We've made some changes to EPA.gov. If the information you are looking for is not here, you may be able to find it on the EPA Web Archive or the January 19, 2017 Web Snapshot.

Indoor Air Quality (IAQ)

Highlighted Analysis of the Building Assessment Survey and Evaluation Study

On this Page:

Building Ventilation Measurements

Ventilation with clean outdoor air is a key component for maintaining good indoor air quality. Ventilation dilutes and removes contaminants generated by occupants and building-related sources. The BASE protocol included collecting data to assess quantity of ventilation air provided to the study spaces. Data collected that informs this assessment included:

  • Collection of HVAC design data.
  • Direct volumetric measurements of airflow quantities for air handlers serving study spaces, such as outdoor air intake, supply airflow, where possible.
  • Carbon dioxide, or CO2, measurements in HVAC air streams to enable estimation of percentage of outdoor air relative to total supply airflow and subsequent calculation of outdoor airflow quantity, where possible.

This information was collected in a standardized manner using methods described in the 2003 Standardized Protocol.

An in-depth analysis of the BASE ventilation data has been completed by the National Institute of Standards and Technology, or NIST, via an Interagency Agreement with the EPA. Outcomes of NIST’s analysis of the BASE ventilation data include:

  • Detailed analysis of ventilation performance parameters (outdoor air intake, supply airflow, etc.) for each BASE study space.
  • Comparison of the ventilation parameters determined by different methods (i.e, direct measurements vs. calculated from CO2 measurements).
  • Comparison of ventilation parameters to design values and industry standards.

The November 2008 NIST report, Analysis of Ventilation Data from the U.S. Environmental Protection Agency Building Assessment Survey and Evaluation (BASE) Study, is a revision of the original report of these data published in 2004. This revision reflects some additional analyses of the data, and results in changes to some of the numerical values reported, but not to the overall conclusions. The additional analyses are discussed in Appendix F of this revised report.

The EPA has identified some key summary findings from the NIST report, however the report should be referenced for additional context and supporting information:

  • Ninety-seven of the one hundred BASE study spaces were mechanically-ventilated; three study spaces were naturally-ventilated. Five BASE study spaces had one hundred percent outdoor air systems.
  • There was a lack of system design information available at many BASE buildings. For example, design minimum outdoor airflows were available for only about one-half of the BASE study space air handlers.
  • Available design data confirmed some common expectations. For example, as might be expected based on thermal load conditions, design supply airflow rates were about 1 cubic foot per minute, or cfm, per square foot of office floor area. Ventilation systems were designed for about ten to twenty percent minimum outdoor air relative to the total supply airflow, on average.
  • Over seventy percent of the BASE study spaces had air handlers equipped with economizers to provide “free-cooling” by increasing outdoor airflow during mild weather. A significant number of BASE measurements occurred during mild or moderate outdoor air temperatures, and the result was increased outdoor air intake rates in systems equipped with economizers to provide free-cooling. Measurements revealed an average outdoor air fraction of nearly forty percent, compared with ten to twenty percent outdoor air that is typical under minimum intake conditions.
  • The average per person outdoor air ventilation rate for all BASE buildings was higher than what might be expected, due primarily due to high outdoor air fractions, or relative to minimum, and actual occupancy being on average slightly less than eighty percent of the design occupancy.
  • Many study spaces had ventilation rates at or below the ASHRAE Standard 62-2001 recommendation of 20 cfm per person for offices. About seventeen percent of the BASE ventilation measurements were less than 20 cfm per person.
  • Many systems had measured airflows significantly different from design values. About 40 percent of the systems with design data available had measured outdoor airflows less than design. Furthermore, forty-four percent of buildings reported not performing HVAC testing and balancing. These results stress the importance of commissioning for new buildings prior to occupancy, and periodic re-commissioning of existing buildings, to ensure that buildings and systems are operating in a manner consistent with the design intent.
  • About twenty percent of air handlers did not have direct measurements of outdoor airflow, usually due to access and physical space limitations. In these cases, outdoor airflow was based on the difference between supply and recirculation airflows. Access to intakes and ductwork for airflow measurements is an important design consideration.

Top of Page

Environmental Monitoring

Biological Contaminants

Airborne biological contaminants were collected in each BASE study space on the third day, or Wednesday, of the five-day BASE study week, using the standard methods described in the 2003 Standardized Protocol. Airborne samples were collected on Wednesday morning and afternoon at one outdoor location and three indoor locations for each study space. Airborne culturable bacteria and fungi samples in all one hundred BASE study spaces were collected for two-minute and five-minute sampling periods using single-stage, multiple-hole impactors, in this case, those manufactured by Anderson Instruments, Smyrna GA. Simultaneous duplicate samples were collected at the outdoor location and one of the three indoor locations.

The one hundred BASE buildings were distributed among 10 different climate regions (see the BASE Regional Map) in the United States. A description of these climate regions, including a profile of the number of BASE buildings in each region and the seasons in which the buildings were studied is summarized in the BASE Building Distribution by Climate Chart. Fifty-two buildings were studied in the summer and 48 buildings in the winter. The climatic conditions and study season can be of particular interest when examining airborne biological contaminants.

