Skip to main content

Table 8 Multiple multinomial logistic regression analysis predicting T4-easy-to-recruit, T4-hard-to-recruit, T4-non-responder, and drop-out since T2 or T3

From: Benefits of extensive recruitment effort persist during follow-ups and are consistent across age group and survey method. The TRAILS study

 

T4-easy-to-recruit (n = 1,359) (ref.)

T4-hard-to-recruit (n = 213)

T4-non-responder (n = 98)

Drop-out since T2 or T3 (n = 150)

 

n(%)

n(%)

OR

95% CI

n(%)

OR

95% CI

n(%)

OR

95% CI

T1 Recruitment Effort

Hard to recruit

42 (3.1)

8 (3.8)

0.98

0.44-2.17

13 (13.3)

2.51*

1.21-5.18

18 (12.0)

2.65*

1.42-4.96

(univariate effects)

  

1.22

0.57-2.64

 

4.80*

2.48-9.28

 

4.28*

2.39-7.64

T1 Socio-demographics

Girl

747 (55.0)

66 (31.0)

0.41*

0.30-0.56

40 (40.8)

0.64*

0.41-1.00

72 (48.0)

0.84

0.59-1.20

(univariate effects)

  

0.37*

0.27-0.50

 

0.57*

0.37-0.56

 

0.76

0.54-1.06

Non-western immigrant

87 (6.4)

23 (10.8)

1.66

1.00-2.76

23 (23.5)

3.11*

1.76-5.47

29 (19.3)

2.67*

1.62-4.39

(univariate effects)

  

1.77*

1.09-2.87

 

4.48*

2.68-7.51

 

3.50*

2.21-5.55

Parents divorced

248 (18.2)

33 (26.8)

1.45

0.99-2.12

33 (33.7)

1.47

0.88-2.47

32 (21.3)

0.81

0.51-1.31

(univariate effects)

  

1.64*

1.17-2.28

 

2.27*

1.46-3.54

 

1.22

0.80-1.84

Lower educ.mother (≤lower tracks of sec. educ.)

422 (31.1)

88 (41.3)

1.30

0.94-1.78

60 (61.2)

2.24*

1.42-3.54

88 (58.7)

2.52*

1.75-3.64

(univariate effects)

  

1.56*

1.16-2.10

 

3.51*

2.30-5.35

 

3.15*

2.23-4.45

Low family income (<€1,135)

157 (11.6)

40 (18.8)

1.23

0.78-1.92

34 (34.7)

1.88*

1.09-3.24

38 (25.3)

1.63*

1.01-2.64

(univariate effects)

  

1.77*

1.21-2.60

 

4.07*

2.60-6.37

 

2.60*

1.74-3.89

Low IQ (wisc < 85)

180 (13.2)

44 (20.7)

1.43

0.96-2.12

37 (37.8)

2.27*

1.40-3.67

39 (26.0)

1.43

0.93-2.20

(univariate effects)

  

1.71*

1.18-2.46

 

3.97*

2.57-6.15

 

2.30*

1.55-3.42

T1 Psycho-pathology

mean (SD)

mean (SD)

OR

95% CI

mean (SD)

OR

95% CI

mean (SD)

OR

95% CI

CBCL Int

0.25 (0.19)

0.24 (0.21)

0.81*

0.67-0.98

0.25 (0.20)

0.75*

0.58-0.97

0.26 (0.21)

0.96

0.78-1.18

(univariate effects)

  

0.97

0.84-1.12

 

1.00

0.81-1.23

 

1.09

0.92-1.28

CBCL Ext

0.22 (0.19)

0.29 (0.21)

1.38*

1.17-1.62

0.31 (0.27)

1.43*

1.14-1.79

0.29 (0.24)

1.29*

1.06-1.56

(univariate effects)

  

1.38*

1.21-1.57

 

1.48*

1.24-1.76

 

1.34*

1.15-1.57

YSR Int

0.37 (0.24)

0.33 (0.21)

0.88

0.75-1.03

0.36 (0.25)

0.88

0.71-1.10

0.33 (0.24)

0.77*

0.64-0.94

(univariate effects)

  

0.85*

0.73-0.99

 

0.96

0.78-1.17

 

0.85

0.71-1.02

  1. *p < .05; Odds ratios (OR) and 95% confidence intervals (CI) from multivariate multinomial logistic regression analysis; CBCL = Child Behavior Checklist (parent report); YRS = Youth Self Report; Int = Internalizing; Ext = Externalizing; CBCL and YSR scores standardized; univariate effects of single multinomial regression analyses italicized. Nagelkerke R² = 0.159.
Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us