Introduction to All of Us Physical Measurement Data Collection and Transformation Methods

  • Updated

Within the All of Us Research Program, physical measurements are obtained from participants in an in-person clinical visit, through online self-reported questionnaires, and via EHR extraction. Measurement data collected via a clinical visit or self-report are referred to ‘All of Us Program Collected Measurements’ and were selected based on factors such as relevance to research, widespread reproducibility, time and training requirements needed to carry them out, and resources required to obtain them.

Methods

Clinical Visits

Participants enrolled through their participating health care providers (HPOs) or a participating direct volunteer location are invited to complete a baseline clinic visit following completion of consent and the Basics survey. The baseline clinic visit includes both physical measurement and biospecimen collection. A selection of physical measurements are collected by trained staff and include blood pressure, height and weight, waist and hip circumference, and heart rate. In total, the physical measurement collection protocol takes about 15 to 20 minutes to complete and is completed by trained program staff. Protocols used by program staff to collect and record each of these measures are linked below.

*Body mass index (BMI) is calculated automatically from measured height and weight. 

Program staff conducting the clinical visit measurements record those data in a dedicated online program application called HealthPro, a web-based application developed and managed by All of Us to record information from participants’ physical measurements and to complete the biospecimen ordering workflow. Clinical visit measurement concepts (e.g., source codes) are deposited into the OMOP Common Data Model’s measurement table and stored in an All of Us specific “PPI” vocabulary. This vocabulary is both browsable and downloadable via Athena, the online OMOP repository. Recorded clinical data are data extracted from the All of Us program’s Raw Data Repository (RDR) and transmitted to the Curated Data Repository (CDR) for the purpose of research-grade dataset creation. Extracted measurement data are mapped according to the codebook source codes, as well OMOP concept codes and IDs. For more information, please see the All of Us Physical Measurement Collection codebook here. Clinic measurements are currently obtained through a single visit though All of Us plans to re-collect through follow-up visits in the future (timeline for launch and data release is TBD). *Note: Clinic visit height and weight data have been cleaned to reduce the likelihood of seeing data entry errors - you can find a list of cleaning applied to these data in the data dictionary.

Self Report (e.g, Remote Collection)

Participants who do not qualify for a baseline clinic visit are invited to remotely self-report height and weight via the online All of Us Participant Portal following completion of consent and the Basics survey. Height and weight data are recorded in an online survey response and corresponding data are mapped to the same height and weight concepts used to record clinic visit height and weight data. Self-reported measures are currently obtained through a single reporting instance though All of Us plans to re-collect through follow-up visits in the future (timeline for launch and data release is TBD). Note: Self-reported height and weight collection forms have validation settings configured to reduce the likelihood of seeing data entry errors - you can find a list of validation settings used in the self-reported physical measurement codebook

To get the counts of participants who self-reported height and weight, the following query can be used: 

#query to load the dataset in workbench
dataset = %env WORKSPACE_CDR

#query to count participants who provide self-reported physical measurements by participant portal
df = pd.read_gbq(f'''
SELECT COUNT(DISTINCT person_id) as n, measurement_source_value, src_id
FROM `{dataset}.measurement`
JOIN `{dataset}.person_ext` USING(person_id)
WHERE measurement_type_concept_id = 32865
GROUP BY 2,3
ORDER BY 2
''')

Electronic Health Records

Physical measurement data can also be derived from EHR records provided on behalf of those participants who have agreed to share. Source and standard concepts for EHR collected physical measurement concepts are stored in the OMOP measurement table. Unlike program collected measurements that are recorded at a single visit, the EHR can contain numerous measurements, recorded at different time points and clinic visits. Researchers can use date information obtained through the observation_date field to distinguish between entry dates.

A note about working with physical measurement data sources and OMOP: To distinguish between measurements recorded by All of Us versus those recorded in a participant’s EHR, use the source concept_ids. See the table below for more detail.

measurement_concept_id Standard measurement_concept_id Source
3004249 Systolic blood pressure 903109

1st systolic blood pressure

3004249 Systolic blood pressure 903114

2nd systolic blood pressure

3004249 Systolic blood pressure 903130 3rd systolic blood pressure
3012888 Diastolic blood pressure 903110 1st diastolic blood pressure
3012888 Diastolic blood pressure 903129 2nd diastolic blood pressure
3012888 Diastolic blood pressure 903106 3rd diastolic blood pressure
3025315 Body weight 903121 Weight
3027018 Heart rate 903112 1st heart rate
3027018 Heart rate 903105 2nd heart rate
3027018 Heart rate 903108 3rd heart rate
3036277 Body height 903133 Height
40759207 Adult waist circumference protocol 903127 1st waist circumference
40759207 Adult waist circumference protocol 903134 2nd waist circumference
40759207 Adult waist circumference protocol 903128 3rd waist circumference
40765148 PhenX- hip circumference protocol 903117 1st hip circumference
40765148 PhenX- hip circumference protocol 903125 2nd hip circumference
40765148 PhenX- hip circumference protocol 903123 3rd hip circumference

 

 

Was this article helpful?

1 out of 1 found this helpful

Have more questions? Submit a request

Comments

0 comments

Article is closed for comments.