Table 2.
Association between cholecystectomy and the risk of colorectal cancer overall and stratified by sex, family history, anatomical location and tumor mismatch repair status
| Cases | Controls | Unadjusted | Adjusted** | ||||
|---|---|---|---|---|---|---|---|
|
| |||||||
| n/N | n/N | OR (95% CI) | P-value | OR (95% CI) | P-value | P-difference* | |
|
| |||||||
| Overall | 521/5847 | 555/4970 | 0.76 (0.66, 0.88) | .001 | 0.88 (0.73, 1.08) | .22 | |
| Sex | |||||||
| Women*** | 395/2959 | 410/2630 | 0.84 (0.72, 0.99) | .0 4 | 1.00 (0.78, 1.28) | .99 | .54 |
| Men | 126/2888 | 145/2340 | 0.74 (0.56, 0.97) | .03 | 0.78 (0.53, 1.13) | .19 | |
|
| |||||||
| Family history | 555/4970 | ||||||
| No | 399/4599 | 0.75 (0.65, 0.88) | <.001 | 0.86 (0.69, 1.06) | .16 | ||
| Yes | 122/1248 | 0.90 (0.71, 1.14) | .39 | 0.98 (0.71, 1.33) | .88 | .64 | |
|
| |||||||
| Anatomical location | 555/4970 | ||||||
| Proximal colon | 216/1968 | 1.02 (0.42, 1.23) | .83 | 0.93 (0.70, 1.24) | .63 | .45 | |
| Distal colon | 151/1811 | 0.70 (0.56, 0.86) | .001 | 0.75 (0.56, 1.02) | .07 | ||
| Colon unspecified | 3/81 | 0.30 (0.09, 0.95) | .04 | 0.59 (0.13, 2.63) | .49 | ||
| Rectum | 151/1987 | 0.66 (0.53, 0.82) | <.001 | 0.97 (0.72, 1.29) | .84 | ||
|
| |||||||
| Tumor MMR status | 555/4970 | ||||||
| Proficient | 314/3790 | 0.74 (0.63, 0.87) | <.001 | 0.86 (0.68, 1.09) | .20 | .54 | |
| Deficient | 71/631 | 0.99 (0.73, 1.34) | .93 | 0.96 (0.63, 1.49) | .87 | ||
OR, Odds ratio; CI, confidence interval, MMR, mismatch repair; n, number of individuals with cholecystectomy; N, total number
for stratified analysis by sex: it was calculated using likelihood ratio tests comparing unconditional multivariable logistic regression models with and without an interaction term; for stratified analyses by first-degree family history, anatomical location and tumor MMR status, it was calculated using likelihood ratio tests on multinomial logistic regression models.
adjusted for sex (women or men, except for stratified analysis by sex), first-degree family history (no, yes; except for stratified analysis by family history), age at study recruitment (continuous), alcohol consumption (per standard drink per day), BMI (WHO categories), BMI at age 20 years (WHO categories), cigarette smoking (never, former, current), aspirin (non-users, <5 years use, ≥5 years use), ibuprofen (non-users, <5 years use, ≥5 years use), diabetes status (no, yes), recruitment center (Cancer Care Ontario, Fred Hutchinson Cancer Research Center, Mayo Clinic or University of Melbourne), regular physical activity (no, yes), and red meat consumption (serving per day)
For women, further adjusted for hormonal contraceptive use (non-users, <5 years use, ≥5 years use) and number of live birth (never, 1, 2, 3, ≥4)