Risk Factors in Heating, Ventilating, and Air-Conditioning Systemsfor Occupant Symptoms in U.S. Office Buildings: the EPA BASE Study

Abstract

Nonspecific building-related symptoms among occupants of modern office buildings worldwide are common and may be associated with important reductions in work performance, but their etiology remains uncertain. Characteristics of heating, ventilating, and air-conditioning (HVAC) systems in office buildings that increase risk of indoor contaminants or reduce effectiveness of ventilation may cause adverse exposures and subsequent increase in these symptoms among occupants. We analyzed data collected by the U.S. EPA from a representative sample of 100 large U.S. office buildings--the Building Assessment and Survey Evaluation (BASE) study--using multivariate logistic regression models with generalized estimating equations adjusted for potential personal and building confounders. We estimated odds ratios (ORs) and 95% confidence intervals (CIs) for associations between seven building-related symptom outcomes and selected HVAC system characteristics. Among factors of HVAC design or configuration: Outdoor air intakes less than 60 m above the ground were associated with approximately doubled odds of most symptoms assessed. Sealed (non-operable) windows were associated with increases in skin and eye symptoms (ORs= 1.9, 1.3, respectively). Outdoor air intake without an intake fan was associated with an increase in eye symptoms (OR=1.7). Local cooling coils were associated with increased headache (OR=1.5). Among factors of HVAC condition, maintenance, or operation: the presence of humidification systems in good condition was associated with an increase in headache (OR=1.4), whereas the presence of humidification systems in poor condition was associated with increases in fatigue/difficulty concentrating, as well as upper respiratory symptoms (ORs=1.8, 1.5). No regularly scheduled inspections for HVAC components was associated with increased eye symptoms, cough and upper respiratory symptoms (ORs=2.2, 1.6, 1.5). Less frequent cleaning of cooling coils or drip pans was associated with increased headache (OR=1.6). Fair or poor condition of duct liner was associated with increased upper respiratory symptoms (OR=1.4). Most of the many potential risk factors assessed here had not been investigated previously, and associations found with single symptoms may have been by chance, including several associations that were the reverse of expected. Risk factors newly identified in these analyses that deserve attention include outdoor air intakes less than 60 m above the ground, lack of operable windows, poorly maintained humidification systems, and lack of scheduled inspection for HVAC systems. Infrequent cleaning of cooling coils and drain pans were associated with increases in several symptoms in these as well as prior analyses of BASE data. Replication of these findings is needed, using more objective measurements of both exposure and health response. Confirmation of the specific HVAC factors responsible for increased symptoms in buildings, and development of prevention strategies could have major public health and economic benefits worldwide.