FHIR Chat · Social History

Hi all, In our ipad app, we are capturing social history such as smoking,alcohol,drug and etc...for example 1) In alcohol abuse, which we have fields like alcohol use,daily amount,frequency,last drinkalcohol.PNG So i'm not able to map each field in observation resource exactly ? 2) In this screen we have cage questions like sought treatment for alcohol abuse, involved in a 12 step program etc ........ These are also unable map in observation..So please give me some suggestions

Ardon Toonstra (May 03 2017 at 10:14):

You can look at these profiles to have some inspiration on how to map this to an Observation resource: https://simplifier.net/Nictiz/zib-AlcoholUse/rendered https://simplifier.net/Nictiz/zib-DrugUse/rendered

Mounika (May 03 2017 at 10:34):

Okay @Ardon Toonstra Thank you. Which tool should i use to create these profiles (which support json format )

Ardon Toonstra (May 03 2017 at 11:58):

I am not aware of a json profiling tool at the moment, but you can use Forge for this (works with xml). Forge can be downloaded for free from Simplifier.net Converting the xml to json shouldn’t be to difficult. You can for example upload your xml to Simplifier and download the json version of it.

Mounika (May 03 2017 at 12:40):

@Ardon Toonstra Oh okay. Thank you

Michelle (Moseman) Miller (May 03 2017 at 13:57):

Refer to prior discussion about representing Social History as observation: https://chat.fhir.org/#narrow/stream/implementers/topic/Social.20History.20Observation.20example

I included a link that shows example tobacco observations (http://fhir.cerner.com/millennium/dstu2/diagnostic/observation/), but there are similar codes available for alcohol, such as....

SNOMED:
228328008 Age at starting drinking (observable entity)
363905002 Details of alcohol drinking behavior (observable entity)
440652002 Alcohol abuse cessation behavior (observable entity)

LOINC:
11287-0 Alcoholic drinks per drinking day - Reported
44940-5 Alcoholic drinks per week - Reported
68518-0 How often do you have a drink containing alcohol

Eric Haas (May 03 2017 at 15:54):

The Ig-publishing +spreadsheet authoring tool will suppport creating profile in both xml and json as well. Here is a template. It designed for publishing but is also a great way to create profiles too. There are some observation profiles created using in here in theresources directory

Mounika (May 04 2017 at 08:59):

Okay @Michelle (Moseman) Miller Thank you. But how to implement these type of cage questions. I'm not getting. please give me some suggestions...

Michelle (Moseman) Miller (May 04 2017 at 12:47):

@Mounika K , Can you elaborate on your question? Each question on your form could be its own Observation where the Observation.code is the question (e.g. Age Started, Daily Amount, etc.) and the Observation.value would be the answer (e.g. valueQuantity of 21 years old for the question of Age Started).

Mounika (May 04 2017 at 12:55):

Okay @Michelle (Moseman) Miller For example, I want to capture psychiatric history of a patient see this , here we have some cage type of questionsCapture2.PNG

Mounika (May 04 2017 at 12:56):

So please explain me how to implement those fields?

Michelle (Moseman) Miller (May 04 2017 at 13:56):

@Mounika K Unfortunately, I don't think I understand your question because I'm responding with the same suggestion as earlier.
Observation.code = "the question" = Suicidal thoughts
Observation.valueBoolean = "the answer" = true or false
or
Observation.valueCodeableConcept = "the answer" = Yes or No / Present or Absent

Mounika (May 05 2017 at 06:15):

Okay @Michelle (Moseman) Miller In observation resource, we don't have that type of code for suicidal thoughts .So we need to create that code or something else?

Mounika (May 05 2017 at 13:13):

In our ipad app, one of the screen we capturing the socialHistory. for example alcohol abuse. in the we are asking the questions like this sought treatement for alcohol abuse, involved in a 12 step program, have you had withdrawl problems, emergency medical attention required due to intoxication, family history of alholism? here my doubt is should i need to add Observation.component.code for each question and value is ans for that question. or is there any other way? and if i don't fined codes for those questions, what to do? please any suggestions?

Mounika (May 05 2017 at 13:19):

Capture11.PNG

Michelle (Moseman) Miller (May 05 2017 at 16:05):

Observation.code is a CodeableConcept, which means it may just have Observation.code.text without any Observation.code.coding. That said, it is better if you can find standard codes, such as SNOMED has Suicidal thoughts (finding) SCTID: 6471006

As I mentioned in https://chat.fhir.org/#narrow/stream/implementers/topic/Social.20History.20Observation.20example, we plan to use Observation.component when there are multiple answers to a single question, such as Alcohol Type

Observation.code = Alcohol Type
Observation.component[1].code = AlcoholType
Observation.component[1].valueCodeableConcept = Beer
Observation.component[2].code = AlcoholType
Observation.component[2].valueCodeableConcept = Wine

Our implementaiton uses Observation.related to link together all of the related Observations. Thus, we have one overall Observation for Alcohol (Risk), which has multiple related Observations for each of the individual questions.

Mounika (May 08 2017 at 07:16):

Okay @Michelle (Moseman) Miller Thank you for that. Should i need to create profiles for each observation ? please clarify me

Mounika (May 08 2017 at 11:14):

In observation B.P. In that profile we have added components like systolic and diastolic like that we have an observation alcohol so we need to create profile for alcohol observation ? or UI will add the components

Lloyd McKenzie (May 08 2017 at 12:34):

It's not clear that alcohol consumption should be handled using "component". Systolic and Diastolic aren't really separable - they pretty much always need to be looked at together. Tobaco use and alcohol use are really independent things

Jenni Syed (May 08 2017 at 12:52):

The only place we use component for Social history is when there are 2 answers to the same question - so not separable. EG: Type of tobacco or Michelle's example above. The rest are all linked/related only

Russell Leftwich (May 08 2017 at 13:58):

Systolic and Diastolic blood pressure are separate measurements. They are separable and you cannot calculate one from the other. There are clinical situations when you can only measure one of the two: if a patient has atrial fibrillation, the systolic blood pressure varies with every beat, only the diastolic has meaning; if a patient is in shock and you can only palpate their blood pressure, you are measuring their systolic pressure only. (I am not sure why this systolic/diastolic comment was in the social history thread)

Jenni Syed (May 08 2017 at 14:04):

@Russell Leftwich I'm guessing it was mentioned because blood pressure is the primary example of a component for Observation. Essentially: when they're measured together, you should interpret them together (you can absolutely have only a systolic or only a diastolic component). http://hl7.org/fhir/stu3/observation-definitions.html#Observation.component

kler (Feb 01 2018 at 11:30):

can any one tell which is the best resource for Social History ??

Lloyd McKenzie (Feb 01 2018 at 15:30):

Look above in this thread - that's the guidance we have at the moment

kler (Feb 02 2018 at 04:28):