Task Team Meeting - Rolling Agenda and Minutes 2020: Difference between revisions

From CCMDB Wiki
Jump to navigation Jump to search
Line 39: Line 39:
**Would not include:  indolent infections such as newly identified TB or histoplasmosis
**Would not include:  indolent infections such as newly identified TB or histoplasmosis
*This decision does NOT influence coding as a comorb an infection which is still present but WAS known pre-admission --- e.g. an osteomyelitis being treated at home with iv antibiotics.
*This decision does NOT influence coding as a comorb an infection which is still present but WAS known pre-admission --- e.g. an osteomyelitis being treated at home with iv antibiotics.
*'''Allan to put this info into the relevant template'''
*Allan to put this info into the relevant template -- DONE.


5. Further discussion of how to code a post-infectious (e.g. post-strep) glomerlonephritis?
5. Further discussion of how to code a post-infectious (e.g. post-strep) glomerlonephritis?
Line 49: Line 49:
*Thereafter, work to obtain RIS data for radiology tests.  Shared Health CIO Charles Conway [204-926-1400; cconway3@sharedhealthmb.ca].
*Thereafter, work to obtain RIS data for radiology tests.  Shared Health CIO Charles Conway [204-926-1400; cconway3@sharedhealthmb.ca].


10.  '''Next Task meeting is July 16, 2020 at 11 am via Zoom.'''  
10.  '''Next Task meeting is July 16, 2020 at 11 am via Zoom.'''


== ICU Database Task Group Meeting – June 11, 2020 ==  
== ICU Database Task Group Meeting – June 11, 2020 ==  

Revision as of 08:54, 2020 June 26

List of items to bring to task meeting

Add to this by adding the following to the article where the problem is documented:

{{DiscussTask | explanation}}
 QuestionModification date"Modification date" is a predefined property that corresponds to the date of the last modification of a subject and is provided by Semantic MediaWiki.
Iatrogenic, mechanical complication/dysfunction, internal prosthetic device or implant or graft NOSAt our last TASK meeting the decision was made to exclude spontaneous rupture of an ETT cuff or cuff leak, but we are wondering if this is correct based on what is listed in the includes section of this page and what is in Iatrogenic, mechanical complication/dysfunction, internal orthopedic prosthetic device or implant or graft or bone device and Iatrogenic, mechanical complication/dysfunction, cardiac or vascular prosthetic device or implant or graft, NOS Lisa Kaita 12:37, 2024 March 20 (CDT)20 March 2024 17:37:32
STB ICUs VAP Rate, CLIBSI Rate Summary
  • IIRC we collected CAM positive (TISS Item) specifically for this, right? If so, can we stop collecting it? And can we make sure a stoppage like this in the future results in reviewing what we collect? Ttenbergen 10:02, 2024 March 20 (CDT)
    • Delirium rate per 1000 days per unit is being reported in the OIT reports. ---JMojica 11:49, 2024 March 20 (CDT)
      • As in Delirium days is reported in Critical Care Program Quality Indicator Report? But that doesn't mention anything about per-1000-days. Ttenbergen 17:00, 2024 March 20 (CDT)
      • The rate is mentioned in the succeeding definition with the delirium days as numerator. Your proposal here is to stop collecting TISS item CAM positive which I disagree because that TISS item is being used and reported as rate in OIT report. Besides, the reason why it was dropped in in the STB VAPCLI report is because the requestor has changed. Brett Hiebert who used to request this was involved in the VAP group and another Delirium group so he asked to have both as one request. Brett had left and the VAP group filled up a new request to continue the VAP data and not on the delirium data. --JMojica 13:58, 2024 March 25 (CDT)
25 March 2024 18:59:38

Also see Task Team Meeting - Rolling Agenda and Minutes 2019

ICU Database Task Group Meeting – June 25, 2020

  • Present: Allan, Iris, Joanna, Julie, Tina, Trish, Pam, Val, Gladys, Sherry, Lisa, Barret
  • Minutes prepared by: AG
  • Action items in BOLD

1. Decisions on TISS going forward

  • We can cease the current TISS audit
  • Allan reports that there is preliminary agreement (Jodi) to routinely only collect a reduced subset of TISS --- with intermittent temporary projects to collect the entire TISS
  • First, we can permanently stop collecting current items 46-49, since those were for the Vital Signs project, which has ended
  • Tentative list of TISS items to collect routinely is: 9, 10, 11, 13-18, 21, 22, 28, 29, 32-34, 36, 40
    • There is also a question of whether to add IABP, as requested by cardiac sciences -- Allan will talk with them about it (email sent 6/26)
  • These would be collected for ICU service patients, wherever they are (ICU, or ‘boarding’ in ED or PACU or elsewhere) starting when the ICU team takes over primary responsibility/care
  • A sheet containing only the ‘reduced TISS’ will be what we would give, going forward, to the ICUs. During the temporary periods when we do the FULL TISS, we determined that we will not swap out the ICU forms, but will have our data collectors do the full TISS forms.

2. Followup on trying to get hospital-level data elements from EPR

  • Allan reported that he and Phil Jarman have a phone conference scheduled in July to discuss this.

3. Continued discussion about the prior decision to begin all (ICU and Medicine) records at the time when that team takes over primary clinical responsibility for the patient, and to do this in a uniform way for all patients on our services.

  • After extensive discussion we recognized that there are a number of interrelated issues that have to be dealt with. So, the next Task group meeting will be exclusively devoted to these. They include items per the May 18 and May 26, 2020 minutes, and also clarification of the definitions and sources of info regarding ‘accept time’ and ‘arrival time’.
  • Allan sent an email 6/26 to Heather Smith asking whether the prior problem of ED staff NOT notifying admitting office of when ICU takes over care of an ED patient has been fixed.

4. Continued discussion about the specific situation of: a diagnosis first identified after admission but which virtually certainly was present pre-admission.

  • After discussion we agreed that a SUBSET of such diagnoses should be recorded as comorbs. This subset includes chronic NONINFECTIOUS conditions only
    • Would include: cancers, collagen-vascular diseases such as Lupus, COPD
    • Would not include: indolent infections such as newly identified TB or histoplasmosis
  • This decision does NOT influence coding as a comorb an infection which is still present but WAS known pre-admission --- e.g. an osteomyelitis being treated at home with iv antibiotics.
  • Allan to put this info into the relevant template -- DONE.

5. Further discussion of how to code a post-infectious (e.g. post-strep) glomerlonephritis?

  • This is more complex than the June 11, 2020 minutes suggests. Allan to contemplate further.

6. Followup about working to reduce collector workload. Things that need to be done:

  • Find a new hosting/software infrastructure that can be in full compliance with privacy requirements. Allan and Tina will followup with CHI and Digital Health to pursue new hosting possibilities.
  • Thereafter, work to obtain CBS TraceLine for all transfusion data. Margaret Ring again (margaret.ring@blood.ca).
  • Thereafter, work to obtain RIS data for radiology tests. Shared Health CIO Charles Conway [204-926-1400; cconway3@sharedhealthmb.ca].

10. Next Task meeting is July 16, 2020 at 11 am via Zoom.

ICU Database Task Group Meeting – June 11, 2020

  • Present: Allan, Iris, Joanna, Julie, Tina, Trish, Pam, Val, Gladys, Sherry, Lisa, Barret
  • Minutes prepared by: AG
  • Action items in BOLD

1. Followup on the TISS form audit to collect information on how the ICU nurses are doing on TISS coding -- the tool is in use and so far it appears that virtually zero TISS sheets are completely accurately completed, including that the last ICU day is almost always missing (because it’s supposed to be done by the night nurses right before midnight. Since Jodi clearly wants this data, we therefore must continue to have the data collectors continue checking them all. Discussion highlighted that having the beside nurses continue to do them does help our collectors.

