benchmarks for managing indoor air quality for formaldehyde

BENCHMARKS FOR MANAGING INDOOR AIR QUALITY

FORMALDEHYDE

2 May 2019

Benchmarks for managing indoor air quality – formaldehyde

French High Council for Public

of 71

The Environmental Health Expert Committee (CSRE) held its meeting on 2 May 2019 and approved the opinion: 12 participants, 0 conflict of interest, votes for: 12, abstentions: 0, against: 0.

Report produced by the HCSP's Environmental Health Expert Committee (CSRE)

2 May 2019

Haut Conseil de la santé publique

14 avenue Duquesne

75350 Paris 07 SP FRANCE

www.hcsp.fr

http://www.hcsp.fr/



of 71

Contents

CONTENTS.................................................................................................................................... 3

1. INTRODUCTION AND BACKGROUND TO THE ISSUE ...................................................................11

2. HEALTH IMPACTS.....................................................................................................................13

2.1 TOXICOKINETICS ................................................................................................................................................13 2.2 ACUTE TOXICITY ................................................................................................................................................14 2.3 CHRONIC TOXICITY ............................................................................................................................................15 2.4 GENOTOXICITY...................................................................................................................................................16 2.5 EFFECTS ON REPRODUCTION ............................................................................................................................16 2.6 CARCINOGENIC EFFECTS ....................................................................................................................................16 2.7 SENSITIVE OR VULNERABLE POPULATIONS .......................................................................................................18

3. OVERVIEW OF FORMALDEHYDE AND ITS INDOOR AIR SOURCES ...............................................20

3.1 PRODUCTION AND USES ...................................................................................................................................20 3.2 INDOOR AIR SOURCES .......................................................................................................................................21 3.3 OUTDOOR AIR ...................................................................................................................................................25

4. DISTRIBUTION OF CONCENTRATIONS MEASURED ACROSS DIFFERENT ENVIRONMENTS ............26

4.1 PASSIVE SAMPLING ...........................................................................................................................................26 4.2 ACTIVE SAMPLING .............................................................................................................................................30 4.3 MEASUREMENTS WITH CONTINUOUS ANALYSERS ..........................................................................................33 4.4 COMPARISON OF THE DIFFERENT SAMPLING METHODS ..................................................................................42

5. EXISTING INDOOR AIR QUALITY GUIDELINES AND BENCHMARKS ..............................................47

5.1 AT NATIONAL LEVEL ..........................................................................................................................................47 5.2 VALUES SET BY INTERNATIONAL ORGANISATIONS OR APPLICABLE IN OTHER COUNTRIES ..................................

6. REGULATORY PROVISIONS .......................................................................................................52

7. PROPOSALS OF BENCHMARKS FOR FORMALDEHYDE ................................................................54

7.1 BENCHMARKS AND MANAGEMENT VALUES FOR INDOOR AIR QUALITY ..........................................................54 7.2 RAPID ACTION VALUE (RAV) ..................................................................................................... 55 7.3 PROPOSAL OF A SAMPLING AND MEASURING STRATEGY ................................................................................56 7.4 OVERVIEW OF VALUES (BENCHMARK AND PROVISIONAL MANAGEMENT VALUE) ................................. 58 7.5 CASE OF NEW BUILDINGS .................................................................................................................................58

8. OCCUPATIONAL EXPOSURE LIMITS (OELS) ................................................................................59

BIBLIOGRAPHY ............................................................................................................................61

APPENDICES ................................................................................................................................66



of 71

List of figures FIGURE 1: CONTRIBUTION OF SOURCES TO EXPOSURE IN JUNIOR SCHOOLS (INCITAIR PROJECT) – LA ROCHELLE, 2016 24

FIGURE 2: VARIATION IN THE CONCENTRATION OF FORMALDEHYDE OVER THE SAMPLING WEEK IN CLASSROOM NO.9 - LCSQA 200933

FIGURE 3: VARIATION IN THE CONCENTRATION OF FORMALDEHYDE MEASURED WITH THE INTERSCAN MONITOR (LCSQA, 2009) ............................................................................................................................................................... 35 FIGURE 4: LAVOISIER SCHOOL - CHANGES IN FORMALDEHYDE CONCENTRATION LEVELS (IN’ AIR SOLUTIONS) - LA

ROCHELLE 2016 ................................................................................................................................ 36 FIGURE 5: LAVOISIER SCHOOL - CHANGES IN FORMALDEHYDE CONCENTRATION LEVELS OVER ONE DAY - LA ROCHELLE

2016 ............................................................................................................................................... 37 FIGURE 6: GRANDES VARENNES SCHOOL - CHANGES IN FORMALDEHYDE CONCENTRATION LEVELS (IN’ AIR SOLUTIONS) -

LA ROCHELLE 2016 ............................................................................................................................ 37 FIGURE 7: FORMALDEHYDE CONCENTRATION LEVELS MEASURED USING NEMO IN ROOM 15 OF SCHOOL 1 – ETHERA

(2015). ........................................................................................................................................... 38 FIGURE 8: FORMALDEHYDE CONCENTRATION LEVELS MEASURED USING NEMO – OLD SCHOOL BUILDING NO.2 - GRENOBLE (2018)39 F IGURE 9: FORMALDEHYDE CONCENTRATION LEVELS MEASURED USING NEMO – OLD SCHOOL BUILDING NO.1 - GRENOBLE (2018).

....................................................................................................................................................... 40 FIGURE 10: FORMALDEHYDE CONCENTRATION LEVELS MEASURED USING NEMO – RECENT SCHOOL- GRENOBLE (2018). 40 F IGURE 11: MONITORING OF INDOOR AND OUTDOOR FORMALDEHYDE LEVELS OVER THE FIRST HALF OF 2015 IN OFFICES– PRIMEQUAL 2016 41 FIGURE 12: COMPARISON BETWEEN ACTIVE SAMPLING (4 x 2-HOUR MEASUREMENTS FOR 3 DAYS) AND PASSIVE SAMPLING (4.5 DAYS) IN CLASSROOM 9 - LCSQA 2009 43 FIGURE 13: COMPARISON BETWEEN THE ACTIVE SAMPLING IN CLASSROOM 9 (2-HOUR TIME INTERVALS) AND PASSIVE SAMPLING (4.5

DAYS ACROSS ALL CLASSROOMS - LCSQA 2009 ...................................................................................... 43

List of tables TABLE I: PHYSICAL PROPERTIES OF FORMALDEHYDE .......................................................................................................... 20 TABLE II: CONCENTRATIONS OF FORMALDEHYDE MEASURED IN THE INDOOR AIR OF HOUSING DURING THE BURNING OF

CANDLES AND INCENSE (EBENE, 2017) .................................................................................................... 24 TABLE III: DISTRIBUTION OF FORMALDEHYDE CONCENTRATIONS IN HOUSING IN MAINLAND FRANCE (OQAI CAMPAIGN)

AND THE PARISIAN REGION (UNIVERSITY OF PARIS-DESCARTES DATA) ........................................................... 27 TABLE IV: FORMALDEHYDE CONCENTRATIONS MEASURED IN INDOOR AIR IN FRANCE IN PRESCHOOLS AND SCHOOLS .. 28 TABLE V: FORMALDEHYDE CONCENTRATIONS IN THE OFFICAIR PROJECT OFFICES – (2010 – 2014) ............................ 30 TABLE VI: CONCENTRATIONS OF FORMALDEHYDE MEASURED IN THE INDOOR AIR OF HOUSING IN FRANCE OVER SHORT

TIME INTERVALS ............................................................................................................................................................ 30 TABLE VII: RESULTS OF THE MEASUREMENTS CONDUCTED IN CLASSROOMS AND COMPARISON WITH THE

MEASUREMENTS TAKEN BY DIFFUSIVE SAMPLERS – ETHERA, 2015 ............................................................... 44



of 71

List of acronyms

AASQA Accredited air quality monitoring association

ACGIH American Conference of Governmental Industrial Hygienists

ADOQ Household activities and indoor air quality project

AFSSET French Agency for Environmental and Occupational Health and Safety

ALARA As low as reasonably achievable

ANSES French Agency for Food, Environmental and Occupational Health & Safety

ATSDR Agency for Toxic Substances and Disease Registry

CEA French Alternative Energies and Atomic Energy Commission

IARC International Agency for Research on Cancer

CNRS French National Centre for Scientific Research

COMEAP Committee on the Medical Effects of Air Pollutants

VOC Volatile Organic Compound

CSRE

CSTB

Environmental Health Expert Committee

French Scientific and Technical Center for Building

CVCA Heating, Ventilation and Air Conditioning

DGPR General Directorate for Risk Prevention

DGS General Directorate for Health

DNPH dinitrophenylhydrazine

DPCs DNA-Protein Crosslinks

ERP Establishment open to the public

HCSP French High Council for Public Health

HPLC-UV High-Performance Liquid Chromatography with UV Detection

INDEX Critical appraisal of the setting and implementation on indoor exposure limits in European Union

