Table 4 MAIHDA models for obesity status.

From: Multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA) to understand how obesity risk varies according to multiple lifestyle behavior recommendations

 

Females

Males

 

Null model

Main effects model

Main effects model+

Null model

Main effects model

Main effects model+

 

OR (95% CI)

OR (95% CI)

OR (95% CI)

OR (95% CI)

OR (95% CI)

OR (95% CI)

Intercept

0.51 (0.44, 0.58)

1.12 (1.03, 1.21)

1.26 (1.12, 1.42)

1.11 (0.96, 1.28)

3.13 (2.78, 3.54)

3.90 (3.37, 4.52)

Smoking

 Previous

 

–

–

 

–

–

 Current

 

0.77 (0.71, 0.84)

0.76 (0.70, 0.83)

 

0.53 (0.47, 0.60)

0.53 (0.47, 0.60)

 Never

 

0.76 (0.72, 0.81)

0.81 (0.76, 0.86)

 

0.50 (0.45, 0.55)

0.51 (0.46, 0.57)

Fruit & vegetable

 Not meets

 

–

–

 

–

–

 Meets

 

0.80 (0.76, 0.85)

0.77 (0.73, 0.82)

 

1.02 (0.92, 1.12)

1.01 (0.92, 1.11)

Physical activity

 Not meets

 

–

–

 

–

–

 Meets

 

0.47 (0.44, 0.50)

0.45 (0.43, 0.48)

 

0.53 (0.48, 0.59)

0.53 (0.48, 0.58)

Alcohol

 Not meets

 

–

–

 

–

–

 Meets

 

1.09 (1.02, 1.15)

1.08 (1.02, 1.14)

 

0.77 (0.70, 0.84)

0.77 (0.70, 0.84)

Sleep

 Not meets

 

–

–

 

–

–

 Meets

 

0.71 (0.67, 0.75)

0.74 (0.69, 0.78)

 

0.69 (0.63, 0.76)

0.70 (0.63, 0.76)

Level 2 variance

0.19 (0.11, 0.27)

0.004 (0.0008, 0.007)

0.004 (0.0006, 0.007)

0.25 (0.14, 0.36)

0.01 (0.007, 0.02)

0.01 (0.007, 0.02)

Level 1 variance

–

–

–

–

–

–

 VPC

5.63%

0.13%

0.12%

7.12%

0.52%

0.52%

 PCV

–

97.77%

97.93%

–

93.08%

93.15%

  1. aConfounders were Age, ethnicity, tdi and employment status.
  2. bAll estimated PCVs are relative to the null model.
Back to article page