A spatial regression model for the disaggregation of areal unit based data to high-resolution grids with application to vaccination coverage mapping

C. E. Utazi, J. Thorley, V. A. Alegana, M. J. Ferrari, K. Nilsen, S. Takahashi, C. J.E. Metcalf, J. Lessler, A. J. Tatem

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The growing demand for spatially detailed data to advance the Sustainable Development Goals agenda of ‘leaving no one behind’ has resulted in a shift in focus from aggregate national and province-based metrics to small areas and high-resolution grids in the health and development arena. Vaccination coverage is customarily measured through aggregate-level statistics, which mask fine-scale heterogeneities and ‘coldspots’ of low coverage. This paper develops a methodology for high-resolution mapping of vaccination coverage using areal data in settings where point-referenced survey data are inaccessible. The proposed methodology is a binomial spatial regression model with a logit link and a combination of covariate data and random effects modelling two levels of spatial autocorrelation in the linear predictor. The principal aspect of the model is the melding of the misaligned areal data and the prediction grid points using the regression component and each of the conditional autoregressive and the Gaussian spatial process random effects. The Bayesian model is fitted using the INLA-SPDE approach. We demonstrate the predictive ability of the model using simulated data sets. The results obtained indicate a good predictive performance by the model, with correlations of between 0.66 and 0.98 obtained at the grid level between true and predicted values. The methodology is applied to predicting the coverage of measles and diphtheria-tetanus-pertussis vaccinations at 5 × 5 km2 in Afghanistan and Pakistan using subnational Demographic and Health Surveys data. The predicted maps are used to highlight vaccination coldspots and assess progress towards coverage targets to facilitate the implementation of more geographically precise interventions. The proposed methodology can be readily applied to wider disaggregation problems in related contexts, including mapping other health and development indicators.

Original languageEnglish (US)
Pages (from-to)3226-3241
Number of pages16
JournalStatistical Methods in Medical Research
Volume28
Issue number10-11
DOIs
StatePublished - Nov 1 2019

Fingerprint

Disaggregation
Vaccination
Spatial Model
Regression Model
Coverage
High Resolution
Grid
Unit
Health
Methodology
Survey Data
Random Effects
Afghanistan
Spatial Analysis
Diphtheria
Whooping Cough
Pakistan
Conservation of Natural Resources
Tetanus
Measles

All Science Journal Classification (ASJC) codes

  • Epidemiology
  • Statistics and Probability
  • Health Information Management

Cite this

Utazi, C. E. ; Thorley, J. ; Alegana, V. A. ; Ferrari, M. J. ; Nilsen, K. ; Takahashi, S. ; Metcalf, C. J.E. ; Lessler, J. ; Tatem, A. J. / A spatial regression model for the disaggregation of areal unit based data to high-resolution grids with application to vaccination coverage mapping. In: Statistical Methods in Medical Research. 2019 ; Vol. 28, No. 10-11. pp. 3226-3241.
@article{f6db55393eda4958a1c46df9561cf476,
title = "A spatial regression model for the disaggregation of areal unit based data to high-resolution grids with application to vaccination coverage mapping",
abstract = "The growing demand for spatially detailed data to advance the Sustainable Development Goals agenda of ‘leaving no one behind’ has resulted in a shift in focus from aggregate national and province-based metrics to small areas and high-resolution grids in the health and development arena. Vaccination coverage is customarily measured through aggregate-level statistics, which mask fine-scale heterogeneities and ‘coldspots’ of low coverage. This paper develops a methodology for high-resolution mapping of vaccination coverage using areal data in settings where point-referenced survey data are inaccessible. The proposed methodology is a binomial spatial regression model with a logit link and a combination of covariate data and random effects modelling two levels of spatial autocorrelation in the linear predictor. The principal aspect of the model is the melding of the misaligned areal data and the prediction grid points using the regression component and each of the conditional autoregressive and the Gaussian spatial process random effects. The Bayesian model is fitted using the INLA-SPDE approach. We demonstrate the predictive ability of the model using simulated data sets. The results obtained indicate a good predictive performance by the model, with correlations of between 0.66 and 0.98 obtained at the grid level between true and predicted values. The methodology is applied to predicting the coverage of measles and diphtheria-tetanus-pertussis vaccinations at 5 × 5 km2 in Afghanistan and Pakistan using subnational Demographic and Health Surveys data. The predicted maps are used to highlight vaccination coldspots and assess progress towards coverage targets to facilitate the implementation of more geographically precise interventions. The proposed methodology can be readily applied to wider disaggregation problems in related contexts, including mapping other health and development indicators.",
author = "Utazi, {C. E.} and J. Thorley and Alegana, {V. A.} and Ferrari, {M. J.} and K. Nilsen and S. Takahashi and Metcalf, {C. J.E.} and J. Lessler and Tatem, {A. J.}",
year = "2019",
month = "11",
day = "1",
doi = "10.1177/0962280218797362",
language = "English (US)",
volume = "28",
pages = "3226--3241",
journal = "Statistical Methods in Medical Research",
issn = "0962-2802",
publisher = "SAGE Publications Ltd",
number = "10-11",

