Patient, procedure and facility predictors of infection within 7 days after colonoscopy or OGD performed at an ambulatory surgery centre, 2014
| Screening colonoscopy (n=462 068) | Non-screening colonoscopy (n=914 140) | OGD (n=873 138) | ||||||||
| % of all procedures | Infection-related unplanned visit rate/1000 procedures | Adjusted OR (95% CI) | % of all procedures | Infection-related unplanned visit rate/1000 procedures | Adjusted OR (95% CI) | % of all procedures | Infection-related unplanned visit rate/1000 procedures | Adjusted OR (95% CI) | ||
| Patient predictors | ||||||||||
| Age* | 0–9 | Not applicable | 0.2 | 4.5 | 2.63 (1.29 to 5.35) | 1.1 | 8.8 | 3.18 (2.50 to 4.05) | ||
| 10–19 | 1.0 | 4.7 | 2.62 (1.89 to 3.65) | 2.3 | 3.1 | 1.20 (0.92 to 1.56) | ||||
| 20–29 | 3.0 | 2.4 | 1.76 (1.34 to 2.31) | 4.6 | 3.4 | 1.43 (1.18 to 1.74) | ||||
| 30–39 | 5.3 | 2.3 | 1.70 (1.36 to 2.12) | 7.4 | 2.9 | 1.15 (0.97 to 1.36) | ||||
| 40–49 | 7.6 | 1.4 | 1.28 (0.93 to 1.76) | 8.9 | 1.7 | 1.29 (1.06 to 1.58) | 12.6 | 2.7 | 1.02 (0.88 to 1.18) | |
| 50–59 | 39.8 | 0.9 | Reference | 31.1 | 1.2 | Reference | 21.9 | 2.6 | Reference | |
| 60–69 | 32.1 | 1.0 | 1.09 (0.88 to 1.37) | 28.9 | 1.3 | 1.05 (0.90 to 1.22) | 24.2 | 2.7 | 0.96 (0.85 to 1.09) | |
| 70–79 | 16.8 | 1.4 | 1.40 (1.09 to 1.78) | 17.1 | 1.7 | 1.28 (1.09 to 1.51) | 18.0 | 2.9 | 1.00(0.88 to 1.13) | |
| 80–100 | 3.8 | 2.5 | 1.96 (1.39 to 2.76) | 4.4 | 3.3 | 2.27 (1.85 to 2.78) | 7.8 | 5.3 | 1.54 (1.35 to 1.77) | |
| Sex | Male | 42.6 | 1.1 | Reference | 47.1 | 1.5 | Reference | 40.7 | 3.3 | Reference |
| Female | 57.4 | 1.1 | 0.94 (0.79 to 1.11) | 52.9 | 1.6 | 1.06 (0.95 to 1.18) | 59.3 | 2.8 | 0.89 (0.83 to 0.97) | |
| Race | White | 66.3 | 1.1 | Reference | 67.7 | 1.5 | Reference | 67.2 | 3.0 | Reference |
| Black | 10.1 | 1.8 | 1.57 (1.22 to 2.01) | 9.2 | 2.0 | 1.32 (1.11 to 1.56) | 8.2 | 4.4 | 1.41 (1.25 to 1.60) | |
| Hispanic | 15.0 | 1.3 | 1.11 (0.87 to 1.42) | 14.1 | 1.6 | 1.04 (0.89 to 1.22) | 15.4 | 2.9 | 0.99 (0.88 to 1.11) | |
| Asian or Pacific Islander | 4.0 | 0.5 | 0.50 (0.27 to 0.96) | 3.8 | 1.2 | 0.77 (0.57 to 1.07) | 3.8 | 2.5 | 0.84 (0.67 to 1.05) | |
| Native American | 0.3 | 4.1 | 3.68 (1.51 to 8.98) | 0.3 | 2.4 | 1.64 (0.78 to 3.45) | 0.3 | 3.3 | 1.09 (0.57 to 2.11) | |
| Other race | 4.3 | 0.6 | 0.70 (0.39 to 1.25) | 4.8 | 1.9 | 1.46 (1.16 to 1.83) | 5.1 | 2.6 | 1.10 (0.57 to 2.11) | |
| Hospitalisation within 30 days prior to procedure | Yes | 0.6 | 15.2 | 9.41 (6.82 to 12.99) | 0.9 | 11.2 | 5.34 (4.31 to 6.63) | 2.2 | 18.3 | 5.11 (4.55 to 5.74) |
| No | 99.4 | 1.0 | Reference | 99.1 | 1.5 | Reference | 97.8 | 2.7 | Reference | |
| Gastrointestinal endoscopic procedure in an ASC within 30 days prior to procedure | Yes | 1.6 | 3.0 | 2.64 (1.71 to 4.07) | 2.4 | 3.0 | 2.20 (1.70 to 2.84) | 3.2 | 4.7 | 1.45 (1.22 to 1.73) |
| No | 98.4 | 1.1 | Reference | 97.6 | 1.5 | Reference | 96.8 | 3.0 | Reference | |
| Inflammatory bowel disease coded on day of procedure † | Yes | Not applicable | 3.9 | 2.9 | 1.72 (1.40 to 2.13) | 1.0 | 4.4 | 1.51 (1.09 to 2.09) | ||
| No | 96.1 | 1.5 | Reference | 99.0 | 3.0 | Reference | ||||
| Elixhauser comorbidities | 0 | 74.6 | 0.9 | Reference | 73.5 | 1.3 | Reference | 63.2 | 2.1 | Reference |
| 1 | 14.5 | 1.4 | 1.23 (0.96 to 1.58) | 15.0 | 2.2 | 1.40 (1.21 to 1.62) | 20.3 | 3.8 | 1.45 (1.31 to 1.60) | |
