Re: [gpc-informatics] #178: determine whether last.visit is the last fulfilled visit or the last scheduled visit

[GPC Informatics]

#178: determine whether last.visit is the last fulfilled visit or the last
scheduled visit
-+---
 Reporter:  mish |   Owner:  dconnolly
 Type:  enhancement  |  Status:  closed
 Priority:  minor|   Milestone:  cohort-char1
Component:  data-stds|  Resolution:  invalid
 Keywords:   |  Blocked By:
 Blocking:   |
-+---
Changes (by dconnolly):

 * status:  assigned => closed
 * resolution:   => invalid


Comment:

 Since our BC finder file QA (#227) is done, this is overtaken by events.

--
Ticket URL: 

gpc-informatics 
Greater Plains Network - Informatics
___
Gpc-dev mailing list
Gpc-dev@listserv.kumc.edu
http://listserv.kumc.edu/mailman/listinfo/gpc-dev

Re: [gpc-informatics] #178: determine whether last.visit is the last fulfilled visit or the last scheduled visit

[GPC Informatics]

#178: determine whether last.visit is the last fulfilled visit or the last
scheduled visit
-+---
 Reporter:  mish |   Owner:  dconnolly
 Type:  enhancement  |  Status:  assigned
 Priority:  minor|   Milestone:  cohort-char1
Component:  data-stds|  Resolution:
 Keywords:   |  Blocked By:
 Blocking:   |
-+---

Comment (by campbell):

 UNMC has saved a copy of our Clarity extract for encounter data mapping to
 the Epic Userweb for access and use by other sites.

 https://datahandbook.epic.com//Reports/Details/9000290

--
Ticket URL: 
http://informatics.gpcnetwork.org/trac/Project/ticket/178#comment:15
gpc-informatics http://informatics.gpcnetwork.org/
Greater Plains Network - Informatics
___
Gpc-dev mailing list
Gpc-dev@listserv.kumc.edu
http://listserv.kumc.edu/mailman/listinfo/gpc-dev

Re: [gpc-informatics] #178: determine whether last.visit is the last fulfilled visit or the last scheduled visit

[GPC Informatics]

#178: determine whether last.visit is the last fulfilled visit or the last
scheduled visit
-+---
 Reporter:  mish |   Owner:  dconnolly
 Type:  enhancement  |  Status:  assigned
 Priority:  minor|   Milestone:  cohort-char1
Component:  data-stds|  Resolution:
 Keywords:   |  Blocked By:
 Blocking:   |
-+---
Changes (by dconnolly):

 * owner:  bokov = dconnolly
 * priority:  major = minor


--
Ticket URL: 
http://informatics.gpcnetwork.org/trac/Project/ticket/178#comment:14
gpc-informatics http://informatics.gpcnetwork.org/
Greater Plains Network - Informatics
___
Gpc-dev mailing list
Gpc-dev@listserv.kumc.edu
http://listserv.kumc.edu/mailman/listinfo/gpc-dev

RE: [gpc-informatics] #178: determine whether last.visit is the last fulfilled visit or the last scheduled visit

[Campbell, James R]

Wendy
KU has not deployed their encounter type variable yet to my knowledge and not 
all deployed data has been moved to Babel for many of our sites.  KU will and 
must deploy encounter type to be compliant with PCORI and participate in GPC.  
So I am simply saying that all of our sites will have the data deployed and it 
should be part of any data management planning.  Therefore the last completed 
encounter as a marker for our last contact with the living patient  will be 
satisfied within our GPC/PCORI common data model as:
the last encounter of type IP, IS, AV, ED or OA for the patient.

