5

Evidence on Health and Safety Impacts of Cabin Temperatures

The committee was charged with assessing the health and safety impacts of temperatures in the passenger cabins of airlines operating under a Part 121 certificate of the Federal Aviation Administration (FAA). Chapter 3 reviewed evidence on the physiological, cognitive, and behavioral effects of cabin temperature exposure, drawing on a broader literature concerning the health and safety impacts of thermal exposures. Importantly, Chapter 3 highlighted key vulnerabilities to temperature effects that vary by age, underlying health conditions, and metabolic rate. Chapter 4 illustrated the application of standards and guidelines on thermal comfort and heat and cold stress to identify cabin temperature and humidity levels at which discomfort and health and safety effects associated with heat and cold stress may be expected. In that discussion it was noted that cabin crew, who are generally more active across all phases of flight, may be at increased risk from temperature exposure compared with the more sedentary passengers. This chapter builds on that foundation by examining the available empirical evidence concerning the impacts of cabin temperature on health and safety, and it concludes with recommendations for future data collection that could enable more reliable assessment of these effects.

Ideally, the committee would have identified a large, national study (e.g., a large-scale survey or medical record analyses of passengers and flight attendants) designed to provide evidence on the health and safety impacts of cabin temperatures. However, a review of published and “gray”¹ litera-

___________________

¹ Gray literature is defined as the broad category of data or information sources (e.g., government and industry reports) not found in typical peer-reviewed published journals (NIH, n.d.).

ture revealed very limited direct evidence on how cabin temperature—and the related factor of humidity—affect health and safety. The scant direct evidence and the limitations of such evidence are reviewed at the start of this chapter. The chapter further examines indirect evidence of cabin-temperature-related health and safety impacts. In the absence of strong direct evidence from large, systematic studies, data from smaller studies, case reports, and incident summaries still provide useful insights. The committee therefore adopted an approach that integrated data from multiple sources by breaking the problem down into the following key questions which guided the committee’s review of the indirect evidence:

• To what extent are passengers and flight attendants exposed to cabin temperatures that are beyond the expected thermal comfort range across all flight phases (e.g., boarding/deplaning, taxiing, and in flight)?
• What are the health and safety impacts of experiencing such temperature exposures?

The empirical evidence reviewed in this chapter is derived from a range of sources, summarized in Table 5-1, including peer-reviewed journal articles (both primary research and literature reviews), government reports, and databases maintained by agencies such as the Occupational Safety and Health Administration (OSHA) and the Department of Transportation (DOT) as well as U.S. airlines, labor unions, and aviation medical consultation organizations. Some reporting systems were developed specifically to capture cabin temperature issues (e.g., the Association of Flight Attendants’ [AFA’s] 2Hot2Cold app and the Association of Professional Flight Attendants’ [APFA’s] hot cabin reporting system), while others—such as the National Aeronautics and Space Administration (NASA) Aviation Safety Reporting System (ASRS)—focus more broadly on aviation safety. Additional details on these data sources are provided in Appendix A.

DIRECT EVIDENCE OF TEMPERATURE IMPACTS ON HEALTH AND SAFETY OF CABIN OCCUPANTS

Evidence from Research Studies

The committee identified only one study—a collaborative effort among the American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE), the FAA, Boeing, Batelle, and Harvard University (see Box 5-1)—that examined the relationship between cabin thermal conditions (temperature and humidity) and self-reported health outcomes (Battelle, 2018; Lordo et al., 2023). This study found a significant positive association

TABLE 5-1 Characteristics of Evidence Sources Used by the Committee

between higher temperatures and several reported symptoms, including back and shoulder pain or stiffness, dry and irritated eyes, ear pain or pressure, nasal congestion, and general discomfort. Conversely, colder temperatures were associated with increased reports of runny nose, sneezing, and dry or irritated throat (Battelle, 2018; Lordo et al., 2023).

Passengers who reported being not “in good health” overall were more likely to experience discomfort and related symptoms (Rankin et al., 2000). This finding highlights a key vulnerability: individuals with compromised health appear more susceptible to temperature-related discomfort—a concern amplified by the aging U.S. population, which faces higher rates of

chronic conditions that impair temperature regulation (as discussed in Chapter 3).

Interpretating the findings from ASHRAE RP-1262 requires taking into account several limitations. First, self-reported health and comfort were collected only from participants aged 18 years and older, excluding infants, children, and adolescents. Second, symptoms related to temperature and humidity during boarding or while on the runway were not captured (surveys were not distributed until the flight was underway). No objective physiological measures—such as heart rate, breathing rate, blood pressure, oxygen saturation, or cognitive function—were recorded. Other FAA-supported studies conducted by the Airliner Cabin Environmental Research Program–National Air Transportation Center of Excellence for Research in the Intermodal Transport Environment independent of the ASHRAE RP-1262 project incorporated physiological measures showing significant health changes in vulnerable and older passengers, including arrythmias and oxygen desaturation linked to environmental stressors such as cabin pressures, even when the symptoms were not reported (McNeely and Watson, 2011). These findings underscore the fact that health consequences are not always apparent.

Evidence from Aviation Industry Reporting Systems

Additional direct evidence on the health and safety impacts of cabin temperatures comes from aviation industry reporting systems, such as the NASA ASRS and data collected through the 2Hot2Cold app. Self-reported health outcome data from these sources are largely limited to symptom descriptions (e.g., nausea, lightheadedness, dehydration), and many minor events likely go unreported. The durations of exposure to cabin temperatures that resulted in health symptoms are generally not reported. There is no predetermined threshold to prompt reporting through these systems. In the absence of an apparent temperature-related illness, injury, or safety event, reporting relies on a subjective determination of whether cabin temperatures are too hot or cold. Crewmember awareness of cabin-temperature-related issues and reporting systems may influence the likelihood of reporting. Additionally, while tools like the 2Hot2Cold app were designed to be used at any time, the committee learned that flight attendants would need to use off-duty (unpaid) time to submit reports (temperature-related or otherwise) to systems such as ASRS, which may have disincentivized reporting. As noted in Appendix A, airlines also maintain their own reporting systems that may include health and safety events, but the committee did not have access to those data.