  • Also, Allan will ask Jodi/Bojan whether they need full TISS year-round, or only intermittently to re-jigger nursing needs (email sent 6/11/2020). If the latter, then we would consider only collecting a partial TISS (see item#4, below) most of the time.

2. Followup on trying to get hospital-level data elements from EPR

  • Allan reported on Dale Gustafson’s rationale for not approving our direct access to COGNOS. Namely that seeking this access was essentially a work-around from regular processes, and he does not support forming our own Decision Support group. He recommends that if the daily extract from the system is not as helpful as we need to continue working with that team to get it enhanced so it does
  • To this end, Allan will directly get involved with Phil Jarman and the Shared Health Decision Support team. Preparatory to that Tina, Julie and Allan will meet by ZOOM to bring him up to speed on: what we need, what we’ve tried in the past to get it and a list of failed attempts.

4. Continued discussion about the prior decision to begin all (ICU and Medicine) records at the time when that team takes over primary clinical responsibility for the patient, and to do this in a uniform way for all patients on our services. (Change to start collection at accept rather than arrive time)

  • Regarding identifying the date/time to start a new record -- Tina indicated that Cognos/ADT provides a “service start date/time” field. And we decided that we will use this as the starting time that “our” team took over care, though the collectors will keep and eye out for whether that is wrong more than a tiny fraction of the time and if so report back on it.
  • Tina/Julie/Pagasa will need change the database machinery to bring all the current and disparate systems into uniformity. As per the May 18, 2020 minutes, this includes converting the “Boarding Location” machinery into a “Physical Location” listing that keeps track of people from the time they first come under our care.
  • Other changes in machinery per the May 18 and May 26, 2020 minutes need to be made.
  • Regarding TISS collection after a record begins but before the patient reaches an actual ICU location:
    • For that interval, the data collectors will record only a subset of TISS items, as follows:
      • Intracranial pressure monitor in place
      • One vasoactive by continuous iv infusion occurred today
      • 2 or more vasoactive by continuous iv infusion occurred today
      • Pulmonary Art. (Swan Ganz) catheter in place today
      • Peripheral arterial catheter in place today
      • Central Venous Catheter in place today
      • Invasive mechanical ventilation via trach or ETT occurred today
      • CPAP or BIPAP (non-invasive) occurred today
      • HD occurred today
      • CRRT occurred today
      • PD occurred today
    • Office staff to consider how to make this work, but one possibility is to use the usual TISS sheet, keep the above-listed items in their current locations (so Teleform will read them as-is) and remove all the other items. Possibly make these special pre-ICU sheets a different color.

5. Followup up about a diagnosis found after admission but which virtually certainly was present pre-admission. Allan was unable to locate Wiki documentation on this so we will discuss at the next meeting. This will need to be documented in Template:ICD10 Guideline Como vs Admit.

6. How to code nonspecific vasculitis. After discussion we agreed to use M31.8, and Tina has renamed it as Vasculopathy/vasculitis, NOS

7. How to code Serotonin syndrome. After discussion, we agreed to use existing code Psychiatric drug NOS, overdose/toxicity.

8. New items:

  • Can the various codes for “chronic abuse/dependence/addiction” of substances be used as admit codes IF the treatment of it is a main reason for admission? Answer: Yes.
  • After discussion, we agreed that Fetal Alcohol Syndrome can apply as a comorbid condition for an adult patient.
    • Fetal alcohol syndrome (applies to baby, not mom) - Tina has changed the name to remove the parenthetical portion that it applies to baby, and Allan has altered that Wiki page accordingly.
  • How to code a post-infectious (e.g. post-strep) glomerulonephritis?
    • GN codes are by type of GN, not by the cause. When a cause is known it should be combined with the GN code.
Our old codeAcute Glomerulonephritis links to Nephritic syndrome, acute, Nephritic syndrome, rapidly progressive - are these what you are referring to or is there a category of dxs that are GNs (in which case we have not encoded it yet)?Ttenbergen 10:20, 2020 June 25 (CDT)
  • SMW


  • Cargo


  • Categories
There are questions in Other specified infectious agents as the cause of diseases classified to other chapters that need answers before I can implement this. Ttenbergen 10:20, 2020 June 25 (CDT)
  • SMW


  • Cargo


  • Categories
  • How to code refeeding syndrome? Answer is to combine the specific fluid/electrolyte disorders (or Electrolyte disorder, NOS) with a new code Tina has added T73 “Effects of starvation”, to be distinguished from E46, Malnutrition itself. Code has been added to CCMDB and wiki but wiki pages still need instructions to differentiate and code.

9. Followup about working to reduce collector workload. Things that need to be done:

  • Find a new hosting/software infrastructure that can be in full compliance with privacy requirements. Allan and Tina will followup with CHI and Digital Health to pursue new hosting possibilities.
  • Thereafter, work to obtain CBS TraceLine for all transfusion data. Margaret Ring again (margaret.ring@blood.ca).
  • Thereafter, work to obtain RIS data for radiology tests. Shared Health CIO Charles Conway [204-926-1400; cconway3@sharedhealthmb.ca].

10. Next Task meeting is June 25, 2020 at 11 am via Zoom.

ICU Database Task Group Meeting – May 26, 2020

  • Present: Allan, Barret, Iris, Joanna, Julie, Michelle, Tina, Trish, Pam, Val, Mindy, Gladys, Sherry
  • Absent: none
  • Minutes prepared by: AG
  • Action items in BOLD

1. Followup on the audit tool to collect information on how the ICU nurses are doing on TISS coding -- Tina rolled out TISS form audit, after which we’ll collect data and work from there.

2. Followup on trying to get hospital-level data elements from EPR

  • Tina is still working to process and make these data available.
  • Regarding trying to get direct access to COGNOS (instead of daily reports generated by others) -- Tina reported that she filled out the request and WRHA approval is on hold --- Allan will talk to Dale Gustafson to try and expedite this. Dale's reply was:
    • inquires led me to believe that this access request was essentially a work-around from regular processes. Thus, to prevent us forming our own Decision Support group, outside of the health system’s Decision Support Team, I did not support the request. 1) If the daily extract from the system is not as helpful as Tina would like – I would kindly suggest that we continue working with that team so the daily extract works for us. 2)In my opinion, it is not appropriate to allow system access to that team, to run other inquires against the health system databases. This is what Decision Support is for. Our team is there to run queries against the CCMDB only.

3. Followup on the ICD10 codes relating to waiting for transfer elsewhere. Names have been changed and Allan has modified all the relevant Wiki pages.

