Table 1.
Descriptive characteristics of the Michigan Genomics Initiative, the UK Biobank, and All of Us. For unweighted metrics, mean (standard deviation) and percent (n) are provided for continuous and categorical/binary variables, respectively. For weighted metrics, mean (standard error) and percent (standard error) are provided for continuous and categorical/binary variables, respectively.
| All of Us (AOU)b | Michigan Genomics Inititative (MGI) | UK Biobank (UKB)c | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
|
| ||||||||||
| NHISa (weighted) | Unweighted | IP-weighted | PS-weighted | Unweighted | IP-Weighted | PS-weighted | Eligible population | Unweighted | IP-weighted | |
|
| ||||||||||
| N | 244,071 | - | - | 81,243 | - | - | - | 401,167 | - | |
| Demographics | ||||||||||
| Age (continuous)d | 47.7 (0.2) | 54.01 (17.27) | 49.8 (0.1) | 51.1 (0.1) | 56.34 (16.99) | 45.5 (0.1) | 45.6 (0.1) | 54.78 (0.01) | 57.67 (8.02) | 54.9 (0.0) |
| Aged | ||||||||||
| 0–9 | 0.0 (0.0) | 0.0 (56) | 0.0 (0.0) | 0.0 (0.0) | 0.0 (1) | 0.0 (0.0) | 0.0 (0.0) | - | 0.0 (0) | 0.0 (0.0) |
| 10–19 | 3.4 (0.2) | 0.8 (2,007) | 1.1 (0.1) | 0.8 (0.0) | 0.6 (472) | 2.2 (0.1) | 2.2 (0.1) | - | 0.0 (0) | 0.0 (0.0) |
| 20–29 | 17.6 (0.3) | 10.2 (24,907) | 13.0 (0.2) | 11.1 (0.1) | 8.7 (7,067) | 21.9 (0.3) | 21.6 (0.3) | - | 0.0 (0) | 0.0 (0.0) |
| 30–39 | 17.1 (0.3) | 14.2 (34,719) | 19.7 (0.2) | 19.1 (0.1) | 10.6 (8,636) | 18.4 (0.3) | 18.7 (0.3) | - | 0.0 (3) | 0.0 (0.0) |
| 40–49 | 16.0 (0.3) | 14.0 (34,080) | 16.6 (0.2) | 20.0 (0.1) | 13.5 (10,928) | 19.7 (0.3) | 20.0 (0.3) | - | 21.0 (84,174) | 34.1 (0.1) |
| 50–59 | 16.4 (0.3) | 19.2 (46,937) | 18.4 (0.2) | 14.8 (0.1) | 19.9 (16,127) | 14.1 (0.2) | 13.9 (0.2) | - | 32.0 (128,374) | 31.5 (0.1) |
| 60–69 | 15.4 (0.2) | 21.4 (52,112) | 17.5 (0.2) | 17.1 (0.1) | 23.1 (18,801) | 13.1 (0.2) | 12.9 (0.2) | - | 46.3 (185,741) | 33.8 (0.1) |
| 70–79 | 9.4 (0.2) | 15.4 (37,532) | 10.8 (0.1) | 13.2 (0.1) | 17.4 (14,108) | 8.0 (0.1) | 8.0 (0.2) | - | 0.7 (2,874) | 0.6 (0.0) |
| 80+ | 4.9 (0.1) | 4.8 (11,721) | 2.8 (0.1) | 3.9 (0.1) | 6.3 (5,103) | 2.6 (0.1) | 2.8 (0.1) | - | 0.0 (0) | 0.0 (0.0) |
| Sex | ||||||||||
| Male | 48.3 (0.4) | 37.8 (92,315) | 45.5 (0.2) | 46.8 (0.2) | 46.2 (37,541) | 48.6 (0.3) | 49.9 (0.3) | 49.2 (0.1) | 44.7 (179,135) | 47.2 (0.1) |
| Female | 51.7 (0.4) | 62.2 (151,756) | 54.5 (0.2) | 53.2 (0.2) | 53.8 (43,702) | 51.4 (0.3) | 50.1 (0.3) | 50.8 (0.1) | 55.3 (222,032) | 52.8 (0.1) |
| Race/Ethnicity | ||||||||||
| Asian | 5.9 (0.3) | 2.7 (6,535) | 2.9 (0.1) | 3.0 (0.1) | 3.7 (2,983) | 11.4 (0.3) | 11.4 (0.3) | - | 2.2 (8,898) | 4.6 (0.1) |
| Black or African American | 11.8 (0.4) | 18.5 (45,243) | 14.6 (0.2) | 15.3 (0.1) | 6.1 (4,959) | 11.1 (0.2) | 11.6 (0.3) | - | 1.6 (6,385) | 2.8 (0.1) |
| Hispanic | 16.5 (0.6) | 18.9 (46,032) | 13.5 (0.1) | 13.4 (0.1) | 2.7 (2,179) | 6.4 (0.2) | 6.5 (0.2) | - | - | - |
| Other/Unknown | 2.6 (0.3) | 4.5 (11,096) | 3.2 (0.1) | 3.6 (0.1) | 4.4 (3,578) | 5.9 (0.2) | 6.1 (0.2) | - | 2.0 (8,094) | 1.7 (0.0) |
| White | 63.2 (0.8) | 55.4 (135,165) | 65.9 (0.2) | 64.7 (0.1) | 83.1 (67,544) | 65.1 (0.4) | 64.4 (0.4) | 87.0 (0.05) | 94.2 (377,790) | 90.9 (0.1) |