NASA ASRS Data

Table 5-2 summarizes the physical, cognitive, and behavioral effects of cabin temperature exposure described in NASA ASRS reports. Because reports of extreme cold were rare, most symptoms listed relate to heat exposure, although the committee did hear during a public meeting an anecdotal account of possible frostbite from exposure of a flight attendant to localized cold conditions (Heiple, 2025). Of the more than 500 unique reports captured from iterative keyword searches of ASRS reports from Part 121 air carrier operations submitted from 1988 to 2026 (see Appendix A), 154 temperature-related reports were identified and analyzed, yielding 13 cases (8.4 percent) in which medical attention was sought,² supporting media reports (Cook, 2017; Fingert, 2025; Grant and Sweeney, 2023) that serious temperature-related health impacts, while uncommon, do occur. Safety concerns are further suggested by behavioral issues (unruliness) and cognitive symptoms (e.g., distraction from duties, procedural errors) noted in ASRS reports. While most cognitive effects were reported by members of the flight crew, similar effects on flight attendants’ cognition and performance of duties are likely, as noted in Chapter 3.

AFA 2Hot2Cold Data

Medical or first aid was provided in 4.5 percent of 3,883 relevant 2Hot2Cold reports, with more extreme temperatures associated with greater risk (see Figure 5-1). For example, 8.8 percent of reports involving cabin temperatures of 40°C (100°F) or more required medical aid. In contrast, there were no reports indicating the provision of medical aid for temperatures between 15 and 20°C (70–79°F). The involvement of medical attention may lead to crew distractions, flight delays, or aircraft diversions for occurrences during flight.

Summary and Synthesis

In summary, evidence from aviation industry reporting systems suggests that health and safety concerns related to cabin temperature exposures are more serious than those captured in the ASHRAE RP-1262 study, which reported associations with only minor health issues (e.g., shoulder pain or stiffness, eye irritation, nasal congestion) (Lordo et al., 2023). The comparison of evidence sources is limited by the fact that the ASHRAE RP-1262 study focused only on in-flight conditions, in contrast to aviation industry

___________________

² The numbers of incidents requiring medical attention are likely higher than reported here, given descriptions of serious symptoms (e.g., seizure, loss of consciousness) in reports for which no mention of medical attention was noted.

TABLE 5-2 Physical, Cognitive, and Behavioral Symptoms Reported in NASA ASRS Reports

^a Numbers in this column do not sum to the total number of committee-identified NASA ASRS reports citing cabin temperature issues (N = 154) because many reports (83) did not include mention of symptoms from temperature exposures, and some reports included symptoms that fall into multiple categories.

^b Of the seven reports of cognitive symptoms that affected performance of duties, all seven involved the flight crew, and one of those also involved flight attendants.

SOURCE: Created with data from NASA (2025).

reporting systems, which include data on in-flight and on-ground occurrences. The absence of information on measurement methods, humidity levels, and duration of exposure limits the inferences that can be drawn from data from incident reporting systems like NASA ASRS and 2Hot2Cold.

INDIRECT EVIDENCE OF TEMPERATURE IMPACTS ON HEALTH AND SAFETY OF CABIN OCCUPANTS

Because direct evidence linking cabin temperature exposure to health and safety outcomes is limited in both quantity and scope, this section examines indirect evidence to assess potential impacts. The analysis draws on available data for cabin temperature and humidity and, where possible, compares these real-world conditions with thermal discomfort or heat and cold stress zones defined by established thermal comfort and stress standards and guidelines (see Chapter 4).

Summary of Evidence from Cabin Temperature and Humidity Data

Data from Research Studies

A limited number of studies have measured and reported aircraft cabin temperature and humidity conditions. In addition to peer-reviewed articles, the committee identified several relevant reports from research studies in the gray literature.³ Table 5-3 summarizes the temperature and humidity ranges reported across these studies,⁴ while Table 5-4 compares the sources that include individual temperature and humidity measurements that could be further reviewed. Comparisons in both tables should be interpreted cautiously, as study protocols were not standardized. For instance, instrumentation varied widely, from inexpensive temperature–hygrometer sensors to sophisticated climate devices. Many data were collected on convenience flights, while other studies aimed for a more representative mix. Some sources recorded conditions only at specific points in a flight (e.g.,

___________________

³ Nearly half of the studies collected data from air carriers based outside the United States (i.e., not operating under an FAA Part 121 certificate). However, this information was not consistently reported. Given the limited availability of cabin temperature and humidity data and the fact that the same commercial aircraft models are used by both domestic and international air carriers, the committee chose not to exclude studies based on whether they involved Part 121 operators. Studies that simulated flight conditions were excluded from the committee’s analysis of cabin temperature and humidity data.

⁴ The article by Wang and colleagues (2024) referenced in Table 5-3 is a review paper that summarizes data from 19 studies that measured cabin temperature and 20 studies that measured relative humidity. Summary data for the primary studies included in that review can be found in Table B-1 in Appendix B.

after boarding and 1 hour after takeoff), whereas others collected continuous measurements throughout the flight. Most measurements were taken during flight, with the exception of the study by Nicholls and Vink (2025), which also captured boarding conditions. In many cases, temperature and humidity were secondary parameters within broader studies focused on air contaminants (e.g., carbon dioxide, carbon monoxide, ozone, volatile organic compounds, particulate matter) or ventilation performance rather than primary endpoints.

1. Nicholls and Vink (2025) publication appendix;
2. ASHRAE RP-1262 (Lordo et al., 2023) project dataset; and
3. ACER program data (unpublished).

