Skip to main content
. 2009 Jul 15;52(9):1755–1765. doi: 10.1007/s00125-009-1453-1

Table 4.

Site-specific cancers by insulin glargine group in the fixed cohort

Site Data presented as Non-glargine insulin Non-glargine plus glargine insulin Insulin glargine only
Breast cancer n (%) 81 (0.5) 5 (0.3) 6 (2.6)
 Model 1a (n = 17,151) HR (95% CI) Reference 0.87 (0.34–2.17) p = 0.76 3.39 (1.46–7.85) p = 0.004
 Model 2b (n = 17,151) HR (95% CI) Reference 0.85 (0.34–2.14) p = 0.73 3.92 (1.58–9.70) p = 0.003
Model 3c (n = 13,635) HR (95% CI) Reference 1.07 (0.42–2.73) p = 0.88 5.04 (1.95–13.03) p = 0.001
Model 4c (n = 10,307) HR (95% CI) Reference 1.10 (0.38–3.16) p = 0.86 3.65 (1.05–12.68) p = 0.042
Prostate n (%) 44 (0.26) 3 (0.17) 1 (0.47)
 Model 1a (n = 19,103) HR (95% CI) Reference 1.76 (0.54–5.74) p = 0.35 1.16 (0.16–8.50) p = 0.88
Colorectal 104 (0.32) 2 (0.06) 3 (0.67)
 Model 1a (n = 36,254) HR (95% CI) Reference 0.36 (0.09–1.45) p = 0.15 1.43 (0.45–4.57) p = 0.54
Lung n (%) 140 (0.43) 5 (0.14) 4 (0.89)
 Model 1a (n = 36,254) HR (95% CI) Reference 0.47 (0.18–1.25) p = 0.13 1.43 (0.53–3.88) p = 0.49
Pancreatic n (%) 38 (0.12) 1 (0.03) 0
 Model 1a (n = 36254) HR (95% CI) Reference 0.54 (0.08–4.32) p = 0.60 No events

aModel 1 adjusts for prior cancer, type of diabetes and calendar year, and is stratified by sex ; timescale is age

bModel 2 further adjusts for metformin, sulfonylurea and other oral hypoglycaemic drugs at baseline

cModel 3 further adjusts for diabetes duration, HbA1c, diastolic BP, systolic BP and deprivation quintile

dModel 4 further adjusts for smoking ever and BMI, but note the reduction in available sample size due to missing covariates