At UNMC at least, NO patient visit occurs that is not stored into i2b2 as an 
Encounter.
We may want to visit this as a quality variable across GPC in terms of 
frequency of non-documented visit activity.
Jim

From: GPC Informatics [d...@madmode.com]
Sent: Wednesday, November 12, 2014 12:15 PM
To: m...@wisc.edu; j...@kumc.edu; dconno...@kumc.edu; Hickman, Hubert B; 
Campbell, James R
Cc: tshire...@kumc.edu; tmcma...@kumc.edu; gpc-dev@listserv.kumc.edu
Subject: Re: [gpc-informatics] #178: determine whether last.visit is the last 
fulfilled visit or the last scheduled visit

#178: determine whether last.visit is the last fulfilled visit or the last
scheduled visit
---+
 Reporter:  mish   |   Owner:  jhe
 Type:  problem|  Status:  assigned
 Priority:  major  |   Milestone:  data-quality2
Component:  data-stds  |  Resolution:
 Keywords: |  Blocked By:  155
 Blocking: |
---+

Comment (by jhe):

 Replying to [comment:7 campbell]:

 Jim,
  Thanks for the clarification. There is a variable called last.visit
 among the demographic variables in Heron. What we wanted to find out is
 whether last.visit is a fulfilled visit or not. If yes, then we know the
 patient was last known alive on that day. if no, then we can't use
 last.visit as last known alive date.

 It is great to know that encounter type can differentiate encounters
 completered and others. But this is not enough for answering the original
 question. When we run a query, we pull encounters related to a specific
 disease only, so the last encounter of the patient with the hospital
 (likely related other diseases) may not be in the pulled data.

 I don't know whether the date of last.visit corresponds to  the last
 encounter (among all encounters with the hospitals). If it is, then we
 need to find that encounter and its type (completed or not). Dan and
 Tamara: do you know whether we can do this?

 If last.visit is not linked to an encounter in the system, then we are
 back to the begining ot the original question: is last.visit fulfilled or
 scheduled only.

 Thanks,

 Wendy



  Wendy
  Encounters are rolled out unequally across GPC right now. Given
 differences in system installations, I suggest you consider the PCORI
 common data model for encounters which we all must implement (and are
 implementing or have implemented.  Encounter types indicate whether
 encounters are scheduled or completed:
   - ENCOUNTR TYPE (LOINC: 75525-6)
 - COMPLETED: AV=AMBULATORY VISIT; ED=EMERGENCY DEPARTMENT; IS=
 NONINSTITUTIONAL INPATIENT STAY; IP=INPATIENT STAY; OA=OTHER AMBULATORY
 VISIT
 - SCHEDULED OR ADMINISTRATIVE: OT=OTHER; UN=UNKNOWN; NI=NO
 INFORMATION
  Our source data at UNMC has considerable more detail regarding
 administrative types of encounters and you should tell us if you have
 needs for tracking those types of events.
  The Completed encounter types are actual face-to-face encounters of some
 type with the patient and are those you presumably what you want for
 research.  The others are administrative or scheduling encounters and
 should be ignored for patient care issues
 
  Have I answered your question?
 
  Jim Campbell

--
Ticket URL: 
http://informatics.gpcnetwork.org/trac/Project/ticket/178#comment:9
gpc-informatics http://informatics.gpcnetwork.org/
Greater Plains Network - Informatics

The information in this e-mail may be privileged and confidential, intended 
only for the use of the addressee(s) above. Any unauthorized use or disclosure 
of this information is prohibited. If you have received this e-mail by mistake, 
please delete it and immediately contact the sender.

___
Gpc-dev mailing list
Gpc-dev@listserv.kumc.edu
http://listserv.kumc.edu/mailman/listinfo/gpc-dev

Re: [gpc-informatics] #178: determine whether last.visit is the last fulfilled visit or the last scheduled visit

[GPC Informatics]

#178: determine whether last.visit is the last fulfilled visit or the last
scheduled visit
-+---
 Reporter:  mish |   Owner:  dconnolly
 Type:  enhancement  |  Status:  assigned
 Priority:  major|   Milestone:  cohort-char1
Component:  data-stds|  Resolution:
 Keywords:   |  Blocked By:  155
 Blocking:   |
-+---
Changes (by dconnolly):

 * owner:  jhe = dconnolly
 * type:  problem = enhancement
 * milestone:  data-quality2 = cohort-char1


Comment:

 Replying to [comment:9 jhe]:
  ... then we need to find that encounter and its type (completed or not).
 Dan and Tamara: do you know whether we can do this?

 Not yet.

 I expect we will be able to as a consequence of work Nathan is doing on
 #145 (and perhaps some other internal HERON work).

