Masters Degrees (Epidemiology and Biostatistics)


Recent Submissions

Now showing 1 - 5 of 16
  • Item
    A secondary analysis using individual patient data of two pragmatic cluster-randomized control trials evaluating 3 monthly and 6 monthly community-based multimonth dispensing of antiretroviral treatment in Southern Africa
    (Stellenbosch : Stellenbosch University, 2021-03) Lopes, John; Fatti, Geoffrey; Lombard, Carl J.; Stellenbosch University. Faculty of Medicine and Health Sciences. Dept. of Global Health. Epidemiology and Biostatistics.
    ENGLISH SUMMARY: Randomized evidence of the effectiveness of community-based multi-month dispensing (MMD) of antiretroviral treatment (ART) is lacking, particularly for 6 months MMD with only annual clinical consultations. Data from two cluster randomized trials (CRTs) were pooled to compare community-based MMD of ART versus non-inferior to standard-of-care facility-based MMD. Methods: Adult people living with HIV (PLHIV) stable on ART for ≥6 months with viral load suppression (VLS) at baseline was included. Community-based 3 monthly (3MC) and 6 monthly (6MC) ART refill dispensing were compared to control facility-based 3 monthly (3MF). Twelve months retention-in-care (primary outcome) was evaluated by intention-to-treat using individual-level regression analyses, with a prespecified non-inferiority margin of -3.25% risk difference (RD). Secondary outcomes evaluated after 12 months were; (1) retention within the study arm (2) VLS, (3) number of unscheduled facility visits, and after 18 months; (4) attrition in ART care. Results: Data pooling yielded a total of 10136 participants with relatively balanced characteristics across the arms, except for age and district. Retention after 12 months was noninferior in the adjusted analysis (3MC: RD=0.3, 95% CI: -0.8 to 1.4 vs 3MF, 6MC: RD=-0.2, 95 % CI: -1.4 to 1.0 vs 3MF). VLS was high (≥97.9%) with no differences between the arms, risk ratio (RR) of 1.0 for 3MC (95 % CI: 1.0 to 1.0, p=0.885) and 6MC (95 % CI: 1.0 to 1.0, p=0.186) compared to 3MF. The incidence rate ratio (IRR) for unscheduled visits showed no difference between 3MC (IRR=0.6, 95 % CI: 0.2 to 2.1, p=0.383) and 6MC (IRR=1.0, 95 % CI: 0.4 to 2.5, p=0.974) compared to 3MF. Participant attrition (0-18 months) for Zimbabwe showed no differences between the arms, facility location or healthcare level. Conclusion: Community-based MMD of ART at 3 and 6 monthly refills for PLHIV stable on ART is safe to scale-up, without increased unscheduled facility visits or compromise in VLS.
  • Item
    The impact of caring for a school-going visually impaired child in Cape Town, Western Cape on the home caregiver and the rest of the family – the health, socio-economic and psychological health burdens
    (Stellenbosch : Stellenbosch University, 2021-03) St. Jerry, Marlyse; Barnes, J. M. (Johanna Maria); Nyasule, Peter S; Stellenbosch University. Faculty of Medicine and Health Sciences. Dept. of Global Health. Epidemiology and Biostatistics.
    ENGLISH SUMMARY : Background: Visually impaired (VI) children need additional care for daily tasks due to their diminished independent living skills (ILS). In learners in Special Educational Needs (LSEN) schools and specially-adapted mainstream schools, occupational therapy and trained assistants help to bridge the gap that diminished ILS pose to their education. However, in impoverished areas of the Cape Metropole, many families cannot afford trained caregivers in their homes. Typically a maternal figure fulfils the responsibilities of caring for the VI child as well as for the rest of the household; such a person then becomes the VI child’s primary home caregiver. Aims and Objectives: The main focus of the study was to determine the caregiving load and its impact on the caregiver's physical, mental and emotional well-being. Methods: The cross-sectional study used a questionnaire consisting of a self-designed questionnaire based on issues from literature on the subject and the validated Caregiver Strain Questionnaire – Short Form (CGSQ-SF). The CGSQ-SF provides qualitative data on the noticeable, emotional, and overall impact of caregiving. During prescheduled home interview sessions, the caregivers completed the questionnaires. The mixed research method used measurable data such as financial status and qualitative data, such as the emotional experiences associated with caring for a VI child. The University of Stellenbosch ethics committee (S13/03/049) and the Western Cape Education Department (20130704-13796) approved the study. Professor M. Kidd of the Centre for Statistical Services at the University of Stellenbosch carried out the statistical analyses, Mann-Whitney U tests and Kruskal-Wallis tests analysed the relationship of variables with CGSQ-SF scores. Results: Of the population of 320 VI learners from the only school for the VI in the Cape Metropole, 150 learners resided at home during the school term. According to the inclusion criteria, only 95 of the 150 home caregivers were eligible. The study obtained consent from 73 caregivers. The CGSQ-SF scoring showed that the majority of the 73 caregivers experienced moderate strain in terms of overall (55/73; 75.3%), objective (57/73; 78.1%) and subjective internalised caregiver strain (55/73; 75.3%). Elevated caregiver strain frequently occurred with these factors: financial difficulty; diminished ILS; and reluctance to spend time away from the VI child. Conclusion: All the caregivers experienced a considerable caregiving load and subsequently substantial caregiver strain level. All the participating households experienced both poverty and caring for a VI child. This bias in the design made it difficult to attribute what strainload was due to either of these variables. It was not possible to determine whether having a VI child increased the household’s strain compared to those who did not have a VI child. The confounding nature of the design made it challenging to unravel the relationship between poverty and caregiver strain due to the presence of a VI child. Qualitative information from the caregivers suggested that they did not view their child’s special needs as the only cause for the financial struggles they face and that their biggest concern was their child’s future as a disabled adult.
