Objective To investigate the relationship between smoking and imaging outcomes over 5 years in axial spondyloarthritis (SpA) and to assess whether socioeconomic factors influence these relationships. Methods Axial SpA patients from the Devenir des Spondylarthropathies Indifferérenciées Récentes cohort were included. The following 4 imaging outcomes were assessed by 3 central readers at baseline, 2 years, and 5 years: spine radiographs (using the modified Stoke Ankylosing Spondylitis Spine Score mSASSS), sacroiliac (SI) joint radiographs (using the modified New York criteria), magnetic resonance imaging (MRI) of the spine (using the Spondyloarthritis Research Consortium of Canada SPARCC score), and MRI of the SI joint (using the SPARCC score). The explanatory variable of interest was smoking status at baseline. Interactions between smoking and socioeconomic factors (i.e., job type blue‐collar or manual work versus white‐collar or nonmanual work and education low versus high) were first tested, and if significant, analyses were run using separate strata. Generalized estimating equations models were used, with adjustments for confounders. Results In total, 406 axial SpA patients were included (52% male, 40% smokers, and 18% blue collar). Smoking was independently associated with more MRI‐detected SI joint inflammation at each visit over the 5 years, an effect that was seen only in patients with blue‐collar professions (β = 5.41 95% confidence interval (95% CI) 1.35, 9.48) and in patients with low education levels (β = 2.65 95% CI 0.42,4.88), using separate models. Smoking was also significantly associated with spinal inflammation (β = 1.69 95% CI 0.45, 2.93) and SI joint damage (β = 0.57 95% CI 0.18, 0.96) across all patients, irrespective of socioeconomic factors and other potential confounders. Conclusion Strong associations were found between smoking at baseline and MRI‐detected SI joint inflammation at each visit over a time period of 5 years in axial SpA patients with a blue‐collar job or low education level. These findings suggest a possible role for mechanical stress amplifying the effect of smoking on axial inflammation in axial SpA.