CO2 Monitors Are Ventilation Triage Tools, Not Stand-Alone Infection-Risk Meters
Carbon dioxide monitoring has become a practical way to notice under-ventilated occupied rooms, especially in schools and workplaces. The same simplicity creates a misuse risk: a single CO2 value can be treated as if it were a direct infection-risk score, even though risk depends on occupancy, activity, source strength, air distribution, filtration, air cleaning, time, and pathogen-specific factors. This paper synthesizes CDC/NIOSH clean-air guidance, EPA school IAQ guidance, ASHRAE Standard 241 summaries, WHO ventilation guidance, NIST CO2-generation work, and peer-reviewed CO2 proxy models. The result is a triage model: CO2 is best used as an inexpensive signal that triggers a structured check of outdoor air, equivalent clean air, filtration, occupancy, and source-control measures. The evidence supports CO2 as a ventilation indicator and model input under explicit assumptions, while rejecting CO2-only thresholds as complete infection-risk metrics. Public-building IAQ programs should therefore publish response protocols, not only sensor dashboards.
Introduction
Carbon dioxide monitoring is attractive because occupants generate CO2 continuously and many rooms can be instrumented at low cost. EPA explicitly says a CO2 monitor is useful for indicating when outdoor-air ventilation may be inadequate in schools [[cite:epa_schools]]. NIST likewise notes that indoor CO2 has long been used to characterize building ventilation and indoor air quality [[cite:nist_persily]].
The problem is that a ventilation indicator can be overpromoted as a direct infection-risk meter. CDC clean-air guidance focuses on maintaining minimum outdoor-air ventilation and aiming for 5 or more air changes per hour (ACH) of clean air through ventilation, natural ventilation, or equivalent devices [[cite:cdc_cleaner_air,cdc_aim5]]. ASHRAE Standard 241 similarly emphasizes infectious-aerosol control and equivalent clean airflow, not CO2 alone [[cite:ashrae_241,ashrae_241_news]].
The distinction is operational. A classroom principal, facilities manager, or office safety lead does not need a perfect epidemiological model to notice that one occupied room repeatedly accumulates rebreathed air faster than comparable rooms. They do need a defensible response path, because a sensor reading that does not lead to airflow verification, filtration checks, or occupancy decisions can become a passive dashboard rather than a public-health control.
This paper asks how CO2 monitors should be incorporated into public-building indoor-air programs without overstating their meaning. The contribution is a three-state triage model that treats CO2 as a signal to investigate, verify, and act, rather than as a stand-alone infection-risk score.
Method
I used a conceptual-synthesis method across official public-health guidance, building-standard summaries, NIST technical work, and peer-reviewed infection-risk proxy papers. Sources were included when they defined a clean-air action, explained CO2 generation or interpretation, or constrained the use of CO2 as an exposure/risk proxy.
The synthesis deliberately separates three evidence classes. The first class is operational guidance: CDC, NIOSH, EPA, WHO, Health Canada, and ASHRAE describe what building operators should measure or change. The second class is physical interpretation: NIST and Persily/de Jonge explain what must be assumed before a CO2 value can be converted into a ventilation estimate. The third class is infection-risk modelling: Rudnick/Milton and Peng/Jimenez show how CO2 can enter risk equations, but only under assumptions that must not disappear when results are communicated to non-specialists.
Triage(CO 2 ) = f( excess CO 2 , occupancy, activity, clean_airflow, filtration, duration, source_control )
Equation (1) is a decision model, not a universal risk equation. It makes the scope explicit: CO2 is one observed signal that must be interpreted alongside occupancy, activity, clean-air delivery, filtration, duration, and source control.
Results
Finding 1: CO2 is an indicator of ventilation adequacy, not a complete IAQ panel. EPA school guidance lists temperature, relative humidity, air movement, and airflow volume as important measurements, and adds CO2 as useful for indicating potentially inadequate outdoor-air ventilation [[cite:epa_schools]]. Health Canada similarly frames office IAQ as an interaction among ventilation, building condition, outdoor air, furnishings, work processes, and occupants [[cite:health_canada_offices]].
Finding 2: CO2 interpretation depends on human generation rates and room assumptions. Persily and de Jonge report example generation rates of 0.0052 L/s for an average adult doing office work and 0.0029 L/s for a child at the same activity level, and they state that peak CO2 approaches require assumptions about known generation, steady state, known outdoor CO2, uniform concentration, and constant ventilation [[cite:persily_pmc]].
