Baseline characteristics of UK Biobank participants
| Self-reported PA | Accelerometer-derived PA | Overlapping PA | |||||||
| No HF | HF | P value | No HF | HF | P value | No HF | HF | P value | |
| n | 387 580 | 1600 | 96 105 | 596 | 80 121 | 190 | <0.001 | ||
| Age (years) | 57 (50–63) | 63 (59–67) | <0.001 | 63 (56–69) | 66 (65–72) | <0.001 | 57 (50–62) | 62 (58–65) | <0.001 |
| Female | 202 169 (52%) | 324 (20%) | <0.001 | 54 293 (57%) | 153 (26%) | 43 727 (55%) | 28 (15%) | ||
| Male | 185 411 (48%) | 1276 (80%) | 41 812 (44%) | 443 (74%) | 36 394 (46%) | 162 (85%) | |||
| Ethnic background | 0.75 | 0.74 | 0.40 | ||||||
| White | 367 547 (95%) | 1526 (95%) | 92 803 (97%) | 579 (98%) | 77 490 (97%) | 186 (98%) | |||
| Mixed or non-white | 18 915 (5%) | 67 (4%) | 2968 (3%) | 15 (2%) | 2413 (3%) | 3 (2%) | |||
| Missing | 1118 (<1%) | 7 (<1%) | 334 (<1%) | 2 (<1%) | 218 (<1%) | 1 (<1%) | |||
| BMI (kg/m2) | <0.001 | <0.001 | <0.001 | ||||||
| <24.9 | 127 352 (33%) | 259 (16%) | 36 815 (38%) | 120 (20%) | 31 033 (39%) | 43 (23%) | |||
| ≥25 and <30 | 165 702 (43%) | 617 (39%) | 39 488 (41%) | 247 (42%) | 33 073 (41%) | 74 (39%) | |||
| ≥30 and <35 | 64 782 (17%) | 466 (29%) | 13 736 (14%) | 144 (25%) | 11 209 (14%) | 50 (26%) | |||
| ≥35 | 24 159 (6%) | 226 (14%) | 4810 (5%) | 77 (13%) | 3783 (5%) | 19 (10%) | |||
| Missing | 5585 (2%) | 32 (2%) | 1256 (1%) | 8 (1%) | 1023 (1%) | 4 (2%) | |||
| Alcohol status | <0.001 | 0.003 | <0.004 | ||||||
| Never | 14 891 (4%) | 94 (6%) | 2790 (3%) | 27 (5%) | 2161 (3%) | 6 (3%) | |||
| Previous | 12 987 (3%) | 156 (10%) | 2620 (3%) | 48 (8%) | 2118 (3%) | 15 (8%) | |||
| Current | 359 513 (93%) | 1349 (84%) | 90 614 (94%) | 521 (87%) | 75 825 (95%) | 169 (89%) | |||
| Missing | 189 (<1%) | 1 (<1%) | 81 (<1%) | 0 | 17 (<1%) | 0 | |||
| Smoking status | <0.001 | 0.002 | <0.002 | ||||||
| Never | 212 488 (55%) | 565 (35%) | 54 839 (57%) | 237 (40%) | 45 656 (57%) | 77 (41%) | |||
| Previous | 134 923 (35%) | 842 (53%) | 34 477 (36%) | 307 (52%) | 28 815 (36%) | 101 (53%) | |||
| Current | 39 197 (10%) | 189 (12%) | 6633 (7%) | 50 (8%) | 5526 (7%) | 12 (6%) | |||
| Missing | 972 (<1%) | 0 | 156 (<1%) | 2 (<1%) | 124 (<1%) | 0 | |||
| Socioeconomic status | 0.06 | 0.87 | 0.13 | ||||||
| Q1 | 77 446 (20%) | 213 (13%) | 21 154 (22%) | 110 (18%) | 17 765 (22%) | 27 (14%) | |||
| Q2 | 77 444 (20%) | 272 (17%) | 20 338 (21%) | 130 (22%) | 16 987 (21%) | 51 (27%) | |||
| Q3 | 77 451 (20%) | 285 (18%) | 19 666 (21%) | 121 (20%) | 16 380 (20%) | 33 (17%) | |||
| Q4 | 77 404 (20%) | 330 (21%) | 18 969 (20%) | 117 (20%) | 15 841 (20%) | 50 (26%) | |||
| Q5 | 77 250 (20%) | 497 (31%) | 15 868 (17%) | 118 (20%) | 13 047 (16%) | 29 (15%) | |||
| Missing | 585 (<1%) | 0 | 110 (<1%) | 0 | 101 (<1%) | 0 | |||
Quantile borders for the Townsend score were –6.26, –3.97, –2.84, –1.44, 1.11, 11.00 based on the participants with self-reported PA data.
Participants with poor wear quality and missing calibration coefficients are not included in the table. P values were calculated using the Kolmogorov-Smirnov test for continuous variables and the χ2 test for categorical variables. The overlapping group refers to individuals who were included in the self-reported analysis and also had sufficient-quality accelerometer recordings.
BMI, Body Mass Index; HF, heart failure; PA, physical activity.