4. Continued discussion about changing the time that an admission begins to when care was actually begun over by the accepting team, and all the consequences of such a change. (Change to start collection at accept rather than arrive time)

  • We generally agreed to make this change and implement the infrastructural modifications per item#4 of the minutes from May 18, 2020. Allan contacted Medicine leadership and they have no problems with making these changes.
  • We discussed and resolved some issues related to this new paradigm:
    • When a person in ED has a change of service while still in ED, this WILL mandate starting a new record.
    • For a person admitted but with “Service Unspecified” (which is an ADT/COGNOS designation) we will assign that person to what we will call the “X Service”
    • For Medicine patients, Locations options will be: Each specific dedicated medicine ward, high obs, COVID ward, ED, other boarding location.
  • Allan to identify the data elements (including specific TISS elements) we need to keep track of even before a patient gets to their ‘home’ location (if they ever do, i.e. EMIP, ECIP). He proposes:
    • Intracranial pressure monitoring
    • One or more vasoactive by continuous iv infusion
    • Pulmonary Art. (Swan Ganz)
    • Peripheral arterial
    • Central Venous Catheter
    • Invasive mechanical ventilation via trach or ETT
    • IABP
    • CPAP or BIPAP (non-invasive)
    • Any form of renal replacement therapy (HD, CRRT, PD)
  • Upon start of a new record, we will handle identification of types of diagnoses (e.g. admit, acquired) just as we did previously.
  • We recognized that Medicine at GH has 3 attendings at a time: 1 on a pure teaching service, 1 on a pure non-teaching service, 1 on a combined teaching + nonteaching service. We will keep track of these as separate services.

5. Followup about working to reduce collector workload. Things that need to be done:

  • Find a new hosting/software infrastructure that can be in full compliance with privacy requirements. Allan and Tina will followup with CHI and Digital Health to pursue new hosting possibilities.
  • Thereafter, work to obtain CBS TraceLine for all transfusion data. Margaret Ring again (margaret.ring@blood.ca).
  • Thereafter, work to obtain RIS data for radiology tests. Shared Health CIO Charles Conway [204-926-1400; cconway3@sharedhealthmb.ca].

6. New issues:

  • It was observed that for ESRD, the diagnostic criteria in ICD10 and APACHE differ. Since these have different purposes, this is OK and we will allow them to be different (which we expect to happen rarely, i.e. only when a person with GFR<15 mL/min is not yet on dialysis). To facilitate this, Tina removed Query check ICD10 ESRD vs AP ARF to fix this issue.
  • Question arose about a diagnosis found after admission but which virtually certainly was present pre-admission. Allan to check on current rule applying to this.
  • An item that was raised by Michelle/Joanna after the meeting: How to code nonspecific vasculitis. Allan's answer it M31.8.
  • Another item raised by Michelle/Joanna after the meeting: How to code Serotonin syndrome. In fact we do not have a current code that covers it well. While the main cause is related to use of selective serotonin uptake inhibitors, many other drugs can cause it. Our best choices are either to extend T43.9 to cover it, or use the US code: T43.22 Poisoning by, adverse effect of and underdosing of selective serotonin reuptake inhibitors, which isn't completely satisfactory either.

ICU Database Task Group Meeting – May 18, 2020

  • Present: Allan, Barret, Iris, Joanna, Julie, Michelle, Tina, Trish, Pam, Val, Mindy, Gladys, Sherry
  • Absent: none
  • Minutes prepared by: AG
  • Action items in BOLD

1. Followup on having the ICU nurses do all TISS coding

  • Again, we heard that things have improved in the past couple of months, especially at HSC, but less so at Grace and St. B. But even at HSC the completion rate by ICU nurses is insufficient to render it unnecessary to continue having collectors check everything. Allan reported that he spoke with Jodi and she indicated that she requires the full TISS data for administrative/clinical purposes.
  • We agreed that Trish will continue to work with unit nursing directors on this, but she needs some concrete data about the magnitude of noncompliance to feed back to them. Therefore, Tina has set up tmp project TISS form audit:
    • For each day for each ICU patient it will allow collectors to identify two binary measures: (a) if that day was/wasn’t done, and (b) if it was done whether there were 2 or more data items either incorrect or missing.
  • We recognized that even if we get to the point where ICU bedside nurses are 100% on TISS, the collectors will still have to deal with the final ICU day, and also (see #4, below) TISS items captured when boarding in ED or PACU, etc.

2. Followup on trying to get hospital-level data elements from EPR.

  • Tina reports that she is now obtaining daily reports from Cognos and is working to process them into a form that will be easy to parse and use by the collectors. These ADT reports do contain EMIP and most (but not all) so-called ECIP patients. Once completed this will become the master source of identifying all patients who should be in our databases, providing their source, hospital admission information, intra-hospital transfers, and (except for patients who spend a prolonged time prior to final hospital disposition in locations other than medicine wards or ICU) their hospital disposition information. We will need to pilot this before moving fully to it. Tina is continuing to work on this, but she indicates that it would be easier if she is able to obtain direct access to Cognos, rather than getting a daily report prepared by others.

3. Followup on the ICD10 codes relating to waiting for transfer elsewhere:

4. Continued discussion about changing the time that an admission begins to when care was actually begun over by the accepting team, and all the consequences of such a change. (Change to start collection at accept rather than arrive time)

  • We did not complete this discussion today and will take it up first at the next Task meeting.
  • Allan suggests this:
    • For Medicine and ICU patients we begin our record when our service takes over care, NOT when the patient reaches their designated “home” location. This means that procedure codes, and counts of things (e.g. transfusions) will start when our service takes over, including any/all time spent on that service (e.g. time in ED or PACU, or other “boarding” location).
    • For ICU patients we will decide on a “minimum dataset” of TISS items to be collected when patients are boarding. These will have to be recorded by collectors.
    • Of note, Tina reports that the DSM data we’re getting DOES include labs from the time in ED, so she will simply need to include the lab data from the time when our service takes over care.
  • Regarding the “machinery” for this -- discuss next time expanding the “Boarding Location” machinery to initial admission and all moves thereafter. In this schema, the name would be changed to something like “Physical Locations”, and the initial one would be wherever the patient was when he/she first began to be cared for by the service/team. This machinery can then easily be used by Julie to report on boarding, lengths of stay and every other aspect of location and timing of care. Because such moves are much more frequent and confusing for Medicine than ICU, as suggested by Michelle, for Medicine patients we would have only 3 possible physical locations: ED, their service location, or a generic boarding location which is not further subdivided.
  • We began to discuss that with the above changes, and the increased boarding that will likely become the norm, it would be simpler to keep track of database records not as we do now (i.e. by home location) but rather by home service. The machinery discussed above will allow Julie to write SAS code to slice and dice the information in any way desired -- e.g. time in each physical location (including high obs). After we discuss this more next time, Allan will talk to Drs. Renner/Hajadiacos if they see any major problems with such a change in process.

5. Followup about working to reduce collector workload. Things that need to be done:

  • Find a new hosting/software infrastructure that can be in full compliance with privacy requirements. Allan and Tina will followup with CHI and Digital Health to pursue new hosting possibilities.
  • Thereafter, work to obtain CBS TraceLine for all transfusion data. Margaret Ring again (margaret.ring@blood.ca).
  • Thereafter, work to obtain RIS data for radiology tests. Shared Health CIO Charles Conway [204-926-1400; cconway3@sharedhealthmb.ca].

ICU Database Task Group Meeting - May 5, 2020

  • Present: Allan, Barret, Iris, Joanna, Julie, Michelle, Tina, Trish
  • Absent: none
  • Minutes prepared by: AG
  • Action items in BOLD

1. Followup about working to reduce collector workload

  • Allan reported that the privacy office is insisting that we bring our software/hardware infrastructure up to full compliance before they will grant us a new PIA and allow us to obtain additional data elements for the databases. Allan and Tina will followup with CHI and Digital Health to pursue new hosting possibilities.
  • Regarding obtaining CBS TraceLine for all transfusion data. Allan will followup with Margaret Ring again (margaret.ring@blood.ca).
  • Regarding obtaining RIS data for radiology tests. Allan will followup again with Shared Health CIO Charles Conway (204-926-1400; cconway3@sharedhealthmb.ca).

