Stream: implementers
Topic: Immunization Schedule
Jose Costa Teixeira (Jul 15 2020 at 21:39):
Just double-checking - the most appropriate resource for a national vaccination plan is PlanDefinition,
right?
Lloyd McKenzie (Jul 15 2020 at 21:45):
For a single individual?
Lloyd McKenzie (Jul 15 2020 at 21:45):
I would say ImmunizationRecommendation
Jose Costa Teixeira (Jul 15 2020 at 21:45):
no, the standard vaccination plan
Jose Costa Teixeira (Jul 15 2020 at 21:46):
or schedule / schema
Lloyd McKenzie (Jul 15 2020 at 21:46):
Yes
Jose Costa Teixeira (Jul 15 2020 at 21:46):
thanks
Lloyd McKenzie (Jul 15 2020 at 21:46):
I would call it "protocol" rather than "plan"
Jose Costa Teixeira (Jul 15 2020 at 21:48):
yes, I got lost in translation between these languages- our local words translate as schedule or schema, what is the common english word for that?
Lloyd McKenzie (Jul 15 2020 at 21:49):
"Immunization schedule" would be pretty typical English terminology.
Jose Costa Teixeira (Jul 15 2020 at 22:05):
I'm trying to express the following:
- Between ages 11 and 12
- Between weeks 24 and 32 of pregnancy
- Every 10 years
Lloyd McKenzie (Jul 15 2020 at 22:06):
Talk to Bryn Rhodes :)
Jose Costa Teixeira (Jul 15 2020 at 22:09):
for 1. I would look at Age but it's a quantity
for 2. I would look at triggerdefinition, but: TriggerDefinition.name is a string (would be better to have a code) - and I don't know how to say "24 weeks of pregnancy" (esp. the "24 weeks from.." bit)
for 3. we don't seem to have anything...?
Jose Costa Teixeira (Jul 15 2020 at 22:09):
@Bryn Rhodes ?
Bryn Rhodes (Jul 15 2020 at 22:26):
- Between ages 11 and 12:
<timingRange>
<lo>
<value value="11"/>
<code value="a"/>
</lo>
<hi>
<value value="12"/>
<code value="a"/>
</hi>
</timingRange>
- Between weeks 24 and 32 of pregnancy
<timingPeriod>
<extension url="http://hl7.org/fhir/StructureDefinition/cqf-expression">
<language value="text/cql"/>
<expression value="Interval[Pregnancy.onset + 24 weeks, Pregnancy.onset + 32 weeks]"
</extension>
</timingPeriod>
Where pregnancy is defined something like:
define Pregnancy: [Condition: PregnancyFindings] C where C.clinicalStatus ~ ActiveClinicalStatus
- Every 10 years
<timingTiming>
<repeat>
<frequency value="1"/>
<period value="10"/>
<periodUnit value="a"/>
</repeat>
</timingTiming>
Jose Costa Teixeira (Jul 16 2020 at 06:59):
I was only looking at r4, didn't see some of these elements.
Is there not a way to say "11 years old" that differs from "11 years"?
Jose Costa Teixeira (Jul 19 2020 at 13:26):
Anyone working on immunization? Is there is a clear/common way to handle this:
image.png
When the patient is 6 years old, they take DTaP for the first time (until then they took DTaP+IPV+Hib+HepB).
What is the protocolApplied.doseNumber in this case?
is it 5 (because for three of those vaccines it's the 5th immunization?
or 1? (because it's the first time they take this drug)
In case it's 5, how do we know whether this means the previous 4 were DTaP or DTaP+IPV+Hib+HepB? Does it matter?
In case it's 1, does it matter?
Lloyd McKenzie (Jul 19 2020 at 14:12):
@Craig Newman
Craig Newman (Jul 20 2020 at 16:22):
It may depend on the system doing the evaluation and forecast, but typically I'd expect it to be counted as the 5th dose of DTaP regardless of the actual product administered in the past. Typically, it won't matter if previous doses were just DTaP alone or DTaP as part of a product that included other vaccine (IPV, Hib, etc).
