Final Report: Healthy High School PRIDE (Partnership in Research on InDoor Environments)

EPA Grant Number: R835638
Title: Healthy High School PRIDE (Partnership in Research on InDoor Environments)
Investigators: Corsi, Richard L. , Kinney, Kerry A. , Horner, Sharon , Novoselac, Atila , Wu, Sarah
Institution: The University of Texas at Austin
EPA Project Officer: Hahn, Intaek
Project Period: February 1, 2015 through January 31, 2019 (Extended to January 31, 2020)
Project Amount: $989,047
RFA: Healthy Schools: Environmental Factors, Children’s Health and Performance, and Sustainable Building Practices (2013) RFA Text |  Recipients Lists
Research Category: Children's Health , Human Health


Past studies of indoor air quality (IAQ) in schools have been deficient in many ways. There has been little progress in determining the actual agents responsible for adverse effects when ventilation is inadequate. Environmental agents responsible for dampness-related health effects have not been determined. Few studies have focused on irritating oxygenated VOCs (OVOCs) and their sources. Schools in hot and humid climates have been under-represented. And the focus to date has been on identifying IAQ problems in schools. Proven low-cost solutions are needed.


The overall goal of the proposed study was to address these research gaps by partnering with six high schools in Central Texas, conducting an intensive field campaign to determine the relationship between environmental factors and student health, and then investigate the efficacy of low-cost solutions. Specific objectives included: (1) identify systematic problems in school HVAC systems that cause poor ventilation rates, increased pollutant concentrations and adverse health symptoms for school occupants and explore low-cost solutions to these problems, (2) utilize molecular techniques to investigate relationships between composition and diversity of the microbial community present in school classrooms, environmental conditions, and health symptoms, (3) determine the role of OVOCs on student and teacher health outcomes, and (4) engage high school student and teacher stewards in the design, data collection and outreach components of the project.

Summary/Accomplishments (Outputs/Outcomes):

The following tasks have been completed during the project:


In the first year of project the team focused on recruiting seven high schools in Central Texas. This included schools from two school districts (five in one school district and one in another school district) and a charter school. All the schools had classrooms that were air conditioned and had some level of mechanical ventilation. Four out of seven schools had portable classroom besides classrooms in permanent structures. Schools were in located in both rural and urban neighborhoods (as indicated in Table 1).


School number

Sampling dates

Site condition



(Fall 2015)


(Spring 2016)


(Fall 2016)


(Spring 2017)

HVAC system






Rural, heavy traffic street nearby














Rural, heavy traffic street nearby

VAV + Wall AC unit a







VAV + Wall AC unit a






Urban, heavy traffic street nearby

VAV + Wall AC unit a













Urban, heavy traffic street nearby


a Portable classrooms used wall AC unit.


Field measurements were completed during the second and third year of the project and consisted of four sampling seasons / academic semesters (as indicated in Table 1) to quantify a suite of indoor air and environmental quality metrics at least 30 classrooms in each academic semester. Sampling events for most metrics were completed over four days, starting on Monday afternoon and ending on Friday afternoon. Measurements included (classroom unless otherwise noted): temperature and relative humidity, carbon dioxide (room and supply air), fine (PM2.5) and coarse (PM10) particulate matter (room and rooftop), speciated volatile organic compounds (VOCs), ozone (room and rooftop), bioaerosols (room air and surfaces), noise, and illuminance. Surveys of perceived air and environmental quality for students and then for teachers across five high schools were also completed.


In parallel with data collection a significant amount of data analysis was completed, particularly for CO2/ventilation, lighting, formaldehyde, ozone, and particulate matter. Also in parallel with the data collection the team generated cumulative distribution plots for aggregated data for several key metrics, analyzed temporal variations in metrics for individual classrooms, to assess correlations between metrics.


After the field campaign, in year four of the project the team analyzed ventilation, CO2 data, and collected PM samples and drafted first two journal papers based on these results. Also, during the year four the team developed and implemented, in spring 2018, a symposium for high school administrators, teachers and students in Central Texas, where we informed/educated school districts and their students to indoor air quality in schools and the most relevant findings of our study. Also in summer 2018, we organized a summer camp at UT at Austin for students from five high schools that participated in both years of the study. During a three-week period a group of high school students (five from each school) were visiting our labs on a daily base where they learned about IAQ, instrumentation, and data processing.


