The Two Cultures for Prevalence Mapping: Small Area Estimation and Spatial Statistics

The emerging need for subnational estimation of demographic and health indicators in lowand middle-income countries (LMICs) is driving a move from design-based methods to spatial and spatio-temporal approaches. The latter are model-based and overcome data sparsity by borrowing strength across space, time and covariates and can, in principle, be leveraged to create yearly fine-scale pixel level maps based on household surveys. However, typical implementations of the model-based approaches do not fully acknowledge the complex survey design, and do not enjoy the theoretical consistency of design-based approaches. We describe how spatial and spatio-temporal methods are currently used for small area estimation in the context of LMICs, highlight the key challenges that need to be overcome, and discuss a new approach, which is methodologically closer in spirit to small area estimation. The main discussion points are demonstrated through two case studies: spatial analysis of vaccination coverage in Nigeria based on the 2018 Demographic and Health Surveys (DHS) survey, and spatio-temporal analysis of neonatal mortality in Malawi based on 2010 and 2015–2016 DHS surveys. We discuss our key findings both generally and with an emphasis on the implications for popular approaches undertaken by industrial producers of subnational prevalence estimates.

[1]  J Lessler,et al.  A spatial regression model for the disaggregation of areal unit based data to high-resolution grids with application to vaccination coverage mapping , 2018, Statistical methods in medical research.

[2]  Jin Rou New,et al.  Global estimation of child mortality using a Bayesian B-spline Bias-reduction model , 2013, 1309.1602.

[3]  Jon Wakefield,et al.  Estimating under-five mortality in space and time in a developing world context , 2017, Statistical methods in medical research.

[4]  A. Tatem,et al.  District‐level estimation of vaccination coverage: Discrete vs continuous spatial models , 2021, Statistics in medicine.

[5]  Haavard Rue,et al.  Constructing Priors that Penalize the Complexity of Gaussian Random Fields , 2015, Journal of the American Statistical Association.

[6]  K. Battle,et al.  A global map of travel time to cities to assess inequalities in accessibility in 2015 , 2018, Nature.

[7]  Peter J. Diggle,et al.  Model-Based Geostatistics for Prevalence Mapping in Low-Resource Settings , 2015, 1505.06891.

[8]  D. Pfeffermann.  The Role of Sampling Weights when Modeling Survey Data , 1993 .

[9]  William C. Gagne-Maynard,et al.  Mapping diphtheria-pertussis-tetanus vaccine coverage in Africa, 2000–2016: a spatial and temporal modelling study , 2019, The Lancet.

[10]  Ramalingam Shanmugam,et al.  Model-based geostatistics for global public health: methods and applications , 2019, Journal of Statistical Computation and Simulation.

[11]  Andrew J. Tatem,et al.  The geography of measles vaccination in the African Great Lakes region , 2017, Nature Communications.

[12]  S. Hay,et al.  Development and validation of a new method for indirect estimation of neonatal, infant, and child mortality trends using summary birth histories , 2018, PLoS medicine.

[13]  Thiago G. Martins,et al.  Penalising Model Component Complexity: A Principled, Practical Approach to Constructing Priors , 2014, 1403.4630.

[14]  Anders Nielsen,et al.  TMB: Automatic Differentiation and Laplace Approximation , 2015, 1509.00660.

[15]  Christian P. Robert,et al.  Statistics for Spatio-Temporal Data , 2014 .

[16]  R. Sugden,et al.  Ignorable and informative designs in survey sampling inference , 1984 .

[17]  Matthew J Ferrari,et al.  Mapping vaccination coverage to explore the effects of delivery mechanisms and inform vaccination strategies , 2019, Nature Communications.

[18]  R Carroll,et al.  Spatial small area smoothing models for handling survey data with nonresponse , 2017, Statistics in medicine.

[19]  L. Alkema,et al.  Global estimation of neonatal mortality using a Bayesian hierarchical splines regression model , 2016, 1612.03561.

[20]  Haavard Rue,et al.  Bayesian Computing with INLA: A Review , 2016, 1604.00860.

[21]  Jon Wakefield,et al.  Harmonizing child mortality data at disparate geographic levels , 2021, Statistical methods in medical research.

[22]  A. Barros,et al.  Monitoring subnational regional inequalities in health: measurement approaches and challenges , 2016, International Journal for Equity in Health.

[23]  Natalia Rojas-Perilla,et al.  From start to finish: a framework for the production of small area official statistics , 2018, Journal of the Royal Statistical Society: Series A (Statistics in Society).

[24]  Jon Wakefield,et al.  Small Area Estimation of Child Mortality in the Absence of Vital Registration , 2014 .

