Abstract
Progress towards government health targets for health areas may be assessed by short term extrapolation of recent trends. Often the observed longitudinal series for a set of health areas is relatively short and a parsimonious model is needed that is adapted to varying observed trajectories between areas. A forecasting model should also include spatial dependence between areas both in representing stable cross-sectional differences and in terms of changing incidence. A fully Bayesian spatio-temporal forecasting model is developed incorporating flexible but parsimonious time dependence while allowing spatial dependencies. An application involves conception rates to women aged under 18 in the 32 boroughs of London.
Highlights
Government health improvement targets to reduce adverse events or health inequity increasingly involve forecasts to assess how far much progress towards the target has been achieved
When the targets are set for health areas, this requires a forecast for a panel data situation, with observations formed by a collection of incidence rates by areas i = 1, . . . , N and times t = 1, . . . , T, and with forecasts required for times t = T + 1, . . . , T + R
This paper sets out a spatio-temporal forecasting model suited to such analysis that allows both for enduring spatial patterning in the health outcome and for spatial clustering of growth or decline in incidence
Summary
Government health improvement targets to reduce adverse events or health inequity increasingly involve forecasts to assess how far much progress towards the target has been achieved. This paper sets out a spatio-temporal forecasting model suited to such analysis that allows both for enduring spatial patterning in the health outcome and for spatial clustering of growth or decline in incidence. Peter Congdon sector interventions, though may still show spatial patterning This is true for the application considered here which involves short term forecasts of conception rates to women under 18 in the 32 London boroughs, using observations over 1992-2003; the percent change in this rate between 1992-95 and 2000-2003 has a correlation of −0.14 with the average rate over the entire 12 year period. The model developed here combines autoregressive and growth curve principles within a spatio-temporal forecasting framework; the model allows for spatial patterning both in the health outcome and in changing incidence of the outcome, with area change profiles modelled using fractional polynomials. Estimation involves repeated sampling via Markov Chain Monte Carlo procedures, and uses the WINBUGS package
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