During the year five, the team focused on processing collected data on volatile organic compounds (VOC), semi volatile organic compounds (SVOC) in classrooms, and collected bio samples. The VOC analysis focused on personal care product while SVOC analysis focused on identifying of sources of SVOCs in schools. The bio samples focused primarily on comparison of portable and permanent classrooms considering various environmental quality data. Also, in collaboration with our partners, The Children's Environmental Health Institute (CEHI), we developed a website for dissemination of project findings (


In year six (during the project extension), the UT Austin team developed improved ventilation strategy that can reduce disease transmission in classrooms and developed the framework to characterize the effect of ventilation-reduced influenza transmission. Also, the impact of thermal comfort on overall perceived environmental quality in classroom is characterized based on students' surveys and data from measurement of thermal comfort parameters, nose level, lighting level, and set of IAQ parameters. Furthermore, a set of measures for improvement of portable and permanent classrooms were identified.  Finally, the study results were disseminated among experts at conferences and meetings as well as with wide audience using social media.


During the project period our team presented at conferences and seminars published and submitted journal papers for publications. Below is the summary of finding for each measured air quality and environmental parameter as well as on comparison of portable and permanent classrooms and possible affordable solution.


CO2 Concentration

A large majority of classrooms studied had average occupied-day carbon dioxide (CO2) concentrations that exceeded generally accepted standards, e.g., 700 ppm above background. Over 90% of classrooms had peak CO2 concentrations greater than generally accepted standards. This is true of either fall or spring CO2 concentrations, across both years of the field campaign. In our discussions with school facility managers after the third of our four sampling phases, we indicated that greater ventilation is needed if they wish to lower CO2 concentrations to acceptable norms. The facilities staff in one high school district opted thereafter to increase fresh air exchange rates in many of their classrooms. The result was noticeable with lower CO2 concentrations in that school district, as well as observable decreases in formaldehyde and VOC concentrations in classrooms.



Four-day passive measurements indicate that formaldehyde concentrations ranged from 5 to 50 ppb across all classrooms, with a median concentration of 23 ppb. A continuous formaldehyde analyzer was used for 38 sampling events in 26 classrooms. These measurements indicated that elevated concentrations at night occurred when mechanical systems were shut down exclusively for purposes of energy conservation. Occupied-day formaldehyde concentrations were much lower than continuous four-day measurements across the two-year field campaign. Four-day average formaldehyde concentrations determined using a relatively inexpensive continuous formaldehyde analyzer were reasonably consistent with four day time-integrated passive formaldehyde measurements over the two-year field campaign.


Particulate matter: PM1, PM2.5, and PM10

Elevated concentrations of coarse particulate matter (PM10) in classrooms appeared to be wholly dependent on student activities, particularly when students entered and left classrooms and re-suspended or shed particles in the process. The average PM2.5 and PM10 concentrations in classrooms were lower than standards developed by WHO and ASHRAE. The indoor-outdoor ratios for PM2.5 and PM10 in this study were much lower than those reported in many other related studies. This may be due to effective use of filtration systems in the schools that participated in this study. A comparison of indoor and outdoor PM concentrations sheds light on the contribution of each environment to indoor PM. The indoor PMconcentration had a significant relationship with outdoor concentration, indicating the importance of the outdoor atmosphere as a source of indoor PM1. Indoor PM2.5 concentration was affected by outdoor PM2.5 as well as some indoor activities. The relationship between indoor and outdoor PM10 during occupied periods was not significant, indicating that the primary source of indoor PM10 indoors is resuspension from indoor flooring and other surfaces. Due to the fact that portable classrooms have higher outdoor air exchange rates (3-4 times that of regular classrooms), the indoor PM concentrations in portable classrooms reacted faster to changes in outdoor concentrations for PM1 and PM2.5. However, significant differences were not observed between portable and regular classrooms for PM1, PM2.5 and PM10.The type of flooring had a significant effect on indoor PM concentration, with carpet leading to higher PM concentrations during occupied periods. The reported PM emission rates reported in this study can be used to better estimate indoor PM concentrations in classrooms.     