Ineris French National Institute for Industrial Environment and Risks

INRS French National Research and Safety Institute for the Prevention of Occupational Accidents and Diseases



of 71

IRCE LYON Lyon Research Institute on Catalysis and the Environment

LCPP Police Headquarters Central Laboratory

LCSQA Central Laboratory for Air Quality Monitoring

LHVP City of Paris Hygiene Laboratory

MAK Maximale arbeitsplatz-konzentration (maximum workplace concentration)

MEEDDM Ministry of Ecology, Energy, Sustainable Development and the Sea

MHLW Ministry of Health, Labour and Welfare

NIPH Norwegian Institute of Public Health

OEHHA Office of Environmental Health Hazard Assessment

OQAI Indoor Air Quality Observatory

PNSE National Health & Environment Plan

ppm Parts per million

ppb Parts per billion

IAQ Indoor Air Quality

RIVM

SCOEL

National institute for public health and the environment (Netherlands)

Scientific Committee On Occupational Exposure Limits

TLV-STEL Threshold limit value – Short term exposure limit

US-EPA United States – Environmental Protection Agency

RAV Rapid action value

IAQG Indoor Air Quality Guideline

STEL Short-Term Exposure Limit

OEL Occupational Exposure Limit

aOEL

IAQB

Average Occupational Exposure Limit

Indoor Air Quality Benchmark

TRV Toxicity Reference Value



of 71

Executive summary Formaldehyde (FOR) is a gas which causes irritation of the airways and ocular mucous membranes. In 2004, the International Agency for Research on Cancer classified it as carcinogenic to humans (Group 1), based on an excessively high number of nasopharyngeal cancer cases observed during occupational exposure. Formaldehyde, which is genotoxic, is characterised by high reactivity with the biological tissues at the contact site, and this explains why it does not enter the bloodstream in any significant amounts.

It is a substance commonly found in enclosed spaces. The main sources of emission are building, decoration and furnishing products (especially particleboards), household products (cleaning products, paint, varnish, glue and cosmetics for example) and combustion activity in all its forms: cookers, boilers, ornamental fireplaces and smoking or burning incense. Outdoor air generally contributes very little to indoor concentrations.

Human exposure mainly occurs by inhalation. The contribution of indoor air predominates in the population's overall exposure owing to the time spent indoors and the low outdoor values typically measured. All of the data available in France shows that, irrespective of the type of building (housing, offices or public venues), the average concentration levels in enclosed spaces are generally in the range of 20 μg/m³. The kinetics of FOR concentrations nevertheless reveal considerable variability in indoor air levels, which can result in exposure exceeding 100 μg/m³ over more than one hour.

For the purposes of monitoring indoor air quality in establishments open to the public, the regulations set certain indoor air quality guidelines (IAQGs). Applicable since 1 January 2015, the IAQG for formaldehyde is set for long-term exposure (> 1 year) at 30 µg/m3 and should be brought down to 10 µg/m3 from 1 January 2023.

Based on the knowledge available at the time its last report on FOR was written (February 2018) and international practices for establishing toxicity reference values (TRVs), the French Agency for Food, Environmental and Occupational Health & Safety (ANSES) now proposes a single IAQG for short-term exposure (assessed over a period of 1 to 4 hours) of 100 µg/m3 to protect the general population from both acute and chronic effects of FOR exposure. This is because the risk of cancer is considered to result from repeated inflammatory processes triggered, over the long term, by bouts of possibly short-term exposure. Consistent with the guideline proposed by the World Health Organisation (WHO) in 2010, this IAQG must be complied with at all times. Ideally, continuous assessment of FOR concentrations should therefore be possible to ensure that this IAQG is never exceeded.

Grounded strictly in health criteria, this guideline does not provide information on "action guidelines", i.e. concentration levels above which action



of 71

is required to protect people's health. This is why the General Directorate for Health (DGS) and General Directorate for Risk Prevention (DGPR) have asked the French High Council for Public Health (HCSP) to establish "benchmarks" for setting the maximum levels in new or renovated buildings and, where necessary, for taking corrective action in establishments open to the public, with such action and its implementation time-limit being adjusted in light of the concentration levels measured.

In the context of this referral, the HCSP began by updating the framework document written in 2009, setting out the common principles guiding establishment of these benchmarks, called "benchmarks for managing indoor air quality". This document corresponds to application of this methodological framework to formaldehyde.

After factoring in the critical effect selected by ANSES for setting its IAQG (eye irritation), considering the population's average exposure levels in the various enclosed spaces and the reported kinetics of FOR in the reports available, and considering the regulatory provisions currently governing FOR and the measurement methods currently available, the HCSP recommends adopting two benchmarks for management where formaldehyde is concerned:

• An indoor air quality benchmark (IAQB) set at 100 μg/m³ (measurements carried out over successive 1 to 4-hour periods, throughout the day, when the building is occupied).

The HCSP recommends that this IAQB of 100 μg/m³ become immediately applicable and complied with in all buildings, with a maximum time-limit for taking corrective action of 6 months.

This 6-month time-limit takes account of the normal requirements associated with performing certain types of work: conducting assessments and surveys, time-limits associated with public procurement, requests for attribution of specific budgets, schedule to be coordinated with the activities of certain establishments and so on. That said, all reasonably feasible measures for reducing exposure of the population (not least the most sensitive groups, such as children or people suffering from respiratory diseases) should be rolled out as swiftly as possible. Examples: enhanced manual airing, installation of additional mechanical ventilation systems, identification of the main emission sources, reorganisation of certain activities, redistribution of people in the building based on the rooms available, etc.

To ensure compliance with the IAQB, measurements must be taken by active sampling at 1- to 4-hourly intervals.

Accordingly, in the specific case of formaldehyde, and owing to its mode of action, the IAQB and IAQG are similar and measured over short time intervals.



of 71

• A provisional management value set at 30 μg/m³ (measurements over a 4.5- to 7-day period).

In order to limit the number, and therefore the cost, of measurements to be taken, given the measuring instruments currently standardised and considering the monitoring studies and different documents analysed, the HCSP recommends setting a provisional management value of 30 μg/m³. In light of the data available, not exceeding the 30 μg/m³ value on average over one week (in practice: from 4.5 to 7 days) would ensure continuous compliance with the IAQB of 100 µg/m3.

This provisional management value will no longer be used once new continuous measuring instruments have been standardised and become available at a reasonable price.

During measurement by diffusive sampling over one week (4.5 to 7 days), where 100 μg/m³ is exceeded – since this necessarily corresponds to at least one instance of the IAQB being exceeded – corrective action will have to be taken within a set time-limit of 6 months.

During measurement by diffusive sampling over one week (4.5 to 7 days), where the provisional management value (30 μg/m³) is exceeded – even though the 100 μg/m³ guideline is not –,compliance with the IAQB (measured over a 1- to 4-hour time interval) is not guaranteed; sources shall have to be identified and an appropriate action plan drawn up. A new measuring campaign will then be carried out to check compliance with the IAQB or, failing that, the provisional management value. In all cases, the time-limit between the initial measurement and new measurement must not be longer than one year.

Regarding new buildings delivered and fitted out from 2020, the concentrations measured must be as low as possible and, in all cases, less than the IAQB or, failing that, the provisional management value. The same applies for buildings where extensive renovations are being carried out. To that end, architects and construction managers shall endeavour particularly to act on indoor FOR emission sources, interior design and construction materials.

• Measuring method

Compliance with the IAQB must be monitored via successive sampling every 1 to 4 hours when the building is occupied.

Compliance with the provisional management value may be monitored via diffusive sampling over 4.5 to 7 days using a standardised method.

In recent years, several direct-read, continuous measuring instruments have been developed, and marketed in some cases, in a bid to provide information on changes in indoor air concentrations. Using such instruments is particularly worthwhile for dynamic measurements that not only enable comparison with the benchmarks, but also identification of episodes of higher exposure over a given period and, as such, support in the search for sources in particular.



of 71

The technical characteristics and performances of these instruments have not been adequately documented or validated to date, however, to be retained in the context of regulatory monitoring.



of 71

1. Introduction and background to the issue In temperate regions, populations spend more than 80% of their time indoors, hence the importance of maintaining good indoor air quality. In these indoor spaces, individuals are exposed to multiple pollutants emitted by the occupants' activities, the building itself, its fixtures or decoration (wall or floor coverings, furniture and so on). Some pollutants also come from the immediate outdoor environment.

The Act bearing on national commitment for the environment has made indoor air quality monitoring a requirement in some establishments open to sensitive groups of people (Articles L. 221-8 and R. 221-30 et seq. of the French Environment Code). The establishments in question are those that are open to the public at large or to vulnerable groups – children in particular.

After the first deadlines set on 1 January 2015, the question of indoor air quality (IAQ) has been subject to regulatory amendments. The recent texts required the authorities to conduct IAQ monitoring by 1 January 2018 in centres receiving children under 6 years old as well as in primary schools. This monitoring must include assessment of the ventilation and airing systems in each establishment, and then either a self-survey followed by an action plan, or the roll-out of indoor air pollutant measuring campaigns which may lead to proposals of corrective action where applicable. Measurements of pollutants should particularly be compared against the indoor air quality guidelines and values activating additional investigations. Whilst these values are statutorily binding in establishments open to vulnerable sections of the public, they may also be used to assess indoor air quality in housing.

