Missed at birth: a cross-sectional analysis of national determinants and subnational changes in birth registration coverage in Kenya, using 2014 and 2022 Demographic and Health Survey (DHS) data.

Introduction: Registration of all births and deaths in sub-Saharan Africa remains inadequate. In Kenya, significant strides have been made, but progress has stalled in recent years. While some studies have examined factors influencing birth registration, national-level analytical assessments of these factors are limited. This study evaluates the progress and determinants of birth registration for children under 3 years in Kenya. Methods: We used cross-sectional data from the 2014 and 2022 Kenya Demographic and Health Surveys complemented by geospatial covariates of travel time and urbanicity. We computed the percentage of children (<3 years) registered with the civil authority at the national and subnational level (counties) and compared temporal changes against the United Nations (UN) ≥ 90% target. Binary and multivariable logistic regressions were used to assess determinants of birth registration in 2022, accounting for survey design. Results: National birth registration coverage improved from 69.0% (2014) to 76.0% (2022). Subnational coverage in 2014 ranged from 20.8% (West Pokot) to 96.4% (Nyeri) and from 46.4% (Marsabit) to 95.5% (Nyeri) in 2022. By 2022, only 17% (8/47) of counties met the UN's ≥ 90% goal, with counties in the central region consistently performing better. Higher odds of birth registration were linked to health-facility births (adjusted OR (AOR)=2.44; 95% CI 2.07 to 2.89), immunisation (AOR=1.76; 95% CI 1.25 to 2.49), maternal age (35-39 years: AOR=1.61; 95% CI 1.22 to 2.12), higher education (AOR=1.39; 95% CI 1.03 to 1.89), media access (AOR=1.47; 95% CI 1.06 to 2.04), affiliation with the Catholic religion and the Kikuyu ethnic group. Households without bank accounts had lower odds (AOR=0.78; 95% CI 0.66 to 0.93). Notably, urbanicity and travel time to civil registration centres were not significantly associated with birth registration. Conclusion: National birth registration coverage has improved, but subnational disparities persist. Our findings show that in 2022, the health, education, financial and media sectors are associated with a higher likelihood of birth registration. These results underscore the need for the government and stakeholders to implement multisectoral strategies to strengthen Civil Registration and Vital Statistics and address socioeconomic and geographic inequalities critical to achieving universal birth registration coverage.