DSM Lab Extract: Difference between revisions

From CCMDB Wiki
Jump to navigation Jump to search
m (flagged as a DT so it won't get lost.)
 
(28 intermediate revisions by 4 users not shown)
Line 1: Line 1:
We are working on a way to get DSM to extract Labs for import to our DB.
The DSM Lab Extract is a set of data we receive from Diagnostic Services of Manitoba. It is requested and processed by [[DSM Labs Consistency check.accdb]] and stored in the [[L_Labs_DSM table]].
It is requested via [[Instructions for requesting a batch of data from DSM]] and imported via  [[Instructions for importing a batch of DSM Data]].
See [[DSM data]] for more info on the data once imported.  


== Outstanding Issues ==
== Things we still want to implement/fix ==
=== additional labs ===
{{DT|
*we want at least NA, K, Lactate but Allan Garland is consulting with colleagues to find out if we need additional
* A possibility to change the current [[Chart]] entry to be the same with SH format  (see [[#DSM Inclusion Criteria/ Process]] for reason why in details).
** Lactacte, NA and K are included with current output.
}}
== DSM Inclusion Criteria/ Process ==
*Must Match using the [[PHIN]] and/or [[Chart]]  (provided and/or modified)
**There is an issue that the [[Chart]] format (i.e.  without leading zeroes) at SBGH  being  provided to  the [[#DSM Contacts]] is not the same with SH format.  Because of this, then [[#DSM Contacts]] needs to add a step of adding 1x 0 or 2x 00 before the chart number to all SBGH records to help improve the matched results and refer to them as modified MRN.
***The database will move to a new platform and the idea of changing the current [[Chart]] to the SH format on the new platform came up. Needs to be discussed further if to be implemented.
*Date range
** The minimum timestamp across all timestamps within the test, and across all tests within the request must >= admitted timestamp minus 2x days
** The maximum timestamp across all timestamps within the test, and across all tests within the request must <= discharge timestamp plus 2x days
*Must include all tests performed.


=== Culture and Sensitivity ===
== Known Issues ==
Get from Group column in 2016-03 import; still testing Ttenbergen 16:29, 2016 September 29 (CDT)
=== Aliases ===
 
There might be some inconsistencies in how the labs are re-linked to patients who arrive as Jon Doe or have aliases in PHIN or Chart for other reason. If we want to include all patients with Aliases, we will need to do something to reconcile with [[Alias ID collection]].
Testing working out well. Pagasa did an audit and results were good with some exceptions we are now working on. Ttenbergen 17:29, 2016 October 6 (CDT)
=== Unmatched Cases ===
 
The patients with aliases when the list was earlier sent will be known after receiving back the extracted Lab because they will be included to the unmatched cases in addition to those who have real no lab data. Below is the action to take:
==== we will miss some ====
* Patients with LOS < 1day will be accepted as no lab data.
Some store the site as a "result" and it's free-form, so we may lose these. EG:
* Patients with LOS >= 1 day - 5 random samples will be cross checked in EPR by the [[Data Processor]] if indeed have no Lab Data. if any lab data  is found, then check the PHIN or CHART if there are any changes (or a case with aliases).
* fungal culture, site ETT, with C&S on same request#
** Re-check some more samples as time permits and do the same step above.
 
** Any unmatched records confirmed having labs present in the crosschecking will be re-send together in the next batch of sending.
=== ABGs ===
The ABGs will not be complete, are only entered at STB and used to be at HSC but flakey lately. More info to come. Ttenbergen 13:17, 2014 August 14 (CDT)
There was also discussion whether we want to include venous blood gasses.
 
=== duplicate records ===
There were 1,865,570 records in the export.  
There were 1,863,373 unique records.
2197 duplicates
Why the duplicates?
Need to confirm if this is still a problem. Ttenbergen 17:09, 2015 October 15 (CDT)
 
2012-03
There are 1685160 lines in table
There are 1684872 lines if I list only unique values.  
288 diff - why?


