Skip to main content
. 2020 Feb 27;17(5):1539. doi: 10.3390/ijerph17051539

Table 5.

Multinomial logistic regression, with an odds ratio (OR) and confidence interval (CI) of 95%, of the association of psychosocial stress at work, measured by models DC-SS, ERI-OC, and DC-ERI, and adjusted glycemic alterations by selected variables, for active men in the second wave of ELSA-Brazil 2008–2014.

Men Prediabetes Maintenance Incident Prediabetes Incident Diabetes
Crude OR
(95% CI)
Adj OR A
(95% CI)
Crude OR
(95% CI)
Adj OR A
(95% CI)
Crude OR
(95% CI)
Adj OR A
(95% CI)
Model 1
DC and low SS absent 1.0 1.0 1.0 1.0 1.0 1.0
DC present 1.61
(1.06–2.43)
1.44
(0.94–2.19)
1.13
(0.79–1.6)
1.04
(0.73–1.48)
0.99
(0.49–2.02)
0.89
(0.44–1.83)
Low SS present 0.98
(0.65–1.48)
1.11
(0.73–1.69)
0.8
(0.57–1.11)
0.88
(0.63–1.23)
1.26
(0.71–2.26)
1.42
(0.79–2.56)
DC and low SS present 0.98
(0.65–1.48)
1.06
(0.7–1.61)
1.02
(0.75–1.4)
1.07
(0.78–1.47)
0.98
(0.53–1.82)
1.01
(0.54–1.88)
p schooling interaction = 0.3236
Model 2
ERI and OC absent 1.0 1.0 1.0 1.0 1.0 1.0
ERI present 0.97
(0.63–1.49)
1.06
(0.69–1.63)
0.95
(0.67–1.35)
0.98
(0.68–1.4)
1.51
(0.84–2.73)
1.59
(0.88–2.87)
OC present 0.99
(0.62–1.57)
0.99
(0.62–1.59)
0.90(0.6–1.33) 0.92
(0.62–1.38)
1.18
(0.58–2.37)
1.21
(0.60–2.47)
ERI and OC present 0.58
(0.37–0.93)
0.65
(0.40–1.05)
0.90(0.65–1.25) 0.98
(0.7–1.37)
0.8
(0.41–1.57)
0.86
(0.43–1.71)
p schooling interaction = 0.2559
Model 3
DC and ERI absent 1.0 1.0 1.0 1.0 1.0 1.0
DC present 1.45
(1.02–2.06)
1.35
(0.94–1.94)
1.17
(0.87–1.57)
1.09
(0.81–1.48)
0.91
(0.5–1.63)
0.82
(0.45–1.49)
ERI present 0.88
(0.53–1.44)
0.99
(0.6–1.64)
0.86
(0.58–1.28)
0.91
(0.61–1.37)
1.2
(0.62–2.29)
1.28
(0.67–2.48)
DC and ERI present 0.87
(0.56–1.36)
0.90(0.58–1.41) 1.09
(0.79–1.51)
1.10(0.79–1.53) 0.93
(0.49–1.74)
0.92
(0.49–1.75)
p schooling interaction = 0.5120

DC: demand–control model. SS: social support. ERI: effort–reward imbalance model. OC: overcommitment. 95% CI: 95% confidence interval. OR: odds ratio. Adjusted A crude model + adjusted by total effect variables (age—weekly workload—work shift—schooling level).