Table 2.
Variables selected in the final model and association with 6-year lung cancer risk: multivariate logistic regression analysis.
| OR with 95% CI- SAIL (development cohort), N = 574,196 | OR with 95% CI-UK Biobank (validation cohort), N = 137,918 | |
|---|---|---|
| Age-continuous | 1.07 (1.06–1.08) | 1.06 (1.04–1.09) |
| Sex—male | 1.09 (1.03–1.15) | 1.24 (1.04–1.48) |
| Socioeconomic status | ||
| SE Quintile 1 least deprived | Reference | Reference |
| SE Quintile 2 | 1.14 (1.03–1.25) | 1.16 (0.83–1.63) |
| SE Quintile 3 | 1.27 (1.17–1.40) | 1.24 (0.89–1.73) |
| SE Quintile 4 | 1.39 (1.27–1.51) | 1.94 (1.44–2.65) |
| SE Quintile 5 most deprived | 1.48 (1.36–1.62) | 2.51 (1.88–3.39) |
| Family history of lung cancer | 1.84 (1.38–2.38) | 1.66 (1.35–2.01) |
| Smoking status | ||
| Never smoked | Reference | Reference |
| Previous smoker | 3.86 (3.49–4.28) | 1.52 (0.38–6.06) |
| Current smoker | 9.94 (9.08–10.89) | 9.17 (2.66–31.40) |
| Body mass index (BMI)-continuous | 1.02 (1.01–1.03) | |
| Presence of LTCs | ||
| Alcohol misuse | 1.31 (1.14–1.48) | 1.44 (0.84–2.35) |
| Coronary heart disease | 1.12 (1.02–1.21) | 0.97 (0.69–1.32) |
| Chronic obstructive pulmonary disease | 2.01 (1.89–2.13) | 2.53 (1.96–3.23) |
| History of previous cancer | 1.37 (1.26–1.47) | 1.40 (1.07–1.81) |
| Peripheral vascular disease | 1.27 (1.16–1.38) | 1.71 (1.23–2.32) |
| Dementia | 0.40 (0.22–0.65) | No output* |
| Painful condition | 1.14 (1.07–1.21) | 1.34 (1.09–1.63) |
| Hypertension | 0.91 (0.85–0.95) | 1.25 (1.04–1.50) |
| Stroke/transient Ischaemic attack | 1.18 (1.06–1.30) | 1.41 (0.94–2.04) |
| History of previous pneumonia | 1.38 (0.73–1.23) | 0.71 (0.30–1.39) |
| BMI: Previous smoker interaction | 0.99 (0.98–0.99) | 0.98 (0.94–1.03) |
| BMI: Current smoker interaction | 0.98 (0.97–0.99) | 0.96 (0.92–1.01) |
SAIL secure anonymised information linkage, OR odds ratio, CI confidence intervals.
WIMD and Townsend score were area-based measures of socioeconomic status in SAIL and UK Biobank, respectively. *No lung cancer cases were recorded in participants with dementia in the UK Biobank cohort.