Analysis of joint spatio-temporal characteristics of multiple diseases

Abstract Purpose: To explore the characteristics and risk factors of the joint incidence of Tuberculosis, Hepatitis B, and AIDS at the provincial level in China. Methods: A Bayesian hierarchical spatio-temporal model of multiple diseases is established. The shared components are used to identify diseases with similar spatio-temporal patterns, and specific components explain the specific spatio-temporal patterns of each disease. Finally, the integrated nested Laplace approximation algorithm is used to estimate the posterior mean of each effect parameter. Results: Tuberculosis and Hepatitis B have more similar shared space mode, and the shared space effect has the highest impact. The shared time risk of the three diseases has little change on the whole, and the change of specific time relative risk and combined time relative risk is quite different. Among them, the combined time relative risk of tuberculosis is almost constant, Hepatitis B first and then stabilizes, and AIDS shows a trend of increasing first and then decreasing. Conclusion: The geographical environment factors may be used as a replacement for the shared Spatial effect. It has a more significant impact on the onset of Tuberculosis and Hepatitis B and has a negligible impact on the onset of HIV. In addition, some risk factors, such as economy and policy, may have different time effects on these three diseases.