Multivariate analyses for DVT, PE and combined deep venous events
| DVT OR (p value) | PE OR (p value) | DVE (DVT and/or PE) OR (p value) | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Age–sex adjusted* | Fully adjusted | Final model | Age–sex adjusted* | Fully adjusted | Final model | Age–sex adjusted* | Fully adjusted | Final model | |
| Age | 1.02 (0.047) | 1.05 (0.162) | – | 1.02 (0.404) | 0.977 (0.621) | – | 1.03 (0.004) | 1.04 (0.202) | – |
| Male sex | 1.01 (0.955) | 0.479 (0.312) | – | 1.41 (0.435) | 0.778 (0.795) | – | 1.15 (0.548) | 0.773 (0.662) | – |
| Caucasian ethnicity | 2.35 (0.003) | 0.754 (0.674) | – | 2.68 (0.083) | 5.77 (0.149) | 2.58 (0.096)† | 2.26 (0.003) | 0.936 (0.913) | – |
| Activity level | 0.827 (0.059) | 0.819 (0.418) | – | 0.771 (0.163) | 0.986 (0.971) | – | 0.846 (0.081) | 0.909 (0.672) | – |
| Heavy smoking (≥40 cigarettes per day during time smoked) | 4.19 (<0.001) | 10.1 (0.007) | 16.6 (<0.001) | 3.50 (0.052) | 5.21 (0.191) | 2.73 (0.121)† | 3.38 (<0.001) | 6.92 (0.013) | 12.5 (<0.001) |
| DBP (per 20 mm Hg) | 1.28 (0.227) | 0.442 (0.445) | – | 2.06 (0.014) | 3.20 (0.419) | 2.29 (0.010)§ | 1.39 (0.077) | 0.764 (0.766) | – |
| SBP (per 20 mm Hg) | 0.996 (0.975) | –‡ | – | 1.59 (0.012) | –‡ | –§ | 1.05 (0.643) | –‡ | – |
| Drinkers who did not specify their maximum alcohol consumption per day | 1.74 (0.170) | 1.46 (0.739) | – | 1.63 (0.534) | 2.70 (0.447) | – | 1.61 (0.235) | 1.09 (0.939) | – |
| History of immobility (>1 day) | 2.17 (0.001) | 5.32 (0.008) | 4.30 (0.006) | 4.44 (0.002) | 4.21 (0.114) | 4.07 (0.004) | 2.23 (<0.001) | 3.80 (0.013) | 3.48 (0.008) |
| Family history of first-degree relatives with superficial venous events | 3.32 (0.024) | 0.719 (0.643) | – | 1.51 (0.600) | 0.730 (0.754) | – | 2.52 (0.036) | 0.626 (0.449) | – |
| Family history of first-degree relatives with DVE | 4.41 (0.003) | 3.51 (0.066) | 3.28 (0.020) | 0.855 (0.889) | 0.431 (0.491) | – | 3.42 (0.010) | 2.85 (0.100) | 2.49 (0.057) |
| Oestrogen use duration for HRT¶ (per 10 years) | – | – | – | 2.73 (0.000) | 3.73 (0.000) | 2.63 (0.001) | |||
| Oestrogen use duration among oestrogen users for HRT (per 10 years) | – | – | – | – | – | – | 4.67 (0.018) | 5.22 (0.018) | 4.74 (0.016) |
Caucasian ethnicity shows significant relationship to DVT if not adjusted for family history of DVE (OR=2.15, p=0.008). If both Caucasian ethnicity and family history of DVE are included, Caucasian ethnicity loses significance (OR 1.29, p=0.659) but family history of deep events retains significance (OR 3.17, p=0.026).
Age (OR=1.02, p=0.035) and Caucasian ethnicity (OR=1.97, p=0.015) show a significant relationship to combined deep events in the final model if family history of venous events is excluded.
For combined DVT and/or PE, oestrogen use for HRT was tested as a predictor in women but was not significant. For oestrogen use duration (per 10 years), ORs (p values) for age-adjusted and final models were 0.532 (0.534) and 0.523 (0.611).
*Age variable was adjusted for sex only; and sex variable was adjusted for age only.
†Although some variables lose significance in the fully adjusted model due to collinearity, variables which were of significance or borderline significance in unadjusted or age/sex adjusted models were added back to the final model. When Caucasian ethnicity and heavy smoking were included without one another in the final model, they resumed significance or borderline significance.
‡Instead of including both SBP and DBP, we used DBP and the residual of regression of SBP on DBP, so as not to double-count the shared (collinear) contribution by the SBP and DBP values.
§In the final model, either one of the blood pressure measures—SBP or DBP—could be included and remained statistically significant (only one of the two blood pressure variables was included because of collinearity).
¶Female-specific analysis. Excludes family history of first deep events (see comment for PE). The final female-specific model included age and history immobilisation, the sole other variables with p < 0.2 (neither <0.1).
DBP, diastolic blood pressure; DVE, deep venous event; DVT, deep venous thrombosis; HRT postmenopausal hormone replacement therapy; OR, odds ratio; PE, pulmonary embolism; SBP, systolic blood pressure.