Observed birth prevalence of congenital anomalies among live births at a regional facility in KwaZulu Natal Province, South Africa.

Congenital disorders (CDs), defined as abnormalities in structure or function present at birth, are an important contributor to the disease burden in developing countries. The size and extent of the problem in South Africa (SA) are unknown due to the lack of recent, reliable, observed data on CDs. To address this empirical data gap, this study aimed to measure the birth prevalence of congenital anomalies (a sub-set of CDs) and to describe the pattern of these anomalies at a regional hospital in KwaZulu Natal (KZN), SA. A retrospective, observational, descriptive review of congenital anomalies diagnosed within the neonatal service at Edendale Hospital (EDH), KZN was undertaken between January and December 2018. All EDH in-house live births diagnosed and notified with congenital anomalies by discharge were included. Stillbirths, other pregnancy losses and out-born neonates were excluded. Data were actively collected from the birth register, neonatal admission register, and the individual paper-based surveillance tool developed by the National Department of Health. The in-facility birth prevalence rate for congenital anomalies was 15.57 per 1 000 live births. The most observed system was musculoskeletal (32%) followed by circulatory system anomalies (19%). When the observed birth prevalence rates of key congenital anomalies were compared with previously published, modelled South African data, no significant difference was found. This study responds to the paucity of birth prevalence data on CDs overall and offers evidence that obvious, structural CDs (congenital anomalies) need to be addressed in the SA public health system.

Hospital-acquired infections in paediatric medical wards at a tertiary hospital in KwaZulu-Natal, South Africa.

BACKGROUND: Hospital-acquired infections (HAIs) impact care and costs in hospitals across the globe. There are minimal data on HAIs in sub-Saharan Africa and data specific to paediatrics are especially limited. OBJECTIVE: To describe the incidence of HAIs in the paediatric medical units at Grey's Hospital, a tertiary government hospital in KwaZulu-Natal, South Africa. METHODS: The Infection Prevention and Control (IPC) team collects data on all laboratory-confirmed infections, including from paediatric patients in two medical units (52 beds), the paediatric intensive/high-care unit (PICU, 8 beds) and the neonatal intensive care unit (NICU, 23 beds). HAIs are defined as infections: (i) not present (active or incubating) at the time of admission, and (ii) with onset >48 h after hospital admission. Daily patient statistics allow calculation of infections per 100 admissions and infections per 1000 patient days. RESULTS: In the non-ICU setting, there were 7.1 and 7.0 HAIs per 100 admissions in 2013 and 2014, respectively. In the PICU, there were 20.4 and 15.3 HAIs per 100 admissions, while in the NICU there were 23.9 and 21.6 HAIs per 100 admissions in 2013 and 2014, respectively. In the non-ICU setting, there were 6.8 HAIs per 1000 patient days in both 2013 and 2014. In the PICU, there were 27.5 and 33.0 HAIs per 1000 patient days, while in the NICU, there were 20.3 and 21.5 HAIs per 1000 patient days in 2013 and 2014, respectively. CONCLUSION: HAIs in non-ICU paediatric wards were consistent with a number of point-prevalence studies performed outside Africa (e.g. Canada, Russia, U.K.). Rates of HAIs in the ICUs were higher than rates reported from the International Nosocomial Infection Control Consortium, and were substantially higher than rates reported in the United States. HAIs are serious and important, especially in ICUs, and may be relatively neglected in low- and middle-income settings. Improved surveillance will allow the development and evaluation of targeted interventions to improve care of patients.

Rapid Accurate Identification of Tuberculous Meningitis Among South African Children Using a Novel Clinical Decision Tool.

BACKGROUND: Early diagnosis of tuberculous meningitis (TBM) is crucial to achieve optimum outcomes. There is no effective rapid diagnostic test for use in children. We aimed to develop a clinical decision tool to facilitate the early diagnosis of childhood TBM. METHODS: Retrospective case-control study was performed across 7 hospitals in KwaZulu-Natal, South Africa (2010-2014). We identified the variables most predictive of microbiologically confirmed TBM in children (3 months to 15 years) by univariate analysis. These variables were modelled into a clinical decision tool and performance tested on an independent sample group. RESULTS: Of 865 children with suspected TBM, 3% (25) were identified with microbiologically confirmed TBM. Clinical information was retrieved for 22 microbiologically confirmed cases of TBM and compared with 66 controls matched for age, ethnicity, sex and geographical origin. The 9 most predictive variables among the confirmed cases were used to develop a clinical decision tool (CHILD TB LP): altered Consciousness; caregiver HIV infected; Illness length >7 days; Lethargy; focal neurologic Deficit; failure to Thrive; Blood/serum sodium <132 mmol/L; CSF >10 Lymphocytes ×10/L; CSF Protein >0.65 g/L. This tool successfully classified an independent sample of 7 cases and 21 controls with a sensitivity of 100% and specificity of 90%. CONCLUSIONS: The CHILD TB LP decision tool accurately classified microbiologically confirmed TBM. We propose that CHILD TB LP is prospectively evaluated as a novel rapid diagnostic tool for use in the initial evaluation of children with suspected neurologic infection presenting to hospitals in similar settings.

Improving paediatric tuberculosis and HIV clinical record keeping: the use of audit and a structured pro forma in a South African regional level hospital.

We report on the impact of revisions made to an existing pro forma facilitating routine assessment and the management of paediatric HIV and tuberculosis (TB) in KwaZulu-Natal, South Africa. An initial documentation audit in 2010 assessed 25 sets of case notes for the documentation of 16 select indicators based on national HIV and TB guidelines. Using the findings of this initial audit, the existing case note pro forma was revised. The introduction of the revised pro forma was accompanied by training and a similar repeat audit was undertaken in 2012. This demonstrated an overall improvement in documentation. The three indicators that improved most were documentation of maternal HIV status, child's HIV status and child's TB risk assessment (all P < 0.001). This study suggests that tailor-made documentation pro formas may have an important role to play in improving record keeping in low-resource settings.