Browsing by Subject "Predictors"

Now showing 1 - 6 of 6
  • Thumbnail Image
    Item
    Open Access
    Household illness, poverty and physical and emotional child abuse victimisation: findings from South Africa's first prospective cohort study
    (Biomed Central Ltd, 2015) Meinck, Franziska; Cluver, Lucie; Boyes, Mark
    BACKGROUND:Physical and emotional abuse of children is a large scale problem in South Africa, with severe negative outcomes for survivors. Although chronic household illness has shown to be a predictor for physical and emotional abuse, no research has thus far investigated the different pathways from household chronic illness to child abuse victimisation in South Africa. METHODS: Confidential self-report questionnaires using internationally utilised measures were completed by children aged 10-17 (n=3515, 56.7% female) using door-to-door sampling in randomly selected areas in rural and urban locations of South Africa. Follow-up surveys were conducted a year later (96.7% retention rate). Using multiple mediation analyses, this study investigated direct and indirect effects of chronic household illness (AIDS or other illness) on frequent (monthly) physical and emotional abuse victimisation with poverty and extent of the ill person's disability as hypothesised mediators. RESULTS: For children in AIDS-ill families, a positive direct effect on physical abuse was obtained. In addition, positive indirect effects through poverty and disability were established. For boys, a positive direct and indirect effect of AIDS-illness on emotional abuse through poverty were detected. For girls, a positive indirect effect through poverty was observed. For children in households with other chronic illness, a negative indirect effect on physical abuse was obtained. In addition, a negative indirect effect through poverty and positive indirect effect through disability was established. For boys, positive and negative indirect effects through poverty and disability were found respectively. For girls, a negative indirect effect through poverty was observed. CONCLUSIONS: These results indicate that children in families affected by AIDS-illness are at higher risk of child abuse victimisation, and this risk is mediated by higher levels of poverty and disability. Children affected by other chronic illness are at lower risk for abuse victimisation unless they are subject to higher levels of household disability. Interventions aiming to reduce poverty and increase family support may help prevent child abuse in families experiencing illness in South Africa.
  • Thumbnail Image
    Item
    Open Access
    Independent predictors of tuberculosis mortality in a high HIV prevalence setting: a retrospective cohort study
    (2015) Pepper, Dominique J; Schomaker, Michael; Wilkinson, Robert J; de Azevedo, Virginia; Maartens, Gary
    BackgroundIdentifying those at increased risk of death during TB treatment is a priority in resource-constrained settings. We performed this study to determine predictors of mortality during TB treatment.MethodsWe performed a retrospective analysis of a TB surveillance population in a high HIV prevalence area that was recorded in ETR.net (Electronic Tuberculosis Register). Adult TB cases initiated TB treatment from 2007 through 2009 in Khayelitsha, South Africa. Cox proportional hazards models were used to identify risk factors for death (after multiple imputations for missing data). Model selection was performed using Akaike’s Information Criterion to obtain the most relevant predictors of death.ResultsOf 16,209 adult TB cases, 851 (5.3%) died during TB treatment. In all TB cases, advancing age, co-infection with HIV, a prior history of TB and the presence of both pulmonary and extra-pulmonary TB were independently associated with an increasing hazard of death. In HIV-infected TB cases, advancing age and female gender were independently associated with an increasing hazard of death. Increasing CD4 counts and antiretroviral treatment during TB treatment were protective against death. In HIV-uninfected TB cases, advancing age was independently associated with death, whereas smear-positive disease was protective.ConclusionWe identified several independent predictors of death during TB treatment in resource-constrained settings. Our findings inform resource-constrained settings about certain subgroups of TB patients that should be targeted to improve mortality during TB treatment.
  • Thumbnail Image
    Item
    Open Access
    Investigating the nature and prevalence of mobile bullying in the rural Eastern Cape Province of South Africa
    (2021) Mtshazi, Sive; Kyobe, Michael
    Limited studies have examined mobile bullying in South African schools in general and in rural schools in particular. The focus on rural schools is important to our understanding of the nature and forms of bullying. Studies conducted in other parts of the world suggest that bullying differs among urban, suburban, and rural schools. Evidence from studies on physical violence suggests that violence also takes place in many rural schools. Since rural areas are reported to have the highest level of mobile phone usage in South Africa, it is possible that mobile bullying may be more predominant in the rural communities and its implications severe. Using a survey of 984 students, the present study investigated the nature of mobile bullying in 7 rural high schools of the Eastern Cape Province of South Africa. It sought to identify the factors that mainly predict this aggression and the data was analysed using quantitative methods. The findings show that males are more involved in mobile bullying than females. Gender, mobile victimization, frequent usage of the mobile phone and retaliation were the main predictors of mobile bullying. This study also revealed that the influence of anonymity of the bully on mobile bullying is not necessarily direct. This influence is moderated by other factors such as the safety risk and the economic environment of school.
