MU2 Progress


(system) #261

harleytuck wrote on Wednesday, December 23, 2015:

ok, thanks for the confirmation. And I haven’t gotten a reply from Milton yet.
I guess we’ll see.

  • HT

(system) #262

harleytuck wrote on Thursday, December 24, 2015:

Ok, I like that plan, if only for its simplicity.

Re: the a8 problems, the pattern has been that Milton and I go along and he notices some problem, “we can’t continue without that”. Then we come back and go til the next speedbump, so I haven’t been able to do a full survey of all the things to correct at once.

The things I know about are the med/ allergy CDS rule that has to be work as meds are entered manually, transferred from NewCrop or imported from a CCDA file. Which sounds like it needs to be one of the default rules that get applied for all new pts. And, being a CDS, must be controlled by the same mechanisms that regulate who can activate and/ or trigger a CDS, and how it gets logged.

The logging requirement is, “4. Verify that the status and timing of the activated clinical decision support interventions is recorded by the EHR (e.g. by viewing a log, report)” so it sounds like that report you made should work just fine. Oh right-- I haven’t played with it yet. Does it record the time the rules are turned off also?

Milton doesn’t like passive reminders- says “They might get missed”. So he has me turn off all the passive reminders and just show the actives, because of the popup. BUT- the reference and citation content, and that popup-y thing that tells the newly added CDS’s are linked to the passive reminders- if turned off we can’t show any of that stuff.

I told him it was a conscious decision to counter alert fatigue and he didn’t object but to be safe, I guess when he says to activate a CDS alert I have to activate both active and passive. It seems like unnecessary klugery to code something to deal with that but now you know the situation.

One thing he wants that we haven’t gotten to and I haven’t been able to test yet is, for e.g., if a rule depends on a diagnosis, when a nurse enters the dx he will see the reminder. But if a Dr who can’t activate that rule enters the dx, she won’t see the reminder. You think that is going to be a result of the fine-grained CDS permissions?

Whew. There endeth my notes.

Last thing: Milton has available next week:
Wednesday – Open from 9:30 am – 1:00pm PST
Thursday – Open from 9:30 am – 12:00pm PST

Nab it all?

  • HT

(system) #263

bradymiller wrote on Thursday, December 24, 2015:

Hi Harley,

Schedule both of them. I’ll begin to work/look at these issues.

-brady


(system) #264

bradymiller wrote on Thursday, December 24, 2015:

This is likely a question for Tony.

Is there any strategic reason to getting a round of Modular Certification before the new year. If so, then I am guessing the testing sessions next week will be too late (or will they?).

-brady
OpenEMR


(system) #265

tmccormi wrote on Thursday, December 24, 2015:

At this point our certification (modular or otherwise) won’t get officially approved and posted until next year, but there is good symbolic and actual value in getting as much as we can to pass testing by the end of the year, I think.

Let’s not stop the train unless the tracks give out.

–Tony


(system) #266

harleytuck wrote on Thursday, December 24, 2015:

This back from Milton:

Hi Harley,
Usually we see one SPM line for one report that covers all three parent_child results for test case 4.1. In your case the reports are separated into three individual reports, therefore it is expected to be in each report.

Happy holidays!
Regards,
Milton Padilla

Sounds like it’s more complicated than I had any idea of.

  • HT

(system) #267

sunsetsystems wrote on Thursday, December 24, 2015:

Hi Harley,

Would you ask him if he was expecting child results to be included in the parent report? So that test case 4.1 would produce 1 report instead of 4?

Oh and also I’m wondering why the 4.1 file includes 3 identical sets of child results…?

Rod


(system) #268

bradymiller wrote on Thursday, December 24, 2015:

Hi Harley,

To address all Milton’s issues.

I just committed a solution to the official codebase to catch allergy conflicts (when a prescription or medication is also on the allergy list). It basically piggybacks the mechanisms of the CDR engine alerts:
-All allergy conflicts always shows in the Active Alert popup when open up the patient summary screen.
-New allergy conflicts show as a popup when refresh the patient summary screen.
These alerts are also logged to the Reports->Clinic->‘Alert Log’ with the other alerts. And it can be turned on/off in Administration->Globals->CDR->‘Enable Allergy Check’; as required with turning on/off CDR rules is only allowed by users with certain (admin) privileges.

