OHDSI Home | Forums | Wiki | Github

Persons and Exposure During Baseline and Total Exposure in Years

Would someone please explain to me what the “Total Exposure in Years” represents in the “Persons and Exposure during baseline period” tab of a full report made in ATLAS? Is it possible to be less than the total number of people in the cohort? Thanks! @Chris_Knoll @t_abdul_basser.

Hey @mattspotnitz, Which report are you specifically referring to? Have you run an estimation study, as designed by ATLAS and executed in R? Or are you looking at incidence Summary report in ATLAS? In either case, ‘Total Exposure in Years’ is the duration of time from the cohort start date to the cohort end date, inclusive of any right-censoring due to person having the outcome. Its in unit of years, so it is quite possible to to see a number of persons smaller than the ‘total exposure’, because persons could be followed for less than 1 year on average.

@mattspotnitz this probably refers to the Hearacles output. This is an obsolete function. It was developed as part of a proof of concept to see if OMOP CDM could support the cost and utilization reports commonly used by health insurance companies. The findings were that, OMOP CDM using the new COST table (of CDM V6.0) was able to support these use cases accurately.

The “Total Exposure in Years” is essentially the denominator in person-time that the persons belonged to the cohort.

Thanks @Patrick_Ryan and @Gowtham_Rao! I am referring to a summary report in ATLAS. Can you please explain the difference between total exposure in years for the “during baseline period” calculation compared to total exposure in years for the “during cohort period” calculation? I am interested in knowing the total amount of time that my cohort was observed in the database prior to the index event. Is that calculated in ATLAS? Also, are there any other obsolete functions in the summary report?

This was used in the cost and utilization pilot. The
“during baseline period” - was upto 1 year prior to cohort_start_date, not including cohort start date when the persons in the cohort had an observation_period
“during cohort period” - was the period between cohort_start_date and cohort_end_date for each person in the cohort.

(I think we only allowed the one record per person)

ps: this is an obsolete function that is no longer supported by the community. We recommend you to please explore, use and contribute to the new ‘Cohort Characterization’ functionality.

t