptive DRGs for all cases with access procedures for dialysis, irrespective of actual dialysis during the same episode. The test groups were:
• M23N ‘Access surgery for hemodialysis, inpatients’ and
• M23O ‘Access surgery for hemodialysis, outpatients’ for cases with constructions of arteriovenous fistulas.
• M24N ‘Catheter for hemodialysis, inpatients’ and
• M24O ‘Catheter for hemodialysis, outpatients’ for cases with insertion of central venous catheters for hemodialysis.
• M25N ‘Access surgery for peritoneal dialysis, inpatients’ and
• M25O ‘Access surgery for peritoneal dialysis, outpatients’ for cases with insertion of catheter for peritoneal dialysis (PD catheter) via laparotomy or laparoscopy
• M26N ‘Catheter for peritoneal dialysis, inpatients’ and
• M26O ‘Catheter for peritoneal dialysis, outpatients’ for cases with percutaneous insertion or exchange of PD catheter.
We also tested to divide some of the groups in different CC levels and to combine some of the groups. The tests are summarized in Appendix_C772.xlsx, sheet “Summary”. Considering the wishes of good medical description, significant cost differences, limited number of groups and relevant sizes of the groups we found the following groups and combinations most appealing:
M23N: Access surgery for hemodialysis, inpatients
M24N: Catheter for hemodialysis, inpatients
M25N + M26N: Open or laparoscopic insertion or exchange of catheter for peritoneal dialysis, inpatients
M23O + M24O + M25O + M26O: All access procedures for dialysis, outpatients
The test showed no serious adverse effects on other existing groups and the result supports the split of DRG 315 that was decided in case #503. The test also supports the clinical judgement that DJ004/JAXX34 ‘Laparoscopy and revision of peritoneal dialysis catheter’ should have OR=1 instead of OR=2.

NPK, Sweden – 2019-01-29
Suggestions
Sweden will introduce three new inpatient DRGs and one outpatient DRG in MDC 11 for cases with access procedures for dialysis:
• DRG M23N ‘Insertion of catheter for hemodialysis’
• DRG M24N ‘Construction of arteriovenous fistula for hemodialysis’
• DRG M25N ‘Insertion of catheter for peritoneal dialysis’
• DRG M26O ‘Access procedures for dialysis, short therapy
Note that the codes don’t have the same meaning as in the analysis above!
The inpatient groups correspond rather well to DRG 315A ‘Kidney and urinary tract disorder, insertion of hemodialysis catheters’, 315D ‘Kidney and urinary tract disorder, construction or closure of arteriovenous fistula’ and 315E ‘Kidney and urinary tract disorder, insertion of peritoneal dialysis catheter’ in Common version but for better conformity we suggest some changes in Common.

