Skip to main content

Table 3 Test of effect modification: differences in the trajectory of daily step counts before, during, and after the lockdown (n = 815)

From: How COVID-19 lockdown and reopening affected daily steps: evidence based on 164,630 person-days of prospectively collected data from Shanghai, China

 

During lockdown

After lockdown

β (SE)a

p-value

β (SE) a

p-value

Age (40+ vs 20–39 years)

−20.46 (5.02)

<.001

−11.07 (4.61)

0.016

Sex (female vs male)

−2.75 (5.29)

0.602

0.31 (4.86)

0.947

Martial status (married vs not married)

−22.62 (6.90)

0.001

−15.89 (6.33)

0.012

Education (university vs no university)

−23.80 (5.24)

<.001

11.88 (4.81)

0.013

Baseline physical activity (insufficient vs sufficient)

−27.52 (5.33)

<.001

−6.01 (4.89)

0.219

BMI (overweight/obese vs normal weight)

−8.81 (5.03)

0.079

−9.05 (4.62)

0.050

Body composition (at risk vs not at risk)

−27.82 (5.35)

<.001

−17.95 (4.91)

<.001

Intervention group allocation (control vs intervention)

−24.94 (5.17)

< 0.001

−3.69 (4.75)

0.437

  1. aInterpreted as the differences in the change of daily step counts. For example: those aged 40+ years increased step counts by 20.46 fewer steps than those aged 20–39 years during the lockdown period
  2. bTest of effect modification was conducted in a linear mixed model with participants as a cluster variable, period (i.e., pre-lockdown, lockdown, and post-lockdown) and day as independent variables and step count as the dependent variable. Effect modification was tested one at a time adjusted for age, sex, education, income, marital status, worksite, intervention group allocation, baseline physical activity and baseline adiposity