Combined, these datasets—referred to as Nicholls and Vink, ASHRAE 1262, and ACER, respectively—provide in-flight temperature and humidity data for 379 flights, offering a broad cross-section of conditions routinely encountered in aircraft cabins. Key characteristics of the datasets are summarized in Table 5-4. In addition to providing data used to characterize distributions of cabin temperature and humidity, these datasets were used in the comparison of cabin conditions with thermal discomfort and heat and cold stress zones as described later in this chapter. As emphasized in Chapter 4, humidity plays a critical role in thermal response, particularly in zones of warm discomfort and heat stress. Therefore, mapping cabin conditions to thermal response zones requires simultaneous temperature and humidity measurements.

TABLE 5-3 Summary of Cabin Temperature and Humidity Data from Published Studies

NOTE: NR = not reported.

^a Wang and colleagues (2024) reviewed 54 studies on in-flight aircraft cabin environmental factors and identified 19 studies (covering a total of 252 flights) that measured cabin temperature and 20 studies (for a total of 273 flights) that measured relative humidity. The reported statistics reflect aggregated data across all applicable flights. As a review article, these figures depend on the data reported in the original studies. Summary data for the primary studies included in the review can be found in Table B-1 in Appendix B.

^b Minimum and maximum values represent averages across flights (which the authors categorized by aircraft type), not the absolute extreme observed across all flights.

TABLE 5-4 Summary of Datasets with Raw Cabin Temperature and Humidity Data

The Nicholls and Vink study employed a well-defined measurement methodology. Cabin temperature and humidity measurements were recorded at two points: upon passenger boarding (once seated) and again 1 hour after takeoff. Measurements were taken using a small temperature and humidity sensor placed on the passenger tray table. Flights were selected by convenience—data were collected on flights taken by project personnel and volunteers—rather than through independent sampling. The device used for these measurements was an inexpensive, consumer-grade device. Given the challenges of accurately measuring humidity (Ma et al., 2023), these low-cost devices likely represent a limitation of the Nicholls and Vink dataset. The device manufacturer’s specifications reported on the online retail site provided in correspondence with the authors (Vink, 2025) are ±1^oC and ±5 percent relative humidity, with operating ranges of -50 to +70^oC (-58 to 158°F) and 10–90 percent relative humidity. The study methods did not

include calibration or verification of sensor accuracy (Vink, 2025). In total, boarding measurements were collected on 113 flights and 1-hour measurements on 101 flights. The study methods provide no information regarding the timeframe for data collection, but correspondence from the authors indicated that data were collected between June 1, 2022, and March 1, 2023 (Vink, 2025). The dates of individual flights are unknown. In contrast to the other datasets used in the committee’s analyses and importantly for a study including data on ground operations, the study by Nicholls and Vink included ambient temperature (but not humidity) data for each of the flights for which cabin temperature and/or humidity measurements were taken. Exterior temperatures varied from -31.1°C to 32.2°C (-24°F to 90°F). Boarding temperatures varied across a smaller range (16.7°C to 30.2°C [62.0°F to 86.6°F]) but were correlated with exterior temperatures (r = .448) (Nicholls and Vink, 2025).

The ASHRAE RP-1262 project, described in Box 5-1, collected data on flights intentionally selected to capture a range of flight conditions (flight lengths, longitude ranges, season) rather than a random sample. Consistent measurement protocols were established and adhered to throughout the study. Instruments were located in an unoccupied seat reserved for this purpose near the middle of the economy section. A key limitation of this dataset is that all measurements were taken only after the aircraft reached 10,000 feet and before descending below that altitude, excluding conditions during boarding and taxiing. Once activated, instruments recorded temperature and humidity measurements every minute. In total, temperature and humidity data were reported for 62 flights.

The ACER dataset was developed through a project aimed at creating a compact instrument package capable of measuring and recording multiple environmental variables—including air temperature, relative humidity, and cabin pressure—while being portable enough to carry on any flight without special accommodations (Chen et al., 2010). Several instrument prototypes were built and tested during the project. Data were collected on flights of convenience by project personnel, colleagues, and volunteers. Due to the developmental nature of the project, measurement protocols varied: some recordings began during boarding, others shortly before takeoff, and some only after reaching altitudes above 10,000 feet. Although records did not specify the exact timing of instrument activation, flight versus ground operation can be distinguished using cabin pressure data. Instrument placement also varied, including being placed on tray tables, hanging outside the seatback pockets, or sitting inside seatback pockets. Once activated, the instruments recorded data continuously until they were turned off. Sensor calibration was conducted and verified for selected instruments by an independent lab, though validation focused primarily on cabin pressure and

CO₂ concentration measurements rather than temperature and humidity (Cao et. al., 2019). In total, data were collected and archived for 210 flights.

The Nicholls and Vink in-flight data were collected 1 hour into each flight. Because neither the ASHRAE 1262 nor ACER data collections recorded the time when data collection began, it was not possible to identify the 1-hour point for comparison. However, continuous data collection in the ACER and ASHRAE 1262 datasets indicate that cabin conditions typically stabilize after takeoff and climb-out, with minimal variation during cruise. To provide a consistent reference point across datasets, measurements from the top of descent—identified by the onset of increasing cabin pressure and, when available, data on changes in altitude—were used to represent in-flight conditions in the ACER and ASHRAE 1262 datasets. An analysis of the longitudinal measurements for a subset of flights (see Appendix B) suggests that the temperatures used in the assessments are representative of an extended exposure and are not the result of short fluctuations in cabin conditions.

