Non-medical risk factors as avoidable determinants of excess mortality in children with chronic kidney disease. A prospective cohort study in Nicaragua, a model low income country

Background: The widely recognized clinical and epidemiological relevance of the socioeconomic determinants of health-disease conditions is expected to be specifically critical in terms of chronic diseases in fragile populations in low-income countries. However, in the literature, there is a substantial gap between the attention directed towards the medical components of these problems and the actual adoption of strategies aimed at providing solutions for the associated socioeconomic determinants, especially in pediatric populations. We report a prospective outcome study on the independent contribution and reciprocal interaction of the medical and socioeconomic factors to the hard end-point of mortality in a cohort of children with chronic kidney disease in Nicaragua. Methods and Findings: Every child (n = 309) diagnosed with chronic kidney disease (CKD) and referred to the tertiary unit of Pediatric Nephrology in Managua (Nicaragua) from a network of nine hospitals serving 80% of the country's pediatric population was registered between January 2005 and December 2013. The three main socioeconomic determinants evaluated were family income, living conditions and the family's level of education. Further potential determinants of the outcomes included duration of exposure to disease, CKD stage at the first visit as suggested by the KDOQI guidelines in children, the time it took the patients to reach the reference centre and rural or urban context of life. Well-defined and systematically collected medical and socioeconomic data were available for 257 children over a mean follow-up period of 2.5\ub12.5 years. Mortality and lost to follow-up were considered as outcome endpoints both independently and in combination, because of the inevitably progressive nature of the disease. A high proportion (55%) of children presented in the advanced stages of CKD (CKD stage IV and V) at the first visit. At the end of follow-up, 145 (57%) of the 257 cohort children were alive, 47 (18%) were lost to follow-up and 65 (25%) had died. Cox regression analysis showed an independent contribution to mortality of CKD stage at diagnosis and of level of education, with overlapping HR values (HR and 95%CI: 2.66; 1.93-3.66 and 2.72; 1.71-4.33, respectively). Conclusions: The unfavourable socioeconomic and cultural background of the pediatric study cohort and the severity of kidney damage at diagnosis were the key determinants of the clinical risk conditions at baseline and of the mortality outcome. Long-term structural interventions on such backgrounds must be adopted to assure effectiveness of medical care and to assure an earlier diagnosis of CKD in these patients. The translation-extension of our results is currently underway with an agenda which includes: 1) better integration of chronic pediatric conditions into primary care strategies to promote prevention and early timely referral; 2) the consideration of socioeconomic conditions as a mandatory component of the packages of best-care; 3) the formulation and flexible adaptation of guidelines and educational programs, based on the information generated by a context-specific, epidemiological monitoring of needs and outcomes, guaranteed by an effective database