Background - Falls cause fear, anxiety and loss of confidence, resulting in activity avoidance, social isolation and increasing frailty. The umbrella term for these problems is ‘fear of falling’, seen in up to 85% of older adults who fall. Evidence of effectiveness of physical and psychological interventions is limited, with no previous studies examining the role of an individually delivered cognitive–behavioural therapy (CBT) approach. Objectives - Primary objective To develop and then determine the effectiveness of a new CBT intervention (CBTi) delivered by health-care assistants (HCAs) plus usual care compared with usual care alone in reducing fear of falling. Secondary objectives To measure the impact of the intervention on falls, injuries, functional abilities, anxiety/depression, quality of life, social participation and loneliness; investigate the acceptability of the intervention for patients, family members and professionals and factors that promote or inhibit its implementation; and measure the costs and benefits of the intervention. Results - Four hundred and fifteen patients were recruited, with 210 patients randomised to CBTi group and 205 to the control group. There were significant reductions in mean FES-I [–4.02; 95% confidence interval (CI) –5.95 to –2.1], single-item numerical fear of falling scale (–1.42; 95% CI –1.87 to 1.07) and HADS (–1; 95% CI –1.6 to –0.3) scores at 12 months in the CBTi group compared with the usual care group. There were no differences in the other secondary outcome measures. Most patients found the CBTi acceptable. Factors affecting the delivery of the CBTi as part of routine practice were identified. There was no evidence that the intervention was cost-effective. Conclusions - Our new CBTi delivered by HCAs significantly improved fear of falling and depression scores in older adults who were attending falls services. There was no impact on other measures. Further work - Further work should focus on a joint CBTi and physical training approach to fear of falling, more rational targeting of CBTi, the possibility of mixed group and individual CBTi, and the cost-effectiveness of provision of CBTi by non-specialists.