Background Rheumatoid arthritis (RA) is known from clinical studies to be associated with a number of different lung conditions, but the relationship with arthritis in the population, where overwhelmingly the most frequent type of arthritis is osteoarthritis, has been less well studied.

Objectives To examine whether arthritis is associated with an increased likelihood of reporting asthma and/or COPD.

Methods Data from the 2011- 2012 Canadian Community Health Survey, a representative survey of the household population were used, restricting analyses to the population age ≥35 years (n=88,734). Participants were asked whether they had specific long term health conditions diagnosed by a health professional: conditions asked about included arthritis (excluding fibromyalgia), asthma, and COPD (specifically chronic bronchitis, emphysema or chronic obstructive pulmonary disease). Multinominal logistic regression was used to examine associations between lung conditions (none, asthma only, COPD only, both asthma and COPD) and arthritis, controlling for covariates: age, sex, education, household income, smoking status (never, former, current), body mass index (BMI: underweight, normal, overweight, obese), and mean number of other chronic conditions (out of 9). As metabolic and inflammatory pathways have been postulated given the strong association of obesity and arthritis, sensitivity analyses were undertaken stratified by BMI category.

Results People with arthritis (23% of the population) more frequently reported lung conditions than those without, respectively asthma 8.2% vs 5.5%, COPD 5.4% vs 2.0%, and asthma+COPD 3.4% vs 0.8%. There was no difference in the proportion of current smokers between those with and without arthritis (20.5%), although those with arthritis were more likely to be former smokers (49.8% vs 36.5%). They were also more likely to be obese (26.8% vs 16.7%) and have more other chronic conditions (mean 1.5 vs 0.5). In regression analyses controlling for covariates, arthritis was significantly associated with reporting of all lung conditions: OR (95%CI): asthma only 1.48 (1.26, 1.73); COPD only 1.60 (1.34, 1.91); asthma+COPD 1.98 (1.65, 2.37). In addition, the odds of all lung conditions were significantly higher for females. Obesity was also a significant risk factor for reporting all lung conditions, as was the number of other chronic conditions. The sensitivity analyses showed the associations between arthritis and the lung conditions were significant and similar for normal, overweight and obese BMI categories. Being a current or former smoking was significantly associated with COPD, but not with asthma only.

Conclusions This study confirms a relationship between arthritis and lung disease, and the multivariate analyses adds to the scant previous population-based literature which was mainly descriptive. That those with arthritis had increased odds of reporting lung disease, asthma, COPD, and asthma+COPD compared to those without arthritis, controlling for a range of other factors, including smoking, raises questions as to the mechanism. The consistency of the association over levels of BMI suggests other pathways in addition to those mediated by obesity/metabolic factors. The findings also suggest that recommendations for the management of arthritis should consider the potential implications of having lung disease, for example, for physical activity.

Disclosure of Interest None declared