Development of a multiple regression model to identify multi-family residential buildings with a high prevalence of sick building syndrome (SBS).

The aim was to develop a multiple logistic regression model to identify multi-family houses with an increase of sick building syndrome (SBS). In Stockholm, 609 multi-family buildings with 14,235 dwellings were selected by stratified random sampling. The response rate was 77%. Multiple logistic regression analysis was applied, adjusting for ownership of the building, building age and size, age, gender, and atopy. Females, subjects with allergy, those above 65 yr, and those in new buildings reported significantly more SBS. Subjects owning their own building reported less SBS, but the relationship between ownership and building age was strong. A regression model, including factors with a high explanatory value was developed. According to the model, 5% of all buildings built before 1961, 13% of those built 1976-1984, and 15% of those built 1985-1990 would have significantly more SBS than expected. In conclusion, SBS is related to personal factors, building age, and ownership of the building. To identify multi-family buildings with more SBS than expected, it is necessary to adjust for ownership and population characteristics.