Mental health symptoms among homeless shelter residents during COVID-19 lockdown in Tshwane, South Africa.

BACKGROUND: In order to contain the spread of COVID-19 in South Africa during the national state of emergency, the Gauteng Department of Social Development established temporary shelters and activated existing facilities to provide basic needs to street-homeless people in Tshwane, which facilitated primary health care service-delivery to this community. AIM: This study aimed to determine and analyse the prevalence of mental health symptoms and demographic characteristics among street-homeless people living in Tshwane's shelters during lockdown. SETTING: Homeless shelters set up in Tshwane during level 5 of the COVID-19 lockdown in South Africa. METHODS: A cross-sectional, analytical study was conducted using a Diagnostic and Statistical Manual of Mental Disorders (DSM-5)-based questionnaire that looked at 13 mental health symptom domains. RESULTS: Presence of moderate-to-severe symptoms were reported among the 295 participants as follows: substance use 202 (68%), anxiety 156 (53%), personality functioning 132 (44%), depression 85 (29%), sleep problems 77 (26%), somatic symptoms 69 (23%), anger 62 (21%), repetitive thoughts and behaviours 60 (20%), dissociation 55 (19%), mania 54 (18%), suicidal ideation 36 (12%), memory 33 (11%) and psychosis 23 (8%). CONCLUSION: A high burden of mental health symptoms was identified. Community-oriented and person-centred health services with clear care-coordination pathways are required to understand and overcome the barriers street-homeless people face in accessing health and social services.Contribution: This study determined the prevalence of mental health symptoms within the street-based population in Tshwane, which has not previously been studied.

Managing acute opioid withdrawal with tramadol during COVID-19 lockdown in a peri-urban setting.

BACKGROUND: The coronavirus disease 2019 (COVID-19) has highlighted the scope of heroin dependence and need for evidence-based treatment amongst marginalised people in South Africa. Acute opioid withdrawal management without maintenance therapy carries risks of increased morbidity and mortality. Due to the high costs of methadone, Tshwane's Community Oriented Substance Use Programme (COSUP) used tramadol for opioid withdrawal management during the initial COVID-19 response. AIM: To describe demographics, route of heroin administration and medication-related experiences amongst people accessing tramadol for treatment of opioid withdrawal. SETTING: Three community-based COSUP sites in Mamelodi (Tshwane, South Africa). METHODS: A retrospective cross-sectional study was conducted. Data were collected using an interviewer-administered paper-based tool between April and August 2020. Descriptive statistics were used to analyse data. RESULTS: Of the 220 service users initiated onto tramadol, almost half (n = 104, 47%) were not contactable. Fifty-eight (26%) people participated, amongst whom most were male (n = 55, 95%). Participants' median age was 32 years. Most participants injected heroin (n = 36, 62.1%). Most participants experienced at least one side effect (n = 47, 81%) with 37 (64%) experiencing two or more side effects from tramadol. Insomnia occurred most frequently (n = 26, 45%). One person without a history of seizures experienced a seizure. Opioid withdrawal symptoms were experienced by 54 participants (93%) whilst taking tramadol. Over half (n = 38, 66%) reported using less heroin whilst on tramadol. CONCLUSION: Tramadol reduced heroin use but was associated with withdrawal symptoms and unfavourable side effects. Findings point to the limitations of tramadol as opioid withdrawal management to retain people in care and the importance of access to first-line opioid agonists.Contribution: This research contributes to the limited data around short-acting tramadol for opioid withdrawal management in the African context, with specific focus on the need for increased access to opioid agonists for those who need them, in primary care settings.