Craig Newman (Jul 20 2020 at 16:24):
There may be a few fringe cases where the product does matter. For example, in the US, the two Menignococcal B vaccine (Trumenba and Bexsero) are not considered to be equivalent and so a series started with one vaccine type cannot be finished with the other vaccine type. But this is pretty rare (at least in the US).
Derek Ritz (Oct 04 2020 at 15:46):
Hi all -- and I'm sorry if this sounds stoopid -- but when we administer combination vaccines, should we not be tracking the "active ingredients" separately for each child? I appreciate fully that, from a supply chain standpoint, I need to track that combo drug ABC was given at time T1. But if, at time T2, the 3 drugs are separately administered (as A, B, and C) -- this is just as valid. The child has now received dose 2 for A, for B, and for C. Right? It would simplify things, it seems to me, to track a child's progress along each separate axis of vaccination -- regardless of the cocktail in which they are delivered.
Jose Costa Teixeira (Oct 04 2020 at 20:30):
you mean if the vaccine is against diphtheria, tetanus, pertussis, poliomyelitis, Haemophilus influenza type B and hepatitis B, you have 6 axis?
Jose Costa Teixeira (Oct 04 2020 at 20:32):
if so, what do you want to track? immunization or immunity? one immunization action addresses 6 immunities.
Derek Ritz (Oct 04 2020 at 23:10):
@Jose Costa Teixeira I guess what I'm thinking is that one administration of a 6-vaccine cocktail provides progress towards immunity against six vaccine-preventable diseases. BTW... I'm surmising that you are taking immunization to mean "the act of administering a vaccine" and immunity to mean "the state of being protected against a vaccine-preventable disease". Is this terminology the accepted practice? (sorry... it makes sense of course... it's just that I've not heard the terms employed that way). And... just because I jabbed a baby and made them cry doesn't necessarily mean I actually made progress towards immunity. The timing of the medication administration is very important. Not adhering to the timing (incorrect spacing between the 3 doses) and the immunity is not achieved.
Jose Costa Teixeira (Oct 05 2020 at 08:50):
I think the tracking of immunity is an interesting area, but not (yet) commonly done. I can think of at least one case where it would make sense, though..
Jose Costa Teixeira (Oct 05 2020 at 08:52):
The problem we have is that our Immunization resources track the paper approach rather closely. It's very close to simply stating "(how) is this person vaccinated according to our rules"?
Jose Costa Teixeira (Oct 05 2020 at 08:52):
E.g. there is one single code for administering the hexavalent vaccine.
Jose Costa Teixeira (Oct 05 2020 at 08:55):
I can imagine that a "Immunity condition" and a "immunization path" for each disease would make sense..
Craig Newman (Oct 05 2020 at 13:06):
Typically, it's the administration event that is captured. If a combination vaccine is used, that's still a single event. In my mind in part this is documentation of what was actually done and is necessary for things like billing and vaccine recall follow up, etc. As you noted, just giving the vaccine says nothing about acquired immunity. I expect that CDS engines that are providing forecasts are able to break apart a combo vaccine into its individual components and forecast on a per disease basis, taking into account that the schedule for each disease may be different even if they are given as part of a single vaccine.
Nathan Bunker (Oct 06 2020 at 13:01):
I don't have a lot to add to this conversation, but having written an immunization forecaster before I have a in-depth view of how you deal with combos. Like Craig says you record immunizations as they are given, but the evaluating system has to see combos as being made up of different vaccinations. There is an idea of a "path to immunity", where each dose (whether combo or not) can lead you towards the goal of immunity. Of course, we don't know if the immunization resulted in immunity for a specific patient but we can indicate that the patient followed the timing and spacing intervals correctly and so is assumed to be immunized.
Last updated: Apr 12 2022 at 19:14 UTC