The French Agency for Food, Environmental and Occupational Health & Safety (ANSES) has been working on establishing indoor air quality guidelines (IAQGs) since 2004. IAQGs have been proposed for a certain number of substances including formaldehyde. An initial report on formaldehyde was published in 2007. In light of new knowledge gained, not least in terms of toxicity, ANSES updated its report in 2018.

The IAQG proposed – henceforth solely for short-term exposure – is a value below which it is reasonable to consider, based on current knowledge, that no one in the general population risks suffering from the adverse effects of the pollutant in question. Although it represents a target value, the IAQG has not been established to indicate a level above which health protection action must be taken.

In a bid to more clearly inform establishment operators or managers about the concentration levels above which such action must be taken, the Ministries for



of 71

Health and the Environment have defined regulatory action values and guidelines for four substances, formaldehyde among them.

The General Directorate for Health (DGS) and General Directorate for Risk Prevention (DGPR) have therefore asked the French High Council for Public Health (HCSP) to update the benchmarks for managing indoor air quality, where formaldehyde is concerned, following ANSES' update of the IAQG. These benchmarks are necessary, not only to determine the maximum levels inside new or renovated buildings, but also to take corrective action in existing buildings, with such action and its implementation time-limit being adjusted in light of the concentration levels measured.

The HCSP's management benchmarks (HSCP, 2019) take the following factors into account:

The type of health risks posed by the pollutant and the critical effect selected for establishing the IAQG;

The levels observed in enclosed spaces,

The levels measured on average in outdoor air,

The types of emission source and methods of reducing or eliminating them.



of 71

2. Health impacts All of the information and toxicological data comes from diverse monographs published by organisations renowned for the scientific excellence of their documents (IARC, WHO, US EPA, ATSDR, ANSES and INRS).

The authors' bibliographic references are indicated to enable direct access to the scientific information, but have not been subject to a new critical review in the context of this report. The effects of formaldehyde on humans have been sufficiently well documented to characterise its dangers. In the paragraphs that follow, only the effects associated with inhalation are described.

2.1 Toxicokinetics

Formaldehyde is an endogenous metabolic intermediate in all cells. It is a product of normal amino acid metabolism. Its physiological blood concentration is around 100 μmol/L (BfR, 2006).

More than 90% of the dose inhaled is retained in the nasal mucosa in rats. In monkeys, it is primarily absorbed via the upper airways but also the trachea and main bronchi. In rats, radioactivity following inhalation of marked formaldehyde (15 ppm/20 µg/m3, 6 hours) is mainly distributed in the œsophagus, kidneys, liver, intestines and lungs. In actual fact, it is not the formaldehyde itself which is distributed, but its metabolites or the products of its reaction with diverse nucleophilic substances. Straight from the respiratory mucosa, it is swiftly oxidised into carbon dioxide and formate by various, widely distributed enzymatic systems particularly requiring the presence of glutathione. The formate is subsequently incorporated into metabolic biosyntheses (INRS, 2011).

Irrespective of the exposure channel, absorption appears to be limited to the layers of cells immediately adjacent to the contact site. This limits systemic leakage and particularly explains why concentrations in the bloodstream vary so little in cases of formaldehyde exposure (ATSDR, 1999).

Following intravenous injection in rats, the plasma half-life of unaltered aldehyde is very short (around 90 seconds). Following inhalation of low or moderate doses, an insignificant amount of formaldehyde is therefore expected at systemic level (Franks, 2005) and no increase in normal blood concentrations has been shown in rats (15 ppm/20 µg/m3 over a 2-hour period), monkeys (6 ppm/7.5 µg/m3, 6 hours/day, 5 days/week for 4 weeks) or humans (1.9 ppm/2.3 µg/m3 for 40 minutes). Nevertheless, caustic lesions at the contact site can facilitate systemic leakage (INRS, 2011).



of 71

Accordingly, formaldehyde is characterised by high reactivity with biological macromolecules. As such, when it is inhaled, the majority of formaldehyde is retained in the nose, the oral mucosa, trachea and proximal bronchi, which are the primary contact sites in humans, together with the eye mucous membranes. This high reactivity at the contact site explains its low systemic leakage, half-life in the blood of seldom more than 90 seconds and the fact that its effects are primarily limited to the respiratory system following inhalation in humans.

2.2 Acute toxicity

Formaldehyde is toxic via inhalation, ingestion and skin contact, with symptoms primarily associated with its irritating properties: it can cause moderate skin irritation but severe eye irritation. Vapours cause irritation of the airways and eye mucous membranes. It is also sensitising for the skin.

Airborne formaldehyde is highly irritating for the eyes, nose and throat at low concentrations in the range of 0.25 to 2 mg/m3 (0.2 to 1.6 ppm). Eye irritation can occur before any odour has been detected.

Several controlled exposure studies involving inhalation have demonstrated the irritating properties of formaldehyde. Symptoms considered are irritation of the eyes with or without tearing, as well as irritation of the nose/throat and a dry mouth. In most studies, these symptoms appear from 0.25 to 0.375 mg/m3 (0.2 to 0.3 ppm); in most cases, patients' discomfort increases at the same time as the exposure concentration, up to 2.5 or 3.75 mg/m3 (2 or 3 ppm) (Andersen and Molhave, 1983; Bender et al., 1983; Kulle, 1993; Day, 1984).

Regarding the effects formaldehyde can have on the respiratory function, the findings are much less clear-cut and sometimes contradictory. A dozen or so studies describe an absence of effects in healthy, sensitised or asthmatic individuals exposed to concentrations of between 0 and 3 ppm (3.75 mg/m3) over intervals ranging from 10 minutes to 4 hours, some whilst doing physical exercise and others not. Only a handful of studies, performed in occupational settings, show a decrease in the FEV1 (Forced Expiratory Volume in 1 Second) for short exposure periods (20 and 30 minutes) at concentrations of 2.4 mg/m3 (1.9 ppm) and 6.4 mg/m3 (5.2 ppm) (INERIS, 2010).

Eye irritation, whether sensory or of the tissues, is an early effect compared with nasal and respiratory irritation. Research on humans has found that eye irritation is the most sensitive effect caused by exposure to formaldehyde. It is observed at concentrations beneath those associated with nasal and respiratory irritation (Paustenbach et al., 1997; AFSSET, 2008; Doty et al., 2004; WHO, 2010; ANSES, 2017). According to ANSES’ report, "Indoor air quality guidelines for formaldehyde", published in 2018, it appears relevant selecting eye irritation as the critical effect.

Two new controlled exposure studies have determined a dose-response relationship associating formaldehyde exposure and the onset of acute effects in humans: Lang et al. (2008) and Mueller et al. (2013).



of 71

ANSES has therefore selected the study by Lang et al. (2008) as a key study for proposing acute reference values for the working and general population.

In the study by Lang et al. (2008), the volunteers were exposed to 10 different concentrations of formaldehyde, continuously over 4 hours, with or without 15-minute peaks in exposure. The concentrations ranged from 185 to 615 μg/m3, corresponding to the lowest tested concentrations among the controlled exposure studies available.

The study by Mueller et al. (2013) rounds off the findings obtained by Lang et al. (2008). This was a larger experimental human study (41 participants), with exposure over a 1-week timeframe, but all of the volunteers were male. Moreover, the division of volunteers into two separate "hypersensitive" and "hyposensitive" groups in terms of sensory nasal irritation is not considered relevant. Indeed, the test of sensitivity to CO2 used for this division is considered to be a pain test, which is not appropriate for identifying individuals who are sensitive to the effects of formaldehyde. Finally, the study was conducted over a 1-week timeframe instead of 2 consecutive weeks as in the study by Lang et al. (2008).

2.3 Chronic toxicity

The irritating effects associated with chronic exposure to formaldehyde are similar to the effects observed during acute exposure.

In humans, irritations of the eyes, throat and airways, tiredness and headaches have been reported in the workplace and the general population, in several studies conducted particularly among mobile home residents (with no indication of how long they have lived in them). These symptoms appear from 120 μg/m3 in the general population (IPCS, 2002; Ritchie et al., 1987).

Effects on breathing, sensitisation phenomena and asthma – which can require hospitalisation – have been observed in epidemiological studies in children aged six months to three years, during home exposure to formaldehyde at concentrations even as low as 60 μg/m3 (Health Canada, 2005). However, it is difficult to conclude on the exclusive role of formaldehyde in these phenomena, owing to the potential confounding by co-pollutants in indoor air (Paustenbach, 1997; IPCS, 2002; AFSSET, 2008; WHO, 2010; Golden, 2011).

Among the recent epidemiological studies bearing specifically on indoor air pollution, four studies have identified a statistically significant relationship between the onset of respiratory symptoms and exposure to the highest concentrations of formaldehyde; five studies have not established such a relationship, however (ANSES, 2018).

Whilst some studies demonstrated sensitisation phenomena and asthma, they presented a certain number of confounding factors, which made interpreting the findings difficult. What is more, these relationships could not be ascertained in several other studies (ANSES, 2018).



of 71

2.4 Genotoxicity

Various in vitro genotoxicity studies have demonstrated positive results in bacteria and mammalian cells. These tests have also revealed formaldehyde's ability to induce primary DNA lesions and DNA-protein crosslinks, whose incomplete repair can lead to mutations (Barker et al., 2005) or clastogenic effects (ANSES, 2011).

