Table 1.
Overview of characteristics of the publications (n = 29) included in systematic review
| Characteristic | Studies, n (%)a |
|---|---|
| Study | |
| Type | |
| Observational | |
| Cohort | 24 (83) |
| Nested Case–Control | 3 (10) |
| Randomized controlled trial | 2 (7) |
| Region (Country) | |
| Asia | |
| Hong Kong | 3 (10) |
| Korea | 1 (3) |
| North America | |
| Canada | 1 (3) |
| United States | 18 (62) |
| Europe | |
| Denmark | 1 (3) |
| France | 1 (3) |
| United Kingdom | 2 (7) |
| Sweden | 1 (3) |
| Europe (10 countries) | 1 (3) |
| Medication categoryb | |
| Anti-hypertensives | 1 (3) |
| Exogenous hormones | |
| Hormone replacement therapy | 9 (31) |
| Oral contraceptives | 4 (14) |
| Metformin | 1 (3) |
| NSAIDs | 13 (45) |
| Statins | 5 (17) |
| Exposure ascertainment | |
| Self- or interview-administered questionnaire | 19 (66) |
| Prescription database | 8 (28) |
| Not applicable (RCT) | 2 (7) |
| Exposure definition | |
| Ever use | 7 (24) |
| Use (defined period) | 7 (24) |
| Current use | 15 (52) |
| Exposure measurement | |
| Fixed | 22 (76) |
| Time-varying | 7 (24) |
| Outcome ascertainment | |
| Registry data | 16 (55) |
| Self-report with medical record verification | 13 (45) |
| Proximal colon cancer definition | |
| Caecum to hepatic flexure | 1 (3) |
| Caecum to transverse colonc | 10 (34) |
| Caecum to splenic flexurec | 10 (34) |
| Not reported | 8 (28) |
| Population | |
| Sample size | |
| < 10,000 | 1 (3) |
| 10,000–99,999 | 18 (62) |
| ≥ 100,000 | 10 (34) |
| Sex | |
| Women onlyd | 15 (52) |
| Men only | 1 (3) |
| Both | 13 (45) |
| Follow-up time years (mean/median) | |
| < 5e | 1 (3) |
| 5–9 | 4 (14) |
| ≥ 10f | 18 (62) |
| Not applicableg | 6 (21) |
NSAID Non-steroidal anti-inflammatory drug, RCT Randomized controlled trial
aSome proportions do not add to 100% due to rounding
bStudies do not total 29 as more than one medication could be investigated within one study
cSome studies excluded the appendix
dNine studies were investigating exogenous female hormones, therefore male participants were not applicable
eMean nor median follow-up reported, but time from recruitment to end of follow-up was less than 5 years
fMean nor median follow-up reported for two studies, but time from recruitment to end of follow-up was ≥ 10 years
gThe outcome measure for three studies was post-colonoscopy proximal colon cancer within 3 years