 I think Jim's diagnosis and suggested direction makes sense, so I'm
 changing this from a problem ticket to an enhancement request ad moving it
 to milestone:cohort-char1. We'll keep you posted on progress, Wendy.

--
Ticket URL: 
http://informatics.gpcnetwork.org/trac/Project/ticket/178#comment:10
gpc-informatics http://informatics.gpcnetwork.org/
Greater Plains Network - Informatics
___
Gpc-dev mailing list
Gpc-dev@listserv.kumc.edu
http://listserv.kumc.edu/mailman/listinfo/gpc-dev

Re: [gpc-informatics] #178: determine whether last.visit is the last fulfilled visit or the last scheduled visit

[GPC Informatics]

#178: determine whether last.visit is the last fulfilled visit or the last
scheduled visit
---+
 Reporter:  mish   |   Owner:  jhe
 Type:  problem|  Status:  assigned
 Priority:  major  |   Milestone:  data-quality2
Component:  data-stds  |  Resolution:
 Keywords: |  Blocked By:  155
 Blocking: |
---+

Comment (by jhe):

 Replying to [comment:7 campbell]:

 Jim,
  Thanks for the clarification. There is a variable called last.visit
 among the demographic variables in Heron. What we wanted to find out is
 whether last.visit is a fulfilled visit or not. If yes, then we know the
 patient was last known alive on that day. if no, then we can't use
 last.visit as last known alive date.

 It is great to know that encounter type can differentiate encounters
 completered and others. But this is not enough for answering the original
 question. When we run a query, we pull encounters related to a specific
 disease only, so the last encounter of the patient with the hospital
 (likely related other diseases) may not be in the pulled data.

 I don't know whether the date of last.visit corresponds to  the last
 encounter (among all encounters with the hospitals). If it is, then we
 need to find that encounter and its type (completed or not). Dan and
 Tamara: do you know whether we can do this?

 If last.visit is not linked to an encounter in the system, then we are
 back to the begining ot the original question: is last.visit fulfilled or
 scheduled only.

 Thanks,

 Wendy



  Wendy
  Encounters are rolled out unequally across GPC right now. Given
 differences in system installations, I suggest you consider the PCORI
 common data model for encounters which we all must implement (and are
 implementing or have implemented.  Encounter types indicate whether
 encounters are scheduled or completed:
   - ENCOUNTR TYPE (LOINC: 75525-6)
 - COMPLETED: AV=AMBULATORY VISIT; ED=EMERGENCY DEPARTMENT; IS=
 NONINSTITUTIONAL INPATIENT STAY; IP=INPATIENT STAY; OA=OTHER AMBULATORY
 VISIT
 - SCHEDULED OR ADMINISTRATIVE: OT=OTHER; UN=UNKNOWN; NI=NO
 INFORMATION
  Our source data at UNMC has considerable more detail regarding
 administrative types of encounters and you should tell us if you have
 needs for tracking those types of events.
  The Completed encounter types are actual face-to-face encounters of some
 type with the patient and are those you presumably what you want for
 research.  The others are administrative or scheduling encounters and
 should be ignored for patient care issues
 
  Have I answered your question?
 
  Jim Campbell

--
Ticket URL: 
http://informatics.gpcnetwork.org/trac/Project/ticket/178#comment:9
gpc-informatics http://informatics.gpcnetwork.org/
Greater Plains Network - Informatics
___
Gpc-dev mailing list
Gpc-dev@listserv.kumc.edu
http://listserv.kumc.edu/mailman/listinfo/gpc-dev

Re: [gpc-informatics] #178: determine whether last.visit is the last fulfilled visit or the last scheduled visit

[GPC Informatics]

#178: determine whether last.visit is the last fulfilled visit or the last
scheduled visit
---+
 Reporter:  mish   |   Owner:  huhickman
 Type:  problem|  Status:  assigned
 Priority:  major  |   Milestone:  data-quality2
Component:  data-stds  |  Resolution:
 Keywords: |  Blocked By:  155
 Blocking: |
---+

Comment (by huhickman):

 Note that our research Clarity instance at UNMC is in SQL Server and our
 i2b2 databases are in Oracle.  This particular view lives in SQL Server
 because of how we have to divide our work up - our Oracle Dynamic Gateway
 for SQL Server has habit of toppling over on queries that are complicated
 so some of our processing is done in SQL Server.  This is one example of
 that.

