| Unique ID issued by UMIN | UMIN000063019 |
|---|---|
| Receipt number | R000072142 |
| Scientific Title | Cumulative antibiotic exposure over five years and the incidence of colorectal cancer: a risk-set-matched nested case-control study in a Japanese health insurance claims database |
| Date of disclosure of the study information | 2026/09/22 |
| Last modified on | 2026/09/22 10:02:05 |
A study of the association between cumulative antibiotic exposure over five years and the subsequent incidence of colorectal cancer, using a Japanese health insurance claims database (risk-set-matched nested case-control study)
Antibiotic Exposure and Colorectal Cancer Study (JMDC)
Cumulative antibiotic exposure over five years and the incidence of colorectal cancer: a risk-set-matched nested case-control study in a Japanese health insurance claims database
Antibiotic Exposure and Colorectal Cancer Nested Case-Control Study (JMDC)
| Japan |
Colorectal cancer (colon and rectal cancer, ICD-10 C18-C20, excluding appendix). The exposure is cumulative systemic antibiotic use.
| Gastroenterology | Hematology and clinical oncology | Infectious disease |
Malignancy
NO
Using the JMDC claims database (2005-2025, restricted to enrollees with a prior systemic anti-infective prescription), to estimate the association between cumulative days of systemic antibiotic coverage during the five years (60 months) ending 24 months before diagnosis and incident colorectal cancer, using a risk-set-matched nested case-control design. Up to five controls per case are matched on sex, birth month, enrollment start month, and age group (under 50 vs 50 or older), and odds ratios adjusted for healthcare use, infections, comorbidities, and co-medications are estimated by conditional logistic regression with person-level cluster-robust variance. Adults under 50 years are also evaluated separately.
Others
Observational study evaluating a risk-factor association. An epidemiological study assessing the association between antibiotic exposure and incident colorectal cancer; it does not aim to establish causality.
Exploratory
Pragmatic
Not applicable
The primary outcome is a first confirmed (non-suspected) colorectal cancer diagnosis (ICD-10 C18-C20, excluding appendix C18.1, neuroendocrine tumours, sarcomas, GIST, melanoma, carcinoma in situ, and anal cancer). Anchored at the diagnosis month, cumulative days of systemic antibiotic coverage over five years with a 24-month lag are categorized as 0 (reference), 1-7, 8-14, 15-30, 31-60, 61-90, and 91 or more days, and odds ratios relative to 0 days are estimated by conditional logistic regression (95% CI; a 6-degree-of-freedom Wald test).
Secondary and sensitivity analyses: analysis in adults under 50 years; analyses by tumour site (colon vs rectum); using treatment courses as the exposure metric; oral agents only and WHO J01 only; a 36-month lag with re-matched sets; exclusion of inflammatory bowel disease or hereditary CRC syndromes; restriction to index months from April 2020; alternative handling of missing records; a linear trend per doubling of days + 1; an interaction with age under 50; and a ten-year window with four categories. Balance is described with standardized differences. Reporting follows STROBE and RECORD-PE.
Observational
| Not applicable |
| Not applicable |
Male and Female
The source population is JMDC claims database enrollees (2005-2025) with a history of systemic anti-infective (EphMRA ATC group J) prescribing. (1) Cases: a first confirmed (non-suspected) colorectal cancer diagnosis (ICD-10 C18-C20, excluding appendix and others). (2) Controls: for each case, enrollees at risk in the diagnosis month, matched on sex, birth month (plus or minus 12 months), enrollment start month (plus or minus 12 months), and age group (under 50 vs 50 or older) at the diagnosis month; up to five controls per case are selected by a reproducible random ordering (controls may be reused and may later become cases). (3) Eligibility: enrollment begins at least 12 months before the exposure window, and the first anti-infective prescription precedes the window start. The exposure window is the 60 months ending 24 months before diagnosis. All ages.
An explicit recurrence code at or before the first colorectal cancer diagnosis, or a code for colorectal-cancer-derived bone metastasis preceding it (excluded as prevalent). Cases with indeterminate exposure due to missing prescription dates or days (the set is dropped), and indeterminate controls (removed individually). Records with only a suspected flag.
32227
| 1st name | Satoshi |
| Middle name | |
| Last name | Kutsuna |
The University of Osaka, Graduate School of Medicine Faculty of Medicine
Department of Infection Prevention and Control
565-0871
2-2, Yamadaoka, Suita city
0836-6879-5111
kutsuna@hp-infect.med.osaka-u.ac.jp
| 1st name | Satoshi |
| Middle name | |
| Last name | Kutsuna |
The University of Osaka, Graduate School of Medicine Faculty of Medicine
Department of Infection Prevention and Control
565-0871
2-2, Yamadaoka, Suita city
0836-6879-5111
kutsuna@hp-infect.med.osaka-u.ac.jp
The University of Osaka, Graduate School of Medicine Faculty of Medicine
The University of Osaka
Self funding
Japan
None
None
Ethics Committee for Observational Research, The University of Osaka Hospital (approval no. 25469)
2-15, Yamadaoka, Suita city
0836-6879-5111
rinri@hp-crc.med.osaka-u.ac.jp
NO
大阪大学大学院医学系研究科 感染制御学 / The University of Osaka, Graduate School of Medicine
| 2026 | Year | 09 | Month | 22 | Day |
Unpublished
32227
Completed
| 2026 | Year | 04 | Month | 14 | Day |
| 2026 | Year | 04 | Month | 14 | Day |
| 2026 | Year | 04 | Month | 14 | Day |
| 2026 | Year | 09 | Month | 22 | Day |
| 2026 | Year | 09 | Month | 22 | Day |
| 2026 | Year | 09 | Month | 22 | Day |
| 2026 | Year | 09 | Month | 22 | Day |
This is a retrospective risk-set-matched nested case-control study using the JMDC claims database (January 2005 to May 2025, restricted to enrollees with a history of systemic anti-infective prescribing). The association between cumulative days of systemic antibiotic exposure over the five years ending 24 months before diagnosis and incident colorectal cancer was estimated by conditional logistic regression. Ethics approval: Ethics Committee for Observational Research, The University of Osaka Hospital (approval no. 25469, 14 April 2026); anonymized data were used and consent was waived.
| 2026 | Year | 09 | Month | 22 | Day |
| 2026 | Year | 09 | Month | 22 | Day |
Value
https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000072142