Reducing inequities in maternal and child health in rural Guatemala through the CBIO+ Approach of Curamericas: 5. Mortality assessment

Abstract

Background: The Curamericas/Guatemala Maternal and Child Health Project, 2011-2015, implemented the Census-Based, Impact-Oriented Approach, the Care Group Approach, and the Community Birthing Center Approach. Together, this expanded set of approaches is known as CBIO+. This is the fifth of 10 papers in our supplement describing the Project and the effectiveness of the CBIO+ Approach. This paper assesses causes, levels, and risk factors for mortality along with changes in mortality.

Methods: The Project maintained Vital Events Registers and conducted verbal autopsies for all deaths of women of reproductive age and under-5 children. Mortality rates and causes of death were derived from these data. To increase the robustness of our findings, we also indirectly estimated mortality decline using the Lives Saved Tool (LiST).

Findings: The leading causes of maternal and under-5 mortality were postpartum hemorrhage and pneumonia, respectively. Home births were associated with an eight-fold increased risk of both maternal (p = 0.01) and neonatal (p = 0.00) mortality. The analysis of vital events data indicated that maternal mortality declined from 632 deaths per 100,000 live births in Years 1 and 2 to 257 deaths per 100,000 live birth in Years 3 and 4, a decline of 59.1%. The vital events data revealed no observable decline in neonatal or under-5 mortality. However, the 12-59-month mortality rate declined from 9 deaths per 1000 live births in the first three years of the Project to 2 deaths per 1000 live births in the final year. The LiST model estimated a net decline of 12, 5, and 22% for maternal, neonatal and under-5 mortality, respectively.

Conclusion: The baseline maternal mortality ratio is one of the highest in the Western hemisphere. There is strong evidence of a decline in maternal mortality in the Project Area. The evidence of a decline in neonatal and under-5 mortality is less robust. Childhood pneumonia and neonatal conditions were the leading causes of under-5 mortality. Expanding access to evidence-based community-based interventions for (1) prevention of postpartum hemorrhage, (2) home-based neonatal care, and (3) management of childhood pneumonia could help further reduce mortality in the Project Area and in similar areas of Guatemala and beyond.

Keywords: Care Groups; Census-Based, Impact-Oriented Approach; Child health; Child mortality; Community birthing centers; Community health; Community-based primary health care; Curamericas Global; Curamericas/Guatemala; Equity; Guatemala; Implementation research; Maternal health; Maternal mortality; Mortality assessment; Primary health care.

Fig. 1

Chain of reporting for vital events

Fig. 2

Causes of maternal mortality in Areas A and B combined, October 2011–May 2015. Note: C-section: Cesarean section

Fig. 3

Location of delivery and location of death for women in Areas A and B whose deaths were registered in the Project’s vital events registration system and classified as maternal, October 2011–May 2015

Fig. 4

Causes of neonatal mortality – Areas A and B, October 2011–May 2015

Fig. 5

Number of deaths by age at death among children who died before 5 years of age in the Project Area, October 2011–May 2015. a Age at death (in days) among neonates. b Age at death (in months) among children 0- < 60 months of age in the Project Area, October 2011–May 2015

Fig. 6

Causes of 1- < 60-month deaths in combined Areas A and B, October 2011–May 2015

Fig. 7

Causes of under-5 mortality for the entire Project Area, October 2011–May 2015. *SIDS: Sudden infant death syndrome

Fig. 8

“Delays” that led to maternal death for both Areas combined, October 2011 – May 2015

Fig. 9

Number of deaths by age at death for maternal deaths, 2011–2015

Fig. 10

Location of birth and death for neonatal deaths – communities of both Areas combined, October 2011–May 2015

Fig. 11

Location of death of children 1- < 60 months of age – communities of both Areas, October 2011–May 2015

Fig. 12

Trends in 12–59-month mortality in Area A, Project Years 1–4

Fig. 13

Estimate of decline in maternal mortality in Area A, 2011 to 2015 using the Lives Saved Tool, compared to estimated historical decline

Fig. 14

Estimate of decline in neonatal mortality in Area A, 2011 to 2015 using the Lives Saved Tool, compared to estimated historical decline

Fig. 15

Estimate of decline in 0- < 60 month mortality in Area A, 2011 to 2015, using the Lives Saved Tool, compared to estimated historical decline