Geospatial analysis of childhood pertussis in Victoria, 1993-97.
OBJECTIVE: To demonstrate the feasibility of using Geographical Information Systems (GIS) in public health research. METHODS: Area-based relationships between the incidence of pertussis and immunisation coverage using data on Victorian children aged 0-4 years who contracted pertussis in 1993-97, were analysed at the Local Govemment Area (LGA) level. DTP3 immunisation uptake by LGA was first stratified into two groups: > or = 90% or < 90% (national target). Those under 90% were then divided into two equal bands. The incidence rates for LGAs were classified into quartiles. Immunisation records of children in the 1996-97 birth cohort were reviewed. RESULTS: When the DTP3 immunisation uptake was > or = 90%, seven LGAs were in the upper quartile and eight LGAs in the lower quartile of childhood pertussis notifications incidence. An equal amount of LGAs were in the upper (n = 6) and lower quartiles (n = 6) of notified childhood pertussis incidence rates with an associated DTP3 coverage rate in the lowest band. Of children bom between 1996-97 who contracted pertussis under the age of two, 29% were not immunised and 78.6% were partially immunised at the time of pertussis onset. CONCLUSIONS: GIS enabled the integration of disparate immunisation-related datasets and identified geographic LGAs where immunisation rates were low and pertussis rates were high. IMPLICATIONS: Future research needs to explore the social and environmental factors associated with those not fully immunised at onset of pertussis disease, particularly those in geographic areas highlighted by this research. GIS has the potential to add value in the monitoring and surveillance of geographical patterns of child health, but investments in data quality are urgently required.