AIM: To test the effectiveness of an organisational change intervention aimed at increasing the offer of nicotine replacement therapy (NRT) in community managed mental health organisations. DESIGN: A pragmatic cluster randomised controlled trial with cluster as the unit of randomisation and six- and nine-month follow-up from baseline. SETTING: Twelve clusters comprising 26 sites providing community based, psychosocial support to people with severe mental illness in New South Wales, Australia, were randomised to control (n = 13 sites, n = 118 consumers) or intervention (n = 13 sites, n = 139 consumers) arms between 2018 and 2019. PARTICIPANTS: Eligible consumers (aged 16 years and older
self-reported daily or occasional cigarette use) completed surveys at baseline (n = 257) and at six- (n = 162, 63%) and nine-month follow-up (n = 144, 56%). INTERVENTION: The intervention included a financial grant, face-to-face and on-line training and proactive monthly support to guide implementation. The active control condition included on-line training and generic, scheduled support via email. MEASUREMENTS: The primary outcome was whether consumers reported receiving an offer of NRT at nine-month follow-up. Secondary outcomes at the consumer, staff and organisational level were also measured. FINDINGS: Consumers in the intervention group had statistically significantly higher odds of being offered NRT at nine-month follow-up compared with control (intention to treat missing = no offer: 38% versus 7%, odds ratio 5.72, 95% confidence interval = 2.2, 14.9). There were no statistically significant differences in seven-day point prevalence or continuous abstinence at six- or nine-month follow-ups. CONCLUSIONS: An organisational change-based program led to an increase in the offer of nicotine replacement therapy (NRT) nine months after program initiation in community managed mental health organisations, compared with active control. There was evidence of greater NRT use in the intervention condition at nine months but no evidence of differences on abstinence measures at six or nine months.