Table 4.
Independent associations of change in MVPA (models fully adjusted using all the variables listed)
| Sample active (≥30 min/week) | Sample inactive (<30 min/week) | |||||
| Before COVID-19 (weighted n=1857) | Before COVID-19 (weighted n=585) | |||||
| Decrease (n=868) versus not | Increase (n=174) versus not | |||||
| aOR | 95% CI | P value | aOR | 95% CI | P value | |
| Female | 1.22 | 1 to 1.49 | 0.047 | 1.56 | 1.02 to 2.38 | 0.04 |
| Age (in decades) | 0.91 | 0.85 to 0.98 | 0.015 | 0.72 | 0.61 to 0.84 | <0.001 |
| White ethnicity | 0.62 | 0.44 to 0.86 | 0.005 | 1.01 | 0.44 to 2.33 | 0.982 |
| Household income ≥50 000 GBP | ||||||
| Income:<50 000 GBP | 1.3 | 1 to 1.67 | 0.048 | 0.49 | 0.26 to 0.93 | 0.03 |
| Income: prefer not to say | 1.22 | 0.81 to 1.85 | 0.335 | 0.36 | 0.14 to 0.93 | 0.034 |
| Education: post-16 years of age or higher | 0.73 | 0.58 to 0.91 | 0.006 | 2.26 | 1.4 to 3.63 | 0.001 |
| Employed | 0.86 | 0.69 to 1.07 | 0.18 | 0.51 | 0.31 to 0.85 | 0.009 |
| Laid-off/furloughed | 1.05 | 0.76 to 1.47 | 0.75 | 1.3 | 0.72 to 2.37 | 0.385 |
| Condition limiting PA | 1.74 | 1.27 to 2.39 | 0.001 | 0.49 | 0.27 to 0.89 | 0.019 |
| BMI ≤24 kg/m2 | ||||||
| BMI: 25–29.99 kg/m2 | 1.03 | 0.82 to 1.28 | 0.816 | 1.54 | 0.94 to 2.53 | 0.088 |
| BMI: obese: 30+ kg/m2 | 0.87 | 0.66 to 1.14 | 0.317 | 0.74 | 0.42 to 1.31 | 0.307 |
| Total isolation | 3.81 | 2.33 to 6.23 | <0.001 | 0.29 | 0.12 to 0.73 | 0.008 |
| Minor/no COVID-19 risk percept | 0.99 | 0.8 to 1.23 | 0.926 | 1.2 | 0.76 to 1.92 | 0.436 |
| Living with children | 0.84 | 0.65 to 1.09 | 0.187 | 2.32 | 1.27 to 4.24 | 0.006 |
| Living with vulnerable | 0.78 | 0.59 to 1.02 | 0.074 | 0.9 | 0.51 to 1.59 | 0.716 |
| Access garden/balcony to exercise comfortably in | 0.75 | 0.6 to 0.94 | 0.011 | 0.72 | 0.46 to 1.13 | 0.157 |
| Access green space within walking distance | 0.82 | 0.67 to 1.01 | 0.059 | 1.1 | 0.72 to 1.67 | 0.665 |
| Smoker | 1.09 | 0.85 to 1.4 | 0.504 | 1.15 | 0.73 to 1.8 | 0.551 |
| Weekly alcohol drinking, none (reference) | ||||||
| Weekly or less | 0.78 | 0.59 to 1.04 | 0.087 | 0.77 | 0.43 to 1.38 | 0.378 |
| More than weekly | 0.78 | 0.6 to 1.02 | 0.068 | 0.88 | 0.52 to 1.5 | 0.644 |
| Deteriorated psychological well-being | 1.4 | 1.15 to 1.71 | 0.001 | 1.09 | 0.72 to 1.66 | 0.682 |
All models were also adjusted for the time of enrolment and time when COVID-19 started to affect individuals in any way. Following the Benjamini-Hochberg false discovery rate (BH FDR) correction, the highest p value that met the threshold for significance was p=0.019 (the significant results are presented in bold).
Segmented analyses of active sample (before COVID-19; MVPA activity ≥30 min; predicting decrease by ≥20 min) and inactive sample (before COVID-19 MVPA; activity <30 min; predicting increase by ≥20 min).
aOR, adjusted OR; BMI, body mass index; MVPA, moderate-to-vigorous aerobic physical activity.