Catatonia is a psychomotor abnormality caused by neurological, medical or severe psychiatric disorders and substances. Its prevalence ranges from less than 10% to just above 60%. Diagnosis may be influenced by the screening tools used. Screening of new admissions to a mental health unit for catatonia was undertaken using three instruments to determine prevalence of catatonia. Participants ranged from age 16 years and over. Recruitment took place from September 2020 to August 2021. The setting was a mental health unit within a general hospital in Nelson Mandela Metro, South Africa. Five assessors were trained by the principal investigator to apply the Bush Francis Screening Instrument (BFCSI), the Bush Francis Catatonia Rating Scale (BFCRS), and the Diagnostic and Statistical Manual 5 (DSM-5), to assess participants. Clinical and demographic data were collected using a specially designed datasheet. Data analysis was performed to identify significant associations between presence or absence of catatonia and clinical and demographic data. Up to 241 participants were screened and 44 (18.3%) had catatonia. All 44 cases were identified through the BFCSI while the DSM-5 identified only 16 (6.6%%) of the 44 cases even though the remaining 28 (63.6%) participants still required treatment for catatonic symptoms. The DSM-5 diagnostic criteria excluded staring, which was the commonest sign of catatonia identified through the BFCSI [n = 33 (75%)]. Close to half (21; 47.7%) of those with catatonia on the BFCSI had schizophrenia. The rest had bipolar disorder (12; 27.3%), substance-induced psychotic disorder (7; 15.9%) and no specified diagnosis in one (1; 2.6%). The BFCSI was very effective at identifying catatonia while the DSM-5 was inadequate, missing close to 64% (28 of 44) of cases. Predictors of catatonia in this sample were a younger age and being male. A prevalence of 18.3%, indicates that assessment for catatonia should be routinely conducted in this and similar settings.