Background: Neuro-imaging is relatively new in psychiatry. Although the actual role of neuro-imaging in psychiatry remains unclear, it is used to strengthen clinical evidence in making psychiatric diagnoses.Aim: To analyse the records of inpatients referred for neuro-imaging (computerised tomography [CT] and/or magnetic resonance imaging [MRI] scans) to determine the proportion of abnormal neuro-imaging results and, if any, factors associated with abnormal neuro-imaging results.Setting: This study was conducted at the Charlotte Maxeke Johannesburg Academic Hospital (CMJAH) situated in Johannesburg, South Africa.Methods: This was a quantitative retrospective record review. All adult psychiatric inpatients who had undergone a CT and/or MRI scan during 01 January 2014 to 31 December 2015 were included. Out-patients or patients admitted in the medical wards were excluded from the study. All neuro-imaging referrals were identified from hospital records and their demographics, scan characteristics and diagnoses were subsequently captured.Results: A total of 1040 patients were admitted to the CMJAH psychiatric unit, of which 213 (20.5%) underwent neuro-imaging tests. Of the 213 scans performed, 74 were abnormal, representing a yield of 34.7%. The most common reported pathology was atrophy (n = 22, 29.7%). There was no statistically significant association between age group (χ2 = 3.9, p = 0.8), gender (χ2 = 1.3; p = 0.5), psychiatric diagnoses and abnormal scans. However, there were trends towards an association with comorbid HIV infection (χ2 = 3.476, p = 0.062) and comorbid substance abuse (χ2 = 2.286, p = 0.091).Conclusion: This study supports the need for clear clinical indications to justify the cost-effective use of neuro-imaging in psychiatry. This study’s high yield of abnormal CT scans, although similar to other studies, advocates that HIV positive testing and the presence of focal neurological signs will improve the yield further.