Objective The aim of the current study is to determine whether baseline serum adiponectin levels predict the development of rheumatoid arthritis (RA). Method The current report includes 3693 individuals from the Swedish Obese Subjects (SOS) study. The original SOS study is a longitudinal non-randomized controlled study aiming to assess the effect of bariatric surgery on obesity-related mortality and morbidity. Participants included in the present report had adiponectin measurement available at baseline and no prevalent RA. The diagnosis of RA was retrieved through the Swedish National Patient Register. Results During a follow-up for up to 29 years, 82 study participants developed RA. Elevated baseline adiponectin levels were associated with a higher risk of developing RA independently of other factors, including C-reactive protein (CRP) and smoking [hazard ratio (HR) 1.70, 95% confidence interval (CI) 1.12–2.60 for an increase in adiponectin of 10 mg/L, p = 0.01]. After stratifying the population according to adiponectin and CRP median at baseline, study participants with both adiponectin and CRP above the median had a higher risk of developing RA compared to subjects with adiponectin and CRP below the median (HR 2.80, 95% CI 1.25–6.31, p = 0.01). Conclusions In this cohort of subjects with obesity followed up for up to 29 years, high serum adiponectin levels at baseline were associated with an increased risk for RA. Moreover, subjects with both high adiponectin and CRP levels at baseline were at particular risk of developing RA. ClinicalTrials.gov Identifier: NCT01479452.