Although treatment default by psychiatric patients or mental health care users is a global challenge, this behavior is reported to be higher in South Africa. The Manguzi District Hospital in rural Kwa-Zulu Natal Province, South Africa, experiences high rates of treatment default by psychiatric patients. The objective of this study was to determine the reasons for treatment defaulting at Manguzi Hospital, KwaZulu-Natal Province, South Africa. An explorative qualitative design, using in-depth interviews, was conducted with mental health care users who had defaulted out-patient psychiatric treatment. Twenty-one mental health care users were interviewed before data saturation was reached. Nvivo version 11 was used to analyze the qualitative data. Major themes that emerged confirmed that social factors are key contributions to treatment defaulting, and these include denial of the mental disorders; belief that they are cured; lack of, or disintegration of social support; preference for traditional medicine; and flaws in the health care system. Social determinants of treatment outcomes for mental disorders require tailor-made support systems for patients in these rural communities, which include increase in health literacy and attention to the cultural understanding of mental disorders.

Jabulile Ndlovu1 and Kebogile Mokwena1, 2 1 Department of Public Health, Sefako Makgatho Health Sciences University, Pretoria, South Africa 2 Research Chair in Substance abuse and population Mental Health, Sefako Makgatho Health Sciences University, Pretoria, South Africa Corresponding author: jabulilendlovu5@gmail.com Abstract An estimated 6% of the world population has serious mental illness, with one in four families having a member with some form of psychiatric disorder, who are mostly cared for by their relatives within a family setting. Although caregiving in a home setting is reported to be associated with significant mental distress, the burden of such distress is rarely measured. The purpose of this study was to quantify the burden of care among family caregivers of relatives with serious mental disorders, as well as to explore possible association between the caregiver burden of care and a range of caregiver and Mental Health Care User (MHCU) variables in a rural district in Kwa-Zulu Natal, South Africa. The Zarit Burden Interview (ZBI) scale was used to collect data from 357 caregivers, and STATA 14 was used to analyze data. The ages of the sample ranged from 18 to 65, with a mean of 50.29, and the majority (86%) were female and unemployed (83%). The ZBI scores ranged from 8 to 85, with a mean of 41.59. The majority (91%) tested positive for caregiver burden of care, which ranged from mild to severe. Using the Pearson Chi-square test of association (p=0.05), variables that were significantly associated with the burden of care were clinically related (caregiver self-reported depression, MHCU diagnosis, recent relapse of the MHCU), socio-economic (caregiver family monthly income, MHCU disability grant status and MHCU employment status) and socio-demographic (MHCU gender and MHCU level of education). The prevalence of burden of care is high and severe, and scarcity of resources in families and communities contribute to the high burden of care in these rural communities.