Figure 5-2 presents a comparative overview of in-flight temperature and humidity distributions across these datasets, grouped into four humidity ranges. All three datasets show similar temperature patterns, dominated by the 20–30^oC (68–86^oF) range, with fewer than 10 percent of flights outside this range. The ACER dataset features a slightly higher mean temperature (25.7^oC, or 78^oF) than Nicholls and Vink or, ASHRAE 1262, which had means of 24.5^oC and 24.9^oC, respectively (76^oF and 77^oF), and than other published datasets (see Tables 5-3 and B-1). This small difference may reflect different instruments or instrument placement practices—some ACER devices were positioned inside the seatback pocket where localized heating could occur.

Relative humidity patterns for the in-flight data differ more noticeably. The ASHRAE 1262 dataset is concentrated in the very low range (less than 15 percent, with an average of 9 percent), while ACER data are more evenly distributed in the low humidity range (less than 30 percent, with an average of 16 percent), similar to published datasets (see Tables 5-3 and B-1). Only the Nicholls and Vink dataset includes some midrange humidity values (30–60 percent), though the average was 20 percent. Given the challenges of accurately measuring low humidity, these differences between the datasets likely reflect instrumentation accuracy limitations. Proper calibration of sensors is essential for accurate humidity measurements (Schellenberg, 2002). None of the reports or publications for the studies associated with these three datasets documented calibration or verification of the accuracy of the humidity sensors used. Some of the data in the datasets appear to be unrealistically low (e.g., less than 5 percent relative humidity). Nevertheless, it is clear from the data collected in these three studies that relative humidity in aircraft during flight is typically low, mostly below 30 percent relative humidity and often below 15 percent.

Data from Aviation Industry Reporting Systems

Data from aviation health and safety and complaint reporting systems, understandably not random, present a markedly different picture from published research data (see Figure 5-3). More than half of the 84 NASA ASRS reports that reported cabin temperatures (out of 154 total reports) cite temperatures of 35^oC (95^oF) or higher, with approximately 13 percent reporting temperatures at or above 40^oC (104^oF). Data from the 2Hot-2Cold app (4,679 relevant reports, all but 4 of which included temperature values) and the AFPA Hot Cabin reporting system (352 reports with 261 including temperature values) show an intermediate range, with the majority of reports toward the warmer end of cabin temperatures, falling between 25^oC (77^oF) and 35^oC (95^oF).⁵ Corresponding humidity data are not available from these sources; however, it is reasonable to assume that humidity

___________________

⁵ The vast majority of reports from these systems concern heat-related cabin temperature issues. In fact, only approximately 5 percent of the relevant NASA ASRS reports identified by the committee involved cold cabin conditions.

could be elevated at higher temperatures during ground operations when ventilation systems are shut off (Horstman, 2025).

Differences in reported temperatures across these systems reflect the different objectives and methods for the reporting systems. NASA ASRS reports are primarily submitted to document safety concerns and are typically generated only in such cases. In contrast, reports from the 2Hot2Cold app and the AFPA Hot Cabin system are better characterized as complaint data, submitted when flight attendants or other cabin occupants perceive conditions as excessively hot (or cold). These reports may or may not be intended to signal a safety issue. Characteristics of these different data sources (including years of available data) are summarized in Table A-2 in Appendix A. Contextualization of the aviation industry reporting data relative to the total number of flights estimated to have occurred over the respective reporting periods is provided later in this chapter (see section on Assessing the Magnitude of the Problem).

Figures 5-4 and 5-5 compare the 2Hot2Cold and NASA ASRS reports with in-flight and ground temperature data from ACER, ASHRAE 1262, and Nicholls and Vink. Figure 5-4 compares in-flight temperatures from all three data sources with the complaint reports, while Figure 5-5 contrasts only Nicholls and Vink on-ground temperatures with those reports. There is an evident distinction between the more typical cabin conditions and the extremes that trigger complaints or safety reports.

Mapping Cabin Temperature and Humidity Data to Thermal Zones Based on Thermal Comfort and Heat and Cold Stress Standards and Guidelines

To explore the available indirect evidence linking cabin temperature and humidity levels to health and safety outcomes, the committee overlaid measured cabin conditions from the Nicholls and Vink, ACER, and ASHRAE RP-1262 datasets onto the thermal response zones that were prepared and described in Chapter 4. These zones represent ranges of environmental conditions associated with comfort, discomfort, and heat or cold stress for typical low and moderate activity levels. Similar methods have been used by others to interpret cabin temperature and humidity data in the context of limits from thermal comfort standards (Gameiro da Silva et al., 2023; Haghighat et al., 1999), although the committee is unaware of any such comparisons that have also incorporated heat and cold stress limits.

To orient readers, Figure 5-6 introduces the thermal zones alone that were developed in Chapter 4. Figures 5-7, 5-8, and 5-9 incorporate data

points from the datasets with linked cabin temperature and humidity measurements. The curvilinear zone boundaries in these figures reflect the effects of humidity on thermal comfort and stress. This visual framework enables comparison of measured cabin conditions with zones that influence physiological, cognitive, and behavioral responses (see Chapter 3). For the mappings, in-flight data from all three datasets (Nicholls and Vink, ACER, and ASHRAE RP-1262) were combined in Figures 5-7 and 5-8. The Nicholls and Vink on-ground measurements collected from 113 flights before departure from the gate were mapped separately (Figure 5-9).

Given the effects of metabolic activity level on thermal comfort, it is instructive to view the measured flight conditions considering the range of activities encountered by flight attendants and by passengers. Figures 5-7 and 5-8 illustrate how measured cabin conditions compare to thermal comfort and stress zones. The figures depict the same cabin temperature and humidity data, but the thermal response zones differ, reflecting two different activity levels described in Chapter 4:

• Case scenario 1: Active flight attendant—representing a moderate activity level and typical clothing (2.3 MET, 0.6 clo)

• Case scenario 2: Seated passenger or flight attendant—representing a low activity level and typical clothing (1.2 MET, 0.60 clo)

Case scenario 1 with 2.3-MET activity level is intended to represent the condition of the sustained activity of a busy flight attendant. Typically, flight attendant activities will be well below this level for much of a flight, especially longer flights. For this scenario (Figure 5-7), conditions from the three data systems fall primarily within the warm discomfort zone, with few observations near cold discomfort and none within heat stress or cold stress zones. This pattern reflects the higher metabolic load of active flight attendants, which increases susceptibility to warm discomfort even at moderate cabin temperatures.

