Abstract Objectives To evaluate 3-year outcomes of following a treat-to-target (T2T) strategy targeting low disease activity for patients with elderly-onset RA (EORA) and to confirm safety profile of T2T. Methods Treatment was adjusted to target low disease activity with conventional synthetic DMARDs, followed by biologic DMARDs (bDMARDs) in 197 MTX-naïve EORA patients (mean age 74.9 years) with moderate-to-high disease activity. Non-implementation of T2T was evaluated at week 12, 24, 36, 52, 76, 104 and 128. To evaluate risks of using MTX, bDMARDs and glucocorticoids, 2122 periods of 3 months each were analysed using Bayesian hierarchical logistic regression models. Results Of the patients, 84.7% received methotrexate, 34.0% glucocorticoids with DMARDs and 41.6% bDMARDs during the observation period. Sixty-nine of the 197 patients failed to adhere to T2T because of comorbidities or the patient’s own decision: 33 failed once, 19 twice, 10 three times and 6 four times or more. Simplified disease activity index (SDAI) remission and HAQ Disability Index (HAQ-DI) ≤0.5 at 3 years were achieved in 57.8% and 70.3% of the 128 patients adhering to T2T, and 34.8% and 43.5% of the 69 patients who did not adhere to T2T, respectively, and these were significantly different. Eighty-nine serious adverse events (SAEs) of any type were reported in 61 patients. MTX, bDMARDs and glucocorticoid were not associated with SAEs when adjusted for mean SDAI during the observation period and comorbidities at baseline. Conclusion T2T strategy for EORA by using MTX and bDMARDs was effective with an acceptable safety profile. Adhering to T2T led to better outcomes.