Objective: To analyze the spatial distribution of AIDS in every city of Jiangsu province, trying to describe the geographic characteristics of AIDS and to develop a prediction model.
Methods: Numbers of patients in Jiangsu province were collected, to establish the database for the geographic information system, then setting up a 'risk map' of the disease. Spatial, autocorrelation. Linear spatial analyses were used to study the patients' numbers.
Results: (1) Results from the autocorrelation analysis showed that the distribution of AIDS was clustered at some places and was at random on the whole. The results also indicated that the distribution of AIDS in Nanjing was of negative correlation, while that in Suzhou and Wuxi were of positive correlation but in Tongshan, Wujiang, Pukou, Nanjing, Lishui, Wuxi and Suzhou showed seven locations of clusters with significantly higher numbers of patients. (2) The trend of 'surface analysis map' indicated that the disease was more severe in the southern than in the northern parts of Jiangsu province. (3) Ordinary Least Squares method was finally used in the linear spatial regression and the results were: t = -1.045 103 (P = 0.299 904); t = -1.443 668 (P = 0.153 714) respectively.
Conclusion: According to the feature spatial distribution of the disease, effective measures should be taken to prevent and to keep the prevalence of AIDS under control.