Estimating micro area behavioural risk factor prevalence from large population-based surveys: a full Bayesian...

Area of Science:

• Spatial statistics
• Public health surveillance
• Geographic information systems (GIS)

Background:

• Public health aims to reduce behavioral risk factors like smoking and obesity.
• Regional survey data lacks granular detail for localized interventions.
• Spatial analysis can provide micro area estimates for targeted public health strategies.

Purpose of the Study:

• To estimate the micro area prevalence of current smoking and excess bodyweight.
• To identify specific geographic areas with high prevalence of these risk factors.
• To demonstrate advanced spatial analysis techniques for public health planning.

Main Methods:

• A spatial Bayesian hierarchical model was applied to Canadian Community Health Survey (CCHS) data.
• Micro area prevalence was estimated for current smoking and excess bodyweight.
• Models incorporated survey cycle, age group, and median household income, with SaTScan used for validation.

Main Results:

• Elevated current smoking rates were found in Sarnia and Windsor for both sexes, and Chatham for males.
• Excess bodyweight was prevalent in Windsor among males.
• Model 2, including income, improved precision for smoking estimates.

Conclusions:

• This study pioneers the use of Bayesian models with complex survey data for micro area risk factor identification.
• Micro area analysis reveals geographic variations in behavioral risk factors.
• Findings support targeted public health planning, surveillance, and chronic disease research.