2. Followup on having the ICU nurses do all TISS coding

  • The data collectors agree that things have improved in the past couple of months after ICU nursing leadership made this a higher priority. But they also agreed that compliance is not close to the 95% we’d need in order to have our collectors cease checking them. It was agreed that Allan will speak to Jodi about whether this can be accomplished, and if so, how. See also TISS form audit, 2020 TISS audit

3. Followup on trying to get hospital-level data elements from EPR. Tina continues to work on this, specifically in the form of reporting from the Cognos interface.

4. Followup on the ICD10 codes relating to waiting for transfer elsewhere (Category:Awaiting/delayed transfer):

  • These currently include: home, LTC/PCH, other acute care facility, transitional care, other care facility NOS. A problem involves difficulty distinguishing the transitional from (low acuity) acute beds at Vic, Conc and Oaks.
  • Allan sent and email to Bojan and Ebi about their needs in this regard, with a suggestion to collapsing the 5 codes to 3: home, LTC/PCH, any other type of care site.
    • Bojan responded that it makes no difference to critical care.
    • Ebi responded with an alternative suggestion of 4 groupings: home, LTC/PCH, other acute site in Winnipeg, low acuity site in Winnipeg (this will include Vic, Oaks and Conc), and any facility outside of Winnipeg.
  • After discussion it was agreed that Dr. Renner’s suggestion is more doable than is the current schema, but it was observed that we would have to do some customizing of ICD-10 codes to accommodate this. Allan will contemplate how to do so.
  • After discussion it was agreed that Dr. Renner’s suggestion is more doable than is the current schema, though combining all facilities outside of Winnipeg into a single category will result in loss of detail. It was observed that we would have to do some customizing of ICD-10 codes to accommodate this.

5. Continued discussion about changing the time that an admission begins to when care was actually begun over by the accepting team, and all the consequences of such a change.

  • There are numerous consequences of such a change that have to be considered regarding collecting the following info for the entire time a patients is under our care, including time spent in ED or elsewhere (e.g. ICU patients with extended time in PACU):
    • counts of resource use elements (e.g. transfusions) done prior to ICU arrival
    • procedures done prior to ICU arrival
    • some elements of TISS -- especially use of life support modalities (but not necessarily every TISS element) prior to ICU arrival
    • we will need to contemplate exactly how Julie’s reporting will take account (or not) of the time spent prior to arrival
    • we will need to be aware of the extra workload involved with collecting data from ED records for those who either never make it to ward/ICU and those who do
    • Tina to check to see if the DSM datastream we get includes labs done prior to actual ICU/unit arrival (i.e. those done after we have accepted care but while still in ED, PACU, etc).
      • Tina confirmed that the test data for the ER portion of the pts stay is available in the DSM data.
  • Also, we will need to consider how to identify patients who are on ICU service but never make it to the ICU (so-called ECIP)
    • For Medicine EMIPs they are identified from the EMR as patients in ED whose discharge service is Medicine. We have to consider more carefully if something akin to this can be done for ICU patients -- we suspect the answer is Yes except for:
      • HSC SICU patients whose service of record in ICU remains the operating surgeon
      • ICU patients who get downgraded to go to wards before ever getting up to an ICU
      • In any case, we believe that the number we miss in this way is likely small
  • We note an major inconsistency on how we currently do things. Specifically, for EMIP patients we collect information (such as counts of resource use) during this time on Medicine in ED, but we do NOT collect those counts for medicine patients who do make it to their medicine wards. For consistency, we should collect such counts and other data items from when “we” took over care.
  • We need to further consider the “machinery” for making this change -- possibly expanding the “Boarding Location” machinery to initial admission and all moves thereafter. In this schema, the name would be changed to something like “Physical Locations”, and the initial one would be wherever the patient was when he/she first began to be cared for by the service/team. This machinery can then easily be used by Julie to report on boarding, lengths of stay and every other aspect of location and timing of care. It also allows us to eliminate the separate data elements of Accept DtTm and Arrive DtTm
    • Such moves are much more frequent and confusing for Medicine than ICU. This is simplified if, as suggested by Michelle, for Medicine patients we only have 3 possible physical locations: ED, their service location, or a generic boarding location which is not further subdivided.
  • Another issue is whether or not to capture patients cared for in ED by subspecialists (e.g. Nephrology, wearing the Nephro hat, not the GIM hat), who then IF they actually get a bed are put in a GIM ward bed and thus included in the Medicine database. We currently DO do so, and it can be tricky to figure out whether that nephrologist caring for the patient while in ED is wearing the Nephro or GIM hat.

6. After discussion about coding CCI Isolation, infectious, we agreed to CHANGE the current rule of only coding isolation if the patient turned out to have the disorder for which they had isolation. Thus, for example, the large number of people being isolated for possible COVID-19 will have the isolation procedure coded. Allan will change this on the wiki -- DONE.

ICU Database Task Group Meeting - March 11, 2020

  • Present: Allan, Barret, Julie, Michelle, Tina, Trish
  • Absent: Joanna
  • Minutes prepared by: AG
  • Action items in BOLD

1. Followup about working to reduce collector workload

  • Regarding obtaining CBS TraceLine for all transfusion data. Allan will followup with Margaret Ring (margaret.ring@blood.ca) in the middle of March.
  • Regarding obtaining RIS data for radiology tests. Allan emailed Shared Health CIO Charles Conway [204-926-1400; cconway3@sharedhealthmb.ca] on 3/4/2020 but has not heard back. He will try him again later this week.

2. Followup on having the ICU nurses do all TISS coding -- Allan reported that there was a phone meeting with Jodi and ICU nurse managers at Grace and St. B in which there was agreement to try and enforce this. After discussion we decided that as a first step in auditing whether or not the ICU nurses are indeed completing these forms, Tina will add some elements (TISS form audit) by which the collectors can identify, for each patient, whether the nurses are or are not completing them. For now, collectors will continue to check and complete whatever needs to be completed. We’ll do this for a few weeks, and then discuss further.

3. Followup on trying to get hospital-level data elements from EPR (Cognos Report Integrator)). Tina will continue following up but also Allan will followup on the PIA.

4. Followup on the ICD10 codes relating to waiting for transfer elsewhere (Category:Awaiting/delayed transfer): These currently include: home, LTC/PCH, other acute care facility, transitional care, other care facility NOS. A problem involves difficulty distinguishing the transitional from (low acuity) acute beds at Vic, Conc and Oaks.

  • Allan sent and email to Bojan and Ebi about their needs in this regard, with a suggestion to collapsing the 5 codes to 3: home, LTC/PCH, any other type of care site.
    • Bojan responded that it makes no difference to critical care.
    • Ebi responded with an alternative suggestion of 4 groupings: home, LTC/PCH, other acute site in Winnipeg, low acuity site in Winnipeg (this will include Vic, Oaks and Conc), and any facility outside of Winnipeg. We’ll discuss this at the next Task meeting.

5. Continued discussion about changing the time that an admission begins to when care was actually begun over by the accepting team, and all the consequences of such a change. We still did not complete this discussion and will further address it at the next Task meeting.