--
Ticket URL: 
http://informatics.gpcnetwork.org/trac/Project/ticket/178#comment:6
gpc-informatics http://informatics.gpcnetwork.org/
Greater Plains Network - Informatics
___
Gpc-dev mailing list
Gpc-dev@listserv.kumc.edu
http://listserv.kumc.edu/mailman/listinfo/gpc-dev

Re: [gpc-informatics] #178: determine whether last.visit is the last fulfilled visit or the last scheduled visit

[GPC Informatics]

#178: determine whether last.visit is the last fulfilled visit or the last
scheduled visit
---+
 Reporter:  mish   |   Owner:  huhickman
 Type:  problem|  Status:  assigned
 Priority:  major  |   Milestone:  data-quality2
Component:  data-stds  |  Resolution:
 Keywords: |  Blocked By:  155
 Blocking: |
---+

Comment (by campbell):

 Wendy
  Encounters are rolled out unequally across GPC right now. Given
 differences in system installations, I suggest you consider the PCORI
 common data model for encounters which we all must implement (and are
 implementing or have implemented.  Encounter types indicate whether
 encounters are scheduled or completed:
  ENCOUNTR TYPE (LOINC: 75525-6)
  COMPLETED: AV=AMBULATORY VISIT; ED=EMERGENCY DEPARTMENT; IS=
 NONINSTITUTIONAL INPATIENT STAY; IP=INPATIENT STAY; OA=OTHER AMBULATORY
 VISIT
  SCHEDULED OR ADMINISTRATIVE: OT=OTHER; UN=UNKNOWN; NI=NO INFORMATION
 Our source data at UNMC has considerable more detail regarding
 administrative types of encounters and you should tell us if you have
 needs for tracking those types of events.
 The Completed encounter types are actual face-to-face encounters of some
 type with the patient and are those you presumably what you want for
 research.  The others are administrative or scheduling encounters and
 should be ignored for patient care issues
 Have I answered your question?
 Jim Campbell

--
Ticket URL: 
http://informatics.gpcnetwork.org/trac/Project/ticket/178#comment:7
gpc-informatics http://informatics.gpcnetwork.org/
Greater Plains Network - Informatics
___
Gpc-dev mailing list
Gpc-dev@listserv.kumc.edu
http://listserv.kumc.edu/mailman/listinfo/gpc-dev

Re: [gpc-informatics] #178: determine whether last.visit is the last fulfilled visit or the last scheduled visit

[GPC Informatics]

#178: determine whether last.visit is the last fulfilled visit or the last
scheduled visit
---+
 Reporter:  mish   |   Owner:  huhickman
 Type:  problem|  Status:  assigned
 Priority:  major  |   Milestone:  data-quality2
Component:  data-stds  |  Resolution:
 Keywords: |  Blocked By:  155
 Blocking: |
---+
Changes (by dconnolly):

 * blockedby:   = 155


Comment:

 In reviewing the details from Keith W. ...
  - [http://listserv.kumc.edu/pipermail/gpc-dev/2014q4/000723.html Oct 28]
  - [http://listserv.kumc.edu/pipermail/gpc-dev/2014q4/000741.html Nov 3]
 with SQL query

 I realize that determining encounter type ETL is actually #155. I expect
 the answer to this derives naturally from that design. Or in other words:
 this is a use case for that design.

--
Ticket URL: 
http://informatics.gpcnetwork.org/trac/Project/ticket/178#comment:5
gpc-informatics http://informatics.gpcnetwork.org/
Greater Plains Network - Informatics
___
Gpc-dev mailing list
Gpc-dev@listserv.kumc.edu
http://listserv.kumc.edu/mailman/listinfo/gpc-dev

RE: [gpc-informatics] #178: determine whether last.visit is the last fulfilled visit or the last scheduled visit

[Wanta Keith M]

All,

This afternoon I got up to speed on visit types.  I have several dozen 
questions that may be relevant for this discussion.  I could probably spend 
some time throwing a query together, but I'd prefer to ask the questions before 
helping with a query since it would get complicated quickly.  This topic might 
drain our time during the next call, so it might be wise to have a side 
conversation before then.  