1. Cases with renal insufficiency due to diabetes (diagnosis codes with pdgprop 11P01) should be included in the groups. This demands rules based on 11P01 + 11S12 for DRG 315A, 11P01 + 11S13 for DRG 315D and 11P01 + 11S14 for DRG 315E and they must be placed before the rules for DRG 292/293 ‘Other endocrine, nutritional or metabolic disease o. r. procedure …’) because the diagnosis codes with pdgprop 11P01 belong to MDC 10 and several of the procedure codes with 11S12, 11S13 or 11S14 also have 10S08 which leads to DRG 292/293.
2. The rules for DRG 315D and 315E are now only in the post-MDC area. Is that intended? Otherwise there should be rules also in the MDC 11 area. The rules from the post-MDC area can be copied to the MDC 11 area but the rule for DRG 315E should be supplemented with a rule with 00X10 (General anesthesia) because JAXX33 ‘Percutaneous introduction of peritoneal dialysis catheter’ with OR=2 should be included in DRG 315E if it is done in general anesthesia.
3. The purpose with post-MDC rules is that they shall catch up cases with the procedure property at issue but with another principal diagnosis because that gives a better medical description than if the cases go to DRG 468 ‘Rare or incorrect combination of diagnosis and extensive procedure’ or DRG 477 ‘Rare or incorrect combination of diagnosis and other procedure’. If this principle is to work, the "mdc" field must be empty. Thus, “11” in the post-MDC rules for DRG 315D and 315E should be deleted. There should also be a post-MDC rule for DRG 315A.
4. TPX10/PHXA20 ‘Implantation of vascular injection port’ is not for dialysis purpose and should not be included in DRG 315A. Thus, procpro 11S12 should be deleted from that code. (Sweden will not use 11S12 but instead introduce a new procedure property in the rules for DRG M23N since we have more specific codes for insertion of catheters for hemodialysis than the other nations.)
5. PFSL90 ‘Construction of arteriovenous fistula in lower leg or foot’ is probably seldom used for dialysis but should be included in DRG 315D for principal reasons and therefore procpro 11S13 should be added.
6. The codes for closure of arteriovenous fistulas, a very simple procedure, should not be included in DRG 315D and therefor procpro 11S13 should be deleted from them. (If this isn’t accepted, Sweden will not use 11S13 but instead introduce a new procedure property in the rules for DRG M24N.)
7. DRG 315E is based on procpro 11S14 but there is only one procedure, JASK10 ‘Laparotomy and insertion of peritoneal dialysis catheter’, with that property. JASK14 ‘Laparoscopy and insertion or revision of peritoneal dialysis catheter’, JAXX34 ‘Laparoscopy and revision of peritoneal dialysis catheter’ and JAXX33 ‘Percutaneous introduction of peritoneal dialysis catheter’ should also have procpro 11S14.
8. According our clinical judgement and supported by the analysis, JAXX34 ‘Laparoscopy and revision of peritoneal dialysis catheter’ should have OR=1 instead of OR=2.

NPK, Sweden – 2019-01-29
Decisions
As mentioned above, Sweden will introduce three new inpatient DRGs:
• DRG M23N ‘Insertion of catheter for hemodialysis’
• DRG M24N ‘Construction of arteriovenous fistula for hemodialysis’
• DRG M25N ‘Insertion of catheter for peritoneal dialysis’
If the suggestions above are accepted these groups will be almost identical to DRG 315A ‘Kidney and urinary tract disorder, insertion of hemodialysis catheters’, 315D ‘Kidney and urinary tract disorder, construction or closure of arteriovenous fistula’ and 315E ‘Kidney and urinary tract disorder, insertion of peritoneal dialysis catheter’ in Common version.
Sweden will also introduce DRG M26O ‘Access procedures for dialysis, short therapy’ that has no equivalent in Common version. DRG M26O is a merger of DRG M23N, M24N and M25N, but for outpatients.
For DRG M23N and M26O we will introduce a new procedure property, 11S16 ‘Insertion of specific hemodialysis catheters’.
For DRG M24N and M26O we will introduce a new procedure property, 11S17 ‘Construction of arteriovenous fistula’. The existing 11S13 can be used instead if 11S13 is deleted from the codes for closure of arteriovenous fistulas as suggested above for Common version.
For DRG M25N and M26O we probably can use the existing procedure property 11S14 ‘Insertion of peritoneal dialysis catheter’ because the suggestions above concerning that property are not controversial.
The procedure codes that will have 11S16, 11S17 or 11S14 added to them are specified in TC_C772.xlsx. Any other existing procpro beginning with 11S will be retained because they may be used in other national versions.
As suggested for Common, we will place the rules with 11P01 in the MDC 10 area just before the rules for DRG L39 and complete rules with MDC=11 in the MDC 11 area just before DRG M29 and also complete rules with MDC and pdgpro fields empty in the post-MDC area.
In each area, the rules will be placed in the order M23N, M24N, M25N, as it is for the DRGs 315A, D and E in Common.
In each area there will be one rule with procpro 11S16 and OR=P (all codes with 11S16 have OR=2) leading to DRG M23N and one rule with procpro 11S17 + OR=S (all codes with 11S17 have OR=1) leading to DRG M24N. For DRG M25N, there will be two rules, one with procpro 11S14 + OR=S and one with with procpro 11S14 + OR=empty but with dgprop1 = 00X10.
The rules for DRG M26O are copies of the rules for DRG M23N, M24N, M25N, placed in the corresponding outpatient MDC areas but with dur = <1. For more details, see TC_C772.xlsx that contains the technical changes for the Swedish version only. Technical changes for the Common version can be written when we know if the suggestions for that version are accepted.

