Background Despite policy pressure and concerns regarding the use of antipsychotics and benzodiazepines, many care home residents with dementia are prescribed psychotropic medication, often off licence. This is the first large study to report psychotropic prescribing and ‘as required’ administration patterns in English care homes. Aims To explore the prevalence and associates of psychotropic prescription in care home residents with dementia and compare the results with national guidance. Method We collected data in a longitudinal cohort study of residents with diagnosed or probable dementia in 86 care homes in England in 2014–2016. We reported the prevalence of psychotropic (antipsychotics, anxiolytics/hypnotics, antidepressants) prescriptions and drug receipt. We explored the associations between resident factors (sociodemographic, agitation [Cohen–Mansfield Agitation Inventory], dementia severity [Clinical Dementia Rating]) and care home factors (type, ownership, size, dementia registration/specialism, quality rating) in prescription and ‘as required’ administration, using multilevel regression models. Results We analysed data from 1425 residents. At baseline, 822 residents (57.7%, 95% CI: 55.1–60.2) were prescribed a psychotropic drug, 310 residents (21.8% 95% CI: 19.7–24.0) were prescribed an anxiolytic/hypnotic, 232 (94.3%, 95% CI: 90.6–96.6) were prescribed one antipsychotic and 14 (5.7%, 95% CI: 3.4–9.4) were prescribed two antipsychotics. The median prescription duration during the study was 1 year. Residents with clinically significant agitation were prescribed more antipsychotics (odds ratio [OR] = 2.00, 95% CI: 1.64–2.45) and anxiolytics/hypnotics (OR = 2.81, 95% CI: 2.31–3.40). Conclusions Antipsychotics and anxiolytics/hypnotics are more commonly prescribed for people with dementia in care homes than in the community, and prescribing may not reflect guidelines. Policies which advocate reduced use of psychotropics should better support psychosocial interventions.