For case scenario 2 (Figure 5-8), the pattern is reversed. More data points fall within the cool discomfort zone, with some approaching cold stress thresholds, while conditions on the warm side remain well below heat stress levels. These differences underscore the challenge of maintaining thermal comfort for both active cabin crew and sedentary passengers in the same environment.

Figure 5-9 presents boarding conditions mapped to thermal response zones for case scenario 1 (active flight attendant in typical clothing) using data from Nicholls and Vink—the only dataset with pre-departure measurements. Boarding conditions are broadly similar to in-flight conditions but show more points near the heat stress threshold, reflecting the absence of full environmental control and higher humidity levels during ground operations. However, no recorded conditions fell within the heat stress or cold stress zones.

Case scenarios 1 and 2 depict two activity levels that generally represent the range of activities expected on aircraft. As seen in Figures 5-8 and 5-9, the fraction of the conditions that falls within each zone, not surprisingly, is much different for low activity than for high activity. In reality, activities will fall on a continuum with most activities not at either extreme, especially for flight attendants. Figure 5-10 examines how activity level across a broader range influences the fraction of cabin conditions falling within each thermal zone, based on combined in-flight data from all three datasets. As activity level increases, the proportion of conditions classified as warm discomfort rises, while the proportion of cool discomfort decreases. This trend illustrates how metabolic rate interacts with cabin conditions to shape thermal experience. It is important to note that zone boundaries represent average responses; individual variability can be substantial. Even though the zones are represented as discrete boundaries in Figures 5-8 and 5-9, the transition from comfort to discomfort and to stress is gradual, and the level of discomfort is greater at conditions near a stress boundary than at conditions near a comfort boundary.

The mapped cabin conditions represent a single point in time for each flight representative of cruise (Figures 5-7 and 5-8) or boarding (Figure 5-9) conditions, but, as discussed throughout this report, duration has a significant influence on the health and safety impacts of temperature exposures. Longitudinal measurements available from the ASHRAE 1262 and ACER datasets suggest that the in-flight cabin conditions depicted in Figures 5-7 and 5-8 represent extended exposures and are not just the result of short fluctuations in cabin conditions (see Appendix B). Longitudinal data are not available for ground conditions depicted in Figure 5-9.

Lacking corresponding humidity measurements, the temperature data from aviation safety and complaint systems could not be similarly mapped onto thermal response zones. However, many of the reports describe incidents occurring during ground operations (see Appendix B for flight phase specific analyses), when humidity would generally be higher than in-flight

levels, particularly in warmer seasons. Temperatures exceeding 35^oC (95^oF) as reported in NASA ASRS submissions, at even moderate humidity levels (30 to 40 percent), begin to approach heat stress conditions for individuals functioning at higher metabolic activity levels (see Figure 4-1). Reports from safety and complaint reporting systems generally lack information regarding the duration of exposure, however, which is a key factor influencing whether temperature exposures change from uncomfortable to unsafe conditions.

Summary and Synthesis

As indicated at the start of this chapter, the committee was guided in its analysis of indirect evidence of cabin-temperature-related health and safety impacts by key questions addressing the characterization of (1) exposure of cabin occupants to cabin temperatures outside the expected thermal comfort range and (2) the health and safety impacts of such exposures.

With regard to exposures, multiple sources indicate that cabin temperatures sometimes reach levels at which health and safety impacts would be expected. Data from the 372 flights included in research studies, when mapped to thermal response zones, show that in-flight temperatures and humidity generally fall within comfort zones, though warm or cool discomfort is common, depending on activity level. There is little evidence that cold thermal stress is widespread, though localized cold discomfort may occur. Boarding conditions in warmer seasons when humidity levels are elevated due to ambient conditions approach the thresholds for heat stress more frequently, underscoring ground operations as a critical phase for thermal management. In contrast, the temperatures reported through aviation safety or complaint systems (e.g., 2Hot2Cold, NASA ASRS) reflect more severe conditions—often far exceeding typical ranges. The reported health and safety impacts of such exposures include a combination of physical, cognitive, and behavioral effects (see Table 5-2), consistent with the published literature on thermal discomfort and stress described in Chapter 3. While these events are infrequent, their severity and potential safety implications warrant targeted mitigation strategies (discussed in Chapter 6), particularly for ground delays and equipment failures.

Conclusion 5-1: The Federal Aviation Administration has jurisdiction over crewmember safety and health. Evidence provided in this report indicates crewmembers continue to experience cabin thermal conditions that result in adverse physiological, cognitive, or behavioral effects.

While the figures depicted in this chapter provide valuable insight into typical cabin conditions and the extremes documented in complaint and safety reporting systems, they do not reveal how often these severe events occur. Understanding the frequency of such conditions is essential for determining the scale of the problem. To assess the magnitude of cabin temperature issues across Part 121 operations, the following section further examines available data sources, including voluntary reporting systems and medical consultation records.

