Abstract Background Behavioural and psychological symptoms are associated with dementia, but are also present in a significant number of the older population without dementia. Here we explore the distribution of behavioural and psychological symptoms in the population without dementia, and their relationship with domains and severity of health and cognitive impairment. Methods The Medical Research Council Cognitive Function and Ageing Study is a two-phase longitudinal study of ageing representative of the population aged 65 and over of England and Wales. A subsample of 1781 participants without a study diagnosis of dementia was included in this study. Information on symptoms including depression, apathy, anxiety, feelings of persecution, hallucination, agitated behaviour, elation, irritability, sleep problems, wandering, confabulation and misidentification, cognitive function, health related factors and socio-demographic information was extracted from interviews with participants and knowledgeable informants. Participants were classified according to the Mini-Mental State Examination and by criteria for subtypes of mild cognitive impairment (MCI). The prevalence of behavioural and psychological symptoms and associations with cognitive function, health and socio-demographics was examined. Co-occurrence of symptoms was tested using factor analysis. Results Most symptoms were reported more frequently in those with more severe cognitive impairment. Subjective memory complaints were the strongest independent predictor of reported symptoms, and most were reported more often in those classified as having MCI than in those with cognitive impairments that did not meet the MCI criteria. The pattern of co-occurrence of symptoms is similar to that seen in dementia. Conclusions Our results highlight that behavioural and psychological symptoms are prevalent in the cognitively impaired older population, and partly explain the variation observed in previous cohorts of individuals with MCI. Behavioural and psychological symptoms offer a target for intervention and so are an important consideration in the assessment of cognitively impaired older people.