This research looks at the health impacts of using gas stoves, unflued gas heaters, open fires and wood burners in New Zealand homes.
The work was commissioned by EECA to contribute to our evidence base relating to household energy use in New Zealand.
About the research
The research was conducted by Emission Impossible Limited, with economic modelling support from Resource Economics.
The study looked at the impacts of nitrogen dioxide (NO2) and particulate matter (PM2.5) emitted indoors from using different types of appliances for heating and cooking.
The study focuses on the direct emission of combustion pollutants into the home. For the gas appliances considered, this happens during normal use and is mitigated by ventilation (from windows and extractor fan use). For wood burning, combustion emissions vary depending on the temperature of the fire, and spike during refuelling, when the door is open.
Summary of report findings
The study found that the use of gas stoves, unflued gas heaters, open fires and wood burners result in the emission of harmful pollutants into the home.
The study found that the amount of exposure to pollutants depends on the appliance, with open fires and unflued gas heaters being the most harmful, followed by gas stoves, older wood burners, and then modern wood burners.
Converting this pollution exposure into health harm, the study estimated the total social cost of indoor air pollution from combustion appliances to be in the order of $5 billion dollars per year. It further estimated the fiscal cost of health treatment and restricted activity at $102 million dollars a year.
Research methodology and considerations
Methodology
The study methodology, and calculation of health impacts, involved three key stages:
- A review of domestic and international evidence to determine how much incremental annual indoor air pollution was associated with each device — NO2 for gas stoves, and PM2.5 for wood burners.
- The application of exposure response functions to determine the health impacts associated with this incremental exposure to NO2 or PM2.5. Exposure response functions are derived from statistical analysis of the impacts of exposure to outdoor air pollution and are widely used to estimate the impacts of indoor air pollution.
- Calculation of the economic (social) and fiscal (health system and restricted activity day) costs associated with the health impacts. Social costs are driven by the Treasury estimates for the value of a statistical life, while fiscal costs are driven by the cost of hospitalisations and treatment, and restricted activity due to illness.
Considerations
This study is about additional exposure connected to appliance use — rather than outdoor air pollution that also impacts air quality in your home. Outdoor air pollution is well known and evidenced, and is principally linked to wood smoke from domestic wood burners and tailpipe emissions from traffic.
- The findings are not applicable to flued gas space heaters, outdoor gas water heater califonts, or outdoor cooking.
- A literature review was carried out to determine the best sources and methods for estimating incremental exposure based on real-world testing in New Zealand and around the world.
- The level of NO2 exposure was estimated using New Zealand evidence, benchmarked against more recent international evidence. The annual exposure level used in this study is in line, or lower, than estimated in these recent international studies - which reflect the current mix of recent appliances.
- This data and modelling include ventilation, window use, and extractor fan use — in line with the real-world efficacy and prevalence of these mitigations.
- The study averages across a diverse set of environments; its findings are an estimate of the overall statistical cost of indoor air pollution from combustion — rather than an estimate of the individual cost to a given household of an appliance.
- The research has been peer reviewed (see below) and benchmarked against international evidence.
- The assessment of the social cost of premature mortality utilises Treasury’s Value of a Statistical Life (VoSL), chosen for consistency with New Zealand’s major outdoor air pollution study HAPINZ 3.0. Alternative assessments for the social cost of mortality, which are also included in Treasury guidance, are lower.
Association versus causality
As with other studies conducted in North America and Europe, the present study directly attributes health impacts and associated costs to harmful pollutants emitted by indoor combustion sources. This implies that the relationship between air pollution and health effects is causal, and that the relationship between outdoor air pollution and health effects applies to indoor air pollution. This is consistent with the approach taken by the World Health Organization.
Assessments of causality rely on multiple lines of scientific enquiry, which include but are not limited to the epidemiological studies used to establish association. The most recent (2021) WHO Global Air Quality Guidelines reflect global consensus that there is a causal relationship between long-term PM2.5 exposure and mortality. WHO (2021) further noted that the relationship between long-term NO2 exposure and mortality is suggestive of causality, and likely causal for respiratory effects. Consequently, all health outcomes assessed in this report are either causal, likely to be causal or suggestive of causality.
Peer reviews
Three formal peer reviews have been conducted as summarised below. Learnings from all of these are incorporated into the current version of the report, and the content of this page.
Dr Nick von Randow
| Background | This peer review focused on the report’s health impact modelling and was commissioned by EECA prior to initial publication of the report. The review focused on credibility, evidence quality, and international context. |
| Summary | The peer review found that the health harm estimates were highly credible. Where this review did disagree with the report, it found that the report likely under-reported the level of harm from indoor air pollution. |
| Resolution | Estimates of harm were left unchanged, noting that this reflects a conservative estimate. |
| Peer review | Peer Review Dr Nick von Randow [PDF 267 KB] |
New Zealand Institute for Economic Research (NZIER)
| Background | This peer review focused on the economic and fiscal valuations of the heath harm estimated in the report and was commissioned by EECA after initial publication of the report. |
| Summary | The peer review found no technical errors in the report. The review noted that the use of the Value of a Statistical Life (VoSL) as a cost measure for premature deaths means that estimates of economic cost are higher than those used to assess other policy interventions which also impact on health outcomes. |
| Resolution | The report authors provided a response to all of the points raised by the peer review and have updated the report to provide further clarification that estimates of costs are based on a continuation of the status quo, rather than a reduction in pollution. Valuations in the report remain unchanged. |
| Peer review | Peer Review NZIER [PDF 731 KB] |
| Response | Response to NZIER Peer Review [PDF 155 KB] |
Dr Brian Cox
| Background | This peer review focused on the evidence for harm from gas combustion in the home and was commissioned by GasNZ. |
| Summary | The peer review asserts that a fundamental requirement of health impact assessments is the existence of a causal relationship, and that the report does not establish this. It also states that a direct relationship between indoor NO2 exposure and illness was not established. |
| Resolution | The report authors provided a response to points raised by the peer review and have updated the report to strengthen the discussion on association versus causation. Estimates of health impacts in the report remain unchanged. |
| Peer review | Peer Review Dr Brian Cox [PDF 267 KB] |
| Response | Response to Dr Brian Cox Peer Review [PDF 176 KB] |
Next steps
EECA will continue to monitor the evidence on indoor air pollution from combustion appliances.
Insights from this report will contribute to shaping EECA programmes, including future public information efforts.