1 / 17

Prepared for 2008 CPHA Conference, June 1-4, 2008 Mikiko Terashima Dr. Pantelis Andreou

Spatial assessment of deprivation and mortality risk in Nova Scotia: Comparison between Bayesian and non-Bayesian approaches. Prepared for 2008 CPHA Conference, June 1-4, 2008 Mikiko Terashima Dr. Pantelis Andreou Dr. Judith Guernsey Department of Community Health & Epidemiology

Download Presentation

Prepared for 2008 CPHA Conference, June 1-4, 2008 Mikiko Terashima Dr. Pantelis Andreou

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Spatial assessment of deprivation and mortality risk in Nova Scotia:Comparison between Bayesian and non-Bayesian approaches Prepared for 2008 CPHA Conference, June 1-4, 2008 Mikiko Terashima Dr. Pantelis Andreou Dr. Judith Guernsey Department of Community Health & Epidemiology Dalhousie University

  2. Objectives • To present basic ideas of Bayesian approaches in disease mapping • Perspectives of a health researcher with minimum statistical background • To demonstrate an example application of Bayesian approaches to spatial analysis of health in comparison with non-Bayesian approaches

  3. Non-Bayesian Parameters are assumed fixed (constant) Use likelihood to calculate parameters “Confidence interval”: Based on hypothetical situations where we were to sample e.g.100 times, 95 times out of them will contain the constant (true) parameter value Standard Error (SE) Chi-square, AIC, -2Log likelihood etc. for Goodness-of-fit Bayesian Parameters are assumed random Use likelihood and prior information to calculate posterior distribution, and estimate parameters from it (Estimated parameters [e.g. mean, median] are used like constant parameter values in non-Bayesian models) “Credible interval”: Based on a number of simulations, we know the distribution of the parameter—posterior distribution—and 95% of them fall between the interval A and B. Markov Chain Standard Error (MCSE) DIC for goodness-of-fit (like AIC but does not require Maximum Likelihood of parameters) Key differences

  4. Why use Bayesian approaches in spatial analysis of health/disease mapping? • It can incorporate spatial random effects in the analysis of spatial variation in health, disease or health/disease risks—Bayesian way is more ‘natural’ and straightforward. • The belief of parameters being random fits assumption that geographical area rates are somewhat different. • Because it “borrows strength” from prior knowledge about factors not in the data, the model is closer to reality than inferring things from only data –reduce noise from maps. • It allows us to estimate values of areas with missing data by sampling the same way as other parameters.

  5. Challenges • “Black box” for non-statisticians • “PC vs. Apple computers” • New and constantly evolving • Application of prior information requires statisticians’ insights and background knowledge about the variables.

  6. Example application: Spatial analysis of deprivation and mortality in Nova Scotia communities

  7. Scenario (Deprivation and morality in Nova Scotia) • A number of deaths were observed in 278 communities of Nova Scotia over the period 1995-2004 (obs). • Counts of events are likely Poisson distributed. Obs~Po() • We want to see the association between deprivation at the community level and observed counts of death. • The community level deprivation was measured by two sets of variables: material (Mat) and psychosocial (Psy). • Spatial autocorrelation might be playing a role (assumed normal). • We want to know if other (unknown) factors are affecting the rates. • We know the expected counts for each community (exp). • We want to map community level risks.

  8. Poisson regression model of counts Offset

  9. Models [+ ei] 1.Log(i)=offseti (+ β0i) + β1i x Mat + β2ix Psy 2.Log(i)=offseti (+ β0i) + β1i x Mat + β2ix Psy β0i = β0+ u0i , u0i ~ N(0, σ0) β1i = β1 + u1i , u1i ~ N(0, σ1) β2i = β2 + u1i , u1i ~ N(0, σ2) ei ~ N(0, σe) + v0i+ u0i (neigh) β0i ~ flat() β1i~ N(0, 1.0E-5) β2i ~ N(0, 1.0E-5) v0i ~ N(0, 1/σ2v) u0i (neigh)  CAR tau.u ~ gamma(0.5, 0.0005) tau.v ~ gamma(0.5, 0.0005) (unstructured random + neighbourhood effect) tau.v Simulated from a similar model template provided within WinBUGS software Conditional Autoregressive model