== Requesting data ==
== Requesting data ==
A file is exported to [[#File Share]], and we ask [[#DSM Contacts]] to provide a new extract (how often? Ttenbergen 17:21, 2015 October 15 (CDT) {{discussion}}).  
A file is exported to [[#File Share]], and we ask [[#DSM Contacts]] to provide a new extract. See [[Instructions for requesting a batch of data from DSM]] for details.
 
see [[Instructions for requesting a batch of data from DSM]]


== Actions by DSM ==
== Actions by DSM ==
Line 40: Line 33:


== Processing data received from DSM ==
== Processing data received from DSM ==
See [[Instructions for importing a batch of DSM Data]]
See  
* [[Instructions for importing a batch of DSM Data]]
* [[Lab identification in the DSM data]]


== File Share ==
== File Share ==
Line 49: Line 44:
* Loretta Eng 9261406 Leng @ dsmanitoba.CA
* Loretta Eng 9261406 Leng @ dsmanitoba.CA


== Log ==
== Related articles ==  
* 2015-10-15 found bug in import, reimported, looks much better now, asked Pagasa, Gail and Trish if we could review a few more discrepancies Monday.
{{Related Articles}}
* 2015-10-08 Pagasa and Gail and Trish reviewed data and returned excel sheet; need to review differences and find out if I can filter the extras
 
* 2015-10-05 pushed Trish on date
* 2015-09-30 emailed Trish to find out when she wants to review and test
* 2015-09-28
* directory issue fixed
*can’t see directory that Alun made;
** asked him to copy content to a diff directory I made.
** Launched ticket with eHealth to fix underlying issue.
* 2015-09-24 Alun emailed that new data is ready
* 2015-09-21 replied busy but will get me a timeline.
* 2015-09-21 emailed Alun f/u
* 2015-08-13 finished first round of http://ccmdb.kuality.ca/index.php?title=DSM_Labs_Importer.mdb waiting for data for further tests
* 2015-08-06 Alun is on vacation until Aug 24; no new import from him
* '''2015-07-22''' requested a new output from Alun because missing time for "arrived" means comparisons will always be off
* 2015-06-29 working...
* 2015-06-25 re-starting; found erroneous line because it causes a short line.
* 2014-12-29 importing new export into test tool; there were problems with the data file, then other things became higher priority
* 2014-12-24 received new export from Alun
* 2014-12-15 fu email to Alun; he figures won’t get to it till after Christmas… * 2014-12-05 provided new extract, updated wiki
*2014-11-20 Loretta emailed me to find out details… emailed her some. We are set to meet Dec 1, will need to review before then.
* 2014-11-06 Allan will book meeting with Lab folk, hopefully over next week or so, snooze for 10days
* 2014/Oct/20 no news from, whatever…
* 2014/Oct/16 emailed Allan for update
* Sept 2014 received extract from Alun carter, some good some bad
* week of 2014-12-05 met with Loretta, Alun and Allan; agreed on next round of tests; Send new test data to Alun today. Ttenbergen 11:28, 2014 December 5 (CST)
* 2014/09/25 met with Allan Garland, discussed. he was going to talk to DSM. Ttenbergen 12:56, 2014 September 29 (CDT)
* email to put this on the Task group's agenda for this Thursday. Ttenbergen 10:39, 2014 September 22 (CDT)
*did an initial analysis which doesn't look good. emailed to Allan and Julie for comment Ttenbergen 17:00, 2014 September 8 (CDT)
*discussed labs list with Allan, now need to do some more validationTtenbergen 17:09, 2014 September 4 (CDT)
*2014/aug/25  started to review data, stuck on matching labs now, will need to talk with Trish when back. Also emailed Allan with some weirdness in the data. Ttenbergen 19:38, 2014 August 25 (CDT)
*2014/Aug/21 we received data last week, but have not formally reviewed it. Ttenbergen 10:39, 2014 August 21 (CDT) Sent list of labs to Allan, he offered to review... Ttenbergen 20:59, 2014 August 21 (CDT)
*2014/Aug/14 phone call with Loretta to confirm what is needed. Requested early sample data so we can start our processing... Ttenbergen 13:17, 2014 August 14 (CDT)
*2014/Aug/07 submitted test data at Loretta’s request
*2014/Jul/31 emailed Loretta again, since no news from Alun; ooomsg says back Tue Aug 5
*2014/Jul/21 We have a file share. Alun will hopefully have test data by end of week, or early next week
*2014/Jul/17 Email from Chad Kitching re. share folder description
*2014-07-14 emailed Alun and Loretta re progress
*2014-06-16 talked with Loretta on Friday. She said Alun Carter is the data guy we will work with. She suggested a different format and requested a list of labs we want, which I gave her. Also emailed Alun to get this moving. However, he is on vacation for 3 weeks, will f/u when he is back
*2014-06-12 emailed Loretta a test extract;
*2014-06-12 emailed Loretta a sample request file. Ttenbergen 17:47, 2014 June 12 (CDT)


