Risk factors for allergy: Evidence from cross sectional analysis of a UK primary care database.

Abstract

Abstract Background As the global prevalence of allergy increases, it is clear that more individuals are presenting with complex and multiple allergies. The impact of various demographic and clinical risk factors on developing allergies has not been previously explored at a population level. Objectives To ascertain if age, gender, ethnicity, geographical region of residence and being diagnosed with another allergy affect the risk of having allergies. Methods The health improvement network database (THIN) is a live primary care database which holds data generalisable to the UK population. A retrospective logistic regression analysis was performed using cross-sectional data for the year 2018 extracted from the THIN database. Demographic data and allergy diagnoses were used as independent variables to estimate the odds of developing allergies. Results Of the 3.03 million records analysed, 49.8% were male and 19.7% were children (aged < 18 yrs). There were gender, ethnicity and region-based differences in the prevalence of Primary Care Physician (PCP) diagnosed allergic diseases in the UK. Individuals with any diagnosed allergy had significantly higher odds of being diagnosed with others. Individuals with eczema and asthma were at higher risk of developing food and drug allergy as well as anaphylaxis. Those with urticaria were also more likely to be diagnosed with anaphylaxis. The odds of developing anaphylaxis are highest in individuals with food allergy [OR: 54.57 (95%CI: 45.65–65.22); p < 0.001 for children and OR:18.05 (95% CI:16.79–19.41); p < 0.001 for adults respectively]. Conclusions Having an allergy significantly increases the odds of being diagnosed with others. There are age, gender, ethnicity and region based differences in the prevalence of allergies. These data suggest that a significant proportion of individuals with allergies have complicated needs related to their ethnic, social, economic as well as clinical phenotype. This should be taken into account when planning for future allergy services.