| BMI (continuous) | 27.8 (0.0) | 30.02 (7.67) | 29.9 (0.0) | 30.0 (0.0) | 29.94 (7.26) | 28.0 (0.1) | 28.1 (0.1) | 27.64 (4.90) | 27.9 (0.0) | |
| BMI | ||||||||||
| Underweight (<18.5) | 1.6 (0.1) | 1.3 (2,977) | 1.3 (0.1) | 1.1 (0.0) | 1.1 (865) | 1.9 (0.1) | 1.7 (0.1) | - | 0.5 (2,052) | 0.6 (0.0) |
| Healthy [18.5, 25) | 31.4 (0.3) | 25.6 (59,653) | 26.7 (0.2) | 26.1 (0.1) | 24.7 (20,025) | 35.9 (0.3) | 34.8 (0.3) | - | 31.0 (123,485) | 30.1 (0.1) |
| Overweight [25, 30) | 33.0 (0.3) | 30.6 (71,174) | 30.5 (0.2) | 30.9 (0.1) | 32.0 (25,954) | 33.9 (0.3) | 33.8 (0.3) | - | 42.5 (169,172) | 41.4 (0.1) |
| Obese [30+) | 31.2 (0.4) | 42.5 (99,024) | 41.5 (0.2) | 41.9 (0.2) | 42.3 (34,285) | 28.4 (0.3) | 29.7 (0.3) | - | 26.0 (103,779) | 27.9 (0.1) |
| Anxiety | 6.1 | 32.0 (78,018) | 30.5 | 23.7 | 35.3 (28,690) | 8.5 | 8.0 | - | 5.5 (22,047) | 6.3 |
| Cancere | 9.5 | 27.1 (66,124) | 22.9 | 24.5 | 49.2 (39,984) | 8.3 | 9.8 | - | 25.9 (103,996) | 23.7 |
| Depression | 7.0 | 25.7 (62,689) | 23.8 | 19.2 | 31.6 (25,634) | 7.5 | 7.3 | - | 0.2 (712) | 0.2 |
| Diabetes | 8.0 | 23.0 (56,210) | 18.7 | 42.6 | 23.1 (18,790) | 8.0 | 7.4 | - | 10.1 (40,353) | 11.4 |
| Hypertension | 31.7 | 47.6 (116,156) | 40.8 | 30.4 | 50.7 (41,182) | 29.9 | 29.9 | - | 33.7 (135,113) | 32.1 |
| EHR characteristics | ||||||||||
| Encounters per person (unique)f | ||||||||||
| Minimum | - | 1 | - | - | 1 | - | - | - | 1 | - |
| Median | - | 19 | - | - | 54 | - | - | - | 3 | - |
| Mean | - | 32 | - | - | 103 | - | - | - | 4 | - |
| Maximum | - | 487 | - | - | 1833 | - | - | - | 92 | - |
| Unique phecodes per personf | - | - | - | - | - | - | - | |||
| Minimum | - | 1 | - | - | 1 | - | - | - | 1 | - |
| Median | - | 52 | - | - | 55 | - | - | - | 12 | - |
| Mean | - | 72 | - | - | 77 | - | - | - | 17 | - |
| Maximum | - | 573 | - | - | 591 | - | - | - | 212 | - |
| Length of follow-up (years)f | - | - | - | - | - | - | - | |||
| Minimum | - | 0.0 | - | - | 0.0 | - | - | - | 0.0 | - |
| Median | - | 7.4 | - | - | 8.1 | - | - | - | 7.0 | - |
| Mean | - | 9.3 | - | - | 9.9 | - | - | - | 7.5 | - |
| Maximum | - | 41.5 | - | - | 43.5 | - | - | - | 22.0 | - |
Abbreviations: -, data unavailable in cohort; AOU, All of Us; IP, inverse probability; MGI, Michigan Genomics Initiative; NHIS, National Health Interview Survey; PS, poststratification; UKB, UK Biobank
2019 NHIS data
All of Us data restricted to those who are present in the phenome.
The UKB weighting procedure and definitions of the UKB-eligible population are described in van Alten and colleagues (doi: 10.1101/2022.05.16.22275048). Further, estimates presentated here of the UKB-eligible population are taken from van Alten and colleagues.
Age defined as age at last EHR encounter in AOU and MGI, age at consent in UKB
Presence of any qualifying cancer phecode in phenome
UK Biobank data only provided first occurrence of each unique phecode. This means that individuals could have encounters that were not reported or extend beyond their last encounter if no unique diagnoses were recorded at that visit.