  • Thumbnail Image
    Item
    Open Access
    Predictors of consent to cell line creation and immortalisation in a South African schizophrenia genomics study
    (BioMed Central, 2018-07-11) Campbell, Megan M; de Vries, Jantina; Mqulwana, Sibonile G; Mndini, Michael M; Ntola, Odwa A; Jonker, Deborah; Malan, Megan; Pretorius, Adele; Zingela, Zukiswa; Van Wyk, Stephanus; Stein, Dan J; Susser, Ezra
    Background Cell line immortalisation is a growing component of African genomics research and biobanking. However, little is known about the factors influencing consent to cell line creation and immortalisation in African research settings. We contribute to addressing this gap by exploring three questions in a sample of Xhosa participants recruited for a South African psychiatric genomics study: First, what proportion of participants consented to cell line storage? Second, what were predictors of this consent? Third, what questions were raised by participants during this consent process? Methods 760 Xhose people with schizophrenia and 760 controls were matched to sex, age, level of education and recruitment region. We used descriptive statistics to determine the proportion of participants who consented to cell line creation and immortalization. Logistic regression methods were used to examine the predictors of consent. Reflections from study recruiters were elicited and discussed to identify key questions raised by participants about consent. Results Approximately 40% of participants consented to cell line storage. The recruiter who sought consent was a strong predictor of participant’s consent. Participants recruited from the South African Eastern Cape (as opposed to the Western Cape), and older participants (aged between 40 and 59 years), were more likely to consent; both these groups were more likely to hold traditional Xhosa values. Neither illness (schizophrenia vs control) nor education (primary vs secondary school) were significant predictors of consent. Key questions raised by participants included two broad themes: clarification of what cell immortalisation means, and issues around individual and community benefit. Conclusions These findings provide guidance on the proportion of participants likely to consent to cell line immortalisation in genomics research in Africa, and reinforce the important and influential role that study recruiters play during seeking of this consent. Our results reinforce the cultural and contextual factors underpinning consent choices, particularly around sharing and reciprocity. Finally, these results provide support for the growing literature challenging the stigmatizing perception that people with severe mental illness are overly vulnerable as a target group for heath research and specifically genomics studies.
  • Thumbnail Image
    Item
    Open Access
    Prevalence, comorbidity and predictors of anxiety disorders in children and adolescents in rural north-eastern Uganda
    (BioMed Central Ltd, 2013) Abbo, Catherine; Kinyanda, Eugene; Kizza, Ruth; Levin, Jonathan; Ndyanabangi, Sheilla; Stein, Dan
    BACKGROUND:Child and adolescent anxiety disorders are the most prevalent form of childhood psychopathology. Research on child and adolescent anxiety disorders has predominantly been done in westernized societies. There is a paucity of data on the prevalence, comorbidity, and predictors of anxiety disorders in children and adolescents in non-western societies including those in sub-Saharan Africa. This paper investigates the prevalence, comorbidity, and predictors of anxiety disorders in children and adolescents in north-eastern Uganda.OBJECTIVE:To determine the prevalence of DSM-IV anxiety disorders, as well as comorbidity patterns and predictors in children and adolescents aged 3 to 19 years in north-eastern Uganda. METHODS: Four districts (Lira, Tororo, Kaberamaido and Gulu) in rural north-eastern Uganda participated in this study. Using a multi-stage sampling procedure, a sample of 420 households with children aged 3-19 years from each district was enrolled into the study. The MINI International Neuropsychiatric Interview for children and adolescents (MINI KID) was used to assess for psychiatric disorders in 1587 of 1680 respondents. RESULTS: The prevalence of anxiety disorders was 26.6%, with rates higher in females (29.7%) than in males (23.1%). The most common disorders in both males and females were specific phobia (15.8%), posttraumatic stress disorder (PTSD) (6.6%) and separation anxiety disorder (5.8%). Children below 5 years of age were significantly more likely to have separation anxiety disorder and specific phobias, while those aged between 14-19 were significantly more likely to have PTSD. Anxiety disorders were more prevalent among respondents with other psychiatric disorders; in respondents with two or more co-morbid psychiatric disorders the prevalence of anxiety disorders was 62.1%. Predictors of anxiety disorders were experience of war trauma (OR=1.93, p<0.001) and a higher score on the emotional symptom scale of the SDQ (OR=2.58, p<0.001). Significant socio-demograghic associations of anxiety disorders were found for female gender, guardian unemployment, living in permanent housing, living without parents, and having parents without education. CONCLUSION: The prevalence of anxiety disorders in children and adolescents in rural north-eastern Uganda is high, but consistent in terms of gender ratio and progression over time with a range of prior work in other contexts. Patterns of comorbidity and predictors of anxiety disorders in this setting are also broadly consistent with previous findings from western community studies. Both psychosocial stressors and exposure to war trauma are significant predictors of anxiety disorders.Prevention and treatment strategies need to be put in place to address the high prevalence rates of anxiety disorders in children and adolescents in Uganda.