Appendix_C772.xlsx (450 KB) Mats Fernström, 2019-01-29 10:01

TC_C772.xlsx (29.6 KB) Mats Fernström, 2019-01-29 10:01

Technical changes case #622.xlsx (3.19 MB) Martti Virtanen, 2019-03-05 14:03

TC_C772_NEW.xlsx (25.8 KB) Mats Fernström, 2019-03-18 13:37

Technical changes case #622-2.xlsx (190 KB) Martti Virtanen, 2019-03-27 14:12

Technical changes case #622-3.xlsx (190 KB) Martti Virtanen, 2019-05-15 12:33

History

#1 Updated by Martti Virtanen 4 months ago

2019-03-05 NCC (MV)
In this case as always in intervention DRG’s a very important issue is the order of the (surgical) intervention DRG’s.

Internal order within the proposal
The intervention group rules are now ordered as follows:

11S16 ‘Insertion of specific hemodialysis catheters’
11S17 ‘Construction of arteriovenous fistula’
11S14 ‘Insertion of peritoneal dialysis catheter’

Performing an intervention from one of these groups does not exclude the possibility to perform an intervention from the other groups during the same contact, at least not when the patient has been admitted to hospital.

11S16 includes 2 NCSP+ codes for catheter insertion or exchange (PHXA50 ‘Implantation or exchange of double lumen dialysis catheter’ and PHXA51 ‘Implantation or exchange of double lumen dialysis catheter in vena femoralis’) linked to several Swedish codes. Usually these interventions are performed without anaesthesia or with small area local anaesthesia at the site of puncture.

11S17 includes 5 NCSP+ codes for construction of arteriovenous fistula from different arteries of upper or lower limb. Usually performed in local anaesthesia.

11S14 includes 4 NCSP? codes for insertion of revision of peritoneal dialysis catheter in by laparotomy, laparoscopy or percutaneously. Presumably almost always demands general anaesthesia or epidural anaesthesia.

Is 11S16 the most resource intensive and 11S14 the least?

Order in relation to other existing DRG-rules.

For outpatients it is very rare that several interventions would be performed at the same contact. Thus the order is less critical and the proposed placing of the new rules is acceptable.

In MDC 10 inpatient grouping the new rules are placed before the rules using procedure property 10S08. The groups at issue include very resource intensive interventions but such combinations should be quite rare. However economic analysis should be performed.

In MDC 11 inpatient grouping the new rules are placed before various ‘other’ intervention groups with procedures with properties 11S07, 11S15, and 11S90 as well as before diabetes cases without dialysis but with various urology related interventions (11S07) and special rules for male genital organ intervention DRG’s.

PostMDC inpatient grouping rules exist currently only for DRG 315E (M25N). These rules as well as the rules for DRG 315D ‘Kidney and urinary tract disorder, construction or closure of arteriovenous fistula’ are placed in postMDC area and demand MDC 11. There are no MDC 10 or MDC 11 rules for these DRG’s. Obviously, the idea has been to allow assignment to DRG 315E and 315D only when no other rules (of MDC 11) apply and only with MD 11 principal dx and avoid any interference with other surgical rules. These are really MDC rules in postMDC area.

In the proposed model all cases of DRG 315E (and 315A & B) with MDC 10 or MDC 11 principal dx are already assigned to the groups by MDC-specific rules in the MDC’areas at issue (10 and 11) . This makes the current type of postMDC rules (that are exceptional) unnecessary. This applies also to the new Swedish rules. I see no reason why any principal dx with these interventions should be accepted as urologic care.