  • Most of the discussion centered around the idea of expanding the “Boarding Location” machinery to initial admission and all moves thereafter. In this schema, the name would be changed to something like “Physical Locations”, and the initial one would be wherever the patient was when he/she first began to be cared for by the service/team. This works if the initial location is the actual service location or not. Any moves thereafter are likewise captured. This machinery can then easily be used by Julie to report on boarding, lengths of stay and every other aspect of location and timing of care. It also allows us to eliminate the separate data elements of “Accept DtTm” and “Arrive DtTm”.
    • Such moves are much more frequent and confusing for Medicine than ICU. This is simplified if, as suggested by Michelle, for Medicine patients we only have 3 possible physical locations: ED, their service location, or a generic boarding location which is not further subdivided.
  • Another issue is whether or not to capture patients cared for in ED by subspecialists (e.g. Nephrology, wearing the Nephro hat, not the GIM hat), who then IF they actually get a bed are put in a GIM ward bed and thus included in the Medicine database. We currently DO do so, and it can be tricky to figure out whether that nephrologist caring for the patient while in ED is wearing the Nephro or GIM hat. We ran out of time to finish this discussion.

Also EMIP/ECIP

ICU Database Task Group Meeting - March 4, 2020

  • Present: Allan, Barret, Iris, Tina, Trish
  • Absent: Julie
  • Minutes prepared by: AG
  • Action items in BOLD

1. Followup about working to reduce collector workload

  • Regarding obtaining CBS TraceLine for all transfusion data. Allan will followup with Margaret Ring (margaret.ring@blood.ca) in the middle of March.
  • Regarding obtaining RIS data for radiology tests. Allan will contact Shared Health CIO Charles Conway [204-926-1400; cconway3@sharedhealthmb.ca] about this.

2. Followup on having the ICU nurses do all TISS coding -- there is a phone meeting scheduled with Jodi, Allan and ICU nurse managers at Grace and St. B next week.

3. Followup on trying to get hospital-level data elements from EPR. Two Cognos licenses were found for us to have. Tina will continue following up but also Allan will followup on the PIA.

4. Continued discussion about changing the time that an admission begins to when care was actually begun over by the accepting team.

  • In principle the ADT system dump (when we start getting it) will help with this but it is, of course, limited by notifications from clinical areas to the hospital admitting offices.
  • We discussed the various difficulties of clearly identifying when this occurs, including: patients who are in an OR or procedure suite prior to arriving in the ICU; in SICU the surgeon remains the official attending of record and therefore the ADT system will not accurately reflect ICU accept timing; at Grace the ED has resisted letting admitting know about such acceptances (Allan will contact Heather Smith to try and get this fixed).
  • Regarding the consequences of this change of when a database record begins on designations of diagnoses as admit/acquired, specifically infections for which we track “where” they occurred, e.g. VAP, CVC-BSI -- we need to clarify exactly the needs of the Program and OIT about this in order to make final decisions about how to do this. We will include Julie in this discussion (regarding how she reports things) at the next Task meeting, and Allan will contact Bojan and Kendiss about it.
  • Tina suggested that instead of creating some terminology such as “ECIP”, that instead this become just a version of Boarding Location and boarding timing infrastructure. This was agreed.
  • We will continue this discussion next time, when Julie is back. And we will need to put this information on the Wiki and widely circulate all the decisions relating to it.

5. New questions:

  • It was agreed that a person CAN have a lesser stage of Chronic Renal Failure (e.g. Stage 3) as a comorbid diagnosis and a higher stage (e.g. Stage 5) as an admit diagnosis. Iris pointed out that this does not currently appear to be allowed by the laptop software. Tina has updated Query check_ICD10_only_1_stage_of_renal_failure to fix this.
  • Discussion about the ICD10 codes relating to waiting for transfer elsewhere:
    • These currently include: home, LTC/PCH, other acute care facility, transitional care, other care facility NOS
    • One problem is that the new transitional facilities (Vic, Conc, Oaks) ALSO have some acute care beds (e.g. Oaks has an ortho ward, and Vic has FP wards) and it’s not always clear, for example, when someone is waiting for a bed at the Vic whether they’re waiting for a transitional or acute care bed.
    • We agreed it would be simpler if those who want/need these data would be contented with only having codes for: home, LTC/PCH, any other type of care site. Allan will contact Bojan and Ebi about their needs in this regard.

ICU Database Task Group Meeting - February 18, 2020

  • Present: Allan, Iris, Joanna, Julie, Trish
  • Absent: Barret, Tina
  • Minutes prepared by: AG
  • Action items in BOLD

1. Followup about working to reduce collector workload

  • Regarding obtaining CBS TraceLine for all transfusion data. Allan will followup with Margaret Ring (margaret.ring@blood.ca) in the middle of February -- DONE, emailed Feb 19, no reply yet.
  • Regarding obtaining RIS data for radiology tests. Allan will followup with Shared Health CIO Charles Conway [204-926-1400; cconway3@sharedhealthmb.ca] at the end of February.

2. Followup on having the ICU nurses do all TISS coding 'PLAN:'Jodi is arranging a conference call including herself, Allan and the ICU nurse managers in all 3 hospitals.

3. Followup on trying to get hospital-level data elements from EPR. Tina discovered that there is an application called Cognos which has capability to extract data elements from existing databases.

  • Tina has opened incident 3845870 and is working with Alex Omsen and Mike Ocko on getting this set up; see Cognos#Phone call with Mike Ocko

4. Followup.

5. After extensive searching we are unable to identify the rationale for the existing “rule” that we not combine sepsis codes with other infection codes (e.g. pneumonia).

  • PLAN: we propose the following: -- Allan has altered Wiki articles to reflect all these items:
  • Rescind this rule, allowing combining any of the 3 sepsis codes with a code for a specific infection -- Allan has altered the text in Template:ICD10 Guideline Sepsis, and
  • Trish will notifying all the collectors.
  • Guidelines for such combination to include:
    • Combine if it is reasonably clear that a specific infection is the source of the sepsis episode. But if it is NOT clear then do not combine.
      • Clear example, so DO combine: sepsis and the only evident infection is pneumonia
      • Unclear example, so do NOT combine: Sepsis with both pneumonia and a UTI.
      • Clear example, so DO combine. Sepsis with pneumonia developing around the same time, and then 5 days later a UTI occurs. Here it’s appropriate to combine the sepsis + pneumonia but not with the UTI.
      • Clear example, so DO combine: Sepsis with pneumonia and Bacteremia, with the same bug isolated from the lungs and blood. Here it’s reasonable to conclude that all 3 are causally related and combine all 3, with the same bug as cause in all 3.
  • Regarding identifying the bug responsible for sepsis -- this also can be unclear, here is some guidance:
    • In the presence of Bacteremia , with or without other infection(s) (e.g. pneumonia) ALL showing the same bug, consider that bug to be the agent for the sepsis
    • Without Bacteremia, with one or more other infections occurring around the same time that all have the same bug, consider that bug to be the agent for the sepsis
    • With multiple infections occurring around the same time as the sepsis, having DIFFERENT bugs, the bug responsible for the sepsis is not clear (even if one of those infections is Bacteremia it’s still not clear), so in this case choose Infectious organism, unknown. Allan has added to that Wiki page to clarify this.