  10. Conditional AutoRegressive model (CAR)-Normal Spatial autocorrelation effect for neighbourhood i Number of neighbours Where Average of surrounding (adjacent) areas (neighbours) Weights are typically 1 (e.g. Besag, York & Mollie, 1991) • A common model to deal with spatial autocorrelation in disease mapping usually results in smoothing • Spatial autocorrelation effect for neighbourhood i depends on the number of neighbours and average of surrounding neighbours

  11. WinBUGS code Poisson regression model model{ for (i in 1 : N) { # Likelihood O[i] ~ dpois(mu[i]) log(mu[i]) <- log(Mrexp[i]) + beta0 + beta1* Mat[i] + beta2 * Psy[i] + b[i] + h[i] RR[i] <- exp(beta0 + beta1* Mat[i] +beta2 * Psy[i] + u[i] + v[i]) Mrexp[i] ~ dgamma (1, 0.1) Mat[i] ~ dnorm (0.0, 1.0) Psy[i] ~ dnorm (0.0, 1.0) # Exchangeable prior on unstructured random effects v[i] ~ dnorm(0, tau.v) } # CAR prior distribution for spatial random effects: u[1:N] ~ car.normal(adj[], weights[], num[], tau.u) for(k in 1:sumNumNeigh) { weights[k] <- 1 } # Other priors: beta0 ~ dflat() beta1 ~ dnorm(0.0, 1.0E-5) beta2 ~ dnorm(0.0, 1.0E-5) tau.u ~ dgamma(0.5, 0.0005) sigma.u <- sqrt(1 / tau.u) tau.v ~ dgamma(0.5, 0.0005) sigma.v <- sqrt(1 / tau.v) } RR for each community Filling missing values Unstructured random effects CAR Other parameter priors

  12. Parameter outputs Model 1 output with MLwiN node mean sd MC error 2.5% median 97.5% start sample beta0 0.1113 0.1676 0.009314 0.07684 0.1276 0.1718 1000 100001 beta1 0.07068 0.05098 0.002561 0.01209 0.06615 0.1272 1000 100001 beta2 0.1364 0.05499 0.002867 0.08638 0.1417 0.1894 1000 100001 sigma.u 0.04025 0.05424 0.002975 0.01283 0.02952 0.09874 1000 100001 sigma.v 0.3742 0.1511 0.008398 0.326 0.3581 0.4014 1000 100001 Log () = offset + 0.152 + 0.052 x Mat + 0.059 x Psy β0 β1 β2 DIC=18146.200 Model 2 output with WinBUGS DIC=7410.290 beta0/β0: Intercept Beta1/β1: Material deprivation Beta2/β2: Psychosocial deprivation sigma.u: Variance due to spatial autocorrelation sigma.v: Variance due to unstructured random effects

  13. SMR map (Model 1) versus RR map (Model 2 ) • Extremely high and low rates were reduced • Rate at each community shrunk to the mean (1.0) • Communities with missing data on the SMR map now have predicted values

  14. Conclusion • Three benefits of Bayesian approaches for spatial health analyses are: 1) random effects due to area variations can be easily incorporated; 2) smoothing effects; and 3) missing values can be inferred using prior knowledge (no holes in the map).  Working closely with knowledgeable statisticians, health researchers can use Bayesian approaches for these benefits (e.g. spatially analyzing and mapping health and health risks).

  15. Acknowledgement • The student is funded by Killam Pre-doctoral Scholarship and CIHR Strategic Training Program in Public Health and the Agricultural Rural Ecosystem (PHARE)

  16. Thank you!

More Related