During today's GPC DEV call, I briefly spoke about release ranges.  This is 
just Epic's term for setting aside id ranges for records and items in 
Chronicles (Epic's DBMS).  We won't care about standard released records since 
analytical data won't be in these masterfiles which records are set aside.  
They are typically used for standard workflows or Model System data (which is 
not relevant to us).  We would however be interested in item category values 
that have a release range.  One example of this would be item EPT-30.  This is 
the encounter type item, which gets extracted to PAT_ENC.ENC_TYPE_C column for 
patient encounters.  Epic releases category values 1 through 119, and 150 
inclusive.  Anything outside that release range, our GPC would not need any 
mapped concepts for since those visit types are not standard within our GPC 
Epic sites.  Then we would need to find the common visit types of Cerner and 
Cattails (spelling?), and only include the shared visit types for all three 
EMRs.

-Original Message-
From: gpc-dev-boun...@listserv.kumc.edu 
[mailto:gpc-dev-boun...@listserv.kumc.edu] On Behalf Of GPC Informatics
Sent: Monday, October 27, 2014 10:39 AM
To: Mish Thomas; huhick...@nebraskamed.com; dconno...@kumc.edu
Cc: tmcma...@kumc.edu; gpc-dev@listserv.kumc.edu; tshire...@kumc.edu; 
j...@kumc.edu
Subject: Re: [gpc-informatics] #178: determine whether last.visit is the last 
fulfilled visit or the last scheduled visit

#178: determine whether last.visit is the last fulfilled visit or the last 
scheduled visit
---+
 Reporter:  mish   |   Owner:  huhickman
 Type:  problem|  Status:  assigned
 Priority:  major  |   Milestone:  data-quality2
Component:  data-stds  |  Resolution:
 Keywords: |  Blocked By:
 Blocking: |
---+

Comment (by dconnolly):

 Tom, you mentioned starting separate calls for quality issues. Do you want  to 
do that right away, or should I continue to track this on the regular  Tuesday 
calls, for now?

 Hubert, please give us an update on your progress.

--
Ticket URL: 
http://informatics.gpcnetwork.org/trac/Project/ticket/178#comment:3
gpc-informatics http://informatics.gpcnetwork.org/
Greater Plains Network - Informatics
___
Gpc-dev mailing list
Gpc-dev@listserv.kumc.edu
http://listserv.kumc.edu/mailman/listinfo/gpc-dev
___
Gpc-dev mailing list
Gpc-dev@listserv.kumc.edu
http://listserv.kumc.edu/mailman/listinfo/gpc-dev

RE: [gpc-informatics] #178: determine whether last.visit is the last fulfilled visit or the last scheduled visit

[Wanta Keith M]

Note, I'm only considering the standard release range category values for 
EPT-30 in this message.

Does last visit need to have some human contact (either physical or via phone 
or electronic communication)?

Since visits in i2b2 are generally captured with a provider (unless you're 
doing genetics sequencing or something), you normally need to have the actual 
provider.  So I think a completed appointment would only be considered with a 
provider, correct?  Since on a canceled, no show, or other that's not 
completed, would not have a confirmed provider.

There are exceptions to this rule.  A nurse can be considered a patient's 
provider for a visit.  In that case, we would have no provider, so those visits 
would likely seem as important as those without a provider.

Visit types such as 20-Treatment Plan or Appointment can turn into other visit 
types.  Since we're dealing with NAACR data, a treatment plan could turn into 
several other visits for chemo treatment (for each visit) for example.  But 
they do not confirm you have an actual visit where the patient was seen.  

116-Episode changes might mean there will be future visits, but doesn't 
necessarily mean we have a visit.  For example, we might have a pregnancy, and 
everything will be billed for that one episode at the end of it.  That first 
encounter might not mean we had actual human contact with the patient, but 
maybe we did.

Are we considering auxiliary (needed for medical staff to access EHR without 
patient) or ancillary system visit types (needed for interfaces/Bridges to move 
data into Epic)?  We don't really have patient/human contact, but there might 
have been from the interfaced medical software.  

