Discordance in postnatal care between mothers and newborns

Measurement artifact or missed opportunity

Bibliographic Data

ID19550915
AuthorsAgbessi Amouzou (0000-0002-6262-3866, Johns Hopkins University, corresponding author), Elizabeth Hazel (0000-0002-9176-3278, Johns Hopkins University), Lara M E Vaz (0000-0002-9072-8091, Save the Children), Lara Vaz, Sanni Yaya (0000-0002-4876-6043, University of Ottawa), Yaya Sanni, Amelia Castillo Morán (0000-0002-4826-1475, World Health Organization), Allisyn Moran
Year2020
Volume10
Issue1
Pages010505-010505
Publication date2020-06-01
Open AccessYes
TypeARTICLE
VenueJournal of Global Health (JOURNAL)
Journal identifiersISSN: 2047-2978 • E-ISSN: 2047-2986
PublisherInternational Society of Global Health (PUBLISHER • GB)
DOI10.7189/jogh.10.010505
PMID32257159
OpenAlexW3011877371
LanguageEN
Citations received4
References cited12

BACKGROUND: Postnatal care (PNC) for mothers and newborns is essential to monitor risks of morbidity and adverse conditions following delivery. Current estimates of the coverage of PNC show substantial discordance between mothers and newborns. We investigate the sources of this discordance in Demographic and Health Surveys (DHS). METHODS: We used DHS data from 48 countries collected since 2011, spanning phases 6 and 7 of the survey program with 32 and 16 surveys, respectively, analyzed. We assessed the distribution of the reported timing of PNC and conducted a sensitivity analysis that excludes/includes PNC reported within 0-1 hour or PNC in the day 2. Agreement in PNC reporting considered four groups: (1) Concordance, neither mother nor newborn received PNC; (2) Concordance, mother and newborn pair received PNC; (3) Discordance, mother received PNC and newborn did not; of (4) Discordance, mother did not receive PNC but the newborn did. We carried out logistic regressions to understand correlates of PNC discordance. All analyses distinguished phase 6 surveys from phase 7. RESULTS: We found substantial differences in the PNC coverage estimated between phase 6 and phase 7 surveys. The phase 7 PNC questions for newborns were improved to increase the understanding of the questions by respondent which probably led to reducing the large PNC gap between mothers and newborns observed in phase 6 surveys. With phase 6 surveys, PNC coverage for mother was estimated on average at 62% compared to only 31% for newborns. No such gap was observed for phase 7 surveys, where for both mothers and newborns, the PNC coverage estimate was similar, at 56%. For both phases, over half of the reported PNC for mothers and newborns occurred during 0-1 hour following delivery, leading to substantial overestimation of PNC coverage, due to confusion between intrapartum care and PNC. There were 37% discordant cases between mother and newborn, largely in favor of the mother in phase 6 surveys, compared to 16% in phase 7 surveys. In phase 6 surveys, discordant PNC cases were observed largely among facility deliveries vs non-facility deliveries (44% compared to 19%). CONCLUSIONS: Current estimates of coverage of PNC from DHS phase 6 surveys appears to include substantial level of measurement noises that could explain substantial part of the mother-newborn discordance in PNC. The PNC estimates appear to capture a substantial number of intrapartum care. Current measurement approaches warrant further validation to ensure accurate monitoring of the PNC programs

Concordance · Logistic regression · Obstetrics · Respondent · Demography · Global Maternal and Child Health · Maternal and fetal healthcare · Maternal and Perinatal Health Interventions · Medicine · Pediatrics

Tools

Open DOISci-HubOpen Access