All of the in vitro and in vivo studies show that formaldehyde seems to be a direct genotoxic compound whose effects are primarily observed at the contact site for high concentration levels (AFSSET, 2008).

Research on micronuclei, chromosomal aberrations or sister chromatid exchange on mouth or nose cells or in lymphocytes has sometimes revealed positive results. That said, the effects observed were not dose-dependent and varied with exposure (INRS, 2011).

The results of the micronucleus tests on circulating lymphocytes, from various studies on workers exposed to formaldehyde, point to a correlation between the level and duration of exposure to formaldehyde and the presence of genetic instability in circulating lymphocytes when these are cultured ex-vivo. It is not possible, however, to distinguish whether the micronuclei observed in this test were caused by the effect formaldehyde has on the lymphocytes circulating in the bloodstream (which would tend to be a marker of exposure to formaldehyde), or by an effect on the lymphoid progenitor cells located in the bone marrow (which, as they undergo increasing mutation, may generate circulating lymphocytes with greater genetic instability). It therefore strikes as difficult to conclude with certainty on a potential systemic genotoxic effect of formaldehyde, since the weight of evidence is considered to be average or low (ANSES, 2018).

2.5 Effects on reproduction

As mentioned in ANSES' 2018 report, Duong et al. (2011) conducted a systematic review of the data on the reproductive and developmental effects of formaldehyde as well as a meta-analysis. The results of this meta-analysis (which were consistent with those of the meta-analysis by Collins et al., 2001) showed that maternal exposure to formaldehyde is associated with a risk of spontaneous abortion.

The authors themselves specified that confounding factors (co-exposure with other compounds that can induce effects on reproduction in the studies, and non-adjusted relative risks/RRs) and recall biases may have caused these RRs to be overestimated, but claimed that they were unable to assess them (Duong et al., 2011).

2.6 Carcinogenic effects

In its 2004 monograph, the IARC concluded that formaldehyde is carcinogenic to humans (classification in Group 1). In 2014, under the European regulations,



of 71

formaldehyde was classified as Category 1B carcinogenic, “presumed” to have carcinogenic potential for humans (ATP 06 -Regulation (EC) No 1272/2008).

Among the datasets available, the two meta-analyses by Blair et al. (1990) and Partanen (1993), which reanalysed the epidemiological evidence (from case-control studies for the most part) according to slightly different techniques, reach the same conclusions. Both consider that a highly probable causal role can be attributed to formaldehyde for nasopharyngeal cancer, and to a lesser extent sinonasal cancer, on account of an exposure-response gradient and the direct action of formaldehyde on these sites (respiratory contact site). In this regard, the meta-analysis by Collins et al. (1997) reveals a relative risk that is significantly higher than 1 of developing nasopharyngeal cancer.

The IARC Working Group behind the monograph (2006) on formaldehyde considers it unlikely that confounding or bias can explain the many positive results for nasopharyngeal cancer recorded in various epidemiological studies. It concludes that the findings of studies conducted in the US on industrial workers, as well as the numerous other positive results of other studies, provide sufficient epidemiological evidence to demonstrate that formaldehyde can cause nasopharyngeal cancer in humans.

The genotoxicity of formaldehyde has only been observed at high concentrations in experimental conditions. The carcinogenic effect on the nasopharynx stems from the cytotoxicity and genotoxicity of formaldehyde. A review by Gaylor et al. (2004) of the results of the study by Monticello et al. (1996) confirmed that the occurrence of nasopharyngeal cancer is the result of two separate events showing a threshold dose-response relationship: a) the cytotoxicity of formaldehyde, responsible for regenerative cellular proliferation, and b) the combined genotoxic effects of formaldehyde including the formation of DPCs (DNA-Protein Crosslinks), which becomes irreversible above a high concentration of formaldehyde (BfR, 2006).

Studies measuring the DPC formation rate in animals conclude that there is a 2.5 mg.m3 threshold above which this rate increases significantly (ANSES, 2018).

The carcinogenic effects of formaldehyde on the nasopharynx are therefore observed in contexts of repeated exposure to high concentrations, first causing cytotoxicity manifested as local irritation.

Numerous studies undertaken in humans have assessed the association between leukaemia mortality and occupational exposure to formaldehyde. The results are ambiguous but tend to show an association between leukaemia and formaldehyde exposure at high concentrations only.

Animal studies provide no evidence of leukaemia occurring at the formaldehyde exposure levels associated with the occurrence of nasal cancers. Indeed, the incidence of leukaemia or lymphoma in animals increased only in the groups with the highest tested concentrations. Experimental studies conducted orally lead to the same conclusion.



of 71

In order to prevent the occurrence of nasopharyngeal cancer, the effect selected is therefore eye irritation (ANSES, 2018).

Despite uncertainties over mechanistic data and the lack of consolidated data in animals, and considering the results of epidemiological studies in humans, the association between formaldehyde exposure and the occurrence of leukaemia in humans cannot be ruled out. Even so, the causal relationship cannot be confirmed (due to confounding factors and uncertainties regarding the characterisation of exposure in particular). Moreover, the association is observed at higher concentrations than those associated with the onset of nasopharyngeal cancer, whose causal relationship with formaldehyde is certain according to WHO.

The carcinogenic effects on the nasopharynx are therefore considered the most sensitive critical effect of chronic airborne exposure to formaldehyde in humans (ANSES, 2018). It is the best described carcinogenic effect of formaldehyde, for which a causal relationship has been established based on wide-ranging human, animal and mechanistic data. The development of nasopharyngeal cancer is linked to repeated and prolonged changes in the nasal epithelium, and therefore to sufficiently high and prolonged exposure first causing irritation. The data on the mode of action enables a threshold dose-response relationship to be determined, with a series of key events leading to the formation of nasopharyngeal tumours, the first symptoms of which are eye and nose irritation.

2.7 Sensitive and vulnerable populations

Inter-individual variability has been reported for irritation effects, but this does not seem to be very high in adult non-smokers, smokers, asthmatics or hypersensitive adults.

Asthma sufferers: an experimental study conducted among 19 asthma sufferers suggests an enhancement effect of pre-exposure to formaldehyde on the immediate and delayed bronchial response during exposure to allergens. But these findings are not corroborated by another study conducted on 12 asthmatics also suffering from hay fever. Based on current knowledge, the irritating or sensitising effects of formaldehyde exacerbating allergies have not been demonstrated.

Children: one comparative study suggests respiratory effects (10% decrease in peak flow rates) in children exposed to formaldehyde concentrations of 37μg/m3 with no effects in adults. Furthermore, in studies investigating a relationship between the onset of respiratory effects in children and formaldehyde exposure at home or school, no conclusions could be drawn with certainty as to whether there was an association, due to exposure co-factors (animal allergens, mould, road traffic, socio-economic factors) (Paustenbach, 1997; IPCS, 2002; AFSSET, 2008; WHO, 2010; Golden, 2011).



of 71

Since it is considered an effect level for all acute and chronic irritating effects due to formaldehyde, eye irritation may be selected as the critical effect for protecting against all the effects of formaldehyde, including the occurrence of nasopharyngeal cancer.

Regarding eye irritation, several ophthalmologists contacted by ANSES as part of their expert assessment on formaldehyde reported inter-individual variability for eye irritation to chemicals, especially formaldehyde (ANSES, 2018). Ocular dryness is one of the aggravating factors and can be correlated with the existence of diseases (e.g. dry eye syndrome) or specific physiological states (e.g. menopause or contact lens users). The study by Wolkoff et al. (2016) lists a number of risk factors associated with ocular dryness including age.



of 71

3. Overview of formaldehyde and its indoor air sources Table I: Physical properties of formaldehyde

Formaldehyde CAS no. 50-00-0

Conversion factor 1 ppm = 1.23 mg/m3

Odour threshold Between 0.1 and 1 ppm

Molar mass 30.03 g/mol

Saturated vapour pressure from 314 to 519 kPa (10 and 25°C),

Solubility 106 mg/l (25°C)

Henry's law constant 3.4.10-5 kPa.m3/mol

octanol-carbon partition coefficient 1.07

(INRS, 2011)

3.1 Production and uses

Formaldehyde is not marketed in gaseous form, but is primarily available in liquid form, formalin:

Formalin contains 30 to 50% (often 37% in weight) of formaldehyde in an aqueous solution and, usually, alcohol to prevent polymerisation (generally 10 to 15% methanol).

Formaldehyde is also available in solid, or polymerised, form:

paraformaldehyde (polymer), as a powder or white crystals, contains 90% to 93% formaldehyde and up to 10% water;

trioxane (trimer), is a crystalline solid.