6. Followup on a set of related issues that revolve around when an an actual admission to a service begins:

  • This is about the need to creating a category for ICU patients akin to “EMIP”. For these we will create new “locations” for ICU patients who never get up to their designated unit. We note that this will include not only those who are held in ED for the entire time, but also those who are held in other locations (such as PACU) for the entire time on ICU service.
  • Regarding terminology we have to decide whether to use “ECIP” (which appears to identify them solely as remaining in ED) or to include other locations as well use something like NCIP as in “NonICU Critical Care Inpatient”.
  • In any case, because of this new phenomenon, we agreed we must from now own consider an ICU record to begin when the patients is ACCEPTED, not when they ARRIVE. This must be put on the Wiki and communicated directly to collectors'. likely in Definition of an ICU admission / Identifying ICU admissions / Identifying patients in boarding locations
  • First though, we must clarify the meaning of “Accept DtTm”. There is currently ambiguity around it because we now refer to it as when an agreement to accept is made, even if that never happens (e.g. patient accepted in principle from Churchill but they die before getting here). Thus, we will now define “accept” as the combination of ALL of the following:
    • (i) the team has said they will accept taking over the primary care responsibility, and
    • (ii) the patient has physically arrived at the hospital in question, and
    • (iii) the team has actually begun providing care in the role of the primary team.
      • An important example is person coming to SICU from Churchill, but they go directly from the plane to OR and are there for 4 hours, only thereafter arriving in ICU where the ICU team is actually caring for the patient. Those 4 hours in OR are not counted and the true accept date/time is when the ICU team is actually providing care as the primary team to the patient.
      • An important example is a patient on a ward who codes and the ICU team comes and runs the code. This by itself is technically an ICU consult and while the ICU team is providing care, they have NOT generally taken over primary team responsibility. Indeed this will never happen if the patient does not survive the code. Only if/when the ICU team agrees to become the primary team and begins providing such care after the code (may be immediately after) does “accept” occur.
    • Tina after reviewing the diagram/table of variables that Julie will do up, she will need to alter the current screen on the laptops which explicitly indicates Accept DtTm as coming solely from ED. Now this can occur coming from ED, PACU, OR, ward -- pursuant to the 3 criteria listed above that must be met.
  • SMW


  • Cargo


  • Categories
    • We tentatively agreed that for the ICU patient who never does arrive at their destination, that the Arrive DtTm will be the Dispo DtTm -- this way elapsed time boarding (interval from accept to arrival) will be the entire stay.
    • Plan: For next Task meeting Julie will bring a diagram of all variables related to Accept DtTm, Arrive DtTm, Dispo DtTm etc -- so we can make sure there are no internal contradictions.
  • Will this mean entering and possibly later changing Service/Location and therefore D_ID of patient records?
  • SMW


  • Cargo


  • Categories
  • Apache II General Collection Guidelines: the “first 24 hrs in ICU” now must begin at the Accept DtTm --> thus collectors will increasingly have to look at labs and vital signs done before ICU arrival. If the entire first 24 hours on ICU service is prior to ICU arrival, then APACHE scores will entirely be based on that pre-ICU time.
    • THIS in turn, has implications for if DSM is even sending us that “prior” data to peruse to generate the correct APACHE scores.
    • PLAN: Julie to check on that.
  • TISS28 Collection Guide -- the collectors will need to complete TISS forms for calendar days prior to ICU arrival. If in a single calendar day there is a partial TISS sheet filled out by collectors from prior to arriving in ICU, PLUS a partial TISS sheet completed by the ICU nurses, then for that calendar day the collector must combine them.
  • Procedures -- procedures that occur before ICU arrival but after ICU service acceptance will generally be considered as Acquired Procedure
  • Diagnoses --- conditions that begin before ICU arrival but after ICU service acceptance will be considered as Acquired Diagnosis / Complication
  • Identification -- this is especially tricky for those who never make it to their ICU. We decided not to spend a lot of effort now on identifying such individuals at HSC or STB, pending obtaining ADT system data dumps.
  • ALL of the above things must be delineated in various places in the Wiki.

7. Followup regarding those admitted initially to the new Medicine “X service” (HSC Unknown Service).

  • PLAN: Tina has to fix some things to go forward with this (see 1/29/2020 minutes).

ICU Database Task Group Meeting - February 12, 2020

  • Present: Allan, Barret, Iris, Julie, Michelle, Tina, Trish
  • Absent: Joanna
  • Minutes prepared by: AG
  • Action items in BOLD

1. Followup about working to reduce collector workload

  • Regarding obtaining CBS TraceLine for all transfusion data. Allan will followup with Margaret Ring (margaret.ring@blood.ca) in the middle of February.
  • Regarding obtaining RIS data for radiology tests. Allan will followup with Shared Health CIO Charles Conway [204-926-1400; cconway3@sharedhealthmb.ca] at the end of February.

2. Followup on having the ICU nurses do all TISS coding -- we have not implemented this yet because:

  • While we have top administrator agreement at HSC. Allan emailed Dan to arrange similar meetings at St. B and Grace, but he responded that he thinks Jodi is in the better position to do so. Allan will bring it up with Jodi.

3. Followup on trying to get hospital-level data elements from EPR. Tina discovered that there is an application called Cognos which has capability to extract data elements from existing databases. Tina has opened incident 3845870 and is working with Alex Omsen and Mike Ocko on getting this set up; see Cognos#Phone call with Mike Ocko.

4. Followup on using DSM data to generate the APACHE lab scores. All reported that he spoke to Bojan and Kendiss about this, and that they would not want to further delay reporting. So, given that Joanna told us that it is not very time-consuming to do it by hand, we will not currently pursue this. BUT we still want/need to know the timing around admission (and more to the point, around Accept Time) of the labs we get from DSM. Julie/Tina will report on this.

5. Request for cognitive impairment as a comorbid. We agreed to add R41.88, whose official ICD10 name is “Other and unspecified symptoms and signs involving cognitive functions and awareness”, but for ease of use we will rename it “Cognitive impairment, NOS”. Tina has added the code to wiki and CCMDB.

6. Regarding sepsis coding, particularly the existing rule NOT to link the sepsis code with a specific infection. Allan sent an email to DCCC and DCMed saying this. That led to some additional collector questions. It also begs the question of why NOT to combine sepsis with a specific infection when are sufficiently certain they’re related. Allan tried to answer this by looking through past minutes, but going back as far as all of 2018 and 2019, he wasn’t able to locate the answer other than what he wrote in Sepsis (SIRS due to infection, without acute organ failure) , i.e. that is is because of “a technical reason having to do with how we combine diagnoses”. Allan will talk with Julie to try and noodle this out, but if we cannot, then we may change this rule. Will discuss at next Task Meeting.

7. Followup on a set of related issues that revolve around when an admission begins.

  • The main new issue is that we have begun to experience (and project this will increase) the phenomenon where ICU patients spend long periods of time admitted to the ICU service BEFORE there is an ICU bed available. Indeed, some at Grace have already been spending their entire time on the ICU service in ED (i.e. ECIP). But more generally this can be in ED, PACU, elsewhere.
  • This latter is akin to “EMIP” patients, but we’ve never really had to deal with them before for ICU. For these we will create “locations” for ICU patients who never get up to their designated unit. We NOTE that this will include not only those who are held in ED for the entire time, but also those who are held in other locations (such as PACU) for the entire time on ICU service.
    • We have to decide on terminology -- whether to use “ECIP” (which appears to identify them solely as remaining in ED) or to include other locations as well use something like NCIP as in “NonICU Critical Care Inpatient”.
  • In any case, because of this new phenomenon, we agreed we must from now own consider an ICU record to begin when the patients is ACCEPTED, not when they ARRIVE. This must be put on the Wiki and communicated directly to collectors - See Definition of an ICU admission.

