Table 6 Coefficient estimates for interactions between household composition and young adulthood as covariates with trajectory membership.

From: Understanding the interplay between social isolation, age, and loneliness during the COVID-19 pandemic

Variable

Trajectory group (ref group = 3)

Estimate

SE

p

OR

95% CI

Constant

1

− 1.38

0.25

< .001

0.25

(0.15–0.41)

2

− 0.55

0.13

< .001

0.58

(0.45–0.75)

4

− 0.62

0.23

.008

0.54

(0.34–0.85)

5

− 2.39

0.31

< .001

0.09

(0.05–0.17)

Young adult

1

− 0.56

0.44

.201

0.57

(0.24–1.35)

2

− 0.92

0.44

.035

0.40

(0.17–0.94)

4

0.65

0.30

.029

1.91

(1.07–3.42)

5

0.91

0.41

.027

2.49

(1.11–5.59)

Living alone

1

− 0.51

0.32

.113

0.60

(0.32–1.13)

2

− 0.02

0.22

.937

0.98

(0.63–1.52)

4

0.48

0.22

.027

1.62

(1.06–2.50)

5

1.38

0.33

< .001

3.99

(2.11–7.56)

Young × living alone

1

− 9.98

11.36

.380

0.00

(0.00–2.17e5)

2

1.02

0.92

.265

2.78

(0.46–16.70)

4

0.12

0.63

.849

1.13

(0.33–3.91)

5

− 2.04

1.22

.096

0.13

(0.01–1.43)

  1. Trajectory groups: 1 = Very low loneliness and stable, 2 = Low loneliness and unstable, 3. = Moderate loneliness and unstable (reference group), 4 = High loneliness and stable, 5 = Very high loneliness and decreasing.
  2. Given the large error associated with the parameters for the interaction term in relation to the first trajectory, the resulting confidence interval is extremely wide.
  3. SE standard error, p p value, OR odds ratio, CI confidence interval.
Back to article page