A Bayesian multinomial model to analyse spatial patterns of childhood co-morbidity in Malawi
Children in less developed countries die from relatively small number of infectious disease, some of which epidemiologically overlap. Using self-reported illness data from the 2000 Malawi Demographic and Health Survey, we applied a random effects multinomial model to assess risk factors of childhood co-morbidity of fever, diarrhoea and pneumonia, and quantify area-specific spatial effects. The spatial structure was modelled using the conditional autoregressive prior. Various models were fitted and compared using deviance information criterion. Inference was Bayesian and was based on Markov Chain Monte Carlo simulation techniques. We found spatial variation in childhood co-morbidity and determinants of each outcome category differed. Specifically, risk factors associated with child co-morbidity included age of the child, place of residence, undernutrition, bednet use and Vitamin A. Higher residual risk levels were identified in the central and southern–eastern regions, particularly for fever, diarrhoea and pneumonia; fever and pneumonia; and fever and diarrhoea combinations. This linkage between childhood health and geographical location warrants further research to assess local causes of these clusters. More generally, although each disease has its own mechanism, overlapping risk factors suggest that integrated disease control approach may be cost-effective and should be employed.
KeywordsChildhood co-morbidity Bayesian multinomial logit model Spatial modelling Multicategorical response data Conditional autoregressive models Malawi
Confidence Interval; Credible Interval
Deviance information criterion
Insecticide treated nets
Markov Chain Monte Carlo
Malawi demographic and health Survey
Relative odds ratio
I would like to acknowledge the research training grant received from WHO/TDR and support from Medical Research Council, Durban, South Africa during my PhD training. I also acknowledge permission granted by MEASURE DHS to use the 2000 Malawi DHS data under the project- Spatial analysis of malariometric indicators in Malawi. We thank the anonymous reviewers for valuable comments on the earlier manuscript.
- 3.National Statistical Office, ORC Macro. Malawi Demographic and Health Survey 2004: Preliminary report. Zomba, Malawi: NSO; 2005.Google Scholar
- 4.World Health Statistics. Country Health System Fact Sheet 2006– Malawi. Url: [http://www.who.int/whosis/en/
- 11.Elliot P, Wakefield J, Best N, Briggs DJ (eds). Spatial epidemiology-methods and applications. London: Oxford University Press; 2000Google Scholar
- 12.National Statistical Office, ORC Macro. Malawi Demographic and Health Survey 2000. Zomba, Malawi: NSO; 2001Google Scholar
- 17.Fahrmeir L, Tutz G. Multivariate statistical modelling based on generalized linear models. New York: Springer-Verlag; 2001.Google Scholar
- 22.Banerjee S, Carlin BP, Gelfand AE. Hierarchical modeling and analysis for spatial data. Chapman and Hall/CRC: London; 2004Google Scholar
- 23.Mugglin AS, Carlin BP, Gelfand AE. Fully model-based approaches for spatially misaligned data. JASA 2000;95:877–87Google Scholar
- 26.Brezger A, Kneib T, Lang S. BayesX: Analyzing Bayesian structured additive regression models. J Stat Soft 2005;14:11Google Scholar
- 27.Joint United Nations Programme on HIV/AIDS (UNAIDS). 2006 Report on the global AIDS epidemic. A UNAIDS 10th anniversary special edition; 2006. Available online at [http://www.unaids.org/en/HIV_data/2006GlobalReport/default.asp.
- 28.Kazembe LN. Spatial analysis of malariometric indicators in Malawi. PhD Thesis. University of KwaZulu-Natal, South Africa; 2006Google Scholar
- 32.World Bank. Profile of poverty in Malawi, 1998. Poverty analysis of the Malawian Integrated Household Survey, 1997–1998. Washington, D.C: The World Bank; 2000Google Scholar
- 33.Spiegelhalter DJ, Thomas A, Best NG. WinBUGS Version 1.2 User Manual. MRC Biostatistics Unit. University of London; 1999Google Scholar
- 34.Lambert PC. Comment on article by Browne and Draper. Bayesian Anal 2006;1:543–6Google Scholar