Bayesian Disease Mapping for Head and Neck cancer in Queensland

Abstract

Aim or Purpose: Bayesian mapping, an effective spatiotemporal approach identifies high risk areas for diseases, has not been used to identify Head and Neck cancer in Australia previously. This study utilises Bayesian mapping to identify communities with higher risk of head and neck cancer so the targeted public health screening and healthcare interventions can be planned and delivered. Materials and Methods: This study analysed the Queensland Cancer Registry data on patients with head and neck cancers recorded between 1982-2018. Domains mapped included the head and neck cancer incidence, overall mortality, 3-year and 5-year mortality in Queensland (QLD). Results: Of the 78 LGAs in QLD, 30 LGAs showed a significantly higher head and neck cancer incidence using higher median smoothed Incidence Risk (SIR), above the state average. Furthermore, mortality risk estimated using smoothed Mortality Risk (SMR), were significantly higher than state average in 38 LGAs. Seventeen LGAs had median SMR above 100% of the state average whilst four LGAs had the highest overall, 3- and 5-year mortality risks – Yarrabah Aboriginal Shire, Cook Shire, Murweh Shire and Mount Isa City. Whilst the reasons for some LGAs exhibiting higher disease incidence are unknown, Bayesian mapping highlights those regions worthy of further investigation. Conclusions: Bayesian disease mapping approach is effective in identification of high-risk communities for head and neck cancer in Queensland. This modelling also identified that incidence and overall mortality due to head and neck cancer were within regional QLD. Findings will aid in designing targeted public health screening and interventions for primary prevention of head and neck cancer in regional and remote communities.