Formaldehyde is produced industrially by vapour-phase catalytic oxidation of methanol. Two main processes are implemented to manufacture formaldehyde, in a batch reactor and with recovery of waste gases:

Partial conversion, with a silver (silver or crystal silver) catalyst heated to a temperature of 600 to 720°C (DGE, 2006). The dehydrogenation reaction happens in the presence of air. Conversion of methanol into formaldehyde is partial, which involves a distillation stage. The waste gases (N2 and H2) are recovered in a boiler.



of 71

Total conversion, with a metal oxide catalyst (modified iron, vanadium, molybdenum) heated to a temperature of between 270 and 380°C. Conversion of methanol into formaldehyde is total. The gases are recycled towards reaction and vents are treated on a catalytic oxidiser (AFSSET, 2009).

Formaldehyde has many uses owing to its physical-chemical properties as a biocide, preservative or fixing agent. It is widely used, in solid or liquid form, across many industrial sectors, not least as a chemical reagent and for manufacturing resins intended for use in everyday consumer and building products (such as DIY products, furniture, cleaning products and cosmetics).

Many industries make use of it, including the veterinary, cosmetics, medical, timber and furnishing, agricultural, metalworking, cleaning (manufacture of detergents), water treatment, textile, leather, photography, adhesive, paint, varnish and coatings and paper sectors (AFSSET, 2009).

Formaldehyde is also a product of incomplete combustion phenomena (smoke from tobacco, candles or hearth fires, etc.).

3.2 Indoor air sources

In indoor air, formaldehyde is a ubiquitous substance. There are three main types of sources responsible for the levels observed (AFSSET, 2009):

building and furnishing products, more particularly particleboard, which is affixed using amino resins (urea-formaldehyde UF, melamine-formaldehyde MF and melamine-urea-formaldehyde MUF) and phenolic resins (phenol-formaldehyde PF and phenol-resorcinol-formaldehyde PRF); around 85% of the formaldehyde used in France goes into making these products;

household products (detergents, cosmetics, paint, varnish and glues for example) in which formaldehyde is often used as a preservative;

combustion sources in the home for cooking food and heating the premises, ornamental fireplaces and smoking or incense burning.

A certain number of studies have been conducted in recent years aimed at specifically identifying the different sources of chemical compounds in indoor air, formaldehyde in particular.

MOBAIR-C and MOBAIR-DE – 2011/2012

The MOBAIR-C project led by the French Institute of Technology for Forest-based and Furniture Sectors (FCBA) and Scientific and Technical Centre for Building (CSTB) between 2010 and 2011, focused on furniture in preschools



of 71

and schools (MOBAIR-C). Another project began at the start of 2012 on furniture equipping babies' and infants' bedrooms (MOBAIR-Domestique Enfants).

The two overarching aims of these projects were to:

deliver relevant knowledge on the contribution of furniture and components in indoor air, primarily furniture in infants' environments,

develop a simple decision support tool to help stakeholders involved in equipping preschool or infant school rooms or infants' bedrooms to choose solutions that emit the least amount of volatile pollutants.

In this study, the formaldehyde emissions of 21 items of furniture for preschools and infant schools and 38 parts making up this furniture were characterised. It was generally observed that furniture emissions remained relatively low, less than 16 μg/m²/h for formaldehyde (with the exception of 2 items of furniture, 51-55 μg/m²/h). The highest-emitting items of furniture where formaldehyde was concerned were the "major import timber chair", particularly because of its glued laminated timber legs, and a chair with a multi-ply seat varnished on one side. For the plastic furniture (bunks, chairs, soft play equipment), the formaldehyde emissions are lower than the emissions from timber furniture (Roux et al., 2011).

ADOQ study; CSTB-INERIS-IRCELYON 2013

The ADOQ project (which stands for DOmestic activities and indoor Air Quality), an applied research project, ran from 2009 to 2013, led by the CSTB, Lyon Research Institute on Catalysis and the Environment (IRCE LYON) and French National Institute for Industrial Environment and Risks (Ineris). ADOQ studied the impact of detergent emissions on indoor air quality, by characterising the pollutant emissions associated with household activity in indoor air and assessing the contribution of secondary compounds on indoor air quality. The method used entailed the combination of measurements in the actual atmosphere (MARIA experimental house) and simulated atmosphere (assessment of household products in a controlled emissions chamber). In this way, the emission factors of 54 products were characterised in the emissions chamber, and then 19 products were tested under realistic conditions, in the MARIA house, on the basis of pre-established scenarios.

Regarding aldehydes, formaldehyde is the compound most commonly emitted by the products tested. The surface emission factors measured range from 0 to 405 μg/m²/h the first half-hour, and from 0 to 615 μg/m²/h the second half-hour. Formaldehyde is emitted by 49 of the 54 products tested and the highest emission factors concern dust collectors and bleaches (Nicolas et al., 2013).



of 71

The formaldehyde concentrations measured under real-world conditions indicate values of between 5 and 41 μg/m3 in the summer and between 12 and 35 μg/m3 in the winter. A significant increase in concentration levels after products had been used was observed for nearly two-thirds of the products tested.

The correlation observed between ozone and formaldehyde highlights the partially secondary character of formaldehyde (Nicolas et al., 2013). The secondary emissions of formaldehyde primarily come from the ozonolysis of terpenes emitted by household products.

INCITAIR study – La Rochelle 2016

The INCITAIR project involved developing a method and tools for taking the air quality criterion (formaldehyde) into account in public procurement contracts where schools are concerned (Cormerais et al, 2017).

Formaldehyde sources need to be ranked via modelling to identify contracts that should take priority. The stages in this identification are as follows:

Listing of potential sources of formaldehyde in schools (permanent sources: materials, furniture; intermittent sources: household products, supplies)

Determination of emission rates by these sources,

Dynamic simulations of the layout of representative classrooms: fitting-out, occupation, activity. In all, 256 simulations enabled weekly profiles of formaldehyde concentrations to be obtained (time variation = 1 min),

Identification of families of sources that contribute the most to occupants' average exposure (period of occupation) and peak exposure (2 hours max.).

With "winter ventilation", emission sources (which contribute the most to average exposure to formaldehyde in schools in La Rochelle) are building materials and furniture (works, supplies and materials, school furniture). Occasional sources such as housework seem to have little impact on average exposure to formaldehyde.



of 71

Figure 1: Contribution of sources to exposure in junior schools (Incitair) – La Rochelle, 2016

EBENE project - 2017; CSTB-INERIS-LCE

In the study by Nicolas M. et al. (2017), on "Exposure to pollutants emitted by candles and incense in indoor environments (EBENE)", the emissions (and associated health risks) from nine incense burners, nine scented candles and one fragrance lamp were assessed. The analytical techniques used for the project combine both indirect methods, requiring sampling then deferred analysis (according to the standardised methods NF ISO 16000-6 and NF ISO 16000-3) and direct methods enabling continuous analysis of the pollutants studied (VOCs and formaldehydes by PTR-ToF-MS). The emission testing was performed in a room of the CSTB's MARIA house, with no furniture and minimum decoration. Formaldehyde is one of the main pollutants emitted by the products tested.

Table II: Concentrations of formaldehyde measured in the indoor air of housing during the burning of candles and incense (EBENE, 2017)

Products

tested

Pollutant

Concentrations measuring during

burning

Concentrations measured 1 hour after

burning

Incense

Formaldehyde Between 10 and 47 µg/m3

Between 8.8 and 34 µg/m3

Candles Between the

detection limit and 5 µg/m3

Between 0.8 and 11.8

µg/m3



of 71

3.3 Outdoor air

Outdoor air can contribute to exposure in certain environments near road traffic or in periods of high atmospheric photochemistry, following reactions involving myriad organic compounds.

According to INERIS' AASQA measurement data inventory, concentrations range between 1.6 and 4.3 µg/m3 in urban environments and between 1.3 and 3.7 µg/m3 in industrial environments (INERIS, 2009).



of 71

4. Distribution of concentrations measured across different environments

Wherever possible, indoor air concentration data corresponds to recent French data. It is presented according to the sampling method used:

Passive sampling is used to estimate exposure levels over a long time interval (4.5 days for schools; 7 days for housing); This sampling method is used to date to comply with the applicable regulations;

Active sampling is used to estimate exposure levels over short time intervals (between 1 and 4 hours);

The sampling method using continuous analysers enables continuous measurement of formaldehyde concentrations and thus observation of their kinetics over the period in question.

4.1 Passive sampling

4.1.1 General case of housing

The French Indoor Air Quality Observatory (OQAI) endeavours to improve knowledge of indoor pollution, its sources and dangers, so as to outline recommendations on indoor air quality. France's first nationwide campaign on air quality in housing focused on 554 homes in 2004- 2006 (OQAI, 2006). For formaldehyde, samples were taken over 7 days using 2,4-DNPH (2,4-diphényl hydrazine) coated Radiello type diffusive samplers. The findings from this measuring campaign indicate formaldehyde concentrations in the bedrooms of 1.3 to 86.3 μg/m3 (average over 7 days), with a median concentration of 19.6 μg/m3.