The volatile organic compound (VOC) concentrations in most classrooms tended to be dominated by scenting agents and cleaning products. A major component of a specific floor cleaner was observed in the air of many classrooms across the two-year field campaign.  Body sprays appeared to be a major source of terpenes in classroom air. We observed dihydromyrcenol in the air of every classroom in our study. Subsequent analyses of 10 popular (amongst high school students) body sprays indicated that this is by far the most dominant scenting agent in every product.


The collected data on semi-volatile organic compounds (SVOCs) show that SVOCs are widely found in high school classrooms; data show that seasons (primarily temperature), building types, and flooring types have impacts on indoor SVOC concentrations. Samples from floor dust, AC's filter dust and indoor air show that - depending on the type of SVOC- concentration vary significantly from classroom to classroom. For example, phthalates (plasticizers) varied by 25 to 50 times among different classrooms (for sample in the air and dust respectively). When considering pesticides, variations between classrooms were also very significant. Results show that restricted pesticides (pentachlorophenol banned in 1980s) are still abundant, suggesting their persistence in indoor environments. For many pesticides, concentrates are higher in portable classrooms than in permanent classrooms. When considering flame retardants, variations between the classrooms are also very large but there is no statistically significant difference between portable and permanent classrooms. The ongoing analysis of the classrooms and SVOCs data will further identify indoor sources and conditions that results in higher exposure to different SVOCs. 



Rooftop ozone concentrations tracked very well with nearest state-run continuous air monitoring stations. Indoor ozone concentrations trended with outdoor concentrations, but as expected, concentrations were lower indoors. The indoor/outdoor ozone concentration was inversely related to occupied-day indoor CO2 concentration, suggesting the important of ozone interactions with classroom occupants.  Ozone decay rates were determined under both occupied and unoccupied conditions in seven portable classrooms. Results indicate that the major 'sink' (removal surface) for ozone is the students themselves (skin lipids and clothing), accounting for approximately 2/3 of total ozone decay in classrooms.



Microbial Communities

Microbial communities were compared for portable and permanent classrooms. When considering the impact of portable buildings and their ventilation systems microbial communities, a set of portable and permanent classrooms were studied in four schools (schools that had both types). In both type of classrooms, human-associated bacteria dominated all the surfaces subject to human touch indicating that additional cleaning of school surfaces (e.g., desks) would be advisable.  Portable classrooms, which are more often associated with a musty (damp) smell, were characterized to determine fresh air exchange rates and the flow of air between hidden spaces (i.e. crawl space and attic space) and the occupied area of the portable classrooms. The hypothesis was that these hidden spaces and microbial transport from these spaces may contribute to microbial exposures in classrooms. Analysis of the dust samples collected from various locations within the portable and permanent classrooms as well as crawl and attic space of portable classrooms were used in this analyses. The results indicate that under certain ventilation conditions (e.g., negative pressurization conditions) in the portable classrooms, microorganisms from the unmaintained attic space can penetrate into the occupied space of the classrooms.  Even though building envelope analysis show that portable structures are particularly susceptible to moisture intrusion, the microbiome results indicate that moisture associated fungi were detected in both portable and permanent buildings. The relative abundances of most moisture-associated fungi were similar in both building types although specific moisture-related fungal taxa were enriched in some classrooms (in both building types) with reported moisture problems.  One significant difference between portable and permanent classrooms is the musty smell in portable classrooms suggesting enriched levels of VOCs associated with microorganisms (mVOCs). One potential reason for this may be the path of outdoor air. In permanent classrooms, outdoor air is supplied directly to the classroom by mechanical system while in portable classrooms, a significant fraction of outdoor air travels through crawl and attic space prior to getting into the classroom.  Research is underway at UT Austin to further test the hypothesis that mVOCs are enriched in classrooms as a result of these uncontrolled air flow patterns. 



