Mixed-methods study assessing the accuracy of verbal autopsy and sociocultural determinants of infant mortality in Gujarat, India

Abstract

Listen

Translate article

BACKGROUND: Accurate determination of infant mortality causes and understanding sociocultural factors influencing care-seeking behaviors are crucial for targeted interventions in resource-limited settings. This mixed-methods study aimed to assess the accuracy of verbal autopsy (VA) in ascertaining infant death causes and explore sociocultural determinants of infant mortality in Gujarat, India. MATERIALS AND METHODS: It was a mixed-method study with a retrospective cohort component for which data from 661 infant records were extracted from the main health office’s database, with a subset of 328 infant deaths selected for verbal autopsy analysis. Delivery characteristics, care-seeking behaviors, access to healthcare, and risk factors were evaluated. Bivariate and multivariate analyses were conducted to identify factors associated with infant mortality. VA diagnostic accuracy was assessed using sensitivity, specificity, predictive values, and receiver operating characteristic curve analysis, with death certificates as the reference standard. Qualitative methods, including in-depth interviews, and focus group discussions were used to explore the sociocultural influences, health system challenges, and stakeholder experiences related to infant deaths and VA implementation. Findings were integrated through triangulation. RESULTS: Institutional deliveries were 583/661 (88%), and the skilled birth attendance was 397/661 (60%). Care-seeking from health facilities was low at 264/661 (40%), with a mean time of 12 hours. Home deaths accounted for 328/661 (49.6%) cases. Low birth weight (aOR 1.81, 95% CI 1.3–2.4, P = 0.002), home delivery (aOR 1.72, 95% CI 1.1–2.8, P = 0.01), early complementary feeding (aOR 1.48, 95% CI 1.1–2.0, P = 0.01), and acute malnutrition (aOR 1.91, 95% CI 1.3–2.6, P = 0.001) were independent risk factors for mortality. Verbal autopsy showed high specificity (87%) but variable sensitivity (70%) in determining causes of death. Qualitative findings revealed barriers to timely care (lack of danger sign recognition, financial constraints, traditional healer reliance), cultural beliefs impacting care practices, gender discrimination, health system constraints (staff shortages, diagnostic limitations), and challenges with VA implementation (recall bias, desire for feedback). CONCLUSION: Enhancing antenatal care, skilled birth attendance, optimal breastfeeding, complementary feeding practices, addressing acute malnutrition, overcoming sociocultural barriers through community engagement, health system strengthening, and culturally sensitive interventions could potentially reduce infant mortality rates. While a verbal autopsy is practical for the cause of death determination in resource-limited settings, its effectiveness relies on addressing the identified challenges through policy measures focused on community participation, health system improvements, and culturally appropriate strategies.