Table 2.

Drug-specific risks of infection requiring hospitalization with nonbiologic DMARD.

Study	Year	Design	Sample, no. patients	Comparison	Duration	Adjustments	Nonbiologic DMARD^*	RR (95% CI)
Doran⁶	2002	Retrospective cohort	609 with RA	Internal controls, ever vs never	1955–1994	Age, sex, smoking status, leukopenia, corticosteroid use, diabetes mellitus	Chemotherapy	5.0 (2.4–10.3)
							Cyclophosphamide	6.1 (3.1–11.8)
							Cyclosporine	2.0 (1.3–3.2)
							Corticosteroids	1.9 (1.5–2.5)
Franklin⁴	2007	Prospective cohort	2108 with new onset inflammation polyarthritis	Internal controls, ever vs never	1990–1999	Age, sex	Steroid use	2.2 (1.5–3.4)
Bernatsky⁷	2007	Nested case-control design within an RA cohort	23,733 with RA	Internal controls, vs 10 randomly selected internal controls per reported infection	1980–2003	Other DMARD medications, no. physician visits	Azathioprine	1.5 (1.2–2.0)
							Cyclophosphamide	3.3 (2.3–4.7)
							Glucocorticoids	2.6 (2.3–2.9)
Smitten²	2008	Retrospective cohort	24,530 with RA	Pts with RA vs pts without RA	1999–2006	Age, sex, other current RA medications, diabetes, chronic lung disease, organic brain disease, cancer, orthopedic procedures, no. hospitalizations between cohort entry, index data and whether or not pts saw a rheumatologist during followup	Oral corticosteroids (any)	1.9 (1.7–2.2)
							≤ 5 mg/day	1.3 (1.1–1.6)
							6–10 mg/day	1.9 (1.5–2.5)
							> 10 mg/day	3.0 (2.4–3.7)
Lacaille⁵	2008	Retrospective cohort	27,710 with RA	Internal controls, vs no DMARD and no corticosteroid	1996–2003	Age, prior infection, no. prior infections, comorbidities, RA duration, and socioeconomic status	Immunosuppressant + CS	1.6 (1.5–1.8)
							Nonimmunosuppressant + CS	1.6 (1.4–1.8)
							CS alone	1.9 (0.06–2.1)

↵* Includes only nonbiologic DMARD shown to be associated with a significant risk of infection in the respective studies. DMARD: disease-modifying antirheumatic drug; RA: rheumatoid arthritis; CS: corticosteroids use.