ASSESSING THE MAGNITUDE OF THE PROBLEM

Comparing research datasets and aviation industry reporting system data reveals that extreme cabin temperature conditions are observed on a very small fraction of flights. The key question, therefore, is how often these more concerning conditions occur. The committee identified a total of 154 relevant NASA ASRS reports from a 36-year period (1990–2025) (see Appendix A), while approximately 5,000 2Hot2Cold complaints were submitted over 8 years (2018–2025). For context, U.S. carriers operated approximately 193 million scheduled passenger flights between 2004 and 2024 (approximately 9 million per year) (DOT, 2025a). Although the time-frames do not align perfectly, the committee calculated that ASRS reports equate to roughly 0.5 reports per 1 million flights (approximately 4 reports per year on average from 2004 to 2024). It is important to note that this number is the frequency of reports, not the frequency of events. Almost certainly it is an undercount of actual events as not every serious hot or cold cabin event is likely to be reported, and relevant reports may have been missed in the keyword search of the database. Unfortunately, no means to estimate the undercount was identified. The frequency calculated is a lower limit. It is not known if it is an accurate estimate, a gross underestimate, or somewhere in between. Similarly, 2Hot2Cold reports are unlikely to represent all occurrences of uncomfortable conditions. Awareness of the reporting app among passengers is likely limited, as most submissions (89 percent) come from flight attendants (AFA, 2025). Even among flight attendants, awareness may vary from airline to airline. Based on available data, committee calculations indicate that 2Hot2Cold reports correspond to approximately 0.8 events per 10,000 flights (approximately 700 events per year). Again, this figure is the frequency of reports and not necessarily the frequency of events. No means to estimate the undercount was identified, and this rate may be an accurate estimate, a gross underestimate or somewhere in between. While complaint conditions appear to occur on a small percentage of flights, they are not negligible, given the scale of

commercial air transport operations.⁶ Also, extreme conditions as described in the ASRS reports and complaints as documented in the 2Hot2Cold data are not expected to occur randomly. They likely are concentrated at certain locations, on certain flight routes, on certain types of aircraft, and on certain types of airlines (e.g., regional versus major). Even though excessively warm or cold conditions may occur on a small fraction of flights, some individuals—certain crewmembers in particular—may experience these conditions repeatedly and with much greater frequency than the overall statistics indicate.

While not linked to health and safety impacts, airline data on the frequency of hot and cold cabin reports also informed the estimations of the magnitude of the problem. Reported numbers of hot cabin events between 2021 and 2025 varied across responding airlines and ranged from 0 to 0.159 percent of total flights during that period (corresponding to a rate of 0 to 15.9 events per 10,000 flights) (A4A, 2026). Cold cabin reports during this same period were generally less frequent and ranged from 0 to 0.109 percent of total flights (corresponding to a rate of 0 to 10.9 events per 10,000 flights).

The airline and aviation safety and complaint system self-reporting data were supplemented by evidence the committee received from two aviation medical consultation organizations, which further suggested the rarity of medically serious health impacts from cabin temperature exposures. A review of two independent databases—MEDAire (2025) and STAT-MD (2025)—yielded very few relevant cases (see Appendix A for description of information requests).

• MedAire: Among 28,072 pre-flight medical consultations for 22 Part 121 carriers over approximately 2.5 years (January 1, 2023–August 31, 2025), only seven cases appeared to involve cabin temperature exposures. Four occurred on previous inbound flights after landing and three during departure prior to takeoff. Of 63,330 in-flight consultations during the same period, none were related to cabin temperature.
• STAT-MD: Across 50,577 consultations over almost 9 years (November 1, 2016–September 17, 2025), only 12 cases were potentially temperature-related: seven linked to heat, two to

___________________

⁶ Data have been collected using the 2Hot2Cold app since 2018, a shorter period than for the NASA ASRS. Additionally, reporting dropped precipitously in 2020, reflecting decreased air travel during the COVID-19 pandemic, making annual estimates from this dataset more uncertain. While characterized as complaint data, it is notable that 173 of the nearly 5,000 reports submitted over the 8-year period noted a need for medical attention, underscoring that reports in this dataset are not limited to complaint conditions.

• cold, and three to dehydration (which may or may not have been temperature-related).

The limitations of the data need to be kept in mind when interpreting them in the context of estimating the frequency of cabin-temperature-related health and safety events. These limitations include:

• Reliance on keyword searches of case notes (e.g., for terms like “hot,” “heat,” “cold”) may have resulted in missed cases.
• Not all airlines contract with medical consultation organizations, and, for those that do, cases managed onboard without third-party involvement would not appear in these databases.
• Heat-related illnesses may have been treated by general medical services after deplaning, bypassing aviation-specific reporting systems.

Consumer complaint data tracked by DOT’s Office of Aviation Consumer Protection (OACP) also suggest that occurrences of cabin temperature issues severe enough to prompt reporting through DOT mechanisms available to passengers are infrequent. Between 2014 and 2023, less than 1 percent of the consumer complaints received by DOT (409 of approximately 200,000) related to cabin temperatures (DOT, 2025b). OACP investigations launched in response to these consumer complaints identified 44 violations of cabin temperature requirements (i.e., the requirement to maintain a “comfortable” cabin temperature during tarmac delays in accordance with 49 USC § 42301, as discussed in Chapter 1). Over the same period, OACP issued 13 warning letters and one enforcement order related to violations of cabin temperature requirements (DOT, 2025b). Of note, passenger complaints to individual airlines would not be included in DOT OACP records.

In summary, while the committee cannot precisely quantify the frequency of extreme hot or cold cabin conditions, available evidence indicates that events that pose serious health and safety risks occur on a very small fraction of flights—less than 1 per million flights. By contrast, conditions causing substantial warm and cold discomfort, though not medically severe, occur more routinely—perhaps 1 in 10,000 flights, equating to, on average, roughly two incidents per day across the U.S. fleet.

Operational factors likely contribute to these exceedances. For example, NASA ASRS reports often cite equipment issues such as inoperative auxiliary power units (APUs) or air conditioning packs (see Chapter 2). While prohibiting flights under these conditions could prevent many extreme events, such policies might disrupt thousands of flights annually to avoid only a handful of severe incidents. Until more comprehensive data are available, broad policy changes should be approached cautiously in order

to balance safety, effectiveness, and operational feasibility. Policy and other mitigation strategies are discussed in Chapter 6.