Thus the postMDC rules for the new DRG’s can be omitted and if the model is accepted by other countries also the existing rules for DRG 315D and 315D.

If the Swedish model is adopted by other countries the DRG 315D (and the property 11S13) becomes unnecessary. However, if nothing else is changed the closure of arteriovenous fistula would be assigned to 477. To avoid this the interventions for closure of arteriovenous fistula (group PBSM codes) must have the properties 11S07 and 11S15. Both because the rules are different in Sweden and other counties.

Procedure property 11S14 is already used by other versions of NordDRG for JASK10 (JAK10) ‘Laparotomy and insertion of peritoneal dialysis catheter’. It is correct that the other proposed additional interventions indicate insertion of peritoneal dialysis catheter:

JASK14 ‘Laparoscopy and insertion or revision of peritoneal dialysis catheter’
JAXX33 ‘Percutaneous introduction of peritoneal dialysis catheter’
JAXX34 ‘Laparoscopy and revision of peritoneal dialysis catheter’

However, why is
JASK12 ’Laparotomy and revision of peritoneal dialysis catheter’
not included.

DRG for insertion of catheter for hemodialysis.

NordDRG system already includes the DRG 315A ‘Kidney and urinary tract disorder, insertion of hemodialysis catheters’ which is not included in NordDRG Swe. The DRG is almost identical in name with the new DRG 315F (M23N) ‘Insertion of catheter for hemodialysis’ proposed by Sweden.

315A uses procedures (with property 11S12):
PHXA20 ‘Implantation of vascular injection port’
PHXA30 ‘Insertion of central venous catheter through external or internal jugular vein’
PHXA32 ‘Insertion of central venous catheter through subclavian or brachiocephalic vein’

315F / M23N uses procedures (with proposed property 11S16):
PHXA50 ‘Implantation or exchange of double lumen dialysis catheter’
PHXA51 ‘Implantation or exchange of double lumen dialysis catheter in vena femoralis’

Codes linked to PHX20, PHXA30 and PHXA32 exist in all countries. Codes linked to PHXA50 exist in Iceland, Latvia, Norway and Sweden. Codes linked to PHXA51 only in Sweden.

It is obvious that these two DRG’s try to catch the same patients. The Swedish codes are more specific for dialysis. Since we cannot directly change the basic classifications the only solution is to combine the two groups by giving the same procedure property (11S12) to all these codes. This will possibly add a few cases to the group in Sweden.

OR=1
In addition Sweden proposes than JAXX34 should have OR=1 instead of OR=2, which I think is reasonable. However, the intervention code exists also in Iceland and at least Iceland should agree.

Summary
This case is very complex and it is obvious that it should be solved together. There are already existing differences between the countries that are not based on real differences in the care. Even the simplest solution for Sweden demands discussion about the effects on other countries.

The technical changes tries to summarize all changes with a possibility to see the effects on other cases.