Regarding ccda file, recommend having one with same drug on the medication and allergy list to flag above (ensure to test the ccda far in advance to ensure it imports these elements along with the medical problem).

The log records the alert time, patient id, user id, list of all alerts, and list of new alerts for each entry. The turning on/off of the rules is not in this log (and should not be) and is instead stored in the general log at Administration->Other->Logs.

When setting whether to select active/passive reminders for a rule, the best way to think about this is that passive is to get the basic functionality and active is to extend the functionality(to be a bit more invasive). So, should never have a rule do only active without passive or wouldl lose the basica functionality of the Clinic Reminder Widget at the right. So, if Milton wants to always use Active for testing, then always use both Active and Passive. It’s my opinion that Milton is going out of his scope by forcing this since Passive does show a popup for any new alerts. I am a physician, and have tried to make the flow natural for both passive and the combination of active/passive, but I can almost guarantee that most physicians will prefer passive only for majority of the rules(and active/passive for only the super important rules, if any). My philosophical differences don’t matter, though, since Milton’s biases and objectives can be achieved by using the active/passive combination.

Regarding the diagnosis with the nurse vs. doctor. The granular functionality is already shown in the physician vs accounting example for the dummy ‘Data Entry - Social Security Number’ previously discussed. Pretty much anything you want can be configured with the right configuration of the ACL and the assignment of the ACO to the rule. If Milton needs a nurse vs. doctor where the nurse sees it but the doctor doesn’t, then could do the following to show this:
Go to Administration->ACL
Click ‘Groups and Access Controls’
Click ‘Physicians-write’->Edit
In the Active column, click ‘Sensitivities->Normal’ and then click ‘>>’ button
(note the ‘Sensitivities->Normal’ item has now moved to the Inactive column)
Now go to Administration->Alerts
For the ‘Hypertension: Blood Pressure Measurement’, for the ACO selection (selector at the right), select ‘Sensitivities: Normal’.
Then click ‘Save’.
Now create 2 new users:
1 user with the ‘Access Control’ setting of ‘Physicians’ - the Doctor
1 user with the ‘Access Control’ setting of ‘Clinicians’ - the Nurse
Now create a new patient and give set a Medical Problem of ‘HTN’
Now log out.
Now log in with the Nurse and the ‘Measurement: Blood Pressure’ alert will show up.
Now log in with the Doctor and the ‘Measurement: Blood Pressure’ alert will NOT show up.
(so it’s simply a matter of configuration and the sky is the limit on this)
(I would recommend even doing these steps in front of Milton so he gets an idea of how granular and configurable this can be).

a8 has really turned into a bear of an item which touches on many things that will be very tough to remember. I’d recommend making a well laid out “script” to go through with Milton of all the elements we have done so far. This “script” can then lend to documentation on the wiki down the road.

Thanks for remaining persistent :slight_smile:
-brady
OpenEMR


(system) #269

bradymiller wrote on Thursday, December 24, 2015:

Also, the listing of the global settings docs for the CDR engine may be helpful (it is up to date with the most recent globals there):
http://www.open-emr.org/wiki/index.php/Administration_Globals#CDR


(system) #270

tmccormi wrote on Thursday, December 24, 2015:

test Site has been updated

commit afc48c4f20b24d4421561c3c97c262e6e95a4cd9
Author: bradymiller bradymiller@users.sourceforge.net
Date: Wed Dec 23 16:32:16 2015 -0800

Incorporated allergy conflicts in addition to some CDR engine improvements.

(system) #271

sunsetsystems wrote on Thursday, December 24, 2015:

Never mind. On studying the test data and Juror Documents, it’s clear that the underlying issue is we are doing nothing special to identify and show parent/child result relationships. This is going to take some work. Just showing the specimen info in child results will not be enough to pass the test.

This was not the clear sailing I expected. How on earth did Ensoftek get it through?

