Stuffy nights: elevated bedroom carbon dioxide concentrations indicate inadequate ventilation in Wellington homes

Adequate ventilation or the supply of fresh outdoor air to indoor spaces is important for maintaining healthy indoor environments.¹ This is particularly relevant in bedrooms, where prolonged occupancy can allow exhaled air and other indoor pollutants to accumulate.² In most New Zealand homes, air exchange relies primarily on natural ventilation (e.g., window and door opening) rather than mechanical systems, and opportunities to ventilate may be constrained by weather, temperature, security concerns, outdoor noise and concerns about outdoor air quality.³

Indoor carbon dioxide (CO₂), primarily generated by occupant respiration, is widely used as a practical proxy indicator of ventilation, with elevated concentrations generally reflecting lower outdoor air exchange per person.⁴ Outdoor CO₂ concentrations in New Zealand are close to global background levels (approximately 423–425ppm).⁵ Indoor concentrations below 800ppm are commonly used as an indicator of adequate ventilation, with higher levels suggesting insufficient air exchange and increased potential for accumulation of indoor pollutants and exhaled bioaerosols, including respiratory pathogens.⁶ New Zealand Standard NZS 4303:1990 Ventilation for acceptable indoor air quality recommends maintaining indoor CO₂ below 1,000ppm.⁷ However, this standard predates more recent evidence on indoor air quality and airborne infection risk, and lower thresholds, such as 800ppm, or alternatively, indoor levels within approximately 400ppm of outdoor concentrations, are now commonly used as indicators of better ventilation.⁶

A growing evidence base indicates that poor bedroom ventilation and elevated overnight CO₂ concentrations are associated with impaired sleep and worse next-day functioning.^8–11 Controlled laboratory studies and real-world bedroom interventions that have trialled increasing ventilation through window or door opening consistently reduce CO₂ concentrations and are associated with small but measurable improvements in sleep quality and efficiency, as well as next-day performance.⁹ For example, a Danish pilot study in student dormitory bedrooms found that increasing ventilation through window opening or fan-assisted airflow reduced CO₂ concentrations and improved self-reported air quality, sleep quality and next-day cognitive performance.⁸ Importantly, several studies use experimental or crossover designs, which strengthen evidence for a causal contribution of ventilation to observed sleep outcomes despite the presence of potential confounding factors.^8–10 Observational evidence similarly links perceived “stuffy” bedroom air and mechanical ventilation type with sleep quality,¹² and more recent real-world research demonstrated that modifying bedroom ventilation rates influenced overnight CO₂ levels and sleep parameters.¹⁰

Despite people spending approximately one-third of their lives asleep, bedroom ventilation in New Zealand homes remains poorly characterised. Existing research is limited to one small study (seven households with two bedrooms monitored per household, n=14) that reported bedroom CO₂ concentrations of up to 4,000ppm.¹³ Therefore, this pilot study aimed to contribute to the evidence base by measuring and evaluating bedroom CO₂ concentrations in a sample of Wellington households.

Methods

Twenty households were recruited through housing support organisations and social media and included at least one consenting adult who self-identified their home as damp. Eligible households were located in Wellington, had access to a mobile smartphone and had a suitable location for monitor installation.

We monitored 20 bedrooms for 2-week periods between July and October 2023 (winter to spring), measuring CO₂, temperature and humidity, using AirSuite monitors (Glance LTE) equipped with non-dispersive infrared CO₂ sensors (range approximately 0–10,000ppm; accuracy ±30ppm, ±3%). In each household one monitor was deployed to one bedroom. Monitors were positioned away from direct breathing zones, windows or direct airflow at a height of approximately 1.5 metres. Households were not provided with any instructions or guidance regarding ventilation or indoor air quality. The sample included a mix of housing types, ages and tenures, comprising owner-occupied dwellings as well as private and social rental properties. Households included people living alone, couples, families with children and shared flats. The study design was approved by the University of Otago Human Ethics Committee (ref: 23/026).

Data were analysed using R (version 4.3.0). CO₂, temperature and relative humidity measurements were aligned by day since monitoring commenced, rather than by calendar date. Data were recorded as 1-minute intervals and aggregated to hourly medians for each household. Night-time was defined as 20:00–08:00 and daytime as 08:00–20:00. In time series figures, individual households are shown as faint lines, with summary statistics presented across households. The proportion of night-time hours with CO₂ concentrations >800ppm was calculated for each household and presented as a ranked dot plot. For exploratory analyses, households were classified as having consistently high overnight CO₂ if ≥80% of night-time hours exceeded 800ppm, and consistently low CO₂ if <20% exceeded 800ppm; households in the intermediate range were excluded. Median night-time temperature and relative humidity were compared between groups using Mann–Whitney U tests.

Results

Most bedrooms were typically occupied by a single person overnight, with a maximum of two occupants. Night-time bedroom CO₂ concentrations were frequently elevated, with the overall median night-time CO₂ 907ppm (75th percentile 1,319ppm; range 420–8150ppm) (Figure 1A). Peak night-time CO₂ concentrations exceeded 2,000ppm in half of the bedrooms, and one bedroom had a maximum concentration of 8,150ppm. The three bedrooms with the highest CO₂ concentrations recorded repeated night-time peaks above 3,000ppm. Day-time bedroom CO₂ concentrations were lower than those recorded at night-time (median 609ppm; range 390–4,020ppm), but several bedrooms still often recorded levels above 800ppm (Figure 1B, 1C).

