Org Web Adapter

ID

7be0ce8c-bbc3-46f3-82bb-20561bee9b73

#+BEGIN: clocktable :scope file :maxlevel 2

#+CAPTION: Clock summary at Tue Mar 19, 201910:46

| Headline | Time | |

|--------------------+--------+------|

| *Total time* | *8:08* | |

|--------------------+--------+------|

| Monday, 03/18/2019 | 8:08 | |

| \_ Tasks | | 8:08 |

#+END:

Monday, 03/18/2019

Tasks

Today's meetings

:LOGBOOK:

CLOCK: Mon Mar 18, 201916:02--Mon Mar 18, 201917:00 => 0:58

CLOCK: Mon Mar 18, 201913:30--Mon Mar 18, 201913:47 => 0:17

:END:

Team IO Standup

Meeting went very quickly today, a testament to both the work I'm doing and the

ability of Ed to finish some parts of the CQM calculator up. Ayhan was also out

sick so a discussion with Ed that would have gone a bit longer due to Ed and

Ayhan working on angora testing.

*E All hands meeting

Select board

[[file:town.org::*Second%20March%202019%20Selectboard%20meeting][Second March 2019 Selectboard meeting]]

DONE Start on draft of budget narrative for town

:LOGBOOK:

CLOCK: Mon Mar 18, 201909:50--Mon Mar 18, 201911:00 => 1:10

:END:

DONE Deposit CMT cash

CLOSED: Mon Mar 18, 201911:15

:LOGBOOK:

CLOCK: Mon Mar 18, 201911:00--Mon Mar 18, 201911:25 => 0:25

:END:

DONE Grab immersion blender from Big house

CLOSED: Mon Mar 18, 201911:35 DEADLINE: Mon Mar 18, 201911:00

DONE Cancel GEICO insurance

SCHEDULED: Mon Mar 18, 2019

:LOGBOOK:

CLOCK: Mon Mar 18, 201911:48--Mon Mar 18, 201912:00 => 0:12

:END:

DONE RSVP to info@manor-inn.com regarding Men's Club on Wednesday

SCHEDULED: Mon Mar 18, 2019

:LOGBOOK:

CLOCK: Mon Mar 18, 201913:07--Mon Mar 18, 201913:13 => 0:06

:END:

I need to check with Emma to make sure Wed. night is free.

After talking with Emma, decided this wasn't worth it.

DONE Check that CQM reports went out on time

SCHEDULED: Mon Mar 18, 2019

:LOGBOOK:

CLOCK: Mon Mar 18, 201911:30--Mon Mar 18, 201911:45 => 0:15

:END:

Was able to verify by checking the prod database that three reprorts were properly created and emails sent.

DONE Figure out why flu vaccine CDS always calculates to the same dates

SCHEDULED: Mon Mar 18, 2019

:LOGBOOK:

CLOCK: Mon Mar 18, 201922:57--Tue Mar 19, 201900:24 => 1:27

CLOCK: Mon Mar 18, 201914:07--Mon Mar 18, 201916:02 => 1:55

CLOCK: Mon Mar 18, 201913:14--Mon Mar 18, 201913:30 => 0:16

CLOCK: Mon Mar 18, 201912:00--Mon Mar 18, 201913:07 => 1:07

:END:

Managed to get this to work by hacking the `code` field on the measure

definitions for 2018 and 2019 CMS147 measures. Very strange.

There appear to be two CDS triggers here, but only one is an ecqm_proportion_measure:

#+BEGIN_SRC python

[{'trigger': 'ecqm_proportion_measure', 'actions': ['immunization', 'alreadydone', 'delayed', 'referred', 'contraindicated', 'declined'], 'summary': '$flu_vaccine[Influenza Immunization] for Patients 6 mo. and older'},

{'trigger': {'$excluded': [{'all': ['$next_flu_shot_not_needed']}], '$pending': [{'none': ['$next_flu_shot_not_needed']}], '$relevant': [{'none': ['$next_flu_shot_not_needed']}]}, 'actions': ['immunization', 'alreadydone', 'delayed', 'referred', 'contraindicated', 'declined'], 'summary': '$next_flu_vaccine[Influenza Immunization] for Patients 6 mo. and older'}]

#+END_SRC

So one kicks out a False on whether it should be shown and the other a True

https://github.com/elationemr/hippo/blob/master/elation/quality_measures/reporting/elation_clinical_reminders.py#L144

09:48 Goals

I still haven't managed to get a mentorship document written up for working with

Ed. That's top of my list today, beyond getting the Flu vaccine CDS working

again.

Work log

Started out the day double checking that CQM reports went out on time. Everything looked good in the database, which was nice to see for a change. We still probably ought to build out some sort of a API for managing task-based reports like that so someone without database access can double check things like that.

I spent most of the day then looking at the flu vaccine issue, which is the same thing I did the past few working days. I started out today by returning the data that I wanted from the API call that returns which CDS should be displayed. This way, I was able to debug the fact that the influence CDS will display if the response to the per-patient CDS call (https://elation.onec.me/patient/64058687489/cds/2019/) is true for the NQF code for the flu vaccine (0041).

After confirming that the CDS works, I went backward to how that true response is generated and discovered something quite curious. Just like with the BMI CDS, which has three different CDS alerts that can be displayed, the flu vaccine CDS has two different alerts. One is for the vaccine, and the other seems to be about when the next vaccine is due.

I honestly couldn't figure out how to trigger the second alert, but it turned out that when we lookup the particular reminder, the first in the sequence was being set to False, i.e. the patient does not need the vaccine (despite the fact that they do), and the second alert in the sequence was being set to True. But the CDS only displays the second alert if the first on is non-False, so it was never appearing.

I need to talk to Sam tomorrow about my findings, because he mentioned something about how the flu vaccine CDS never actually worked as designed. I think this may be the culprit. If I remove the second reminder (for the next vaccine date) from the measure definition file, everything displays as you'd expect, and addressing the alert properly adds the patient to the numerator and all that jazz. So I think this is a solution, but I'm still unhappy with the fact that I don't know WHY it's the solution specifically.