Skip to main content

Table 4 Multivariable model: Difference between men and women in exposure to determinants of physical functioning

From: Exploring the difference between men and women in physical functioning: How do sociodemographic, lifestyle- and health-related determinants contribute?

Variable

Beta sex (women)a

95% CI

 

% Change Beta sexb

Full model (includes all variables)

-3.28

(-4.13

-2.43)

Reference

Socio-demographic determinants

 - Education (low/medium vs high)

-3.67

(-4.52

-2.82)

- 11.9%

 - Work status (no vs yes)

-3.72

(-4.57

-2.87)

- 13.5%

 - Living situation (alone vs together)

-3.30

(-4.15

-2.46)

- 0.6%

Lifestyle-related determinants

 - Alcohol consumption (> 1 glass/day vs ≤ 1 glass/day)

-3.45

(-4.28

-2.62)

- 5.2%

 - Current smoking status (yes vs no)

-3.28

(-4.14

-2.43)

- 0.1%

 - Physical activity (h/week)

-2.97

(-3.80

-2.14)

 + 9.5%

 - Sleep hours (long/short vs normal)

-3.26

(-4.11

-2.40)

 + 0.8%

 - BMI (kg/m2)

-2.81

(-3.68

-1.94)

 + 14.2%

Health-related determinants

 - Chronic diseases (one/two/more vs none)

-3.26

(-4.14

-2.39)

 + 0.5%

 - Pain (mild/medium to severe vs none)

-5.41

(-6.41

-4.42)

- 65.0% c

 - Mental health (poor vs good)

-3.41

(-4.26

-2.55)

- 3.9%

  1. Abbreviations: 95%CI = 95% Confidence Interval, BMI = Body Mass Index
  2. Note: Each row represents the full model including all variables, age, age2 and sex excluding the indicated variable (-)
  3. Note: Bold is significant (percentage change > 5%)
  4. aRepresents the sex difference (women versus men) in physical functioning
  5. bPercentage change = 1—(Beta “sex” full model minus indicated variable / Beta “sex” full model), interpretation: A negative percentage change means this factor (partly) explains the lower physical functioning score among women compared to men (i.e. the sex difference decreases when adjusted for this determinant) and a positive percentage change suggest that this factor supresses the lower physical functioning score among women compared to men because after the adjustment the sex difference increases
  6. cTo illustrate: A larger sex difference (-5.41) was observed in the multivariable model including the risk factor pain compared to the model without pain (-3.28). So, adjusting for pain decreased the lower physical functioning score among women compared to men by (2.13/3.28) 65.0%