I'd personally choose to throw away the following since we're dealing with 
things outside patient care, or they are just based on Epic Hyperspace 
workflows, or because they involve potential future visits (e.g. Wait List) or 
other unrelated patient activity.

5-Canceled   
6-Unmerge   
7-Contact Moved   
10-EMPTY   
30-Update   
97-Committee Review   
98-Post Mortem Documentation   
99-Billing Encounter   
109-History   
110-Referral   
112-Rx Refill Authorize   
115-Resolute Professional Billing Hospital Prof Fee  

Then we have the gray area visit types of whether there was human contact or 
not.  This would require more discussion.  For example, 60 is the precursor for 
something scheduled for the future, but we don't know if there was human 
contact.  And Abstract is very abstract!

30-Update   
60-Health Maintenance Letter   
64-Questionnaire Series Submission   
71-Nurse Triage   
102-Consent Form   
105-Letter (Out)   
150-Abstract   

Then we should consider if the patient is deceased or not.  That can change 
whether or not we have a last visit.  But a consent form (visit type 102) from 
a family member for example for organ donations would technically be considered 
a last visit in some doctor's books, even if the patient is deceased.

I'd like to build a query, but I'd like to weed out what is needed by PCORI 
first.


-Original Message-
From: Dan Connolly [mailto:dconno...@kumc.edu] 
Sent: Tuesday, October 28, 2014 5:13 PM
To: Wanta Keith M; gpc-dev@listserv.kumc.edu; Mish Thomas; 
huhick...@nebraskamed.com
Cc: Tamara McMahon; Theresa Shireman; Jianghua He
Subject: RE: [gpc-informatics] #178: determine whether last.visit is the last 
fulfilled visit or the last scheduled visit

Several dozen questions... for example?

i.e. please let's try to have the bulk of the technical discussion in email.

Please do throw a query together, even if only to point out why a simple one 
isn't adequate.

-- 
Dan


From: Wanta Keith M [kwa...@uwhealth.org]
Sent: Tuesday, October 28, 2014 4:12 PM
To: gpc-dev@listserv.kumc.edu; Mish Thomas; huhick...@nebraskamed.com; Dan 
Connolly
Cc: Tamara McMahon; Theresa Shireman; Jianghua He
Subject: RE: [gpc-informatics] #178: determine whether last.visit is the
last fulfilled visit or the last scheduled visit

All,

This afternoon I got up to speed on visit types.  I have several dozen 
questions that may be relevant for this discussion.  I could probably spend 
some time throwing a query together, but I'd prefer to ask the questions before 
helping with a query since it would get complicated quickly.  This topic might 
drain our time during the next call, so it might be wise to have a side 
conversation before then.

During today's GPC DEV call, I briefly spoke about release ranges.  This is 
just Epic's term for setting aside id ranges for records and items in 
Chronicles (Epic's DBMS).  We won't care about standard released records since 
analytical data won't be in these masterfiles which records are set aside.  
They are typically used for standard workflows or Model System data (which is 
not relevant to us).  We would however be interested in item category values 
that have a release range.  One example of this would be item EPT-30

Re: [gpc-informatics] #178: determine whether last.visit is the last fulfilled visit or the last scheduled visit

[GPC Informatics]

#178: determine whether last.visit is the last fulfilled visit or the last
scheduled visit
---+
 Reporter:  mish   |   Owner:  hhickman
 Type:  problem|  Status:  assigned
 Priority:  major  |   Milestone:  data-quality2
Component:  data-stds  |  Resolution:
 Keywords: |  Blocked By:
 Blocking: |
---+
Changes (by dconnolly):

 * owner:  dconnolly = hhickman
 * status:  new = assigned


Comment:

 Hubert,

 How are you determining visit type from Epic clarity data?

 Would you please share your SQL code? (or point me/us to it, if you
 already have)

--
Ticket URL: 
http://informatics.gpcnetwork.org/trac/Project/ticket/178#comment:1
gpc-informatics http://informatics.gpcnetwork.org/
Greater Plains Network - Informatics
___
Gpc-dev mailing list
Gpc-dev@listserv.kumc.edu
http://listserv.kumc.edu/mailman/listinfo/gpc-dev