Encounter Infrastructure - Capture longitudinal data in data model

[lopierra]

Currently we have Conditions but these are not linked to time points. How to capture this? Record IDs, Encounters, Observations, ??

We have various datasets that have timepoints, e.g.

• HTP - multiple visits per participant (currently capture by Age at Biospecimen collection, since a sample is collected at each visit).
  • Conditions are shown as "ever observed" for that participant - not linked to specific visits
  • non-Condition data (BMI, weight, height) varies by visit
• DS-Connect - multiple surveys per participant, could indicate new/updated information. Also many conditions have "Age at" columns
• ABC-DS - 3 timepoints - baseline, Cycle 2, Cycle 3, ~1 year apart

[lopierra]

See #69 for Participant fields related to Age at enrollment/registration

Here are the other age or time-related fields:

• HTP: BMI, Weight_kg, Height_cm at each visit
• Nexus: TBD when I get Data Service pull from Meen
• Lupo: TBD
• Sherman: TBD
• PCGC: TBD
• DSC: deceased_year, survey_time, age at various conditions, basically every survey is a time point
• ABC-DS: baseline/cycle 2/cycle 3,
• BRI-DSR: n/a

[putmantime]

@lopierra added an eventID and event field but this needs to be reviewed.
Relates to modeling efforts for Conditions/Observations/Measurements

[lopierra]

We will use Encounters as the grouping for longitudinal data - work in progress here

Also want to try to align with FHIR if possible

[lopierra]

Add Clinical Trial stuff - https://docs.google.com/spreadsheets/d/15zaNUyosidzlQp-gVXImckjo661c7q8K9yfKHw1ApXM/edit?gid=0#gid=0