Characteristics of the study cohorts
| Men | Women | |||||||
|---|---|---|---|---|---|---|---|---|
| Barrett’s | Oesophagitis | Reflux | Standard ref | Barrett’s | Oesophagitis | Reflux | Standard ref | |
| UTS, up to standard. | ||||||||
| No of subjects (n = 27 813) | 1033 | 3897 | 3828 | 8264 | 644 | 2495 | 2500 | 5152 |
| (% in each cohort) | 62% | 61% | 60% | 62% | 38% | 39% | 40% | 38% |
| Mean age (y) | 61.0 | 61.3 | 61.1 | 61.0 | 67.8 | 68.2 | 67.9 | 67.8 |
| Available UTS records in GPRD (y) | 7971 | 32260 | 31657 | 65446 | 5071 | 21147 | 20933 | 41277 |
| Follow up used in the analysis (y) | 1628 | 4505 | 4383 | 20511 | 987 | 3001 | 2860 | 12579 |
| Total oesophageal cancer (n = 137) | 22 | 35 | 20 | 15 | 12 | 16 | 9 | 8 |
| Already diagnosed at the start (n = 54) | 4 | 21 | 7 | 3 | 1 | 9 | 5 | 4 |
| Within first year of follow up (n = 45) | 9 | 9 | 9 | 4 | 7 | 4 | 2 | 1 |
| Beyond first year of follow up (n = 38) | 9 | 5 | 4 | 8 | 4 | 3 | 2 | 3 |
| Having endoscopy record | 62% | 36% | 22% | 7% | 59% | 36% | 22% | 8% |
| Smoking | ||||||||
| Never or non-smoker | 48% | 48% | 50% | 44% | 64% | 62% | 63% | 59% |
| Ex-smoker | 16% | 15% | 14% | 12% | 9% | 9% | 9% | 8% |
| Current smoker | 17% | 20% | 20% | 16% | 9% | 12% | 12% | 10% |
| Missing | 19% | 18% | 17% | 28% | 17% | 18% | 16% | 24% |
| Alcohol | ||||||||
| Non-drinkers | 11% | 9% | 10% | 8% | 21% | 19% | 20% | 16% |
| Infrequent drinkers (rarely or occasionally) | 14% | 17% | 17% | 14% | 23% | 24% | 24% | 22% |
| Regular drinkers | 17% | 17% | 17% | 14% | 8% | 10% | 9% | 10% |
| Missing | 58% | 56% | 56% | 64% | 49% | 47% | 47% | 51% |
| Body mass index | ||||||||
| Normal weight (18–24) | 29% | 30% | 30% | 27% | 28% | 32% | 32% | 33% |
| Overweight (25–29) | 37% | 37% | 38% | 31% | 31% | 29% | 30% | 25% |
| Obese (>30) | 10% | 11% | 11% | 10% | 17% | 15% | 15% | 12% |
| Missing | 24% | 22% | 21% | 31% | 25% | 24% | 23% | 29% |