We previously reported that 24% of 113 rheumatoid arthritis (RA) patients had hearing impairment (HI). We investigated if disease activity was a predictor of incidental HI. One hundred and four patients completed three consecutive 6 months-apart rheumatic evaluations and concomitant audiometric evaluations which included at least an interview, an otoscopic evaluation, and a pure tone audiometry. HI was defined if the average thresholds for at least one of low-, mid-, or high-frequency ranges were ≥25 decibels (dB) hearing level in one or both ears. Appropriated statistics was used. Internal review board approval was obtained. Patients were most frequently middle-aged (43.4 ± 13.3 years), female (89.4%), and had median disease duration of 5 years and low disease activity. All were receiving RA treatment. At inclusion, 24 patients had HI which was sensorineural in 91.7% of them. Among the 80 patients without HI at baseline, 10 (12.5%) developed incidental HI, and they had more disease activity either at baseline ([median, range] disease activity score-28 joints evaluated-C-reactive protein [DAS28-CRP], 3.9 [1.6-7.3] vs. 2.1 [1-8.7], p = 0.006) or cumulative previous incidental HI (3.4 [1.8-4.8] vs. 2 [1-6.2], p = 0.007) and were more frequently on combined methotrexate and sulfasalazine (20 vs. 1.4%, p = 0.05) than their counterparts. In the adjusted Cox proportional model, cumulative DAS28-CRP was the only variable to predict incidental HI (odds ratio, 1.8; 95% confidence interval, 1.1-2.7; p = 0.01). Almost 13% of RA patients with short disease duration and low disease activity developed incidental HI during 1 year. Cumulative disease activity predicted incidental HI.