Estimation of district-level under-5 mortality in Zambia using birth history data, 1980-2010.

Birth history data-the primary source of data on under-5 mortality in developing countries-are infrequently used for subnational estimates due to concerns over small sample sizes. In this study we consider different methods for analyzing birth history data in combination with various small area models. We construct a simulation environment to assess the performance of different combinations of birth history methods and small area models in terms of bias, efficiency, and coverage. We find that performance is highly dependent on the birth history method applied and how temporal trends are accounted for. We estimated trends in district-level under-5 mortality in Zambia from 1980 to 2010 using the best-performing model. We find that under-5 mortality is highly variable within Zambia: there was a 1.8-fold difference between the lowest and highest levels in 2010, and declines over the period 1980 to 2010 ranged from less than 5% to more than 50%.

[1]  A. Flaxman,et al.  Error and bias in under-5 mortality estimates derived from birth histories with small sample sizes , 2013, Population Health Metrics.

[2]  B P Carlin,et al.  Spatio-temporal models with errors in covariates: mapping Ohio lung cancer mortality. , 1998, Statistics in medicine.

[3]  N. Cressie,et al.  Statistics for Spatial Data. , 1992 .

[4]  Deborah Balk,et al.  The Global Distribution of Infant Mortality: A subnational spatial view. , 2008, Population, space and place.

[5]  L. Fahrmeir,et al.  Analyzing infant mortality with geoadditive categorical regression models: a case study for Nigeria. , 2004, Economics and human biology.

[6]  C B Dean,et al.  Hierarchical Bayesian spatiotemporal analysis of revascularization odds using smoothing splines , 2008, Statistics in medicine.

[7]  L. Kazembe,et al.  Quantifying Spatial Disparities in Neonatal Mortality Using a Structured Additive Regression Model , 2010, PloS one.

[8]  D. Clayton,et al.  Bayesian analysis of space-time variation in disease risk. , 1995, Statistics in medicine.

[9]  Ngianga-Bakwin Kandala,et al.  A Geo-Additive Bayesian Discrete-Time Survival Model and its Application to Spatial Analysis of Childhood Mortality in Malawi , 2006 .

[10]  Bradley P. Carlin,et al.  Bayesian measures of model complexity and fit , 2002 .

[11]  Martin Bangha,et al.  Spatial differentials in childhood mortality in South Africa: evidence from the 2001 census , 2013 .

[12]  Mahmoud Torabi,et al.  Spatio‐temporal modelling of disease mapping of rates , 2010 .

[13]  A. Lawson,et al.  Joint spatial survival modeling for the age at diagnosis and the vital outcome of prostate cancer , 2008, Statistics in medicine.

[14]  Sw. Banerjee,et al.  Hierarchical Modeling and Analysis for Spatial Data , 2003 .

[15]  Alan D. Lopez,et al.  Age-specific and sex-specific mortality in 187 countries, 1970–2010: a systematic analysis for the Global Burden of Disease Study 2010 , 2012, The Lancet.

[16]  A. Cross,et al.  Zambia Demographic and Health Survey 1992 , 1993 .

[17]  L. Kazembe,et al.  Spatial analysis of the relationship between early childhood mortality and malaria endemicity in Malawi. , 2007, Geospatial health.

[18]  Y. MacNab,et al.  Autoregressive Spatial Smoothing and Temporal Spline Smoothing for Mapping Rates , 2001, Biometrics.

[19]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[20]  H. Rue,et al.  Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations , 2009 .

[21]  S. Martino.  Approximate Bayesian Inference for Latent Gaussian Models , 2007 .

[22]  Zambia.  Zambia Demographic and Health Survey 2007 , 2009 .

[23]  Bradley P. Carlin,et al.  Structuring Correlation within Hierarchical Spatio-temporal Models for Disease Rates , 1997 .

[24]  Mike Rees,et al.  5. Statistics for Spatial Data , 1993 .

[25]  Alan D. Lopez,et al.  Measuring Under-Five Mortality: Validation of New Low-Cost Methods , 2010, PLoS medicine.

[26]  Mohsen Naghavi,et al.  Progress towards Millennium Development Goals 4 and 5 on maternal and child mortality: an updated systematic analysis , 2011, The Lancet.

[27]  Leonhard Held,et al.  A primer on disease mapping and ecological regression using $${\texttt{INLA}}$$ , 2011, Comput. Stat..

[28]  Spatial modeling of geographic inequalities in infant and child mortality across Nepal. , 2011, Health & place.

[29]  Stephen S. Lim,et al.  Prevalence of physical activity and obesity in US counties, 2001–2011: a road map for action , 2013, Population Health Metrics.

[30]  B. Carlin,et al.  Identifiability and convergence issues for Markov chain Monte Carlo fitting of spatial models. , 2000, Statistics in medicine.

[31]  W. Pan,et al.  Infant and Child Mortality in India in the Last Two Decades: A Geospatial Analysis , 2011, PloS one.

[32]  Andrew B. Lawson,et al.  Bayesian Disease Mapping: Hierarchical Modeling in Spatial Epidemiology , 2008 .

[33]  Y. MacNab,et al.  Spatio‐temporal modelling of rates for the construction of disease maps , 2002, Statistics in medicine.

[34]  I Kleinschmidt,et al.  Spatial patterns of infant mortality in Mali: the effect of malaria endemicity. , 2004, American journal of epidemiology.

[35]  Zambia,et al.  Zambia Demographic and Health Survey 2001-2002 , 2003 .

[36]  J. Besag,et al.  Bayesian image restoration, with two applications in spatial statistics , 1991 .

[37]  Bradley P. Carlin,et al.  Hierarchical Spatio-Temporal Mapping of Disease Rates , 1997 .

[38]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[39]  Taha Noorollah.  Modified Logit Life Table System: Principles, Empirical Validation, and Application , 2013 .

[40]  Jon Pedersen,et al.  Child Mortality Estimation: Appropriate Time Periods for Child Mortality Estimates from Full Birth Histories , 2012, PLoS medicine.

[41]  A. Mbonye,et al.  Examining small area estimation techniques for public health intervention: Lessons from application to under-5 mortality data in Uganda , 2011, Journal of public health policy.

[42]  David G. Clayton,et al.  Generalized linear mixed models , 1996 .

[43]  Alan D. Lopez,et al.  Equity and Geography: The Case of Child Mortality in Papua New Guinea , 2012, PloS one.

[44]  L Knorr-Held,et al.  Bayesian modelling of inseparable space-time variation in disease risk. , 2000, Statistics in medicine.