Table 1.

Selected studies associating passive cigarette smoking with risk of rheumatoid arthritis.

Reference	Study design	Population Sample (n) RA outcomes (n)	Passive smoking exposure methods	RA outcome methods	Personal cigarette smoking and other adjustment variables	Effect size (95% confidence interval) for passive smoking and RA risk	Comments
Jaakkola JJ, Int J Epidemiol (2005) [24]	Prospective cohort	Finland, national registries, singleton births, <7 years of age n=58,841 n=44 incident RA	Finnish Medical Birth Registry, categorical (no smoking, <10 cigarettes/day, >10 cigarettes/day)	Hospitalization and/or billing code	No personal smoking adjustment (but population was <7 years old) Sex, maternal parity, maternal age, marital status, maternal occupation, birth weight, gestational age	RA and other polyarthritis in first 7 years of life: OR 2.10 (1.30-3.40)	Effect of maternal smoking on RA risk limited to girls Higher exposure to smoke increased the risk of RA and other polyarthritis in girls All cases were juvenile-onset; may not be generalizable to adult RA
Costenbader KH, Am J Med (2006) [25]	Prospective cohort	United Status, Nurses’ Health Study, female nurses n=103,818 n=453 incident RA	Self-report, categorical (per 10 years lived with smoker)	1987 ACR criteria	Stratified by ever/never smokers and ever smoker analysis adjusted for personal smoking (continuous, pack-years) BMI, alcohol use, paternal occupation, age at menarche, parity, duration of breastfeeding, postmenopausal hormone use	All RA among ever smokers, >30 years lived with smoker (reference: 0 years): RR 1.59 (0.92-2.74) All RA among never smokers, >30 years lived with smoker (reference: 0 years): RR 1.46 (0.92-2.32)	Living with smoker for >30 years associated with increased RA risk, although not statistically significant No dose effect Measured smoke exposure at home and work
Hedström AK, Ann Rheum Dis (2018) [21]	Case-control	Sweden, never-smokers aged 18-70 years (EIRA) n=2,353 n=589 incident RA	Self-report, categorical (per 10 years passive smoke exposure)	1987 ACR criteria	All never smokers Age, sex, residential area, ancestry	ACPA-positive RA: OR 1.0 (0.8-1.2) ACPA-negative RA: OR 0.9 (0.7-1.2)	No trend between duration of passive smoking and RA risk No significant age- or sex-related differences
Seror R, Rheumatology (2019) [23]	Prospective cohort	France, females (E3N Cohort) n=71,248 n=371 incident RA	Self-reported (≤a few hours a week vs >a few hours a day of childhood smoke exposure, <1 hour/day vs ≥1 hour/day of adult smoke exposure)	Self-report and billing code	No personal smoking adjustment (stratified analysis as never/ever smoking) Age	Childhood passive smoking and all RA: HR 1.43 (0.97-2.11) Adult passive smoking and all RA: HR 0.96 (0.69-1.34)	RA onset earlier in smokers also exposed to smoke in childhood No association between adult passive smoking and RA
Kronzer VL, Arthritis Rheumatol (2019) [22]	Case-control	Minnesota and Florida, Mayo Clinic Biobank participants n=4,084 n=1,023 prevalent/incident RA	Self-report, categorical (/10 years smoke exposure, exposure in home vs workplace)	Self-report, billing code, and/or 2010 ACR criteria	Personal smoking status (never/past/current) Age, sex, BMI, race, education, year, residential area	Home exposure and all RA: OR 1.06 (0.91-1.23) Workplace exposure and all RA: OR 1.01 (0.86-1.17) Combined home and workplace and all RA per 10 pack-years: OR 1.04 (1.00-1.09)	Also investigated age at first exposure, duration, and packs/day Potential dose effect observed No difference in effect of passive smoking on non-smokers vs smokers Also investigated asthma and allergies for RA risk

Statistically significant results are bolded.

ACPA, anti-citrullinated protein antibody; ACR, American College of Rheumatology; BMI, body mass index; E3N, Etude Epidémiologique auprès de femmes de la Mutuelle Générale de l'Education Nationale; EIRA, Epidemiological Investigations of Rheumatoid Arthritis; HR, hazard ratio; OR, odds ratio; RA, rheumatoid arthritis.