An assessment of voting knowledge and related decisions amongst hospitalised mental healthcare users in South Africa

BACKGROUND: The South African Constitution protects the right to vote for every citizen. The Electoral Act (No. 73 of 1998) limits registration on the voter’s roll on the basis of being declared of ‘unsound mind’ or ‘mentally disordered’ by the high court or detention under the Mental Health Care Act (No. 17 of 2002). There is limited information regarding voting knowledge and subsequent voting-related decisions amongst South African involuntary mental healthcare users (MHCUs). AIM: To compare voting knowledge and related decisions between hospitalised MHCUs and non-psychiatric hospitalised patients (controls). SETTING: Participants were recruited from Sterkfontein Psychiatric Hospital (MHCUs) and Chris Hani Baragawanth Academic Hospital orthopaedic wards (controls) in Gauteng, South Africa. METHOD: A cross-sectional survey was conducted using a modified Cognitive Assessment Tool for Voting (MCAT-V) questionnaire. Scores on the MCAT-V were compared between the MHCU and control groups, along with socio-demographic variables and clinical variables. RESULTS: There was a significant association between group (MHCU vs. control) and HLOE (p = 0.016). Although the median overall score for the controls (11; interquartile range [IQR] 10–12) was significantly higher than that for the MHCUs (10; IQR 8–12) (p = 0.043), when controlling for education level, there was no significant association between group (MHCU/control) and MCAT-V scores (p = 0.011). The MCAT-V scores of the ‘Doe questions’ between the MHCUs and controls were not significantly different (p = 0.063). There was a difference in ‘reasoning scores’ between MHCUs and controls (p = 0.0082) and this was associated with level of educational attainment (p = 0.013). CONCLUSION: The limitations regarding voter registration legislated in the South African Electoral Act, are not supported by the findings of this study. The MCAT-V demonstrates a possible educational bias and therefore is not recommended as a screening tool for assessing voting competency.