  • Thumbnail Image
    Item
    Open Access
    Rate of psychiatric readmissions and associated factors at Saint John of God Psychiatric Hospital in Mzuzu, Malawi
    (2019) Msiska, Manson Mwachande; Sorsdahl, Katherine; Holzer, Stefan
    Background: Globally, studies have established that 40-50% of psychiatric patients with SMDs are readmitted within one year of discharge from the acute hospital admission. Lowand middle-income countries (LMICs) such as Malawi have also reported high rates of psychiatric readmissions. This poses challenges when providing psychiatric care to patients. Most of Malawi`s health institutions, including Saint John of God Psychiatric Hospital (SJOG), rely primarily on donor funding. In order to maximise the available donor funding, there is a need to reduce readmissions resulting from modifiable or controlled factors. There are no studies in Malawi which have investigated these risk factors. The study aimed to establish the frequency of readmissions and the associated factors among patients at SJOG Psychiatric Hospital in Mzuzu, Malawi. The specific areas examined were sociodemographic and clinical-related factors associated with readmission. Methods: This was a retrospective cohort case record review study. Two hundred and seventy five clinical files of patients admitted for the first time at SJOG Psychiatric Hospital Mzuzu, Malawi between 1 January, 2014 and 31 December, 2015 were extracted. Data on socio-demographics and clinical information were collected using an extraction sheet at 3, 6 and 12 months post-discharge from the acute (first) hospital admission. Logistic regression models were developed to investigate the associations between socio-demographics, clinicalrelated factors and readmissions. Ethical approval for this study was granted by the Faculty of Health Sciences Human Research Ethics Committee at the University of Cape Town. Approval to conduct this research in Malawi was obtained from the National Health Sciences Research Ethics Committee. Results: Readmission rates of 1.5%, 4.4%, and 11.3% were found within the 3, 6 and 12 months of discharge from the acute hospital admission respectively. None of the independent variables predicted readmission within the 3 month of discharge from the acute hospital admission. In the unadjusted logistic regression model, having children (OR=0.26, 95% C.I 0.07-0.96) protected against readmissions within the 6 month of follow-up period. In the unadjusted logistic regression model, having children (OR= 0.40, 95% C.I 0.18-0.88), staying outside the hospital catchment area (OR=0.44, 95% C.I 0.20-0.96), and having insight (OR=0.22, 95% C.I 0.10-0.49) into their illness were protective factors to readmission, while taking SGAs (OR=4.67, 95% C.I 1.33-16.39) predicted readmission within the 12 month follow-up period. After adjusting for age and gender in the multivariable analysis, staying outside catchment area (OR=0.33, 95% C.I 0.14-0.79) and having insight (OR=0.19, 95% C.I 0.08-0.46) to their illness were protective factors, while taking SGAs (OR=5.29, 95% C.I 1.43-19.51) remained a predictor of readmission within 12 months of discharge from the acute admission. Conclusion: The findings of this study demonstrated that readmissions are associated with socio-demographic and clinical factors such as catchment area, patient insight into their condition and type of antipsychotics. The study identifies the need to develop interventions targeting the groups at risk of being readmitted.