Rod


(system) #272

tmccormi wrote on Saturday, December 26, 2015:

Re: This was not the clear sailing I expected. How on earth did Ensoftek get it through? Rod

Ensoftek has full lab interfaces to several national labs (as does MI2 and WMT and, I suspect, ZHH), those labs already required that kind of thing.

In MU1 we certified using MI2 Labcorp results interface, I was/am hoping to not have to do that so there would be more in the FOSS code and not be dependent on proprietary interfaces.

If we have to we can certify with one of our commercial interfaces, but it makes the user dependent on the lab we choose and the Labs are in control of who gets to have those interfaces, not the vendor or the user.


(system) #273

sunsetsystems wrote on Monday, December 28, 2015:

Anybody want to sponsor the lab code updates? Tony, does OEMR have funds for this? I figure about $700 for analysis and programming to handle child results and get the missing information into the reports. This would include some reworking of code to keep it clean.

Rod


(system) #274

harleytuck wrote on Monday, December 28, 2015:

This just in from Milton- might affect plans for Ensoftek’s testing f2, also the correction needed to the cds logging and Rod’s labs corrections: his testing time has changed- he only has Tuesday (tomorrow) 10am - 3p.
I reserved it.

  • HT

(system) #275

harleytuck wrote on Monday, December 28, 2015:

Hi Brady-
Was looking at the logging function and noticed that by way of stating the status of a rule it lists current rules, , which we might be able to get away with. But when I turn off just one rule it doesn’t state that; it just shows an empty line. as in the top few lines of the pic.

And please note my other entry this am: Milton changed the ONLY testing time he has to tomorrow 10a - 3p.

Rgds- HT


(system) #276

harleytuck wrote on Monday, December 28, 2015:

Hi Brady-
Was looking at the logging function and noticed that by way of stating the status of a rule it lists current rules, , which we might be able to get away with. But when I turn off just one rule it doesn’t state that; it just shows an empty line. as in the top few lines of the pic.

And please note my other entry this am: Milton changed the ONLY testing time he has to tomorrow 10a - 3p.

Rgds- HT


(system) #277

harleytuck wrote on Monday, December 28, 2015:

Hi Folks-
To recap the latest MU testing situation which I’ve separately mentioned scattered across other postings:

  • Brady’s CDS logging system needs to specify date / time of when a rule is activated and deactivated, and which rule it is. Otherwise we can’t test a8-CDS.
  • Rod’s lab display has to do the specimen type thing Milton described in his email I posted last week. Otherwise we can’t test b5a-Labs.
  • have called Vamsi and he absolutely has no time to do the f2- Immunization Registry testing. He took me on a quick walkthrough of the test requirements but it’s clear that we do not have time for him to show me how to do it myself.
  • as of Milton’s email to me this morning we have only tomorrow 10a - 3p pacific time for testing, not Wed and Thu.
  • otherwise OpenEMR will be certified with the modules we have passed at this point in time.
    Regards- HT

(system) #278

bradymiller wrote on Monday, December 28, 2015:

Hi Harley,

There are two very different things going on here that are logged in two separate places.

The Alert Log will log/record all alerts that a user sees, which is required in the testing criteria. I added the “New Alerts” column in the report to clearly show the new alerts (albeit this could be deduced from the “All Alerts” column by looking at the prior similar log entry, of which same would go for alerts that have gone away…).

The triggering of the alert is different than the turning on/off of a rule (this is where the user would explicitly turn the actual rule off/on for entire clinic or a selected patient). And this behavior will be logged in OpenEMR’s general log at Adminstration->Logs.

-brady


(system) #279

bradymiller wrote on Monday, December 28, 2015:

Also,

If you just simply need a column in the Alert Log that states “Removed Alerts” to show the alerts that no longer exist(when compared to prior alert log entry), I could easily add that into the codebase this evening, if needed.

-brady


(system) #280

harleytuck wrote on Monday, December 28, 2015:

Hi Brady-
Ok, I appreciate the clarification. Discovered that my problem re: finding the turning off/ on of the rules was I was not looking for the right sql statement in the logs.

You don’t need to add a ‘removed’ alerts now I know how to locate them in the logs. Thanks anyway.

  • HT