Table 2.
Main characteristics of cohort studies of insulin use and cancer risk
| Source | Country | Average follow-up period (year) | Gender | age at enrollment (year) | Study population | Cancer site: number of cases | RR (95 % CI) | Controlled variables | |
|---|---|---|---|---|---|---|---|---|---|
| Yang et al. [6] | UK | 5.6 | M/F | 74.9 | UK nested case–control study: 24,918 patients with medical recorded type 2 DM (mean age, 61) | Colorectal: 125 | 2.1 (1.2, 3.4) | Age, gender, smoking | |
| Exposed group: 3,160 | |||||||||
| Comparison group: 21,758 men and women without insulin use | |||||||||
| Yang et al. [10] | Hong Kong | 3.0 | M/F | 35–49 | The Hong Kong Diabetes registry: 4,623patients with type 2 DM | All cancers: 152 | 0.17 (0.09, 0.32) | Age, gender, smoking, alcohol, duration of diabetes, systolic BP, LDL, HDL, triglycerides, use of anti-hypertensive drugs | |
| Exposed group: 971 using insulin | Digestive system: 119 | 0.19 (0.08, 0.46) | |||||||
| Comparison group: 1,935 not using insulin | Other cancers: 87 | 0.20 (0.10, 0.41) | |||||||
| Bowker et al. [11] | Canada | 5.4 | M/F | 63.4 | Population-based retrospective cohort study: 10,309 new users of metformin or sulfonylurea | Cancer- related mortality: 84 | 1.90 (1.50, 2.40) | Age, gender, metformin use, sulfonylurea use, chronic disease score | |
| Exposed group: 1,443 | |||||||||
| Comparison group: 8,866 men and women not insulin user | |||||||||
| Hemkens et al. [1] | Germany | 1.63 | M/F | ≥18 | The German statutory health insurance fund: 127,031 insulin-treated diabetic patients | All cancers: 5,009 | 1.19 (1.09, 1.29) | Age, gender, BMI | |
| Comparison group | |||||||||
| Human insulin: 95,804 | |||||||||
| Aspart: 4,103 | |||||||||
| Lispro: 3,269 | |||||||||
| Glargine:23,855 | |||||||||
| Colhoun, et al. [13] | Scotland | 4.0 | M/F | ≥65 | The Scottish Diabetes Research network Epidemiology Group: 49,197 patients with insulin-treated DM | All cancers: 715 | 1.73 (0.98, 3.06) | Age, gender, type of diabetes, calendar time prior cancer, metformin, diabetes duration, HbA1c, BP, BMI, smoking | |
| Comparison group: insulin glargine users alone: 447 | Breast: 92 | 3.65 (1.05, 12.68) | |||||||
| Insulin glargine other insulin users: 3,512 | Prostate: 48 | 1.16 (0.16, 8.50) | |||||||
| Other insulin users: 32,295 | Colorectal 109 | 1.43 (0.45, 4.57) | |||||||
| Lung: 149 | 1.43 (0.53, 3.88) | ||||||||
| Jonasson et al. [25] | Sweden | 2.0 | M/F | 35–84 | The population-based follow-up study in Sweden: 114,841 | All cancers: 2,348 | 1.06 (0.90, 1.25) | Age, gender, BMI, smoking | |
| Comparison group: insulin glargine users alone: 5,970 | All cancers and insitu tumors: 2,509 | 1.04 (0.88, 1.22) | Age at onset of DM and CVD, calendar period | ||||||
| Insulin glargine and other insulin users: 20,316 | Breast: 208 | 1.97 (1.30, 3.00) | |||||||
| Other insulins users: 88,555 | Gastrointestinal: 454 | 0.91 (0.61, 1.38) | |||||||
| Prostate: 464 | 1.26 (0.88, 1.80) | ||||||||
| Currie et al. [14] | UK | – | M/F | >40 | Retrospective cohort study of people treated in UK general practice: 62,809 | All cancers: 2,106 | 1.42 (1.27, 1.60) | Age, gender, smoking, prior cancer, antidiabetic treatment | |
| Insulin:10,067 | Breast: 305 | 1.07 (0.79, 1.44) | |||||||
| Comparison group | Colorectal: 292 | 1.69 (1.23, 2.33) | |||||||
| Metformin monotherapy:31,421 | Pancreas: 89 | 4.63 (2.64, 8.10) | |||||||
| Sulfonylurea monthearp:7,439 | Prostate: 301 | 1.10 (0.79, 1.52) | |||||||
| Metformin plus sulfonyurea:13,882 | |||||||||
| Campbell et al. [15] | USA | 15 | M/F | 50–74 | Cancer Prevention Study II Nutrition Cohort: 73,312 men and 81,663 women | Colorectal cancer | 1.02 (0.79, 1.30) | Age, BMI, Education, physical activity, drug use, alcohol use, NSAID use, family history of colorectal cancer | |
| Men: 218 | 1.36 (1.05, 1.78) | ||||||||
| Women: 103 | 0.95 (0.64, 1.41) | ||||||||
| Monami et al. [17] | Italy | 4.3 | M/F | 63.4 | City Registry Office: 3,002 patients with type 2 DM | Cancer-related mortality: 87 | 2.11 (1.01, 4.50) | Age, gender, BMI, smoking, HbA1c, | 1.01, 4.50 |
| Exposed group: 764 | |||||||||
| Comparison group: 2,553 | |||||||||
| Baur et al. [16] | Germany | 4–5 | M/F | 70.4 | Diabetes Cardiovascular Risk and Evaluation: Targets and Essential Data for Commitment of Treatment (DETECT) study from German primary care practices: 6,789 | Cancer-related mortality: 78 | 3.87 (1.53, 9.81) | Age, gender, HbA1c, smoking, and BMI | |
| Insulin users:329 | |||||||||
| Comparison group: noninsulin users, 6,460 | |||||||||
* RR relative risk; CI confidence interval; DM diabetes mellitus