Table 2

Role of material, psychosocial and behavioural factors and childhood socioeconomic conditionsa in explaining associations of adulthood SEPb with CVD mortality (adjusted for age), among men (n=5395)
Adulthood SEP
1- low (n=1270) 2- middle low (n=1795) 3- middle high (n=1088) 4- high (n=1242)
Explanatory modelsc HR (95% CI) %attentuationd(95% CI) HR (95% CI) %attentuationd(95% CI) HR (95% CI) %attentuationd(95% CI) HR (95% CI)
Died from CVD (n (%)) 184 (14.5) 156 (8.7) 97 (8.9) 80 (6.4)
Model 1: 1.84 (1.41-2.39) 1.32 (1.01-1.73) 1.31 (0.97-1.76) 1.00
adulthood SEP
Model 2: 1.71 (1.29-2.26) −15% (−40 to 5) 1.26 (0.95-1.66) −19% (−120 to 14) 1.27 (0.94-1.71) −13% (−112 to 29) 1.00
adulthood SEP
+ childhood conditions
Model 3: 1.49 (1.12-1.98) −42% (−73 to −20) 1.16 (0.88-1.54) −50% (−252 to −7) 1.22 (0.91-1.65) −29% (−259 to 87) 1.00
adulthood SEP
+ material factorse
Model 4: 1.70 (1.30-2.21) −17% (−39 to −12) 1.28 (0.98-1.68) −13% (−92 to 9) 1.33 (0.99-1.79) - 1.00
adulthood SEP
+behavioural factorsf
Model 5: 1.76 (1.34-2.29) −10% (−21 to −1) 1.32 (1.00-1.72) - 1.32 (0.99-1.78) - 1.00
adulthood SEP
+psychosocial factorsg
Model 6: 1.40 (1.05-1.86) −52% (−94 to −33) 1.17 (0.88-1.54) −47% (−249 to −1) 1.26 (0.94-1.71) −16% (−137 to 46) 1.00
adulthood SEP
+ material + behavioural
+ psychosocial factors
Model 7: 1.34 (0.99-1.82) −60% (−109 to −32) 1.13 (0.85-1.51) −59% (−291 to −2) 1.24 (0.92-1.68) −23% (−202 to 64) 1.00
adulthood SEP
+ childhood conditions
+ material + behavioural
+ psychosocial factors
Direct contribution of childhood conditions (i.e. independent of adulthood risk factors) 8% (60–52) 12% (59–47) 7% (23–16)
Direct contribution of adulthood risk factors (i.e. independent of childhood conditions) 45% (60–15) 40% (59–19) 10% (23–13)
Indirect contribution of childhood conditions, i.e. via adulthood risk factors 7% (15+52-60) 7% (19+47-59) 6% (13+16-23)

SEP, socioeconomic position; CVD, cardiovascular diseases; HR, hazard ratio; CI, confidence interval; BMI, body mass index (kg/m2).

a Childhood socioeconomic conditions were determined by the occupation of the respondent’s father when the respondent was 12 years of age, with 1=professional, 2=white collar, 3=blue collar.

b Adulthood socioeconomic position was determined by the respondent’s highest attained educational level, with 1= primary, 2= lower secondary, 3= higher secondary, 4=tertiary.

c Only factors that were significantly associated with CVD mortality and unequally distributed across adulthood SEP groups were included in the explanatory models, and all models were adjusted for age.

d The percentages show the reduction in harzard ratio (HR) compared to model 1, per SEP group. For instance, the reduction in the OR for the lowest adulthood SEP group when adding childhood socioeconomic conditions to the first model, is [(1.84-1.71)/(1.84-1.00)]*100 = 15%.

e Material factors: car ownership, housing tenure, and financial problems.

f Behavioural factors: smoking, physical activity.

g Psychosocial factors: marital status, use of sleep/anxiety drugs.

Kamphuis et al.

Kamphuis et al. BMC Public Health 2012 12:1045   doi:10.1186/1471-2458-12-1045

Open Data