In a different context, the Public Health & Environment Laboratory of Paris-Descartes University and Directorate for Child Welfare and Health of Paris City Council monitored a cohort of more than 4,000 newborn babies in the Parisian region in 2003. The home environment of 200 of these babies was characterised (Dassonville et al, 2009). The concentrations measured, varying between 6.6 and 66.5 μg/m3, are similar to the levels measured by the OQAI.



of 71

Table III: Distribution of formaldehyde concentrations in housing in mainland France (OQAI campaign) and the Parisian region (OQAI, 2006)

Formaldehyde concentrations in

μg/m3

OQAI French homes (n=554), 2003-

2005

Newborn babies’ homes in the

Parisian region (n=206), 2003

Minimum 1.3 6.6

Median 19.6 19.4

75th percentile 28.3 27.0

Maximum 86.3 66.5

From 2012, as part of the OQAI's efforts dedicated to energy-efficient buildings, concentration data was published for 63 homes in buildings forming part of the Research and Experimental Platform on Energy in Buildings (PREBAT). Measurements were taken in the bedroom and living room over two different periods of the year (when the heating was on and when it was not being used). The median concentration is 17.2 µg/m3 (maximum at 38.4 µg/m3) (OQAI 2016), with a significant difference when the heating was on (p <0.001).

Papers published over the past decade (2009-2016) in the literature report median formaldehyde concentrations measured over 7 days in the indoor air of housing in France of between 7 and 29 µg/m3 (ANSES, 2018). The maximum values mentioned range between 61.3 µg/m3 (new individual low-energy houses) and 113 µg/m3 (PELAGIE cohort: 150 homes: 120 in rural areas and 30 in urban areas – between September 2012 and October 2013 in Brittany) (Dallongeville, 2015).

4.1.2 Schools and preschools

Formaldehyde is the most widely studied pollutant in France in schools and childcare centres. In general, the measurements are usually integrated over periods of around 4.5 days and carried out using radial diffusive samplers (Radiello®).



of 71

Table IV: Formaldehyde concentrations measured in indoor air in France in schools and preschools

Source study

Study site

Study description Sampling

details

Concentrations measured

[min – max]

OQAI pilot study

(OQAI, 2004)

Strasbourg, Aix-Marseille and Nord-Pas-de- Calais

9 schools between March and July 2001

from Monday to Friday

Passive sampling (Radiello®),

13 to 67 μg/m3

Aspa, 2005 Strasbourg and Childcare Passive sampling (Radiello®), measurements carried out over 48h

(24%) > 30 μg/m3 (n = 127), (5 %) > 50 μg/m3 (n = 27) > 100 μg/m3 (n = 3)

its centres (143), infant conurbation schools (157) and junior schools (222) between 2004 and 2005

ATMO Rhône Alpes, 2009

Rhône– Alpes Region

50 centres, preschools and infant schools 3 rooms/centre 4 weeks in 2006: June, October, December, March

Passive sampling (Radiello®), 2,4 DNPH coated Florisil cartridge

Median: 16.4 µg/m3

max: 35.9 µg/m3

Over 5 days

Roda et al.

(2011)

Paris (Parisian Region,

28 preschools Passive sampling (Radiello®),

Winter: 10.2 μg/m3 (median) Summer: 14.6 μg/m3 (median)

[4.8 – 40.1]

Birth Florisil Cartridge cohort coated in 2,4 PARIS) DNPH, over 5 days Acetonitrile desorption 13 regions N =431

15.9 μg/m3 (median) [3.7–75.1] Average difference between the "summer" and "winter" measurement: 4 μg/m3

- 160 schools and preschools

– 431 classrooms Michelot et al. (2011)

September 2009 to May 2010

Passive sampling

France (Radiello®) over pilot campaign 4.5 days of

regulatory monitoring

Analysis by HPLC-UV Measurements over 2 periods (summer and winter)

Ramalho et al. (2015)

310 schools and preschools – 896 classrooms

2009-2011

Infant schools n=297 19 μg/m3 (median)

[2–98]



of 71

Source study

Study site

Study description Sampling details

Concentrations measured

[min – max] Junior

schools n=769 17 μg/m3 (median) [1.6 - 70]

Annesi-Maesano et al. (2012)

Bordeaux, Clermont- Ferrand, Créteil,

108 schools – 401 classes (years 4 & 5 [ages 9-11]) March 1999 and October 2000

Passive sampling (Radiello®) over 5 days (Diffusion rate: 20.4 mL.min-1)

Analysis by HPLC

n=540 21.54 μg/m3 (average; SD 15.54)

[12-55]

Banerjee et al. (2012)

Marseille, Strasbourg and Reims

(ISAAC study - French

part) Verriele et 10 schools with low

energy consumption (junior, secondary and higher education) 2013

Passive al. (2014) sampling

(Radiello®)

(Nord and Alsace; MERMAID study)

coated in 2,4 DNPH over 4.5 days Acetonitrile

13 μg/m3 (median) [9 – 37]

extraction HPLC-UV analysis

Canha, 2016 OQAI France

Campaign

600 classes in France studied between 2013 and 2016

Passive sampling (Radiello®) coated in 2,4 DNPH over 4.5 days

19.2 μg/m3 (median)

66.2 μg/m3 (maximum)

Ineris, 2018 Report on indoor air quality monitoring

Passive sampling (Radiello®)

n = 8524 16.1 μg/m3

France

in establishments open to the public.

coated in 2,4 DNPH over 4.5 days

(median)

[0.3 – 161]

n = 19 %) > 100 µg/m3

4.1.3 Offices

After housing, offices are the second environment where most workers spend most of their time. Office spaces have not been as extensively studied as homes or childcare centres and schools.



of 71

OFFICAIR – 2010-2014

OFFICAIR is a European collaborative project, which has received funding from the European Union, as part of the 7th Framework Programme.

This project began on 1 November 2010 and ended on 31 October 2013. Eight European countries got involved, including France (Mandin, 2017). The French office blocks were located in Paris, Meylan and Strasbourg. Measurements were carried out during one week in the summer (2012), then one week in winter (2012-2013), from Monday to Friday, in 37 buildings.

Table V: Formaldehyde concentrations in the OFFICAIR project offices – (2010 – 2014)

Pollutant

Ventilation

Summer

Winter

Median, all buildings (n=37) in

µg/m3

Median, French

buildings (n=9)

in µg/m3

Median, all buildings (n=35) in

µg/m3

Median, French

buildings (n=9)

in µg/m3

Formaldehyde

80% of buildings, mechanical

5%, natural

100% of French buildings

14.0

16.0

7.5

6.4

4.2 Active sampling

4.2.1 General case of housing

Sampling was carried out over a short time interval (from 20 to 90 minutes) in housing in the city of Strasbourg and its suburb (Marchand, 2006, 2008). The measuring method used in these studies was the one that has been validated for comparison with ANSES' IAQGs established in 2018: active sampling pump with a 2,4-DNPH coated cartridge, solvent desorption and analysis by HPL/UV.

Table VI: Concentrations of formaldehyde measured in the indoor air of housing in France over short time intervals.

Source study

Town

Study description Data on the measuring technique

Concentration measured

[min – max]

Marchand et al. 2006

Strasbourg and its suburb

22 homes located in the countryside

4 categories defined

Active sampling over 20 to 90 minutes by pumping on

n = 16

35.7 - 46.1 µg/m3 on average



of 71

Source study

Town

Study description Data on the measuring technique

Concentration measured

[min – max] by ground surface area: cartridge [13.3 – 123.4]

During the smoking

experiment (5 cigarettes)

217 µg/m3 on average

[213 – 221.1]

6 x 20 to 50 m² homes; 6 coated in 2,4- x 50 to 80 m² homes; 3 x DNPH, solvent

80 to 120 m² homes and 7 with a surface area of more

than

desorption and

120 m² analysis by

September 2004 and January HPL/UV

2005

Marchand et al. 2008

Strasbourg

and its suburb

(Part of a case-control study on asthma sufferers)

112 homes

2 simultaneous samples in the bedroom and living room of each home

February 2004 to May 2004 and

October 2004 to May 2005

Active sampling

over 30 to 95 minutes

Tube with 2 x 2,4-DNPH

coated cartridges,

solvent desorption and

analysis by HPL/UV

n= 162 (2nd series of

measurements)

29.2 µg/m3 (SD ± 14.6)

(median) [7-83]

It is also interesting to note that the statistical analysis of data bearing on the socio-economic variables of households identifies the employment status as a predictor of formaldehyde concentrations (Brown, 2015). Accordingly, low-income households are more likely to be exposed to higher concentrations of formaldehyde. The study cited also found that formaldehyde concentrations indoors were also associated with new builds.

SPSE study

In the context of complaints reporting respiratory symptoms, samples were taken by the Parisian Environmental Health Service (SPSE) in housing, over 1-hour time intervals (generally speaking, one or two measuring points per home, depending on the occupants' complaints). Between 2012 and 2018, 64 samples were taken (unpublished data communicated by the SPSE). The concentrations measured ranged between 3.4 and 79.8 µg/m3. The concentration median was 27.3 µg/m3 and the 90th percentile was 58 µg/m3. The maximum concentration (79.8 µg/m3) was measured in a living room/lounge area.



of 71

Studies analysing the data produced in ADOQ – INERIS 2015, 2019

From concentrations measured under real-world conditions in the MARIA experimental house, exposure associated with domestic uses of the products tested were assessed.

