We are currently in middle of ETL process of our EHR data. We have Inpatient, Outpatient and Emergency patient records. We have the patient visit start and visit end datatime captured in our EHR data.
Should we consider visit_start_date and visit_end_date as observation_period_start_date and observation_period_end_date? I mean we do not have info about the patients before they visit hospital. We don’t have any claims data or other types of patient data. As of now, we only have clinical data
For example -
The below table shows the hospital visit journey of few patients
Here, you can see that Person_id = 1 has followed IP–>OP—>IP visit journey.
So to record his observation period start and end dates, should I have one unique record per person which captures his whole window. I mean Minimum (Visit_start_date) and Max (Visit_end_date), this will result in one unique record per person. Please note that the date difference between patient’s journey can be in years as well
Or, should I have each visit record as an entry in observation_period as well?
But is this all called as Observation periods? I am quite confused
In Atlas, we have the observation period window (Cohort Entry criteria), where we have to provide the condition (days before and after index date), but hospital only contains data of their visit. Is visit dates and observation dates the same?
Can you please help us understand how to map observation period table and any simple example to help us understand would be very helpful.