Doctoral Degrees (Paediatrics and Child Health)

Permanent URI for this collection

https://scholar.sun.ac.za/handle/10019.1/3786

Browse

Browsing Doctoral Degrees (Paediatrics and Child Health) by Author "Dramowski, Angela"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Determinants of healthcare-associated infection among hospitalized children
    (Stellenbosch : Stellenbosch University, 2017-03) Dramowski, Angela; Cotton, Mark F.; Whitelaw, Andrew; Stellenbosch University. Faculty of Medicine and Health Sciences. Dept. of Paediatrics and Child Health.
    ENGLISH SUMMARY: Healthcare-associated infection (HAI) is the most frequent complication of hospitalization resulting in suffering, excess mortality and increased healthcare costs. Although paediatric HAI burden and impact is well-described in high-income countries, it is largely unquantified in Africa. Our research aimed to: (i) comprehensively describe the epidemiology and impact of HAI and HA-bloodstream infections in an African cohort of hospitalized children; (ii) establish appropriate HAI surveillance methods for our setting; (iii) investigate selected local determinants of paediatric HAI (including healthcare worker HAI-related knowledge, attitudes and practices, isolation facility utilization and terminal cleaning practices). In a retrospective analysis of paediatric bloodstream infection trends at Tygerberg Hospital, we reported the highest estimate of HA-bloodstream infections rates among African children to date. HIV-infection, HA-bloodstream infection, fungal and Gram-negative pathogens were important predictors of bloodstream infection-associated mortality. We conducted prospective clinical surveillance in paediatric wards and the intensive care unit at Tygerberg Children’s Hospital. HAI incidence (31/1000 patient days) exceeded published rates in high and low-middle income countries, with highest infection density in the paediatric intensive care unit (94/1000 patient days). Children experiencing HAI events were young (median 8.4 months), more likely to be malnourished, HIV-exposed uninfected or HIV-infected and to have pre-existing co-morbidities. Hospital-acquired pneumonia, bloodstream and urinary tract infections predominated. The increased odds for HAI in HIV-exposed uninfected and HIV-infected children is a novel association. Two-thirds of in-patient mortality was associated with HAI and patients with any HAI event had a 6-fold increase in crude mortality. Patients experiencing HAI had 3-fold higher rates of re-hospitalization within 30 days. Direct costs of HAI were substantial; mean duration of hospitalization, bed availability, antimicrobial consumption and laboratory investigation usage were significantly impacted by HAI. Although prospective clinical HAI surveillance is considered the reference standard, its use in Africa is limited by lack of resources and expertise. We compared the performance of three alternate HAI surveillance methods (point prevalence surveys [PPS], laboratory surveillance and tracking of antimicrobial prescriptions) using the prospectively collected paediatric HAI dataset as the reference standard. Although repeated PPS, laboratory and antimicrobial prescription tracking were demonstrated to be feasible HAI surveillance methods, a combination of laboratory-antimicrobial surveillance achieved best sensitivity (85%) and positive predictive value (97%), and required fewer resources to perform. South African paediatric healthcare facilities should individualise HAI surveillance, selecting a method suited to available resources and practice context. We identified a shortage of isolation facilities and sub-optimal identification of patients requiring isolation as potential contributors to infection transmission. The need for negative-pressure ventilation and airborne isolation facilities on children’s wards in TB-endemic settings was also highlighted. For terminal cleaning of paediatric isolation rooms we investigated three evaluation methods; fluorescent markers emerged as most cost-effective and feasible for our resource-limited setting. Finally, we surveyed two-thirds of our paediatric department’s staff regarding their knowledge, attitudes and practices related to HAI, identifying several knowledge gaps and opportunities for improved infection prevention practice. Owing to the extreme paucity of data, paediatric HAI in Africa remains an underappreciated and underfunded public health problem. We believe that this doctoral research dissertation provides unequivocal justification for greater resource allocation to HAI surveillance and prevention programmes for hospitalized African children.