#2 Updated by Mats Fernström 3 months ago

Mats Fernström, NPK Sweden, 2019-03-18 (NPK ID C772)
At the meeting last week I answered to Martti’s questions. Here are my answers and comments in writing.
Yes, 11S16 is the most resource intensive and 11S14 is the least. That is reported in the sheet “Summary” in Appendix_C772.xlsx. Rounded to the nearest thousand SEK, it looks like this (trimmed):
•11S16 ‘Insertion of specific hemodialysis catheters’ 139 000 SEK
•11S17 ‘Construction of arteriovenous fistula’ 56 000 SEK
•11S14 ‘Insertion of peritoneal dialysis catheter’ 45 000 SEK
It was somewhat unexpected that the cases with insertion of catheter for hemodialysis were more expensive than the cases with construction of arteriovenous fistulas. A possible explanation may be that these patients have a more severe kidney failure so the need for dialysis is more urgent.
The answer to Martti’s comment on the positioning of the new rules in the MDC 10 inpatient grouping area is that there are economic data in Appendix_C772.xlsx (but perhaps not easily found). After the suggested change:
• Mean cost DRG L39C/292 ‘Other endocrine, nutritional or metabolic disease o. r. procedure w cc’ is 143 000 SEK
• Mean cost DRG L39E/293 ‘Other endocrine, nutritional or metabolic disease o. r. procedure w/o cc’ is 71 000 SEK.
Thus cases in DRG 292 ‘Other endocrine, nutritional or metabolic disease o. r. procedure w cc’ are the most expensive but if we place the rules with 11P01 and 11S17 after the rules for 10S08 they will be impossible because the codes with 11P01 belong to MDC 10 and the codes with 11S17 also have 10S08.
The answer to Martti’s comment on the positioning of the new rules in the MDC 11 inpatient grouping area is that the positioning is intentional. The ‘other’ intervention groups are residual groups that usually are lowest down in the hierarchy.
About the post-MDC rules for the new DRG’s we agreed on that they are not needed so I will withdraw them from our suggestion.
I will also withdraw our suggestion that closure of arteriovenous fistula isn’t included in our DRG M24N ‘Con-struction of arteriovenous fistula for hemodialysis’. Then DRG M24N will be identical to the Finnish DRG 315D and there is no need for the suggested procedure property 11S17 (luckily, for now I discovered that we have suggested 11S17 for something completely different in another case).
JASK12 ’Laparotomy and revision of peritoneal dialysis catheter’ was not included in the suggestion because I didn’t know that it exists. It is not present in the Swedish surgical classification.
Finally, we really want just the more specific codes for hemodialysis catheters in the new DRG M23N and that is why we need the new procpro 11S16.
When updating our technical changes I noticed that the outpatient rules with 11P01 were placed wrong in our original suggestion and I also noticed that there is no need to move any existing rules as I thought before.
Corrected technical changes for the Swedish version is in the file TC_C772_NEW.xlsx.

#3 Updated by Martti Virtanen 3 months ago

2018-03-27 Martti Virtanen, Nordic Casemix Centre
The changes in this case are:
1) Cases with insertion of peritoneal dialysis catheter are to be assigned to DRG 315E/M25N ‘Insertion of catheter for peritoneal dialysis’. This applies to all interventions of this type. The property 11S14 is given in addition to JASK10 also to JASK12 (only Den and Ice), JASK14, JAXX33 and JAXX34. Rules for 315E/M25N are created also in MDC10 inpatient area. These changes are for all versions of NordDRG.
2) Cases with ‘Insertion of catheter for hemodialysis’ are assigned to DRG 315A/M23N. However, in Sweden only cases where hemodialysis catheter is used (PHXA50) are assigned to this DRG by the new procedure property 11S16 ‘Insertion of specific hemodialysis catheters’. In other countries the old rules using the property 11S12 apply.
PHXA50 is the only NCPS+ code with the property 11S16 and all necessary Swedish national codes are linked to it. Rules using 11S16 are only active in NordDRG Swe.
3) For short therapy/outpatients a new DRG 316P/M26O ‘Access procedures for dialysis, short therapy’. It includes cases with interventions with properties 11S13, 11S14 and 11S16.
4) A new NCSP+ code PHXA60 ‘Implantation or exchange of other multi lumen catheter’ is created and Swedish codes SP211-SP299 are linked to this code. PHXA60 will have the same properties that PHXA50 had before the changes (OR=2, PROCPRO 40V11, 99O11, 00X99 not 11S16). Thus the new mapping will not affect DRG assignment.
5) The code PHXA51 actually is not needed. DP002, the only national version code linked to it, stands for Implantation or exchange of double lumen dialysis catheter in vena femoralis’ but SP351 and SP352 indicate the same intervention (with or without tunnelling). They are linked to PHXA50. NCSP+ has no codes for other specific veins and thus the PHXA51 will be removed and DP002 linked to PHXA50.
6) The rules for DRG 315D/M24N ‘Construction of arteriovenous fistula for hemodialysis’ and 315E/M25N ‘Insertion of catheter for peritoneal dialysis’ are placed in the MDC 10 and MDC11 instead of the postMDC area they are now as proposed by Sweden. This is valid for all versions and should not have any effect on DRG assignment in other versions than Swe where the DRG’s are new.
The English names of DRG 315D/M24N and 315E/M25N are changed as proposed by Sweden. This does not affect other versions.
7) In MDC 11 the ord codes proposed by Sweden are identical with some existing rules (both Swe and others, green marked lines in technical changes) and can thus not be used. The rules are now place together and the Swedish rule for DRG 315M/M29A ‘Other kidney & urinary tract o. r. procedures w mcc’ is kept together with the other 11S07 based rule for the same DRG after the new rules.
OBS. The latest Swedish proposal leaves both these rules between 315A/M23N ‘Insertion of catheter for hemodialysis‘ and 315D/M24N. That can also be arranged if so wished by Sweden. However, there are two additional rules for DRG 315M/M29A based on 11S15 placed later in the table.
8) JAXX34 (Den, Ice, Swe) is changed from OR=2 to OR=1 as originally proposed by Swe.
The technical changes summarize (hopefully all changes). Sheets dg and princ. dg prop have now changes, they only indicate the related dx’s and properties. In other sheets the most relevant rows are indicated with yellow background. Some name changes are highlighted with red text.
I hope I got all necessary changes included. The changes in DRG assignment for other than Swe version are minimal.