Detailed analyses of the BASE data for airborne biological contaminants have been conducted by the California Department of Health Services under an agreement between the U.S. EPA and Public Health Foundation Enterprises, Inc, City of Industry, California. The results of these analyses are described in several technical documents, including:

  • Macher, J.M.; Tsai, F.C.; Burton, L.E.; Liu, K.S.; Waldman, J.M. 2001. "Prevalence of culturable airborne fungi in one hundred U.S. office buildings in the Building Assessment Survey and Evaluation (BASE) study." In: Indoor Air Quality 2001. Moisture, Microbes, and Health Effects: Indoor Air Quality and Moisture in Buildings. November 4-7, 2001. San Francisco, CA. Atlanta, GA: ASHRAE.
  • Tsai, F.C.; Macher, J.M. 2005. "Concentrations of airborne culturable bacteria in one hundred large U.S. office buildings from the BASE study." Indoor Air 15 (Suppl 9):71-81.

    Results for the airborne fungi and bacteria samples are summarized in the following sections, and were derived from these two references.

Airborne Fungi

For the airborne samples collected in the one hundred BASE study spaces using the Andersen Instruments impactors, fungi were collected on malt extract agar, or MEA, that was incubated at room temperature with 12-hour cycles of light and darkness. After sampling, the agar plates were sent to an analytical laboratory using overnight shipment. The analytical laboratory reported the results from these Andersen samples as the number of colony-forming units, or CFUs, of each fungal group per plate.

Airborne samples for total fungi were also collected in 44 of the BASE study spaces using a second type of multiple-hole impactor, in this case, those manufactured by Burkard Manufacturing Co, Hertfordshire, England. These total fungi results were determined by direct microscopic examination and reported as the number of spores of each fungal type per slide, while factoring in the area of the slide that was examined.

Results are presented for the prevalence of fungi, specifically, the percentage of samples in which a fungus was identified. Ninety-five percent confidence intervals were calculated for the prevalence of the fungal groups and categories that are described in subsequent sections. Comparisons, or differences, between prevalence were considered statistically-significant if the 95 percent confidence interval for a pair-wise comparison did not include zero. Analyses were performed using a beta version of the BASE dataset.

Comparison of Airborne Fungi Sampling Methods: Thirty groups of fungi were identified in 1% or more of the Andersen or Burkard samples. This comparison is made to show that there are similarities and differences in the two methods, but not to provide a direct comparison of the results. A total of 58 fungal groups were reported in the BASE study, of which 28 (48%) were identified only by culture, 13 (22%) were identified only by direct examination, and 17 (30%) were identified by both methods.

Indoor/Outdoor Sampling Comparison: The fungal groups identified in 1% or more of the culturable (Andersen) samples were analyzed by separating the outdoor and indoor samples. The frequency with which individual fungal groups were found was higher outdoors than indoors in all but seven groups (Sporobolomyces, Tritirachium, Exobasidium-like, Thysanophora, Rhinocladiella-like, Botryosporium, and Wardomyces spp.), with the latter five of these groups being isolated only from indoor air samples. None of these seven fungal groups were found in more than 2% of the culturable samples.

Combining Fungal Groups by Common Characteristics: Based on the fungi identified in the BASE buildings from the culturable samples, the fungi were combined into four categories based on features such as typical sources and growth preferences: leaf-surface (phylloplane) fungi, soil fungi, water-requiring (hydrophilic) fungi, and potentially toxic fungi. The purpose of combining fungi into categories rather than examining each of many individual fungal groups facilitates observing associations with other factors including sampling season (summer vs. winter) and geographic climate conditions.

Seasonal Comparisons: Pair-wise comparisons were made of the frequencies in which the four fungal categories were identified in samples collected in the summer versus winter, at the indoor and outdoor sampling locations. The analyses revealed that nearly all pair-wise differences for summer versus winter were statistically-significant, indicating a higher prevalence of these fungal categories during the summer for both indoor and outdoor samples. The exception was water-requiring fungi collected outdoors which was not statistically-significant.

Climate Comparisons: Comparisons were made among the BASE climate regions with regard to the regions’ summer and winter outdoor weather characteristics (temperature and humidity).

Summer climate comparisons were made of the frequencies in which the four fungal categories were identified in samples collected in regions with damp summer climates vs. dry summer climates. Data from three of the EPA climate regions (D, G, and H) were excluded because they have mixed summer outdoor humidity characteristics. The only statistically-significant pair-wise differences were observed for leaf-surface and water-requiring fungi in indoor samples, with an increased prevalence of these fungal categories in damp summer climate regions.

A similar climate comparison based on summer outdoor temperature characteristics was also made (cool/moderate vs. hot summer temperatures), and there were no statistically-significant results for any fungal category either indoors or outdoors.

