Application of spatial Besag, York and Mollie method in estimating interprovincial neck pain prevalence in the National Disease and Health Survey in Iran

Background Aim: Geographical analysis of the frequency of disease incidence can have an important role in the allocation of resources, facilities, and manpower in addition to the formulation and evolution of etiological assumptions. The main objective of this study was to estimate the prevalence of neck pain interprovincially, and set a disease mapping using spatial Besag, York and Mollie (BYM) with regard to surrounding neighborhoods. To reduce the incidence of neck pain in adulthood, identification of risk factors that predict the onset, and continuation of pain in the patients is important. Methods Materials: The population examined in this study was extrapolated from records of the “National Disease and Health Survey in Iran,” which had a data plan of a general population survey conducted during 1999-2000, in which adults responded on the incidence of neck pain. The participants were guided by a questionnaire that had an image on which they could identify the exact location of the pain. Results: Explanatory variables in the model included sex, education level, area of residence, smoking, age and body mass index, and all of them showed a significant relationship with neck pain. To have a better model for a more reliable prediction, we grouped the provinces into divisions to have a more regular shape since the spatial BYM model cannot simultaneously account for population and spatial patterns. In neck pain, prevalence estimated by spatial BYM, Lorestan province with 7.85% had the lowest prevalence while Kurdistan province with 17.27% had the highest prevalence. Furthermore, in the male population, Ghazvin province with 5.53% had the lowest prevalence, whereas Kurdistan province with 10.33% had the highest prevalence of neck pain. Besides, in the female population, the Lorestan province with 10.33% had the lowest prevalence, while the province of Yazd with 22.45% has the highest prevalence of neck pain. Conclusion: In this study, the model assumed included measurable and immeasurable factors to provide reliable estimates for each province. The application of spatial BYM method with the inclusion of the location of disease occurrence is a more efficient and reliable method for diseases mapping with a higher power of predictability.