  • Item
    Influence of previous tuberculosis treatment on time to culture conversion for patients receiving a bedaquiline-containing regimen at Sizwe Tropical Disease Hospital, South Africa
    (Stellenbosch : Stellenbosch University, 2021-03) Saimen, Amashnee; Esterhuizen, Tonya; Padanilam, Xavier; Stellenbosch University. Faculty of Medicine and Health Sciences. Dept. of Global Health. Epidemiology and Biostatistics.; Stellenbosch University. Faculty of Medicine and Health Sciences. Dept. of Global Health. Epidemiology and Biostatistics.
    ENGLISH SUMMARY : Tuberculosis remains one of the leading causes of death worldwide. There is a growing crisis concerning the number of drug resistant TB cases. New drug regimens were urgently needed to improve mortality and morbidity among drug resistant TB patients. Bedaqualine is a newly developed diarylquinoline with a unique mechanism of action. . Studies have reported varying time to culture conversion regarding RR/MDR-TB patients with history of previous TB treatment. The Rv0678 mutation found in patients with prior Rifampicin exposure has demonstrated a more than four-fold increase in the minimum inhibitory concentrations of Bedaquiline. Previously treated TB patients may be more likely to have higher bacterial load due to extensive parenchymal damage. This study will provide information on effective Bedaquiline treatment duration for those previously exposed to TB treatment. Aim: To determine whether previous exposure to TB treatment influences the time to culture conversion as compared to no previous TB treatment exposure in patients receiving a DR- TB regimen containing Bedaquiline. Primary Objective: To compare the time to culture conversion for previously treated and new DR-TB patients receiving the Bedaquiline containing regimen. Secondary Objectives: 1.To evaluate treatment regimens at time of culture conversion for previously treated andnew DR-TB patients 2.To compare the rate of relapse in previously treated and new DR-TB patients receivingthe Bedaquiline containing regimen. 3.To compare the duration of Bedaquiline therapy in previously treated and new DR-TBpatients 4.To establish the severity of disease of previously treated and new DR-TB patients. Setting and Study Population: The study will be conducted at Sizwe Tropical Disease Hospital in Gauteng Province. Sizwe Tropical Disease Hospital serves as a referral centre for complicated MDR/XDR-TB cases in Gauteng. Study Design: A retrospective cohort study will be undertaken for confirmed DR-TB patients who were initiated on DR-TB regimen containing Bedaquiline from April 2016 to March 2019. Inclusion Criteria: Bacteriologically confirmed DR-TB All patients receiving DR-TB Regimens containing Bedaquiline including new DR-TB with no previous history of TB treatment Documented culture conversion Variables: Culture Conversion; Time to culture conversion; Duration of Bedaquiline treatment Sampling Technique: The group sample sizes were determined to be 76 (Previous treatment group) and 304 (New treatment group), power of 80% with a level of significance of α 0.05. Time Frame: The study will be conducted from June 2020 to November2020.
  • Item
    Identifying gaps using the EPICOT+ framework and exploring the association between funding sources and author conclusions in primary nutrition research addressing non-communicable diseases from Cochrane nutrition reviews : a descriptive-analytical cross-sectional study
    (Stellenbosch : Stellenbosch University, 2021-03) Ruzive, Sheena; Naude, Celeste E.; Stellenbosch University. Faculty of Medicine and Health Sciences. Dept. of Global Health. Epidemiology and Biostatistics.