Winter climate comparisons were made of the frequencies in which the four fungal groups were identified in indoor and outdoor samples for three climate regions based on the winter outdoor temperature characteristics (cool, moderate, or hot). Results are provided for outdoor samples and indoor samples. Pair-wise comparisons were made for three cases: hot vs. moderate winters, hot vs. cool winters, and moderate vs. cool winters.

Outdoors: All winter comparisons for leaf-surface fungi were significant. The comparisons between hot vs. moderate and hot vs. cool regions were significant for soil and potentially toxigenic fungi. These results predominantly show increased frequency of the fungal groups in warmer climate regions. No comparisons were significant for water-requiring fungi.

Indoors: Comparisons between hot vs. moderate regions were significant for soil and water-requiring fungi, with higher frequency in moderate regions. Comparisons were significant between hot vs. cool regions for leaf-surface and soil fungi, with higher frequency in hot regions. Comparisons between moderate vs. cool regions were significant for leaf surface, soil, and water-requiring fungi, with higher frequency in moderate regions. No comparisons were significant for potentially toxigenic fungi.

Airborne Bacteria

For the airborne samples collected in the one hundred BASE study spaces using the Andersen Instruments impactors, bacteria were collected on tryptic soy agar (TSA) that was incubated at two temperatures, 30oC (mesophilic bacteria) and 55oC (thermophilic bacteria). Seven bacterial groups were reported based on Gram stain reaction (positive or negative), cell shape (coccus or rod), distinguishable type of Gram-positive rod (actinomycetes and Bacillus species), and Unknown isolates. Culture results were reported as the number of colony-forming units (CFUs) for each bacterial group per sample and further adjusted by the air sample volume to obtain bacterial concentration in air (CFU/m3). During the analyses conducted by the California Department of Health Services, three groups of Gram-positive rods were reported separately and combined, and the seven bacterial groups from both incubation temperatures were reported separately and summed to obtain total bacterial concentrations. Of the bacterial samples collected in the BASE study, 42.7% of the indoor samples and 36.4% of the outdoor samples were below the detection limits. To characterize the concentrations systematically in the one hundred BASE buildings, multiple air samples for each building were aggregated into one indoor and one outdoor concentration by averaging.

Seasonal Comparisons: The average concentrations of the seven bacterial groups in the one hundred BASE buildings were determined by season, for indoor and outdoor samples. Total outdoor bacterial concentrations (for all bacterial groups combined) were higher in winter than summer (194 vs. 165 CFU/m3), although the differences were not significant, while total indoor concentrations were significantly higher in summer (116 vs. 87 CFU/m3, p<0.001). Increased concentrations of Unknown bacteria and Gram-positive rods contributed to the higher winter bacterial concentration outdoors, whereas elevated Unknown bacteria and Gram-positive cocci contributed to the higher indoor concentration in summer. A major seasonal difference was observed for Gram-positive cocci indoors (summer: 48 CFU/m3; winter: 29 CFU/m3), and was the only group for which the mean concentration was higher indoors than outdoors in both seasons.

Information regarding the various classes of bacteria (Gram-positive cocci, Gram-negative rods, etc.) can be useful. Gram-positive cocci are shed by people. High numbers of Gram-positive cocci could be an indication of overcrowding and/or poor ventilation. Furthermore, although not seen in these analyses, high numbers of Gram-negative rods indoors could suggest that organisms have amplified somewhere in the building and that there may be a moisture problem.

Comparisons of Total, Mesophilic, and Thermophilic Bacteria: The average concentrations of total bacteria (sum of mesophilic and thermophilic bacteria) were determined, combining seasons and incubation temperatures. For all bacteria groups combined, the average outdoor total concentration was significantly higher than the average indoor total concentration (179 vs. 102 CFU/m3, p=0.003). For the individual bacterial groups, the majority had significantly higher outdoor total concentrations, with the exceptions of Gram-positive cocci for which indoor concentrations were significantly higher (39 vs. 24 CFU/m3, p<0.001), and Gram-negative cocci which showed no significant difference. The concentrations for mesophilic bacteria were similar to those for the total bacteria, thus indicating that the thermophilic bacteria are very rare in the indoor and outdoor environments.

Climate Comparisons: A comparison was also made for mean total airborne bacteria concentrations for the different EPA climate regions, also considering sampling location and study season. When summer and winter data were combined, the average outdoor concentrations of total culturable bacteria were higher than indoors in all ten climate regions. Stratification by season showed that this pattern was observed in winter in all ten regions and in summer in all but two regions, in which indoor concentrations were slightly higher than outdoor concentrations. The concentrations of culturable bacteria in outdoor air showed a relatively consistent, although not statistically-significant, seasonal pattern (higher in winter) across most of the climate zones even though the summer and winter climate conditions varied widely among the regions. Indoors, factors other than seasonal changes across the climate regions may have affected bacterial concentrations (higher in winter in five regions, and higher in summer in five regions), although again these differences were not statistically-significant.

Top of Page