Browsing by Author "Jacobs, Solomon M"
Now showing 1 - 1 of 1
Results Per Page
Sort Options
- ItemOpen AccessHealth service utilisation prior to out-of-hospital natural deaths among children under five in Metro West, Cape Town in 2018: a retrospective analysis of data from the Child Death Review(2022) Jacobs, Solomon M; Mathews, Shanaaz; Hendricks, Michael; Phillips, TammyBackground In the Metro West geographical service area within the City of Cape Town district the under-five mortality rate in the Metro West GSA decreased from 25 per 1000 live births in 2010 to 22 deaths per 1000 live births in 2013, but the rate of decrease slowed down in part because of the amount of child deaths outside of health facilities. Fifty-five percent of under-five deaths occur out-of-hospital in South Africa, with a similar percentage in the Metro West (49-52% in 2010 to 2015). Describing factors that enable or prevent health service usage among natural under-five deaths is an important precursor for effective interventions. Objectives and Methods A retrospective cross-sectional design utilised secondary, routinely collected data from 1 January to 31 December 2018 on under-five out-of-hospital natural deaths reported to Salt River Mortuary. We used the data, which included routine interviews with the caregivers, together with social and medical data collated by the Child Death Review, to describe the cause of death, the socio-demographic profile, and the routine and prior-to-death health service usage. Dimensions of health service access according to the WHO and Anderson, respectively, were used as a heuristic lens to describe the applicable variables and to formulate a priori multivariable logistic regression models to compare those who did and did not seek care before death. These dimensions include physical accessibility, financial affordability, the health needs of the child as well as the existing health behaviour of the caregiver. Results Of the 187 cases described, 68% died of lower respiratory tract infections and 8% of diarrhoea. Fifty four percent of cases were younger than 3 months, and 40% were born prematurely. In terms of the residential health sub-district, 37% resided in Mitchells Plain, 29% in Klipfontein, 18% in Western and 14% in Southern; 52% resided in needy or very needy areas. Mothers were single (69%), unemployed (73%) and lived in informal housing (46%). Of the cases who were alive at the time, immunisation coverage was 79%, 70% and 68% at the 6-, 10- and 14-week visits. However, only 23% of mothers sought health care prior to the child's death. Overall, 51% of the mothers recognised symptoms of illness prior to death and symptoms were recognised in 95% of the mothers who sought health care compared to 37% of the mothers who did not seek health care (p-value < 0.001). Multivariable logistic regression models showed the importance in recognising symptoms in seeking health care (aOR 18.28, 95% CI 3.67-90.93), and that, while not statistically significant, the recognition of symptoms was less likely at younger ages (aOR 0.28, 95% CI 0.07-1.14 for cases younger than 3 months compared to those older than 6 months) Conclusion The study identified key risk factors implicated in the out-of-hospital deaths in Metro West and the need for mothers to identify and seek health care when their child is symptomatic. There should be focused support during the first 1000 days for mothers identified as being at-risk, namely: those who are single, unemployed, lack social support and abuse substances, and for babies with prematurity and HIV-exposure. Counselling should emphasise the recognition of LRTI symptoms and health care seeking to prevent child deaths and reduce the under-five mortality. Further research is needed to consolidate which provincial geographical areas should be prioritized for targeted interventions that impact on health care seeking behaviour, as well as the most effective child health education and messaging. Data paucity on medical history may be addressed by data augmentation from the Provincial Health Data Centre and improvements should be made for capturing children's anthropometry.