Skip to main content
. 2014 Dec 10;14:350. doi: 10.1186/s12888-014-0350-4

Table 3.

Logistic regression models of the association between the extent of chronic pain and probable bipolar disorder, relative to probable major depressive disorder

Model OR (95% CI*) P value^ P value^^
1. No adjustments (n = 149,612) No chronic pain 1 <0.001
1 site 1.04 (0.92, 1.18) 0.520
2-3 sites 1.25 (1.10, 1.43) 0.001
4-7 sites 1.42 (1.13, 1.77) 0.002
Widespread pain 1.92 (1.44, 2.57) <0.001
2. Adjusted for sex, age, ethnicity, deprivation and employment status (n = 147,463) No chronic pain 1 <0.001
1 site 1.02 (0.90, 1.17) 0.724
2-3 sites 1.18 (1.03, 1.34) 0.018
4-7 sites 1.16 (0.92, 1.47) 0.210
Widespread pain 1.34 (0.99, 1.83) 0.061
3. Adjustments as for 2. plus BMI, smoking status and frequency of alcohol consumption (n = 145,945) No chronic pain 1 <0.001
1 site 1.01 (0.88, 1.15) 0.898
2-3 sites 1.15 (1.01, 1.32) 0.039
4-7 sites 1.11 (0.88, 1.41) 0.372
Widespread pain 1.32 (0.97, 1.81) 0.079
4. Adjustments as for 3. plus additional long term condition morbidity count (n = 145,518) No chronic pain 1 <0.001
1 site 1.01 (0.88, 1.15) 0.881
2-3 sites 1.16 (1.01, 1.33) 0.035
4-7 sites 1.12 (0.88, 1.43) 0.339
Widespread pain 1.34 (0.97, 1.83) 0.072

*Relative to the probable MDD group.

^P-value for each level of chronic pain.

^^P-value for overall model.

OR – odds ratio; 95% CI – 95% confidence interval; BMI – body mass index.