The study collected detailed information about on lighting conditions in 28 classrooms, in both portable and permanent buildings. Overall the study found that all classrooms used artificial light all the time. The study found that a vast majority of classrooms had more than sufficient light, considering the purpose of the classroom. When considering variation in light level between the classrooms, the mean light level in permanent classrooms had a wider distribution than that in the portable classrooms. One explanation for the large variation is that multiple types of classrooms were surveyed in this study, including computer labs, science labs, and regular classrooms. Another reason for large variation of light level in permanent classrooms is the geometry of the space. The surveyed permanent classrooms had different room geometries, height and different arrangements of electric lighting systems and windows. In contrast, all the surveyed portable classrooms had similar daylighting systems and electrical lighting systems. Large variations in light levels inside of the classroom space were present in all portable classrooms, but in only 1/3 of permanent classrooms. This indicates that most of the permanent classrooms had more uniform light distribution compared to the portable classrooms.

Student Survey Results

A set of collected data that characterize thermal comfort and noise level in classroom (not previously processed) were analyzed (together with lighting and IAQ parameters) when considering student survey on perceived environmental quality. The data show that satisfaction of environmental quality in classrooms dominates primarily with thermal comfort, and then on noise and lighting level. Preliminary results that the classroom air quality was mostly concern for teachers and not so much for students. The ongoing analysis of the survey results should help to see if the parameters that contribute to the poor IAQ (such as increased, CO2, formaldehyde, ozone, and/or VOC level in specific classrooms) are detectable by students.


Overall Comparison of Environmental quality in Portable and Permanent Classrooms

Our comparison of permanent and portable classrooms suggests that, for our subject schools, indoor air quality is generally no different between the two types of classrooms for nearly all metrics (analyses of biological samples are not yet complete).


Ventilation control as economical and adorable solution for better IAQ

As an affordable solution for improvement of air quality in schools, we investigated the benefits from improved ventilation strategy that increased outdoor air supply to reduce student absences due to illness during cold and flu season. Benefits are measured in economic terms as reduced school district absences and hence greater state appropriations. Costs are measured in terms of increased energy use due to increased outdoor air supply rates. Both, benefits and cost, vary for different States, and our study for Texas and California shows that, regardless on the appropriation rate and climate conditions, this ventilation strategy results in increased air quality, reduced number of illnesses, and money saving for school district. We developed a model that for different appropriation mechanisms associated with student absenteeism and climate condition provides optimum increase in ventilation rate.


Future Activities Plans

·         Finish and publish journal papers listed in the section above.

·         Continue work with Austin's school districts on procedures for improvement of air quality in portable and permanent classrooms. In collaboration with the school district develop guidance for keeping good IAQ in School; publish this guidance at our partners' website: as well as at the ASHRAE Journal which is trade journal for building operators and practitioners.

·         Prepare a summary manuscript on air quality in high school classrooms for an educational journal. The paper will provide prospective on perceived and measured air quality with the focused on how teacher and school administrators can affect air quality in learning environments.  


·         Expand EPA school study by comparing IAQ in High school with IAQ in University classrooms. Based on EPA study results, the funding is secured for the extension of the EPA project related research. Specifically, the EPA study results were used to (apply for and) receive the NSF student fellowship for Ms Sangeetha Kumar who will continue to work on processing the EPA data in the following 3 years. Besides processing existing data, she is working with the school districts to collect data which were not initially collected (i.e. student absenteeism during the flu season). Also, the additional funding from UT is secured to support a set of graduate and undergraduate students to replicate part of measurement conducted for EPA's PRIDE school project; but this time in a set of university classrooms. The data collection in university classrooms focuses mostly on CO2 levels in classrooms, ventilation rate, and absenteeism due to the flu. This data (collected before the classroom closure due to the COVID 9) together with EPA project data will help to identify potential benefits from improved ventilation and filtration in classrooms.


Journal Articles on this Report : 1 Displayed | Download in RIS Format

Other project views: All 25 publications 1 publications in selected types All 1 journal articles
Type Citation Project Document Sources
Journal Article Ren J, Wade M, Corsi R, Novolelac A. Particulate matter in mechanically ventilated high school classrooms. Building and Environment 2020;184.
abstract available   full text available
R835638 (Final)
  • Full-text: Science Direct - Full Text HTML
  • Other: Science Direct - Full Text PDF
  • Supplemental Keywords:

    children's respiratory health, community partnership, school practice, mediators, particulates, surveys, attendance

    Relevant Websites: Exit

    Progress and Final Reports:

    Original Abstract
  • 2015 Progress Report
  • 2016 Progress Report
  • 2017 Progress Report
  • 2018 Progress Report