Regarding the concentrations measured – even though the levels of formaldehyde were initially high in the rooms studied, owing to the presence of other sources such as building and furnishing products – a significant increase commonly appears after household products have been used. For example, levels exceeding 100 μg/m3 were measured when using a window cleaning wipe. Moreover, twice the usual levels can be noted in the case of two bathroom sprays, with values reaching 17 and 30 μg/m3. Under more controlled conditions (in winter), the impact of household activities is more limited, but does not completely go away. Overall, the levels noted after cleaning range between 10 and 35 μg/m3.

To recreate actual conditions of use as far as possible, scenarios involving "multiple products in one room" were tested. Measurements were carried out over the 2 hours following the end of the scenarios. Regarding formaldehyde, increases in levels are observed in some cases, with values ranging from 12 to 23 μg/m3 immediately following cleaning (Nicolas et al., 2013).

On the basis of the concentrations measured following use of the products tested, average daily concentrations (ADCs) were extrapolated according to different scenarios of use; they range between 0.06 and 8.6 µg/m3 for the average; the 90th percentile can be as high as 17 µg/m3 for a very high exposure scenario. Application of conventional good practices (airing after use, rinsing the cleaned surfaces) in a high-exposure scenario can bring the calculated risks back down to levels below those of concern. Furthermore, concerning uses and buildings which can be considered similar to average conditions, in light of the available data, the concentrations attributable to household products – calculated in the context of ADOQ and bearing on products assessed individually – account for a low proportion of the average concentrations measured in French housing (less than 1%). That said, there is little data available for assessing exposure: high uncertainty therefore surrounds the results presented above. One study by INERIS (currently being finalised) is set to update these initial assessments by proposing complete housework sessions.


Measurements over short time intervals, around 1 to 4 hours, are uncommon in schools and childcare centres.

Central Laboratory for Air Quality Monitoring - 2009

For the purposes of gaining a clearer idea of the uncertainty surrounding the formaldehyde sampling strategy in indoor air, not least in terms of spatial representativeness, a



of 71

measuring campaign was conducted in one school in June 2009 at the request of the Ministry of Ecology, Energy, Sustainable Development and the Sea (MEEDDM); four classrooms were selected.

Active sampling was carried out using pumps on Sep Pack® type cartridges containing as the adsorbent a derivatising agent coating the adsorption cartridge. Active sampling was carried out daily, every two hours or so, at a fixed rate of 100 mL.min-1, i.e. two samples in the morning from 8.30 to 11.00 and from 11.00 to 13.30 and two in the afternoon from 13.30 to 15.00 and from 15.00 to 16.30.

The concentrations measured by active sampling over the different days varied between 17 and 56 µg/m3. Even though the daily variations are not repeatable from one day to the next, the daily averages are similar overall, in the region of the concentration level that had been measured with the Interscan monitor (continuous analyser based on an electrochemical detection method) during the preliminary visit: 30 µg/m3.

Figure 2: Variation in the concentration of formaldehyde over the sampling week in classroom no. 9 - LCSQA 2009

4.3 Measurements with continuous analysers

For formaldehyde, in light of the technical and economic considerations, measurements integrated over several days were carried out (4.5 days for schools; 7 days for housing) (Kirchner, 2006). But unlike continuous measurements, it is not possible to identify the kinetics of formaldehyde concentrations or peaks in pollution and associated emitting activities from these integrated measurements.

Different continuous measuring devices are currently available on the market or are being developed, including:



of 71

Interscan: measurement is based on an electrochemical detection method. The formaldehyde molecules pass through a diffusive medium and react electrochemically on the surface of an electrode. This reaction generates a current whose intensity is directly proportional to the concentration of formaldehyde.

• AL4021 Aerolaser analyser: The principle of this device is to generate a reaction between the formaldehyde present in the air and a specific reagent, to form a derivative that can be analysed in liquid phase by fluorescence spectroscopy.

• In’Air Solutions analyser: air is continuously sampled at a constant rate of 20 mL./min-1. The method is based on the same three stages as the AL4021 Aerolaser analyser: trapping gaseous formaldehyde in an aqueous solution, chemical reaction between the formaldehyde and a selective derivative agent, and detection by fluorescence of the product. Response time is 10 minutes. The formaldehyde analyser detects a concentration of between 0.8 ppb (1 µg/m3) and 400 ppb (500 µg/m3).

The NEMo® sensor is grounded in measuring technology developed by Ethera under a CEA/CNRS licence (Chevallier, 2012). The measuring device comprises a nanoporous material reacting with formaldehyde whose optical density increases during this reaction. This is associated with an optical detection system enabling measurement of the variation in the material's optical density over time. Thanks to this technology, direct optical reading of sensors is possible throughout exposure for concentrations in the region of μg/m3 (ppb).

These analysers have been used in a range of studies:



As part of the air quality monitoring campaign (2009-2011) in French schools and preschools, the LCSQA launched a study aimed at organising a measuring campaign in a school to gain a clearer idea of the uncertainty surrounding the formaldehyde sampling strategy outlined in the protocols.

To specify the time representativeness, continuous measurements were taken in a room throughout a day in class in a bid to identify variations in concentration depending on the pupils' activities. Continuous sampling was carried out using the Interscan monitor.

http://lodel.irevues.inist.fr/pollution-atmospherique/index.php?id=6016&ftn1



of 71

Figure 3: Changes in formaldehyde concentration levels measured using the Interscan monitor (LCSQA, 2009)

Formaldehyde concentration levels rise as the school day progresses, over about a 2-hour period. This increase in formaldehyde concentrations in the classroom is because of the presence of pupils and their activities, but may also be explained by the starting-up of the monitor, which can take a long time as shown by LCSQA tests in 2008.

The formaldehyde levels reached a relatively stable value of around 40 μg/m3 for the rest of the day.

With the exception of a sharp fall in concentration levels around 15.00, possibly owing to the door and windows being opened during break time (although note that the opening of doors at 11.30 for the lunch break did not bring about a significant drop in the concentrations measured), no particular shift – based on the class's activity (colouring or writing for example) – could be observed (LCSQA, 2009).

IMPACTAIR study – La Rochelle 2016

The IMPACTAIR project is part of an overall programme primarily aimed at addressing a key public health issue, by improving indoor air quality (IAQ) in preschools, infant schools and junior schools in the city of La Rochelle (Cormerais, B., 2017). The methodological plan was to take passive measurements, active measurements pursuant to the reference standard ISO 16000-3 and continuous measurements.

The continuous measurements were taken using the In’Air Solutions analyser in two classrooms. The samples were taken over a 5-week period according to different hypotheses:



of 71

Week 1: contribution of the building (classroom empty);

Week 2: contribution of furniture (classroom furnished);

Week 3: ventilation protocol with children present (classroom + furniture + usual ventilation guidelines);

Week 4: ventilation protocol with children present (classroom + furniture + specific ventilation guidelines - OQAI);

Week 5: ventilation protocol with children present (classroom + furniture + occupation with light indicators for indoor air stuffiness).

Lavoisier School

Figure 4: Lavoisier School - Changes in formaldehyde concentration levels (In’Air Solutions) - La Rochelle 2016.



of 71

The day of 16 March 2016 was focused on in more detail.

Figure 5: Lavoisier School - Changes in formaldehyde concentration levels over one day - La Rochelle 2016.

The door being opened at 06.45 gradually enabled the formaldehyde concentration levels to be brought down. The concentration levels decreased more significantly after a window was opened at about 10.30. Thirty minutes later (at 11.00), the concentration levels began to climb back up after the doors and windows were closed. The effect of the windows being opened at 12.30 is not immediately noticeable. This may be due to cleaning chores being carried out. The formaldehyde concentration levels then fell again around 14.00-14.15, once cleaning had finished – only to begin another steady upward trend up to a peak during the night, since the window had been closed.

Grandes Varennes School

Figure 6: Grandes Varennes School - Changes in formaldehyde concentration levels (In’Air Solutions) - La Rochelle 2016



of 71

The kinetics of formaldehyde concentrations vary from one week to the next, but it can be noted that levels generally range between 1.1 and 41.1 µg/m3 for Lavoisier School, and between 1.6 and 58.6 µg/m3 for Grandes Varennes School.

The results do not reveal any significant rise in pollutants associated with potentially polluting activities, and tally with the "Incitair" findings – that pollution mainly stems from the building and the furniture. Furthermore, usual ventilation practices are not enough to ensure sufficient renewal of air, and opening windows enables a swift albeit short-term fall in formaldehyde concentrations.

ETHERA -2015

As part of the programme validating its continuous analyser NEMo®, the company Ethera implemented continuous measurements across two sites. The 2 sites studied are primary schools with different ventilation systems: the first is a recent Low-Energy Building, fitted with a system that is activated when people are present. The second school is older, and relies on natural airing. Measurements were taken over the last week of the 2014-2015 school year, so from Monday 29/06/2015 to Monday 06/07/2015. The main sources of formaldehyde were the building materials (floors, walls, ceilings) and the furniture (tables, chairs and cupboards mostly). It should be noted that there were also creative art materials in the classrooms (crayons, felt-tip pens, glues, etc.). Cleaning was carried out every morning before the children arrived, and the rooms were aired while this was being carried out. Figure 7: Formaldehyde concentration levels measured using NEMo in room 15 of school 1 – Ethera

(2015).

