OK-AIR study protocol: a longitudinal cluster-randomised 2x2 factorial trial of portable air purification and upper-room UVGI on sick-related absences, indoor air quality, environmental pathogens and social-emotional development in early care and education classrooms (birth-5 years)

This protocol is reported in accordance with the SPIRIT 2025 guidelines for clinical trial protocols. IntroductionYoung children, from birth to age 5 y are particularly vulnerable to indoor air pollutants and respiratory pathogens. Portable air purifiers (or filtration) and upper-room ultraviolet germicidal irradiation (UVGI) are two widely used interventions with the potential to improve indoor air quality (IAQ) and reduce sick-related absences. However, a review of the literature revealed no real-world randomised studies evaluating their effectiveness in reducing young childrens sick-related absences in early care and education (ECE) classrooms. Methods and AnalysisThe OK-AIR study is a longitudinal, cluster-randomised 2x2 factorial trial conducted in Head Start centers using two implementation cohorts: Cohort 1 (five Head Start centers and 20 classrooms from 2023 to 2024) and Cohort 2 (11 centers and 59 classrooms from 2025 to 2026), with expanded inclusion of rural areas. Cohort 1 enrolled 204 children, 48 teachers and 5 site directors, and Cohort 2 enrolled 462 children, 97 teachers and 11 site directors. Within each center, four classrooms are randomised to: (1) control; (2) portable filtration; (3) upper-room ultraviolet germicidal irradiation (UVGI); or (4) both interventions. Cohort 2 was initially planned as a second factorial trial but was amended to a purifier-only design due to funding changes; details are provided in the protocol amendments section. We collect continuous IAQ data, including particulate matter (PM) with aerodynamic diameters [≤]1 {micro}m (PM1), [≤]2.5 {micro}m (PM2.5), [≤]4 {micro}m (PM4), and [≤]10 {micro}m (PM10); total volatile organic compounds (TVOCs) index; nitrogen oxides (NOx) index; carbon monoxide (CO), noise; temperature; and relative humidity, alongside daily child absences. Seasonal environmental surface swabs (dining tables and toilet flooring) are tested by Reverse-Transcriptase quantitative Polymerase Chain Reaction (RT-qPCR) for Influenza A/B, Respiratory Syncytial Virus (RSV), Human Parainfluenza Virus Type 3 (HPIV3), Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), and Norovirus. IAQ monitoring is structured across Winter, Spring, Summer, and Fall, including designated baseline/off-period weeks to characterize temporal and seasonal variability in environmental measures across classrooms and centers. Multi-informant surveys (Director, Teacher, Parent) capture contextual factors, and childrens social-emotional development is assessed using teacher ratings on the Devereux Early Childhood Assessment (DECA). The primary outcome is the sick-related absence rate, analyzed as cumulative absences over the attendance year while accounting for clustering by school and classroom using generalized mixed-effects models. Secondary outcomes include childrens social-emotional ratings, IAQ metrics and pathogen detection rates; analyses of IAQ incorporate time/seasonal structure, and season-stratified absenteeism analyses will be treated as secondary/exploratory refinements. An economic evaluation will estimate incremental intervention costs and cost-effectiveness/cost-benefit (such as cost per sick-related absence day averted). Ethics and DisseminationThis study was approved by the Institutional Review Board (IRB) at the University of Oklahoma. Findings will be shared through peer-reviewed publications; presentations at local, state, and national conferences; research briefs developed for lay and policy audiences; and community briefings prioritizing the participating early childhood programs and communities. DisclaimerThe views expressed are those of the authors and do not reflect the official views of the Uniformed Services University or the United States Department of War. Strengths and Limitations of This StudyO_LIReal-world longitudinal cluster RCT: The study uses a rigorous longitudinal cluster-randomised 2x2 factorial design in real-world ECE settings. C_LIO_LICombined interventions: Interventions target both air filtration and disinfection, allowing for combined and comparative evaluation. C_LIO_LIObjective air-quality monitoring: Continuous monitoring of IAQ metrics provides objective and reliable data on environmental change. C_LIO_LIEnvironmental pathogen surveillance: qPCR on surface swabs yields an objective biological outcome to triangulate with IAQ and absences. C_LIO_LIComprehensive context and child measures: Multi-method and multi-reporter data collection includes Head Start attendance records, continuous air monitoring, pathogen detection, contextual surveys completed by center directors, teachers, and parents, and standardized social-emotional assessments (DECA) completed by classroom teachers. Head Start program records providing childrens longer-term health data available through Health Insurance Portability and Accountability Act (HIPAA) authorization. C_LIO_LIClustered/temporal complexity: Seasonal design accounts for variation over time but may introduce complexity in modeling temporal effects. C_LIO_LIPractical Implications: Study findings will have practical implications for Head Start and other ECE programs striving to maximize child attendance with cost effective strategies. C_LI