Inconsistent with Acquired Diagnosis / Complication

  • SMW


  • Cargo


  • Categories

May be inconsistent with Admit Procedure, what do we want?

  • SMW


  • Cargo


  • Categories
    • Identifying ICU admissions -- this is especially tricky for those who never make it to their ICU. Lisa at Grace has come up with a way to find them for now, but a separate method would likely be needed for the other sites. Luckily, this phenomenon is currently rare except at Grace. Tina points out that if/when we get a data dump from the ADT system (which she is currently working on, and it appears might actually happen), then she will be able to process this data to identify such individuals and then (after the fact of course) it will be possible for collectors to go back and obtain the data for them. Thus, we decided not to spend a lot of effort now on identifying such individuals at HSC or St. B, pending that plan.

8. After discussion, we agreed that Julie will do away with the designation of Parked in ER/Direct admit for direct admissions which spend a bit of time in ED. Tina identified that there are many connected issues, further work/discussion required.

9. Followup regarding those admitted initially to the new Medicine “X service” (HSC Unknown Service). Tina has to fix some things to go forward with this (see 1/29/2020 minutes).

ICU Database Task Group Meeting - January 29, 2020

  • Present: Allan, Barret , Joanna, Julie, Michelle, Tina
  • Absent: Trish
  • Minutes prepared by: AG
  • Action items in BOLD

1. Followup about working to reduce collector workload

  • Regarding obtaining CBS TraceLine for all transfusion data. Allan will followup with Margaret Ring (margaret.ring@blood.ca) in the middle of February.
  • Regarding obtaining RIS data for radiology tests. Allan will followup with Shared Health CIO Charles Conway [204-926-1400; cconway3@sharedhealthmb.ca] at the end of February.

2. Followup on having the ICU nurses do all TISS coding -- we have not implemented this yet because:

  • While we have top administrator agreement at HSC, Allan has emailed Dan to arrange similar meetings at St. B and Grace

3. Followup on trying to get hospital-level data elements from EPR. Tina discovered that there is an application called Cognos which has capability to extract data elements from existing databases. Tina has opened incident 3845870 and is working with Alex Omsen and Mike Ocko on getting this set up; see Cognos#Phone call with Mike Ocko

4. Followup on pulling in certain lab results in addition to current counts via re-parsing all DSM data from 1/1/2019.

  • We will soon have Julie work to do the programming needed to automatically do the lab-related APACHE II APS score.

5. New items:

  • Request for cognitive impairment as a comorbid -- there is one to discuss next time: R41.88 = Other and unspecified symptoms and signs involving cognitive functions and awareness
  • Question of whether to code Diabetes newly found during hospitalization as a comorbid --- we agreed Yes, as the general EXISTING rule for comorbids is to code it as such even when it was just newly discovered but MUST have been present previously, even if that wasn’t known.
  • Regarding sepsis coding, particularly the existing rule NOT to link the sepsis code with a specific infection. Allan sent an email to DCCC and DCMed saying this. That led to some additional collector questions. It also begs the question of why NOT to combine sepsis with a specific infection when are sufficiently certain they’re related. To answer this, we need to search through the prior minutes to remember exactly why we made this rule.
  • Regarding those admitted initially to the new Medicine “X service”. This is an extra medicine attending who takes patients before a regular service can deal with them. 90% do eventually get to a “permanent” service, and when that occurs they are officially considered to have always been on that permanent service. The question is how to handle the 10% who remain on X for their entire stay. There are a variety of practical issues related to this -- such as assignment of their DID (as it includes their location), and which collector covers which Medicine service. We decided not to change anything now, but Tina has to fix some things to go forward with this.
  • Regarding defining start of an ICU admission (Definition of an ICU admission, Identifying ICU admissions). After discussion we agreed to use the Accept DtTm for this. Tina had a look at this on wiki and identified many contingencies; further discussion/work required.
    • As a consequence, we also will need (given that ICU admissions are now waiting substantial times to actually get to ICU) to consider that the “first 24 hrs in ICU” are from accept date/time, not arrival. THIS in turn, has implications for if DSM is even sending us that “prior” data to peruse to generate the correct APACHE scores. Julie to check on that, and clearly we need to discuss this more and clearly document on Wiki and notify all collectors.
  • There is a new phenomenon of extended “boarding” of ICU patients. This can be in ED, PACU, elsewhere. We now have to deal with this and so we probably need to make an ECIP designation, similar to EMIP. The problem is that they’re hard to capture if they never get to ICU we’re dependent on accidentally hearing about them. So, we need to devise a process at each of the 3 hospitals to do this. Trish will ask Lisa at Grace to work on it. Allan will talk to Bojan about having the ICU fellow or attending be responsible for adding these people to the paper ICU admission logs at each site.

ICU Database Task Group Meeting - January 23, 2020

  • Present: Allan, Barret , Joanna, Julie, Michelle, Tina
  • Absent: Trish
  • Minutes prepared by: AG
  • Action items in BOLD

1. Followup about working to reduce collector workload

  • Regarding obtaining CBS TraceLine for all transfusion data. Allan reported that he followed up with Margaret Ring (margaret.ring@blood.ca) on 1/20/2020 and her reply was to expect a response in a few weeks. Allan will follow up with her in the middle of February.
  • Regarding obtaining RIS data for radiology tests. Allan reported that he followed up with Shared Health CIO Charles Conway [204-926-1400; cconway3@sharedhealthmb.ca] on 1/20/2020, who replied that he’ll get back to Allan. Allan will follow up with him at the end of February.

2. Followup on having the ICU nurses do all TISS coding -- we have not implemented this yet because:

  • While we have top administrator agreement at HSC, we are waiting for Dan to arrange similar meetings at St. B and Grace

3. Followup on trying to get hospital-level data elements from EPR. Tina discovered that there is an application called Cognos which has capability to extract data elements from existing databases. Tina has opened incident 3845870 and is working with Alex Omsen and Mike Ocko on getting this set up; see Cognos#Phone call with Mike Ocko

4. Follow up on how to code diagnostic sampling of pericardial fluid or the pericardium. See Jan 2, 2020 minutes for decisions. Tina has added (D) Pericardium with code 2.HA .

5. Followup on pulling in certain lab results in addition to current counts via re-parsing all DSM data from 1/1/2019.

  • Barret proposed a schema, but after discussion we recognized that Julie is already pulling ALL the actual results into a SAS format, and that there is no impediment from doing this going forward. It means that for any specific question/project, Julie will have to write SAS code to pull and format the data elements of interest.
  • We will then start to discuss having Julie work to do the programming needed to automatically do the lab-related APACHE II APS score.

6. Following up on fact that some collectors are applying priority 0 to unconfirmed diagnoses with priority 0. The concern about this is that Julie uses the lowest priority diagnosis for reporting of cause of readmission, and this is done with incomplete data (i.e. before the unconfirmed diagnoses are all figured out). But after discussion we realized that this will not be a problem if Julie adds to this analysis the requirement that there Dx Primary also be checked.

7. Followup on coding fistulas

  • We already have codes for these: J95.03 for T-E, K31.6 for stomach or duodenum, K60.5 for anorectal, K63.2 for intestinal, N32.2 for bladder, N82 for female genital tract.
  • We agreed NOT to add any others, which are rarer, and in those cases use a NOS code -- e.g. Gallbladder or bile duct disorder, NOS
  • There are other issues related to fistulae -- e.g. that there are always 2 parts involved. So we decided that the rule will be to combine the two portions together, except when one end of the fistula is to outside (i.e. the skin). Tina has created Template:ICD10 Guideline Fistulas, and Allan will populate it with this information -- DONE.