| 2–14 | 10.9 | 2.3 | 1.70 (1.32 to 2.19) | 11.5 | 2.8 | 1.70 (1.46 to 1.99) | 16.5 | 5.6 | 1.82 (1.63 to 2.02) | |
| Procedure predictors | ||||||||||
| Invasiveness‡ | Yes | 16.6 | 0.8 | 1.16 (0.82 to 1.65) | 83.5 | 1.6 | 1.03 (0.87 to 1.22) | 89.0 | 2.9 | 0.82 (0.74 to 0.91) |
| No | 83.4 | 1.2 | Reference | 16.5% | 1.3 | Reference | 11.0% | 4.3 | Reference | |
| Concurrence with another gastrointestinal endoscopic procedure | Yes | 18.0 | 1.7 | 1.40 (1.14 to 1.71) | 20.9 | 2.5 | 1.62 (1.44 to 1.82) | 35.3 | 2.5 | 0.78 (0.72 to 0.85) |
| No | 82.0 | 1.0 | Reference | 79.1 | 1.3 | Reference | 64.7 | 3.3 | Reference | |
| Facility predictors | ||||||||||
| ASC state | Florida | 40.4 | 1.4 | Reference | 39.8 | 1.7 | Reference | 42.4 | 3.0 | Reference |
| California | 20.3 | 1.2 | 0.83 (0.64 to 1.07) | 18.0 | 2.0 | 0.96 (0.82 to 1.12) | 17.4 | 4.4 | 1.03 (0.92 to 1.14) | |
| Georgia | 5.1 | 1.6 | 0.78 (0.53 to 1.13) | 3.7 | 2.0 | 0.88 (0.68 to 1.15) | 4.7 | 4.3 | 0.81 (0.68 to 0.95) | |
| New York | 31.5 | 0.7 | 0.52 (0.38 to 0.71) | 34.9 | 1.2 | 0.67 (0.58 to 0.78) | 32.8 | 2.1 | 0.63 (0.57 to 0.70) | |
| Nebraska | 1.6 | 1.8 | 1.18 (0.64 to 2.18) | 2.1 | 1.4 | 0.67 (0.44 to 1.00) | 2.0 | 3.9 | 0.82 (0.64 to 1.06) | |
| Vermont | 1.1 | 0.2 | 0.16 (0.02 to 1.13) | 1.6 | 0.9 | 0.58 (0.33 to 1.02) | 0.7 | 2.5 | 0.62 (0.38 to 1.01) | |
| Annual procedure volume (tertile)§ | Low | 2.8 | 2.2 | Reference | 2.8 | 2.6 | Reference | 2.8 | 5.7 | Reference |
| Middle | 16.7 | 1.2 | 0.86 (0.57 to 1.31) | 17.7 | 1.9 | 0.83 (0.63 to 1.08) | 18.2 | 3.8 | 0.69 (0.58 to 0.84) | |
| High | 80.5 | 1.0 | 0.77 (0.51 to 1.16) | 79.5 | 1.4 | 0.82 (0.63 to 1.06) | 79.0 | 2.8 | 0.69 (0.58 to 0.82) | |
| Freestanding | Yes | 46.2 | 0.9 | 0.67 (0.48 to 0.92) | 44.1 | 1.2 | 0.84 (0.69 to 1.02) | 41.5 | 1.8 | 0.76 (0.65 to 0.89) |
| No | 53.8 | 1.3 | Reference | 55.9 | 1.8 | Reference | 58.5 | 3.9 | Reference | |
| Multispecialty | Yes | 63.6 | 1.2 | 0.77 (0.58 to 1.02) | 64.2 | 1.8 | 1.07 (0.90 to 1.28) | 65.8 | 3.7 | 1.16 (1.00 to 1.36) |
| No | 36.4 | 1.0 | Reference | 35.8 | 1.2 | Reference | 34.2 | 1.8 | Reference | |
| Proportion of procedures performed on patients with infection coded on day of procedure | 0% | 1.2 | 1.5 | Reference | 1.6 | 1.5 | Reference | 1.3 | 2.0 | Reference |
| >0%–5% | 80.9 | 1.0 | 0.72 (0.35 to 1.48) | 79.6 | 1.4 | 0.76 (0.50 to 1.16) | 76.7 | 2.7 | 1.15 (0.75 to 1.74) | |
| >5%–50% | 17.9 | 1.6 | 0.80 (0.38 to 1.68) | 18.7 | 2.3 | 0.86 (0.55 to 1.34) | 22.0 | 4.4 | 1.24 (0.81 to 1.89) | |
NA: not applicable. Adjust ORs were calculated from multivariable models adjusting patient, procedural and facility predictors.
*Screening colonoscopies coded for patients younger than 40 years old were reclassified as non-screening colonoscopies.
†Screening colonoscopies coded for patients with inflammatory bowel disease were reclassified as non-screening colonoscopies.
‡A procedure was defined as invasive if gastrointestinal mucosa was likely disrupted during the procedure.
§Screening colonoscopy volume tertiles: lowest 1–90, middle 91–370, highest 371–6645. Non-screening colonoscopy volume tertiles: 1–165, 166–730, 731–7597. OGD volume tertiles: 1–163, 164–693, 694–5944.
ASC, ambulatory surgery centre; OGD, osophagogastroduodenoscopy.