Finding 3: clean-air targets require airflow and filtration metrics. CDC recommends aiming for 5 or more ACH of clean air when possible, and NIOSH explains that filtration and air cleaning can contribute equivalent ACH [[cite:cdc_cleaner_air,cdc_aim5]]. NIOSH separately recommends MERV-13 or better central HVAC filtration when compatible, especially where outdoor-air delivery is constrained [[cite:cdc_merv13]].
Finding 4: infection-risk proxy models exist, but their assumptions must travel with the number. Rudnick and Milton use CO2 as a marker for rebreathed-air fraction in an alternative infection-risk equation, and Peng and Jimenez derive CO2-based risk proxies for COVID-19 settings [[cite:rudnick_milton,peng_jimenez]]. NIST notes that many studies use CO2 as an indicator or proxy, but some present the link without explaining the basis and there is no direct evidence correlating CO2 concentration with virus-containing aerosol levels [[cite:nist_co2_application]].
A CO2 Triage Model
The triage model has three states. A watch state records normal occupied baselines and checks sensor placement. An investigate state starts when CO2 rises above the locally defined baseline or remains high during occupied periods; staff then check occupancy, outdoor-air delivery, ventilation schedules, and whether demand-controlled ventilation is suppressing air supply. A correct state documents interventions such as changing schedules, increasing outdoor air when feasible, improving filtration, adding equivalent clean air, reducing occupancy, or repairing HVAC faults.
Schools require particular care. NIOSH recommends avoiding demand-controlled ventilation reductions that keep air supply from remaining constant during the day and considering maximum outside airflow before and after occupancy [[cite:cdc_schools]]. CO2 dashboards should therefore be paired with building-operation records; otherwise users may see a number without knowing which operational lever to pull.
This metadata requirement is the practical difference between monitoring and governance. A building can collect thousands of sensor readings while still lacking evidence about whether a room was over capacity, whether the outside-air damper was open, whether a portable air cleaner was operating, or whether occupants were exposed for five minutes or five hours. The proposed protocol therefore treats CO2 data as incomplete until paired with at least enough context to decide whether the next action is measurement, maintenance, or management.
Discussion
The synthesis supports a strong but narrow claim: CO2 monitoring is a practical triage layer for occupied public buildings. It can identify rooms where outdoor-air ventilation may be insufficient under current occupancy and activity. It cannot by itself account for filtration, air cleaners, pathogen source strength, mask use, room air distribution, or exposure duration.
A second implication is that CO2 monitoring programs should publish response protocols, not only sensor readings. ASHRAE 241 frames infectious-aerosol control around air-system design, operation, maintenance, equivalent clean airflow, and filtration/air-cleaning technologies [[cite:ashrae_241,ashrae_241_news]]. A sensor dashboard without actions can generate anxiety without improving clean-air delivery.
A third implication is that CO2 should be integrated with whole-building IAQ governance. WHO frames ventilation assessment as part of broader steps to improve indoor air quality and reduce transmission risk [[cite:who_roadmap]]. Health Canada emphasizes that IAQ management must balance ventilation, outdoor air, building condition, contaminants, occupant activities, and energy considerations [[cite:health_canada_offices]].
The response-protocol framing also reduces equity risk. Public buildings with limited maintenance capacity may buy visible sensors faster than they can rebalance HVAC systems or improve filtration. A triage protocol makes that gap visible: it records which rooms repeatedly trigger investigation, which interventions were attempted, and which capital or staffing constraints prevent correction. This keeps the monitoring program connected to building stewardship rather than shifting responsibility onto occupants who merely see the readings.
This paper is limited by its synthesis design. It does not validate a new sensor algorithm or prescribe universal CO2 thresholds. Instead, it identifies what evidence must travel with a CO2 interpretation: occupancy assumptions, activity assumptions, outdoor concentration, clean-air delivery, filtration, sensor placement, and action protocol.
Conclusion
CO2 monitors are useful because they make invisible ventilation problems visible enough to investigate. Their best public-building role is triage: identify rooms that need operational attention, then verify and correct clean-air delivery. Treating CO2 as a stand-alone infection-risk meter overstates the evidence and hides the actions that actually improve indoor air: outdoor air, equivalent clean air, filtration, air cleaning, occupancy management, and maintenance.
The strongest operational recommendation is therefore simple: publish the response rule with the sensor reading. A public dashboard should show what level triggers investigation, who checks the ventilation system, what clean-air intervention is available, and how follow-up is documented. Without that protocol, CO2 monitoring can create a false sense of precision. With it, the same low-cost sensor becomes a useful entry point into building-technology maintenance, public-health prevention, and accountable indoor-air governance.