8. Followup on the uncertainty about what to do when in CCI, both diagnostic and therapeutic procedures are done on the same body part. We discussed this previously in the minutes, and under CCI_Collection#Related_Imaging.2C_Diagnostic_and_Therapeutic_procedures_performed_at_the_same_time it states when both are done to code them both.

9. Followup on how to code miscellaneous neuromuscular disorders? The solution(s) are:

10. Followup on issues of ESRD and Acute renal failure.

  • Can ESRD be both and Admit and a Comorbid diagnosis? --> Answer is YES, if it was present prior to hospital admission AND it satisfies the criteria for an Admit diagnosis.
  • Can a patient have both acute renal failure and ESRD (Stage 5)? --> Answer is YES, with limitation:.
    • If the criteria for Stage 5 is being on dialysis, then the answer is No -- this is the limitation
    • But if the criteria is creatinine clearance <15 ml/min in someone NOT yet on dialysis, then YES.
    • The implications of these is that Tina has removed the checks currently in existence relating to each of these 2 items (they turn out to be 1 check, both dxs could not be present at same ward/unit admission).

11. New items:

  • Coding procedures for control of bleeding -- The issue here is that in CCI one currently has two separate options for the “what was done” part of this procedure, i.e. Control of Bleeding of the body part bleeding, or Occlusion of the bleeding vessel (e.g. via clipping or embolization or sclerosis).
    • After discussion we agreed that the most informative and favored option is to combine the body part bleeding with Control of Bleeding. Thus for control of bleeding esophageal varices, the body part to go along with it is (T) Esophagus. Allan will add to these wiki articles to make this clear -- DONE.

ICU Database Task Group Meeting – January 2, 2020

See Task Team Meeting - Rolling Agenda and Minutes 2019#ICU Database Task Group Meeting – December 11, 2019

  • Present: Allan, Barret , Joanna, Julie, Michelle, Tina, Trish
  • Absent:
  • Minutes prepared by: AG
  • Action items in BOLD

1. Followup about working to reduce collector workload

  • Regarding obtaining CBS TraceLine for all transfusion data. Allan will follow up with Margaret Ring (margaret.ring@blood.ca) and Tony Loewen (anthony.loewen@blood.ca)
  • Regarding obtaining RIS data for radiology tests. Allan will follow up with RIS administrators (Angela Charbonneau 926-9874; Randy Roels 926-9871, rroels@sharedhealth.mb.ca).
  • Regarding obtaining automated ABG data at HSC and St. B -- this has been implemented, getting that data from those sites from DSM data.
  • Regarding consolidating some of the “what was done” components of CCI therapeutic interventions -- the plans from the 11/11/2019 meeting have been implemented.

2. Followup on having the ICU nurses do all TISS coding -- we have not implemented this yet because:

  • While we have top administrator agreement at HSC, we are waiting for Dan to arrange similar meetings at St. B and Grace

3. Followup on trying to get hospital-level data elements from EPR. Tina discovered that there is an application called Cognos which has capability to extract data elements from existing databases.

4. Follow up on how to code diagnostic sampling of pericardial fluid or the pericardium. Answers are:

  • for therapeutic pericardial drainage combine (T) Pericardium with Drainage, Evacuation
  • for diagnostic pericardiocentesis -- we agreed today to add the “body part” item for diagnostic tests of 2.HA, (D) Pericardium. Then this is combined with the appropriate Biopsy (endoscopic) or Biopsy (non-endoscopic) depending how it was sampled. In other words, we are not distinguishing between a “fluid biopsy” of the pericardium and a true tissue biopsy of it. Tina added this.

5. Followup on pulling in certain lab results in addition to current counts vis re-parsing all DSM data from 1/1/2019

  • Barret with help from Tina will generate a draft list of which tests to do this for, which we’ll discuss at the next Task meeting. It should include the tests needed to calculate APACHE 2. Care must be taken to balance usefulness with data storage issues.
  • As part of this, we will work so that the labs values that are part of APACHE 2 no longer need to be dealt with by the data collectors.

6. Item we didn't discuss yet -- unconfirmed diagnoses with priority 0. We'll discuss next time.

7. Followup on query about coding for the myraid of other fistulas out there

  • There are separate codes for all of these when nontraumatic.
  • We already have codes for these: J95.03 for T-E, K31.6 for stomach or duodenum, K60.5 for anorectal, K63.2 for intestinal, N32.2 for bladder, N82 for female genital tract.
  • We don’t have codes now for: K82.3 for gallbladder, M25.1 for joint, or less common ones (e.g. lacrimal duct) and have to decide whether to add either of them. We also have to decide how to handle the fact that there are always TWO things connected by the fistula. We already have a “Category:Fistula” but probably need to create a template for it too that references the category but also discusses the various issues. If Tina will do this, Allan will populate it.

8. New items:

  • There is some uncertainty about what to do when in CCI, both diagnostic and therapeutic procedures are done on the same body part. We discussed this previously in the minutes, and under CCI Procedures it states when both are done to code them both.
  • How to code miscellaneous neuromuscular disorders?
    • This is problematic because although they are often discussed as if they were a single category of disorders, in fact they are two separate categories, comprising nervous system disorders and muscular disorders. This is why there is no ICD10 code for miscellaneous neuromuscular disorders.
    • The example used, Kennedy’s disease, also called “Spinal and bulbar muscular atrophy”, and is a degenerative disease of the CNS that results in muscle atrophy.
    • We already have sufficient “NOS” codes within the neuro and muscular disorders to handle this when the disorder is known but we don’t have a specific code for it (as in the Kennedy disease example, where one should use Degenerative nervous system disorder, NOS). The other potentially useful NOS codes are Muscle disorder/myopathy (primary or secondary), NOS and Disorder of nervous system (any part), NOS, and for when it’s a movement disorder Movement disorder, NOS.
    • The bigger problem is when it’s not clear whether it’s a primary nervous system vs. muscle disorder. In this case, one can:
      • wait to see whether the medical teams decide on nervous vs. muscular disorder
      • use a code that represents the symptom(s) not the disease per se. For example, if the symptoms are movement-related, one can use existing code Involuntary movements, NOS. And lastly when none of this is good enough, we might consider adding the code R29.8, to be called “Other and unspecified symptoms and signs involving the nervous and musculoskeletal systems”. We’ll discuss this next meeting.
  • Can ESRD be both and Admit and a Comorbid diagnosis? The answer is YES, if it was present prior to hospital admission AND it satisfies the criteria for an Admit diagnosis. This decision also answers the issue of having acute on chronic renal failure.
  • What should be done when creatinine clearance is <15 ml/min but the patient is admitted for uremic symptoms? This is an issue because the definition of Chronic kidney disease (end-stage renal/kidney disease, ESRD), Stage 5, GFR LT 15 is either on dialysis or with clearance<15. But of course, there are some patients who don’t start dialysis until they have GFR<10 or sometimes even lower. The answer is to follow the definition, such a person qualifies for Stage 5 as a comorbid condition, and as directly above, that same code can be used as an Admit diagnosis, along with the specific reason for admission (e.g. hyperkalemia).

ICU Database Task Group Meeting – December 11, 2019

See Task Team Meeting - Rolling Agenda and Minutes 2019#ICU Database Task Group Meeting – December 11, 2019