An adaptive spatio-temporal smoothing model for estimating trends and step changes in disease risk

Alastair Rushworth, Duncan Lee, Christophe Sarran

Research output: Working paper

Abstract

Statistical models used to estimate the spatio-temporal pattern in disease risk from areal unit data often represent the risk surface for each time period in terms of known covariates and a set of spatially smooth random effects. The latter act as a proxy for unmeasured spatial confounding, whose spatial structure is often characterised by a spatially smooth evolution between some pairs of adjacent areal units while other pairs exhibit large step changes. This spatial heterogeneity is not consistent with a global smoothing model in which partial correlation exists between all pairs of adjacent spatial random effects, and a novel space-time disease model with an adaptive spatial smoothing specification that can identify step changes is therefore proposed. The new model is motivated by a new study of respiratory and circulatory disease risk across the set of Local Authorities in England, and is rigorously tested by simulation to assess its efficacy. Results from the England study show that the two diseases have similar spatial patterns in risk, and exhibit a number of common step changes in the unmeasured component of risk between neighbouring local authorities.
LanguageEnglish
Number of pages27
Publication statusPublished - 4 Nov 2014

Fingerprint

Smoothing
Random Effects
Adjacent
Partial Correlation
Spatial Heterogeneity
Unit
Spatio-temporal Patterns
Model
Confounding
Spatial Pattern
Spatial Structure
Statistical Model
Efficacy
Covariates
Space-time
Trends
Specification
Estimate
Simulation

Keywords

  • adaptive smoothing
  • Gaussian Markov random fields
  • spatio-temporal disease mapping
  • step change detection

Cite this

@techreport{76a2a652bc794da3883f98e06125bb4c,
title = "An adaptive spatio-temporal smoothing model for estimating trends and step changes in disease risk",
abstract = "Statistical models used to estimate the spatio-temporal pattern in disease risk from areal unit data often represent the risk surface for each time period in terms of known covariates and a set of spatially smooth random effects. The latter act as a proxy for unmeasured spatial confounding, whose spatial structure is often characterised by a spatially smooth evolution between some pairs of adjacent areal units while other pairs exhibit large step changes. This spatial heterogeneity is not consistent with a global smoothing model in which partial correlation exists between all pairs of adjacent spatial random effects, and a novel space-time disease model with an adaptive spatial smoothing specification that can identify step changes is therefore proposed. The new model is motivated by a new study of respiratory and circulatory disease risk across the set of Local Authorities in England, and is rigorously tested by simulation to assess its efficacy. Results from the England study show that the two diseases have similar spatial patterns in risk, and exhibit a number of common step changes in the unmeasured component of risk between neighbouring local authorities.",
keywords = "adaptive smoothing, Gaussian Markov random fields, spatio-temporal disease mapping, step change detection",
author = "Alastair Rushworth and Duncan Lee and Christophe Sarran",
year = "2014",
month = "11",
day = "4",
language = "English",
type = "WorkingPaper",

}

An adaptive spatio-temporal smoothing model for estimating trends and step changes in disease risk. / Rushworth, Alastair; Lee, Duncan; Sarran, Christophe.

2014.

Research output: Working paper

TY - UNPB

T1 - An adaptive spatio-temporal smoothing model for estimating trends and step changes in disease risk

AU - Rushworth, Alastair

AU - Lee, Duncan

AU - Sarran, Christophe

PY - 2014/11/4

Y1 - 2014/11/4

N2 - Statistical models used to estimate the spatio-temporal pattern in disease risk from areal unit data often represent the risk surface for each time period in terms of known covariates and a set of spatially smooth random effects. The latter act as a proxy for unmeasured spatial confounding, whose spatial structure is often characterised by a spatially smooth evolution between some pairs of adjacent areal units while other pairs exhibit large step changes. This spatial heterogeneity is not consistent with a global smoothing model in which partial correlation exists between all pairs of adjacent spatial random effects, and a novel space-time disease model with an adaptive spatial smoothing specification that can identify step changes is therefore proposed. The new model is motivated by a new study of respiratory and circulatory disease risk across the set of Local Authorities in England, and is rigorously tested by simulation to assess its efficacy. Results from the England study show that the two diseases have similar spatial patterns in risk, and exhibit a number of common step changes in the unmeasured component of risk between neighbouring local authorities.

AB - Statistical models used to estimate the spatio-temporal pattern in disease risk from areal unit data often represent the risk surface for each time period in terms of known covariates and a set of spatially smooth random effects. The latter act as a proxy for unmeasured spatial confounding, whose spatial structure is often characterised by a spatially smooth evolution between some pairs of adjacent areal units while other pairs exhibit large step changes. This spatial heterogeneity is not consistent with a global smoothing model in which partial correlation exists between all pairs of adjacent spatial random effects, and a novel space-time disease model with an adaptive spatial smoothing specification that can identify step changes is therefore proposed. The new model is motivated by a new study of respiratory and circulatory disease risk across the set of Local Authorities in England, and is rigorously tested by simulation to assess its efficacy. Results from the England study show that the two diseases have similar spatial patterns in risk, and exhibit a number of common step changes in the unmeasured component of risk between neighbouring local authorities.

KW - adaptive smoothing

KW - Gaussian Markov random fields

KW - spatio-temporal disease mapping

KW - step change detection

UR - http://arxiv.org/abs/1411.0924

M3 - Working paper

BT - An adaptive spatio-temporal smoothing model for estimating trends and step changes in disease risk

ER -