#4 Updated by Martti Virtanen 3 months ago

  • Status changed from Active to Accepted

#5 Updated by Mats Fernström 2 months ago

Mats Fernström, NPK Sweden 2019-04-16 (Swe ID C772)
I have compared Marttis file “Technical changes case #622-2.xlsx” with our file “TC_C772_NEW.xlsx” and found some errors in Marttis file.
• In the sheet drglogic it says that a rule with ORD = 411D140001 for DRG 315M/M29A shall OUT (and then the same rule is instead inserted with ORD = 411D140005). I cannot find any rule with ORD = 411D140001, neither in the SOS version for 2019, nor in Ndrg_SWE2019PR0_ Excel transport file base.xlsx. I assume that Martti means the rule with ORD = 411D1410001 but we don’t have to move that rule. There is plenty of space for the new rules with ORD = 411D140001, 411D140002, 411D140003 and 411D140004. (We had suggested the ORD values 411D140000, 411D140010, 411D140020 and 411D140030 but accept Marttis values for the sake of simplicity.)
• In the sheet proc the codes PESM10/ PEM10 ‘Closure of arteriovenous fistula of femoral artery’/ ‘Slutning av a-v fistel från a. femoralis’ and PFSL90/ PFL90 ‘Construction of arteriovenous fistula in lower leg or foot’/ ‘Anläggande av a-v fistel i underben eller fot’ should be included with insertion of PROCPR 11S13. (The latter is probably very seldom performed for dialysis but should anyhow be included for principal reasons.)

#6 Updated by Martti Virtanen about 1 month ago

2019-05-15 Nordic Casemix Centre/ Martti Virtanen
1) Since we are moving to the new NDMS system the ORD-codes will not be stable any more. For the transfer to the new system we are already forced to totally rearrange the ord-codes. In the future they will numeric characters only. No other types of charactes are allowed since the ord is maintained as a number. Exported strings will not have leading zeros. For the transfer all other than numeric characters are replaced by numbers and the first character is rearranged. Please follow the ID's to find the rule you are dealing with.
Thus in this situation only order is important and the resulting order seems to be OK.
2) We add the property 11S13 'Construction or closure of arteriovenous fistula' to the codes PESM10 and PFSL90. This is a correction to the accepted case.

Also available in: Atom PDF