BACKGROUND: Osteoarthritis (OA) is considered an established risk factor for falls. Published studies evaluating secondary falls prevention strategies among individuals with OA are limited. OBJECTIVE: To evaluate the effect of a personalized home-based exercise program to improve postural balance, fear of falling, and falls risk in older fallers with knee OA and gait and balance problems. DESIGN: Randomized controlled trial. SETTING: University of Malaya Medical Centre. PARTICIPANTS: Fallers who had both radiological OA and a Timed Up and Go (TUG) score of over 13.5 seconds. MAIN OUTCOME MEASURE: Postural sway (composite sway) was quantified with the Modified Clinical Test of Sensory Interaction on Balance (mCTSIB) under 4 different sensory conditions: eyes open on firm surface, eyes closed on firm surface, eyes open on unstable foam surface, and eyes closed on unstable foam surface. Participants were asked to stand upright and to attempt to hold their position for 10 seconds for each test condition. The average reading for all conditions were calculated. METHODS: Participants randomized to the intervention arm received a home-based modified Otago Exercise Program (OEP) as part of a multifactorial intervention, whereas control participants received general health advice and conventional treatment. This was a secondary subgroup analysis from an original randomized controlled trial, the Malaysian Falls Assessment and Intervention Trial (MyFAIT) (trial registration number: ISRCTN11674947). Posturography using a long force plate balance platform (Balancemaster, NeuroCom, USA), the Knee injury and Osteoarthritis Outcome Score (KOOS) and the short-form Falls Efficacy Scale-International (short FES-I) were assessed at baseline and 6 months. RESULTS: Results of 41 fallers with radiological evidence of OA and impaired TUG (intervention, 17; control, 24) were available for the final analysis. Between-group analysis revealed significant improvements in the Modified Clinical Test of Sensory Interaction on Balance (mCTSIB), Limits of Stability (LOS), and short FES-I scores by the intervention group compared to the control group at 6 months. No significant difference in time to first fall or in fall-free survival between the intervention and control groups was found. CONCLUSION: Home-based balance and strength exercises benefited older fallers with OA and gait and balance disorders by improving postural control, with no observable trend in reduction of fall recurrence. Our findings will now inform a future, adequately powered, randomized controlled study using fall events as definitive outcomes. LEVEL OF EVIDENCE: I.