}

A spatial regression model for the disaggregation of areal unit based data to high-resolution grids with application to vaccination coverage mapping. / Utazi, C. E.; Thorley, J.; Alegana, V. A.; Ferrari, M. J.; Nilsen, K.; Takahashi, S.; Metcalf, C. J.E.; Lessler, J.; Tatem, A. J.

In: Statistical Methods in Medical Research, Vol. 28, No. 10-11, 01.11.2019, p. 3226-3241.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A spatial regression model for the disaggregation of areal unit based data to high-resolution grids with application to vaccination coverage mapping

AU - Utazi, C. E.

AU - Thorley, J.

AU - Alegana, V. A.

AU - Ferrari, M. J.

AU - Nilsen, K.

AU - Takahashi, S.

AU - Metcalf, C. J.E.

AU - Lessler, J.

AU - Tatem, A. J.

PY - 2019/11/1

Y1 - 2019/11/1

N2 - The growing demand for spatially detailed data to advance the Sustainable Development Goals agenda of ‘leaving no one behind’ has resulted in a shift in focus from aggregate national and province-based metrics to small areas and high-resolution grids in the health and development arena. Vaccination coverage is customarily measured through aggregate-level statistics, which mask fine-scale heterogeneities and ‘coldspots’ of low coverage. This paper develops a methodology for high-resolution mapping of vaccination coverage using areal data in settings where point-referenced survey data are inaccessible. The proposed methodology is a binomial spatial regression model with a logit link and a combination of covariate data and random effects modelling two levels of spatial autocorrelation in the linear predictor. The principal aspect of the model is the melding of the misaligned areal data and the prediction grid points using the regression component and each of the conditional autoregressive and the Gaussian spatial process random effects. The Bayesian model is fitted using the INLA-SPDE approach. We demonstrate the predictive ability of the model using simulated data sets. The results obtained indicate a good predictive performance by the model, with correlations of between 0.66 and 0.98 obtained at the grid level between true and predicted values. The methodology is applied to predicting the coverage of measles and diphtheria-tetanus-pertussis vaccinations at 5 × 5 km2 in Afghanistan and Pakistan using subnational Demographic and Health Surveys data. The predicted maps are used to highlight vaccination coldspots and assess progress towards coverage targets to facilitate the implementation of more geographically precise interventions. The proposed methodology can be readily applied to wider disaggregation problems in related contexts, including mapping other health and development indicators.

AB - The growing demand for spatially detailed data to advance the Sustainable Development Goals agenda of ‘leaving no one behind’ has resulted in a shift in focus from aggregate national and province-based metrics to small areas and high-resolution grids in the health and development arena. Vaccination coverage is customarily measured through aggregate-level statistics, which mask fine-scale heterogeneities and ‘coldspots’ of low coverage. This paper develops a methodology for high-resolution mapping of vaccination coverage using areal data in settings where point-referenced survey data are inaccessible. The proposed methodology is a binomial spatial regression model with a logit link and a combination of covariate data and random effects modelling two levels of spatial autocorrelation in the linear predictor. The principal aspect of the model is the melding of the misaligned areal data and the prediction grid points using the regression component and each of the conditional autoregressive and the Gaussian spatial process random effects. The Bayesian model is fitted using the INLA-SPDE approach. We demonstrate the predictive ability of the model using simulated data sets. The results obtained indicate a good predictive performance by the model, with correlations of between 0.66 and 0.98 obtained at the grid level between true and predicted values. The methodology is applied to predicting the coverage of measles and diphtheria-tetanus-pertussis vaccinations at 5 × 5 km2 in Afghanistan and Pakistan using subnational Demographic and Health Surveys data. The predicted maps are used to highlight vaccination coldspots and assess progress towards coverage targets to facilitate the implementation of more geographically precise interventions. The proposed methodology can be readily applied to wider disaggregation problems in related contexts, including mapping other health and development indicators.

UR - http://www.scopus.com/inward/record.url?scp=85060451398&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85060451398&partnerID=8YFLogxK

U2 - 10.1177/0962280218797362

DO - 10.1177/0962280218797362

M3 - Article

C2 - 30229698

AN - SCOPUS:85060451398

VL - 28

SP - 3226

EP - 3241

JO - Statistical Methods in Medical Research

JF - Statistical Methods in Medical Research

SN - 0962-2802

IS - 10-11

ER -