Childhood Exposure to Parental Smoking Increases Risk of Rheumatoid Arthritis in Adulthood

A hand and wrist x-ray showing severe arthritis of the wrist or carpus and Boutonniere deformity of the thumb.

Active cigarette smoking is the best-established modifiable risk factor for the development of rheumatoid arthritis (RA), but whether passive smoking can be linked to RA is unclear.

Kazuki Yoshida, MD, ScD, instructor in Medicine, and Jeffrey A. Sparks, MD, MMSc, assistant professor of Medicine, both of the Department of Medicine and Division of Rheumatology, Inflammation and Immunity at Brigham and Women’s Hospital, and colleagues recently conducted a study to examine the influence of passive smoking in a large prospective cohort at several stages of the life course, including in utero, while accounting for pack-years of active smoking.

In Arthritis & Rheumatology, they report an increased risk of seropositive RA associated with exposure to childhood parental smoking, especially when combined with active smoking in adulthood.

Key Takeaways

  • This cohort study examined prospectively collected data on 90,923 participants in the Nurses’ Health Study II to examine how passive smoking over the life course influenced the risk of rheumatoid arthritis (RA) development in adulthood
  • Exposure to parental smoking during childhood was associated with adult-onset seropositive RA (HR, 1.75 [95% confidence interval (CI) 1.03, 2.98]; P<0.05), suggesting a direct influence of early-life exposures on RA risk
  • The risk of seropositive RA was even higher in participants exposed to parental smoking in childhood who became smokers themselves (HR, 2.18 [95%CI 1.23, 3.88] compared with participants not exposed to parental smoking who nonetheless became smokers)
  • Exposure to passive smoking in utero did not influence adult RA risk. Passive smoking during adulthood (from age 18 to 52) showed a statistically non-significant trend (HR 1.30 [95%CI 0.97, 1.74]).

Study Methods

The Nurses’ Health Study II, the data source, enrolled 116,429 female registered nurses ages 25–42 years in 1989. For this analysis, the researchers identified 90,923 nurses who did not report RA at baseline. Participants are asked to complete health questionnaires every two years, including questions about passive smoking in 1999, when the women were 35–52 years old.

Over a median of 28 years of follow-up, 532 cases of incident (newly diagnosed) RA were reported and confirmed to meet research criteria on medical record review; 352 of them were determined to be seropositive.

In Utero Exposure

Maternal smoking during pregnancy was not associated with RA after adjustment for later-life active smoking.

Childhood Exposure

Exposure to childhood parental smoking was associated with seropositive incident RA:

  • HR, 1.41 ([95%CI 1.08, 1.83]; P<0.05) in women exposed vs. unexposed after adjustment for confounders
  • HR, 1.75 ([95%CI 1.03, 2.98]; P<0.05) for women exposed vs. unexposed after adjustment for pack-years of active smoking and other confounders
  • HR, 2.18 ([95%CI 1.23, 3.88]; P<0.05) after restricting to 32,216 ever-smokers and adjusting for confounders, pack-years of active smoking, and other confounders

Adult Exposure

Passive smoking in adulthood was defined as years lived with a smoker beginning at age 18. No length of exposure was significantly associated with all RA or seropositive RA with or without adjustment for active smoking.

Life-course Epidemiology

Epidemiologists who adhere to the strict notion of the “social trajectory” model of how the adult disease develops might say early-life passive smoking is important only because it promotes later-life uptake of active smoking—it does not cause harm by itself.

In contrast, the findings of this study support the “cumulative exposure” model, in which both passive and active smoking increases the risk of RA by cumulative dose-response. Understanding that exposure to early-life risk factors influences later-life health may aid in developing effective interventions.