Conclusion 5-2: Conditions involving cabin temperatures that pose serious health and safety risks in the form of physiological effects are not routinely encountered in Part 121 operations. However, temperatures eliciting thermal discomfort are likely to occur more frequently and are capable of producing cognitive and behavioral effects. Although the frequency of health and safety incidents involving physiological, cognitive, or behavioral effects cannot be reliably estimated with currently available data, reports of cabin-temperature-related health and safety incidents suggest a need for improved monitoring, reporting, and assessment of hazards and risk mitigation options.

ADDRESSING CURRENT DATA LIMITATIONS

Limitations of Existing Data

A key takeaway from the analyses presented in this chapter is the limited availability of reliable data from which conclusions regarding the health and safety impacts of cabin temperatures can be drawn. Direct evidence linking health outcomes to aircraft cabin temperatures was confined to a single study that relied on subjective reporting. The committee’s analyses of indirect evidence were based largely on studies involving small convenience samples of flights and variable measurement methods as well as voluntary reporting systems with self-reported data, none of which provided a comprehensive or representative picture. Despite these limitations, the available data provide useful insights for guiding mitigation strategies. For example, multiple sources indicate that hot cabin events are more frequent in summer and during ground operations (see Figures B-1 and B-5 in Appendix B), reinforcing the need for seasonal planning (discussed further in Chapter 6).

The committee explored additional sources of health and safety data that do not rely on subjective reporting, including OSHA injury data, emergency medical services (EMS) data, and records from organizations providing medical consultation services to the aviation industry (discussed earlier in this chapter). However, these sources proved largely uninformative for the following reasons:

• OSHA injury data: As airlines must meet OSHA recordkeeping requirements,⁷ the committee commissioned an analysis to explore associations between outdoor ambient temperature—which has been found to correlate with boarding temperatures (Nicholls and Vink, 2025)—and reported flight attendant injuries occurring on aircraft. However, airlines are not required to list the actual injury location in OSHA logs, and the available data typically reflect the flight attendants’ home base rather than the site of the incident, precluding meaningful linkage to ambient temperature.
• EMS and medical consultation data: National-level EMS data can identify encounters at airports or on airplanes but cannot distinguish whether incidents involved cabin occupants versus other scenarios (e.g., ground crew working outside on the ramp). While state-level data may offer greater detail, collecting and analyzing these data was not feasible within the committee’s timeframe. Similarly, medical consultation databases yielded very few cases potentially related to cabin temperature, and keyword-based searches may have missed relevant events (e.g., cases of heat-related conditions that are not recognized heat illnesses).

Although these data were not useful for the committee’s analyses, it is conceivable that changes to data collection and reporting methods for these sources could increase their utility in the future.

Conclusion 5-3: There are no systematically collected data with which to assess the health and safety impacts of aircraft cabin temperature on flight attendants and passengers. Available data on cabin temperature and humidity are fragmented, incomplete, and based on extremely small samples relative to the millions of commercial flights conducted annually. The lack of health and safety data—including cognitive and behavioral effects and data specific to vulnerable populations—impedes the assessment of associations between outcomes and temperature-related exposures.

___________________

⁷ In 2023, OSHA implemented a final rule (88 FR 47254) requiring establishments with 100 or more employees to electronically submit detailed records of every work-related injury and illness through the agency’s Injury Tracking Application. Publicly available records on flight attendant injuries can be found at https://www.osha.gov/Establishment-Specific-Injury-and-Illness-Data (accessed March 6, 2026).

Future Data Needs for Estimating Frequency of Cabin Temperature and Humidity Hazards

It follows from Conclusions 5-2 and 5-3 that additional data collection efforts are needed to improve assessment of cabin temperature and humidity hazards. Analyses of newly captured data could advance risk mitigation approaches, although such actions need not wait on data collection efforts (see Chapter 6). Data obtained from a research effort would ideally be made publicly available following completion of the research and publication of the findings, consistent with general practice for government-funded research.

Recommendation 5-1: The Federal Aviation Administration, in collaboration with airlines and aircraft manufacturers, should establish a research program to systematically monitor, record, and assess temperature and humidity conditions in a representative sample of Part 121 passenger aircraft cabins.

The objectives of the recommended research program would be primarily to develop an initial evidence base for predicting the frequency and severity of cabin-temperature-related health and safety events in higher-risk settings. An appropriately robust estimate of the frequency and distribution of measurements falling across the categories represented in Figures 4-1 and 4-2 is essential for such predictions. A feasible research effort should be restricted to a sample of high-risk settings. To identify these settings and design temperature and humidity data collection instrument packages and procedures for the distribution and collection of appropriate data in such settings, a multi-step process is suggested. Below the committee offers, as one of several reasonable options, a possible approach to the recommended research.

The committee identified a number of issues that would need to be considered either in the research effort or in the case of a conclusion that indicates more widespread ongoing collection of data is necessary.

While onboard monitoring systems offer the most reliable environmental data, older aircraft generally lack these capabilities and are unlikely to be retrofitted. Where suitable onboard sensors and data recording capabilities already exist, the program should take advantage of those systems immediately. For aircraft lacking these capabilities, portable instrument packages (sensors and data loggers) that measure, at a minimum, air temperature, humidity, and cabin pressure (to identify flight phase) and feature onboard recording and remote download capability would likely need to be developed. Such tools may be particularly important for capturing cabin humidity levels, as the committee heard that onboard systems do not currently monitor and record such data (A4A, 2026; AVA, 2026; RAA, 2026) and have been used in previous research on cabin environmental factors (Batelle, 2018).