    ENGLISH SUMMARY : Background: With the rise in non-communicable diseases (NCDs) globally, we aimed to summarise the research gaps and describe the adequacy of the reporting of future research recommendations in Cochrane reviews of nutrition interventions addressing NCDs. We also aimed to explore the influence of funding sources and author- sponsor financial ties on author conclusions in a subset of primary studies included in these reviews. Methods: Two researchers independently screened a Cochrane nutrition reviews database (n=470, July 2015) to identify reviews addressing four NCDs (cardiovascular diseases, cancer, chronic respiratory diseases and diabetes). The “implications for research” section of eligible reviews was analysed using the evidence, population, intervention, comparison, outcome, timeframe, study design and burden of disease (EPICOT+) framework to describe the extent of reporting of research recommendations and to summarise gaps. A purposive sample of English full-text studies included in reviews addressing alternative nutrition supplements were analysed to assess reporting of conflict of interest (COI), funding sources and author-sponsor financial ties, and to explore influences of funding sources and author-sponsor financial ties on author conclusions. Results: Ninety-eight eligible reviews were analysed. The EPICOT+ reporting was as follows: evidence 34/98 (33.7%), population 68/98 (69.4%), intervention 90/98 (91.8%), comparison 26/98 (26.5%), outcomes 78/98 (79.6%), study design 85/98 (86.7%), time frame 52/98 (53.1%), and burden of disease 7/98 (7.1%). Studies requiring better quality, different interventions, and outcomes in low- and middle-income countries (LMICs) were highlighted. Seven reviews addressed alternative nutrition supplements, including 51 eligible primary studies. Conflicts of interest were disclosed in 10/51 (19.2%); funding in 27/51 (51.9%), of which, 11/27 (40.7%) were industry and 16/27 (59.3%) were non-industry sponsors; and author-sponsor financial ties in 9/51 (13.4%), of which 1/9 (11.1%) was industry and 8/9 (88.9%) were non-industry. There was no association between authors making favourable conclusions and having industry sponsors and author-sponsor financial ties (8/12) compared with non-industry sponsors and no author-sponsor financial ties (10/24), (Fisher exact p =0.289). Conclusions: EPICOT+ items were not well reported in most reviews. Future studies of better quality, different interventions, outcomes or populations in LMICs are needed. Authors should disclose all COI, funding sources and author-sponsor financial ties. Possible influences of funding sources and author-sponsor financial ties on author conclusions needs further investigation.
  • Item
    Traumatic brain injury in the intensive care unit : association between the Glasgow coma score and the intensive care unit mortality; the Botswana experience
    (Stellenbosch : Stellenbosch University, 2021-03) Mkubwa, Jack Joseph; Esterhuizen, Tonya; Bedada, Alemayehu; Stellenbosch University. Faculty of Medicine and Health Sciences. Dept. of Global Health. Epidemiology and Biostatistics.
    ENGLISH SUMMARY : Background: Traumatic brain injury prevalence in Botswana is high and this, coupled with a small population, may reduce productivity. There is no previous study on the association between mortality in traumatic brain injury and the Glasgow coma score in Botswana although global literature supports its existence. Objectives Our primary aim was to determine the association between the initial Glasgow Coma scale score and the time to mortality of adults admitted with TBI at the Princess Marina Hospital, Botswana, between 2014 and 2019. Secondary aims were to assess the risk factors associated with time to mortality and to estimate the mortality rate from TBI. Methods This was a retrospective cohort design, medical record census conducted from 1st January 2014 to 31st December 2019. Results: In total, 137 participants fulfilled the inclusion criteria and majority, 114 (83.2%) were male with mean age 34.5 years. The initial GCS score and time to mortality were associated (aHR: 0.69; 95% CI: 0.508-0.947). Other factors associated with time to mortality included constricted pupil (aHR: 0.12; 95% CI: 0.044-0.344), temperature (aHR: 0.82; 95% CI: 0.727-0.929), and subdural haematoma (aHR: 3.41; 95% CI: 1.819-6.517). Most cases of TBI, 74 (54%) were due to road crashes. Mortality was 48, 35% (95% CI: 27.1% to 43.6%) and entirely from severe TBI. Conclusion: The study confirmed significant association between Glasgow coma score and mortality. Males were mainly involved in TBI. These findings lacked external validity due to a small sample size and therefore a larger multi-centre study is required for validation.