Indigenous Ethnicity and Low Maternal Education Are Associated with Delayed Diagnosis and Mortality in Infants with Congenital Heart Defects in Panama

Abstract

BackgroundThis is the first study in Panama and Central America that has included indigenous populations in an assessment of the association between socioeconomic variables with delayed diagnosis and mortality due to congenital heart defects (CHD).MethodsA retrospective observational study was conducted. A sample calculation was performed and 954 infants born from 2010 to 2014 were randomly selected from clinical records of all Panamanian public health institutions with paediatric cardiologists. Critical CHD was defined according to the defects listed as targets of newborn pulse oximetry screening. Diagnoses were considered delayed when made after the third day of life for the critical CHD and after the twentieth day of life for the non-critical. A logistic regression model was performed to examine the association between socioeconomic variables and delayed diagnosis. A Cox proportional hazards model was used to assess the relationship between socioeconomic features and mortality.ResultsAn increased risk of delayed diagnosis was observed in infants with indigenous ethnicity (AOR, 1.56; 95% CI, 1.03–2.37), low maternal education (AOR, 1.57; 95% CI, 1.09–2.25) and homebirth (AOR, 4.32; 95% CI, 1.63–11.48). Indigenous infants had a higher risk of dying due to CHD (HR, 1.43; 95% CI, 1.03–1.99), as did those with low maternal education (HR, 1.95; 95% CI, 1.45–2.62).ConclusionInequalities in access to health care, conditioned by unfavourable socioeconomic features, may play a key role in delayed diagnosis and mortality of CHD patients. Further studies are required to study the relationship between indigenous ethnicity and these adverse health outcomes.

Socio‐economic inequalities in mortality in children with congenital heart disease: A systematic review and meta‐analysis

“…The definitions for SES measures used in individual studies are given in Table . The included articles were based on US (n = 22), UK (n = 5), and Panama (n = 1) populations. Most articles included all CHD subtypes combined (n = 21), with the rest including cases of HLH or single ventricle (SV) defects (n = 5), atrioventricular septal defect (AVSD, n = 1), and a composite group of “critical” subtypes (n = 1) .…”

“…The included articles were based on US (n = 22), UK (n = 5), and Panama (n = 1) populations. Most articles included all CHD subtypes combined (n = 21), with the rest including cases of HLH or single ventricle (SV) defects (n = 5), atrioventricular septal defect (AVSD, n = 1), and a composite group of “critical” subtypes (n = 1) . Cases were ascertained from congenital anomaly registers (n = 11), the Kids Inpatient Database (KIDS, n = 4), the SV reconstruction (SVR) trial (n = 2), clinical databases (n = 2), the Paediatric Health Information system (n = 2), medical records, or hospital discharges (n = 7) .…”

“… Area‐based education was analysed as a continuous variable (n = 1, derived from the education domain of the English IMD) or categorised into deciles (n = 1, derived from the US census‐based score) . Individual level education was categorised into maternal years in education (n = 4) or categorised into paternal years in education (n = 1) …”

“…Articles that analysed all CHD subtypes combined reported significant associations between parental education with neonatal and infant mortality, but not with long‐term mortality (Table ). One of those articles measured area‐based education using the census score, reporting a stronger effect with infant mortality than in neonatal mortality (OR 1.75 vs 1.49) .…”

“…f Area-based education was analysed as a continuous variable (n = 1, derived from the education domain of the English IMD) 43 or categorised into deciles (n = 1, derived from the US census-based score). 32 Individual level education was categorised into maternal years in education (n = 4) 29,31,34,42,45,46 or categorised into paternal years in education (n = 1). 33 g It states a lower census-based z-scoreas opposed to a higher census based z-score that are designed to be applied when reading and appraising research.…”