Table 2. Risk of Bias and Study Quality Assessment for Included Studies Using NOS.

Source	Assessment of quality of a cohort study–NOS Domaina										Total NOS score (maximum, 9; AHRQ standard)
	Selection					Comparability and Outcome
	Representativeness of exposed cohort	Selection of nonexposed cohort	Ascertainment of exposure	Demonstration that outcome of interest was not present at start of study	Total score (maximum, 4)	Comparability of cohorts on basis of design or analysis (maximum, 2)	Assessment of outcome	Was follow-up long enough for outcomes to occur? (≥6 mo)	Adequacy of follow-up of cohorts	Total score (maximum, 5)
El-Matary et al,12 2020	★ (a): All persons in University of Manitoba IBD Epidemiology Database who received diagnosis of IBD before age 18 y	★ (a): Manitoba universal health insurance	★ (a): Validated administrative case definition of IBD	(b): No statement of exclusion of prevalent cancer cases at point of follow-up	3	★ (a): Age, sex, and region of residence matched	★ (b): Linkage to Manitoba cancer registry	★ (a): More than 34-y study period	(d): No statement on adequacy of follow-up	3	6 (Good)
Kjaergaard et al,14 2020	★ (a): All persons in Danish national patient registry who received diagnosis of IBD before age 18 y	★ (a): Reference population derived from Danish national patient registry	★ (a): ICD-8 and ICD-10 diagnosis; minimum 2 diagnoses from Danish national patient registry	★ (a): Individuals who received diagnosis of cancer before IBD diagnosis date were excluded	4	★★ (a): Reference population sex and age matched; confounders adjusted to include sex, age, and calendar year of diagnosis	★ (b): Linkage to Danish Cancer Registry	★ (a): More than 40-y study period	(d): No statement on adequacy of follow-up	4	8 (Good)
Malham et al,15 2019	★ (a): All persons who received diagnosis of IBD from Special Reimbursement Registry (social insurance institution) before age 18 y	★ (a): Background national population in corresponding age groups	★ (a): ICD-8, ICD-9, and ICD-10 diagnosis in Finnish national registry	(b): No statement of exclusion of prevalent cancer cases at point of follow-up	3	(c): No matching to reference population; no reporting of confounding	★ (b): Linkage to Finnish Cancer Registry; use of NORDCAN age-specific cancer rates for reference population	★ (a): Median 9.6-y follow-up time	(d): No statement on adequacy of follow-up	2	5 (Fair)
Olen et al,16 2017	★ (a): All persons who received diagnosis of IBD in National Swedish Patient Register before age 18 y	★ (a): Derived from national Swedish Population Register	★ (a): ICD-10 diagnosis in National Swedish Patient Register; minimum 2 diagnoses; diagnostic procedural codes (eg, disease-specific surgery)	★ (a): Individuals in pediatric IBD or general population who had cancer before start of follow-up were excluded	4	★★ (a): General population were matched by sex, age, year of birth, and county; confounders adjusted for were sex, age, birth year, and region of residence	★ (b): Linkage to National Swedish Cancer Register	★ (a): More than 50-y study period	★ (a): Estimated register completeness of >96%	5	9 (Good)
Peneau et al,17 2013	★ (a): All persons in EPIMAD—retrospective population-based study of incidence cases of IBD in northern France since 1988 (<17 y)	(c): Reference population is not described	★ (b): Confirmation by 2 gastroenterologists; recorded as definite, probable, or possible	★ (a): No statement of exclusion of prevalent cancer cases at point of follow-up	3	★ (b): No matching to reference population; confounders adjusted to include age and sex	★ (b): Diagnosis from clinical record and confirmed using administrative health database	★ (a): Median 11.4-y follow-up time	(d): No statement on adequacy of follow-up	3	6 (Good)

Abbreviations: AHRQ, Agency for Healthcare Research and Quality; EPIMAD, Epidémiologie des maladies inflammatoires de l’Intestin; IBD, inflammatory bowel disease; ICD-8, International Classification of Diseases, Eighth Revision; ICD-9, International Classification of Diseases, Ninth Revision; ICD-10, International Statistical Classification of Diseases and Related Health Problems, Tenth Revision; NOS, Newcastle-Ottawa Scale; NORDCAN, Nordic Cancer Registry.

^a Each star totals 1 point on the NOS; (a), (b), (c), and (d) are NOS assessment form descriptors (available in eFigure 1 in the Supplement).