Leveraging the All of Us Database for Primary Care Research with Large Datasets

Daniel J. Parente

doi:10.3122/jabfm.2023.230453R2

Article Figures & Data

Figures

Tables

Download figure
Open in new tab
Figure 1.
Schematic diagram of the All of Us research platform.
Download figure
Open in new tab
Figure 2.
Recommended workflow for defining concept sets, cohorts, and datasets.

Tables

Figures

View popup

Table 1.

Glossary of Terms

Term	Meaning	Examples
Concept	A piece of information about a participant. Concepts may include both the information itself (e.g., a survey response) and metadata about the information (e.g., time of collection, survey version).	- A participant’s age at time of consent - A survey question about experience of food insecurity
Concept set	A group of related concepts.	- The social determinants survey asks questions about whether a respondent experienced food insecurity, or only worried about experiencing food insecurity. A concept set containing both questions would represent a response to either (or both) of these questions.
Value	A specific data element (a number, a text string, etc.) from a concept.	- The participant’s answer to a survey question - The time and date the participant completed a survey - The version of the survey administered
Cohort	A population of interest, built using Boolean logic (“AND” and “OR” relationships) between concepts and concept sets.	- All participants who had both a Vitamin B12 measurement AND answered at least one of the two food insecurity questions on the social determinants of health survey. - All adults with diagnosis of diabetes
Dataset	A set of tables extracted from the All of Us database, chosen by specifying concept sets, and then selecting values from the concept sets to extract.	- A demographics table containing the participant’s code number and age, plus a measurements table containing a participant’s code number, Vitamin B12 serum concentration, and the time the laboratory test was collected.