Current technologies for low-power compact sensors, batteries, data storage, and communication make continuous measurement feasible over multiple days without recharging or downloading. The primary challenge lies in logistics—deploying, maintaining, tracking, and managing the sensors and the large volumes of data collected. Procedures will need to be developed to ensure consistent deployment, recharging, data retrieval, management, analysis, and interpretation. Because temperature and humidity are not uniform throughout the cabin, more than one sensor may be needed on some aircraft to assess spatial variation (e.g., between classes of service, near open doors while aircraft are on the ground, and measurements on solar side of aircraft).

Ideally, a sufficient number of sensor packages would be deployed to collect data across a representative cross-section of aircraft, flights, routes, and environmental conditions with enough flights monitored to capture infrequent events. Consultation with statisticians may help inform the number of sensor packages that would be sufficient to achieve a representative cross-section. Data on ambient thermal conditions would ideally also be collected when capturing cabin thermal data during ground phases to better understand high-risk conditions. Depending on the types of infrequent events to be captured, the magnitude of the monitoring required can become infeasible with a random sampling approach. With dozens of different aircraft makes and models, dozens of different types of flights, dozens of different types of routes, and a variety of different types of environmental conditions, capturing infrequent events—for example, that occur on perhaps one in 10,000 flights—and parsed by each of these factors would require an astronomical number of flights to be monitored. Identifying and targeting high-risk conditions—such as routes that operate within hot or

cold environments, airports where cabin temperature issues are identified more frequently, identified problem routes, or aircraft with deferred APU repairs—should increase the likelihood of capturing infrequent events of interest where health and safety are clearly at risk of being compromised. Such targeting is essential for a practical, limited scale research data collection effort.

At least two possible outcomes of the proposed FAA-funded research program can be envisioned:

1. There is sufficient evidence that incidents of temperature exceedances occur such that cabin-temperature-related health and safety incidents are predicted to occur with such frequency and wide distribution to warrant an ongoing monitoring program.
2. The evidence suggests that problem exceedances are sufficiently localized that targeted mitigation measures (targeting specific aircraft, routes, seasons) would sufficiently address cabin-temperature-related hazards without the need for an industry-wide monitoring program.

In the first scenario, there are both potential benefits and challenges to an ongoing monitoring program that would need to be considered. Such a program could benefit industry by supporting ongoing quality-assurance efforts, benchmarking, and predictive risk modeling. Similar approaches are already used by organizations such as the International Air Transport Association for other operational parameters (IATA, 2026). The resulting data from such a program would also enable the ongoing evaluation of circumstances and conditions that predispose critical medical emergencies and inform strategies to prevent them.

In contrast to a research program, however, reliance on temporary sensors and data loggers for routine monitoring of cabin environmental conditions is not feasible. A long-term monitoring program would therefore require onboard sensor systems for recording and downloading cabin temperature and humidity, the feasibility of which (e.g., technological, logistical) would need to be evaluated. Some data collection challenges may resolve over time as older aircraft models are replaced with newer ones that include temperature recording and download capabilities. However, as noted above, aircraft used in Part 121 operations do not currently include equipment for monitoring humidity. Achieving such capabilities would require manufacturers to offer retrofit kits to airlines and/or incorporate integrated fixed humidity sensors into the aircraft design for new production aircraft, which, the committee acknowledges, would be accompanied by nontrivial tradeoffs (e.g., added weight, data storage needs, power requirements, maintainability). Moreover, assurance of the reliability of the

data (precision and accuracy) and calibration requirements need to be considered, given the difficulty with humidity measurement at the low ranges typical for in-flight conditions (between 10 and 20 percent) and elevated ranges during ground operations (i.e., greater than 50 percent).

Because collecting temperature and humidity data on all flights may be prohibitive, a systematic sampling of airlines and flights could be done annually to collect the necessary information to serve as the basis of tracking of thermal conditions, their burden, and their consequences, either industry-wide or within sub-groups, if indicated. A program that might serve as a model for such a tracking system is the U.S. Department of Labor’s Bureau of Labor Statistics annual national sample of enterprises to report occupational injuries and illnesses, which makes it possible to provide national estimates of the distribution, frequency and burden of occupational injury and illnesses (BLS, 2024).

Environmental factors beyond temperature and humidity are recognized as important to the health and safety of cabin occupants. Therefore, such a program, if pursued, could be integrated with monitoring efforts for other cabin air quality parameters, achieving greater efficiency than independent monitoring programs. The FAA-funded Cabin Air Quality and Bleed Air Contamination Detection project (FAA, 2025) outlines such a program and may serve as a guide.

Separately, FAA should also explore (e.g., through a workshop) methods for systematic and direct collection of data on health outcomes from exposure to hot or cold cabin temperatures, particularly for vulnerable populations (e.g., infants and young children, individuals with chronic conditions, and others with increased susceptibility). Changes in passenger and cabin crew demographics, particularly the aging of the U.S. population, likely affect vulnerability to cabin environmental conditions and should be considered. Health-related data to supplement those from existing complaint and safety reporting systems are more challenging to collect at scale, but examples of potential methods include simple surveys to collect self-reported data on physiological, cognitive, and behavioral effects of cabin thermal conditions and digital health technologies (e.g., smartwatches), which provide objective physiologic measures to complement environmental monitoring and self-report data. Most airlines and airline associations periodically conduct passenger surveys to gain insights into passenger satisfaction, behaviors, and expectations. Stakeholders across diverse segments (e.g., air carriers, airports, manufacturers) can use the resulting data and insights to make informed decisions that guide research projects, enhance strategies for operations, or inform policy decisions with the goal of improving passenger experiences and aligning more closely with evolving passenger and crewmember needs. Surveys for health, safety, or cognitive/behavioral experiences that carry minimum burden for airlines and passengers could be

modeled on these. By using a mix of closed- and open-ended questions, air carriers could gather both quantitative and qualitative data on experiences of thermal discomfort and any associated physical, cognitive, and mental health symptoms. Air carriers can distribute surveys through a variety of channels, such as email, mobile apps, in-flight entertainment systems, or social media platforms.

REFERENCES