View Figure 1–2.

At the household level, the percentage of night-time recordings higher than 800ppm ranged from 0.15% to 100% (Figure 2). Six bedrooms spent ≥80% of night-time above 800ppm, while two spent <10% of night-time above 800ppm. The highest-exposure bedrooms showed sustained CO₂ elevations across most nights.

The median night-time bedroom temperature across households was 15.8 degrees Celsius (range 12.7–19.4 degrees Celsius), and median relative humidity was 71.0% (range 55.9–80.4%). Median night-time bedroom temperature and relative humidity did not differ between households with consistently high versus consistently low overnight CO₂ levels (temperature 17.7 degrees Celsius, interquartile range [IQR] 13.9–19.1 versus 13.8 degrees Celsius, IQR 13.6–14.1, p=0.64; humidity 71.4%, IQR 66.4–78.8 versus 72.7%, IQR 71.7–73.7, p=0.86).

Discussion

In this pilot study of Wellington homes, bedroom CO₂ concentrations were frequently elevated overnight, with many bedrooms exceeding the commonly used 800ppm ventilation threshold for much of the night. Half of homes experienced night-time peak concentrations above 2,000ppm, and several recorded repeated peaks above 3,000ppm. Because households were not provided with any instructions regarding ventilation or indoor air quality, the concentrations observed reflect usual, real-world bedroom conditions. The values in our study exceed those typically reported in most international bedroom field studies, where night-time CO₂ concentrations in poorly ventilated rooms more commonly range between approximately 900ppm and 2,500ppm.^8–10 Our findings indicate very low overnight outdoor air exchange in a substantial proportion of bedrooms and suggest that inadequate overnight bedroom ventilation may represent an under-recognised indoor public health issue in New Zealand housing.

In our study elevated CO₂ occurred despite similar bedroom temperature and relative humidity across households, indicating that ventilation practices and building characteristics, rather than thermal conditions, are the primary drivers of overnight CO₂ accumulation. In the context of New Zealand’s predominantly naturally ventilated housing stock, this likely reflects limited and inconsistent overnight air exchange in many bedrooms, where windows and doors are often closed during sleep. Variation between households may also reflect differences in occupancy and ventilation behaviours as supported by field notes indicating that some bedroom doors and windows were commonly kept closed overnight. In addition, there is a wide range of ventilation rates across New Zealand housing, where older, more “leaky” homes tend to have higher background air exchange than newer, more airtight dwellings,¹³ which may have contributed to variability in bedroom CO₂ levels.

These are important findings because multiple studies indicate that bedroom ventilation influences sleep quality and next-day functioning.^8–11 Poor ventilation also increases the likelihood that indoor pollutants and dampness-related exposures, including mould, bacteria and dust mites, could accumulate indoors and contribute to respiratory symptoms, asthma and a range of adverse health outcomes.^14–16 Conditions associated with low ventilation, reflected by elevated CO₂, are often associated with higher indoor humidity, which can be further increased by moisture generated through breathing. Persistent indoor relative humidity above approximately 65% increases the likelihood of mould growth. Mould is a well-recognised problem in New Zealand housing, particularly in rental properties—where mould is reported in over half of homes.^15,17

Practical strategies to reduce overnight CO₂ concentrations in bedrooms include increasing natural ventilation (e.g., opening windows or doors), ensuring vents are unobstructed. However, reliance on occupant behaviour alone may not be sufficient, as ventilation achieved through window opening can be variable and may not adequately control indoor temperature, humidity or pollutant accumulation, even when windows and doors are opened. Mechanical systems can provide more consistent air exchange, although their installation and operating costs may limit uptake, particularly in existing housing.

Current regulatory frameworks, including the Building Code, NZ 4303:1990 and the Healthy Homes Standards, do not specifically address ventilation in bedrooms or overnight conditions. While based on a small sample, our findings suggest that elevated CO₂ concentrations may be common in New Zealand bedrooms, indicating a potential gap in current standards. This indicates that current standards may warrant review and highlights the need for practical, low-cost and retrofit-friendly solutions to improve ventilation in existing housing. Strengthened design specifications for new homes are also needed, given the trend towards increasing airtightness.

This study has several limitations. Firstly, CO₂ was used as a proxy for ventilation rather than as a direct measure of indoor air quality or health outcomes. Elevated CO₂ concentrations therefore reflect conditions of low air exchange, under which indoor pollutants may accumulate. The sample size was modest and the monitoring period was relatively short, and we did not directly measure air exchange rates or occupant behaviours.

Despite the limitations mentioned above, this pilot study provides novel, real-world evidence of substantial variation in overnight bedroom ventilation conditions in Wellington homes. Our study was not designed or sufficiently powered to assess differences by building age, and future research should include larger samples, details about occupant behaviours, weather conditions, longer monitoring periods and direct assessment of associations between bedroom ventilation, CO₂ concentrations, sleep quality and next-day functioning. Further work is also needed to evaluate practical interventions to improve overnight bedroom ventilation.

Conclusion

Elevated bedroom CO₂ concentrations were common across separate households in this sample, indicating widespread inadequate overnight bedroom ventilation and highlighting an opportunity to improve ventilation and indoor air quality, and support sleep, performance and health in New Zealand households.