[25]  R. Lehtonen,et al.  Chapter 31 - Design-based Methods of Estimation for Domains and Small Areas , 2009 .

[26]  C. Darker,et al.  Is the Urban Child Health Advantage Declining in Malawi?: Evidence from Demographic and Health Surveys and Multiple Indicator Cluster Surveys , 2018, Journal of Urban Health.

[27]  Roderick J. A. Little.  The Bayesian Approach to Sample Survey Inference , 2003 .

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

[29]  Jon Wakefield,et al.  Changes in the spatial distribution of the under-five mortality rate: Small-area analysis of 122 DHS surveys in 262 subregions of 35 countries in Africa , 2019, PloS one.

[30]  Joseph Hilbe,et al.  Data Analysis Using Regression and Multilevel/Hierarchical Models , 2009 .

[31]  Haavard Rue,et al.  Spatial modelling with R-INLA: A review , 2018, 1802.06350.

[32]  Jon Wakefield,et al.  A Statistical Introduction to Template Model Builder: A Flexible Tool for Spatial Modeling , 2021 .

[33]  Norman E. Breslow,et al.  Estimation of Disease Rates in Small Areas: A new Mixed Model for Spatial Dependence , 2000 .

[34]  Tatsuya Kubokawa,et al.  Estimation of mean squared error of model-based small area estimators , 2009 .

[35]  Session,et al.  Resolution Adopted By The General Assembly , 1984, International Legal Materials.

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

[37]  Peter J. Diggle,et al.  PrevMap:An R Package for Prevalence Mapping , 2017 .

[38]  Thomas Lumley,et al.  Analysis of Complex Survey Samples , 2004 .

[39]  Andrea Riebler,et al.  An intuitive Bayesian spatial model for disease mapping that accounts for scaling , 2016, Statistical methods in medical research.

[40]  Domingo Morales,et al.  Small area estimation with spatio-temporal Fay-Herriot models , 2013, Comput. Stat. Data Anal..

[41]  Haavard Rue,et al.  Penalised Complexity Priors for Stationary Autoregressive Processes , 2016, 1608.08941.

[42]  Rachel M. Harter,et al.  An Error-Components Model for Prediction of County Crop Areas Using Survey and Satellite Data , 1988 .

[43]  D. Horvitz,et al.  A Generalization of Sampling Without Replacement from a Finite Universe , 1952 .

[44]  Andrew J. Tatem,et al.  High resolution age-structured mapping of childhood vaccination coverage in low and middle income countries , 2018, Vaccine.

[45]  D. Basu,et al.  An Essay on the Logical Foundations of Survey Sampling, Part One* , 2011 .

[46]  H. Rue,et al.  Scaling intrinsic Gaussian Markov random field priors in spatial modelling , 2014 .

[47]  P. Diggle,et al.  Geostatistical inference under preferential sampling , 2010 .

[48]  John Bryant,et al.  Fully Bayesian Benchmarking of Small Area Estimation Models , 2020 .

[49]  D. Rhoda,et al.  Geospatial variation in measles vaccine coverage through routine and campaign strategies in Nigeria: Analysis of recent household surveys , 2020, Vaccine.

[50]  Finn Lindgren,et al.  Bayesian Spatial Modelling with R-INLA , 2015 .

[51]  Jon Wakefield,et al.  Design- and Model-Based Approaches to Small-Area Estimation in A Low- and Middle-Income Country Context: Comparisons and Recommendations , 2019, Journal of Survey Statistics and Methodology.

[52]  Robert D. Tortora,et al.  Sampling: Design and Analysis , 2000 .

[53]  R. Fay,et al.  Estimates of Income for Small Places: An Application of James-Stein Procedures to Census Data , 1979 .

[54]  Manfred S. Green,et al.  Mapping 123 million neonatal, infant and child deaths between 2000 and 2017 , 2019, Nature.

[55]  Jon Wakefield,et al.  Pointless spatial modeling. , 2018, Biostatistics.

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

[57]  Jon Wakefield,et al.  Space-Time Smoothing of Demographic and Health Indicators using the R Package SUMMER , 2020, 2007.05117.

[58]  I. Molina,et al.  A Map of the Poor or a Poor Map? , 2021, Policy Research Working Papers.

[59]  Kidane Tadesse Gebremariam,et al.  Mapping routine measles vaccination in low- and middle-income countries , 2020, Nature.

[60]  T. Vos,et al.  Guidelines for Accurate and Transparent Health Estimates Reporting: the GATHER statement , 2016, PLoS medicine.

[61]  Jean D. Opsomer,et al.  Model-Assisted Survey Estimation with Modern Prediction Techniques , 2017 .

[62]  A. Scott,et al.  Estimation in Multi-Stage Surveys , 1969 .