[[Category: DSM Labs Extract]]
[[Category: DSM Labs Extract]]
[[Category: Data Sources]]

Latest revision as of 12:36, 2023 July 5

The DSM Lab Extract is a set of data we receive from Diagnostic Services of Manitoba. It is requested and processed by DSM Labs Consistency check.accdb and stored in the L_Labs_DSM table. It is requested via Instructions for requesting a batch of data from DSM and imported via Instructions for importing a batch of DSM Data. See DSM data for more info on the data once imported.

Things we still want to implement/fix

  • SMW


  • Cargo


  • Categories

DSM Inclusion Criteria/ Process

  • Must Match using the PHIN and/or Chart (provided and/or modified)
    • There is an issue that the Chart format (i.e. without leading zeroes) at SBGH being provided to the #DSM Contacts is not the same with SH format. Because of this, then #DSM Contacts needs to add a step of adding 1x 0 or 2x 00 before the chart number to all SBGH records to help improve the matched results and refer to them as modified MRN.
      • The database will move to a new platform and the idea of changing the current Chart to the SH format on the new platform came up. Needs to be discussed further if to be implemented.
  • Date range
    • The minimum timestamp across all timestamps within the test, and across all tests within the request must >= admitted timestamp minus 2x days
    • The maximum timestamp across all timestamps within the test, and across all tests within the request must <= discharge timestamp plus 2x days
  • Must include all tests performed.

Known Issues

Aliases

There might be some inconsistencies in how the labs are re-linked to patients who arrive as Jon Doe or have aliases in PHIN or Chart for other reason. If we want to include all patients with Aliases, we will need to do something to reconcile with Alias ID collection.

Unmatched Cases

The patients with aliases when the list was earlier sent will be known after receiving back the extracted Lab because they will be included to the unmatched cases in addition to those who have real no lab data. Below is the action to take:

  • Patients with LOS < 1day will be accepted as no lab data.
  • Patients with LOS >= 1 day - 5 random samples will be cross checked in EPR by the Data Processor if indeed have no Lab Data. if any lab data is found, then check the PHIN or CHART if there are any changes (or a case with aliases).
    • Re-check some more samples as time permits and do the same step above.
    • Any unmatched records confirmed having labs present in the crosschecking will be re-send together in the next batch of sending.

Requesting data

A file is exported to #File Share, and we ask #DSM Contacts to provide a new extract. See Instructions for requesting a batch of data from DSM for details.

Actions by DSM

DSM will take our file and generate an output of all labs for those patients for the admission. They will return it to us via the file share.

Processing data received from DSM

See

File Share

\\ad.wrha.mb.ca\wrha\REGION\Shared\CCMDB_Lab_Exchange

DSM Contacts

  • main contact: Alun Carter
  • Loretta Eng 9261406 Leng @ dsmanitoba.CA

Related articles

Related articles:
Retrieved from "https://ccmdb.kuality.ca/index.php?title=DSM_Lab_Extract&oldid=159798"