The times at which children were present are shown by the light blue columns. Note that the measurement is 0 on the graph when the concentration of formaldehyde in the classroom is less than the quantification limit – so between 2 and 4 µg/m3 depending on the environmental conditions



of 71

The concentration of formaldehyde in the classroom is not constant throughout the week. The presence of day/night cycles – where concentrations are low during the day and high at night – can be explained by the fact that ventilation is activated by a presence sensor (100% of rated output when staff and pupils are present (day), 10% of rated output when they are not (night)).

This phenomenon is enhanced by the good practices in place in the school, namely that the rooms are aired during cleaning, which takes place before the pupils arrive. The impact this airing has on the formaldehyde concentration levels leads to a fall in their levels every morning at around 07.00 (except at the weekend and on Monday 06/07/2015, the first day of the school holidays).

The presence of a polluting activity can also be observed on 03/07/2015 between 10.00 and 12.00. During this period, two classes of pupils gathered in the same classroom to engage in creative art activities, which generated a peak in formaldehyde levels (Francois R, 2016).

City of Grenoble –2018

Grenoble performed 4 measuring campaigns across schools using the ETHERA laboratory's NEMo sensor.

The measurements carried out correspond to continuous periods of time encompassing the presence and absence of pupils in the classrooms. The measuring campaigns were performed in two old school buildings, one new school building and one recent one. Their aim was also to identify differences in formaldehyde concentration levels depending on the fixtures in the rooms.

Average concentrations measured throughout the period ranged from 2 µg/m3 (for the 2nd old school building, on the ground floor, from 26/02/18 to 02/03/18), to 51 µg/m3 (for the 1st old school building, from 07/05/18 to 11/05/18). Considering the times during which the children were present, average concentrations ranged between 4 and 25 µg/m3.

Figure 8: Formaldehyde concentration levels measured using NEMo – old school building no. 2 –

Grenoble (2018)



of 71

Figure 9: Formaldehyde concentration levels measured using NEMo – old school building no. 1 - Grenoble (2018).

The highest concentrations – in the region of 70 µg/m3 – were detected in the 1st old school building, as well as in the recent school. These concentration peaks result in average concentration levels of more than 30 µg/m3.

Figure 10: Formaldehyde concentration levels measured using NEMo – recent school - Grenoble (2018).

4.3.2 Offices

PRIMEQUAL – 2016

As part of the inter-organisation research programme for better air quality (PRIMEQUAL), the findings were presented of the formaldehyde dynamic monitoring carried out between January 2012 and June 2015.

The approach involved monitoring (over a short time interval, of 1 to 20 minutes) the change in concentration levels of such target pollutants as formaldehyde in an office space. Note that this office space was fitted with an air extraction system with no sweeping. The space was also equipped for real-time monitoring of occupancy, management of windows and doors, indoor and outdoor climate parameters and other parameters of interest.



of 71

The device used to measure formaldehyde concentrations was the AL4021 analyser (Aerolaser). In this study, formaldehyde was measured indoors during extraction (AL4021 Aerolaser analyser) for 1 min/20 min and outdoors. Ventilation was permanent, by extraction (230 m3/h) with 6 to 12 occupants, varying throughout the day. Formaldehyde was measured over the course of 437 days between 2012 and 2015.

Figure 11: Monitoring of indoor and outdoor formaldehyde levels over the first half of 2015 in offices – Primequal 2016.

The variability in terms of indoor formaldehyde concentrations was characterised at different time scales.

These levels can vary widely throughout the same day, partly due to the opening of doors and windows and to occupation, but also owing to fluctuations in humidity and temperature.

Opening windows can bring about moderate decreases in indoor formaldehyde concentrations – by around 20% on average for one open window and up to 40% for more than 3 open windows. The influence of opening windows is not systematic, however, and depends on other factors (Ramalho.O et al. 2016).

Regarding the influence of occupants, their presence tends to slightly increase formaldehyde levels when no doors or windows are open (concentrations of 17 ppb (21 µg/m3) under unoccupied conditions and 19 ppb (23 µg/m3) under occupied conditions). That said, the presence of occupants does not to have an influence on formaldehyde levels when doors or windows are open.



of 71

4.4 Comparison of the different sampling methods

Some of the aforementioned studies have also compared the concentrations measured using the different sampling methods available for formaldehyde, partly in a bid to validate the methods employed and partly so as to improve knowledge of occupants' actual exposure in indoor spaces.

4.4.1 Comparison of passive sampling/active sampling


One of the aims of this study was to compare the average concentration levels measured in classrooms identified as per the protocol's recommendations, with the average concentration levels measured across all of the school's classrooms, with a view to assessing the impact that the choice of classroom to be equipped has on the representativeness of the measuring campaign.

Four classrooms to be equipped were selected for comparison with ten classrooms presenting a similar layout (in terms of volume, furniture and coverings).

In addition, the concentrations measured over a 4.5-day period using passive and active type sampling methods were compared.

Active sampling was carried out, using pumps on Sep Pack® type cartridges containing as the adsorbent a derivatising agent coating the adsorption cartridge, every two hours or thereabouts, so two samples in the morning from 8.30 to 11.00 and from 11.00 to 13.30 and two in the afternoon from 13.30 to 15.00 and from 15.00 to 16.30.

Passive sampling was carried out using Radiello® type diffusive samplers. The diffusive samplers were positioned in the middle of the classroom and in each corner, 2 metres above ground and in double to guarantee the reproducibility of the results. Concentrations vary between 14.4 µg/m3 and 20.7 µg/m3 for the 8 classrooms tested.

Concentrations measured by active sampling over 2-hourly intervals for 3 days were compared with the measurement taken by diffusive samplers over a 4.5-day time interval in the same classroom.



of 71

Figure 12: Comparison between active sampling (4 x 2-hour measurements for 3 days) and passive sampling (4.5 days) in Classroom 9 - LCSQA 2009.

Average concentration levels measured by active sampling are higher than the average levels measured by passive sampling. Indeed, the average daily concentrations calculated using the 4 active measurements are around 31.1 to 31.4 µg/m3, higher than the average measured by diffusive sampler (18.9 µg/m3). It is also worth noting that, if we consider the 30% uncertainty associated with the diffusive sampler measurement, the average concentration remains lower than that obtained by active sampling. This difference can nevertheless be explained by the fact that the two sampling methods were not implemented over exactly the same periods.

It should, however, be noted that, as mentioned above, it appears that an average concentration over 4.5 days of less than 30 µg/m3 would be associated with concentrations over 2 hours of less than 100 µg/m3.

Figure 13: Comparison between the active sampling in classroom 9 (2-hour time intervals) and passive sampling (4.5 days) across all classrooms - LCSQA 2009



of 71

The formaldehyde concentrations measured using passive methods over 4.5 days across all classrooms are 18 ± 2 µg/m3. Note that if the uncertainty over the measurement, estimated to be around 30%, is taken into account, this dispersion between the concentrations measured across different classrooms is not significant. This campaign therefore seems to indicate that the representativeness of the spatial sampling strategy proposed in the formaldehyde monitoring protocol is satisfactory for this school, under these operating conditions.

4.4.2 Comparison between continuous analyser/passive sampling

ETHERA -2015

As part of the programme to validate its continuous analyser NEMo®, the company Ethera conducted a study aimed at comparing the information obtained with integrated measuring methods over 4.5 days, such as Radiello®, with the results obtained with the NEMo analyser. As a reminder, measurements were carried out in 2 schools: the first is a recent Low-Energy Building, fitted with a system that is activated when people are present; the second school is older, and relies on natural airing.

In the first school, the data recorded by three NEMo® devices placed in one classroom enables the repeatability of the measurements under real-world conditions to be studied. For the campaign run in school no. 1, a very good repeatability of the concentration levels over 7 days, can be observed – at 2%.

The table below summarises the results of the recordings compared with the results of conventional passive measurements using Radiello® cartridges (François et al. 2016).

Table VII: Results of the measurements conducted in classrooms and comparison with the measurements taken by diffusive samplers – Ethera, 2015

School no. 1 with a ventilation system activated when people are present

Class-room

Start

date/time

End

date/time

Dur-ation (days)

Average T°

[min – max] °C

Average RH [min – max]

%

Type of measure-ment

Average °C [min-

max] µg/m3

Rela-tive dev-iation

Class-room 20, NEMo 1

29/06/2015 14.45

06/07/2015 8.45

6.8 27.5

[25.5; 29.5] 55.4

[44; 61]

Radiello 61.7 23.9%

NEMo 76.5 [0; 159.8]


29/06/2015 14.45

06/07/2015 8.45

6.8 27.1

[25.5; 29.0] 54.9

[44; 61]

Radiello 61.7 21.8% NEMo 75.1

[0; 157.5]


29/06/2015 14.45

06/07/2015 8.45

6.8 27.8

[26.0; 29.5] 54.5

[45; 60]

Radiello 61.7 27.4%

NEMo 78.6 [0; 153.0]



of 71

School No.2 with natural airing

Class-room

Start

date/time

End

date/time

Dur-ation (days)

Average T°

[min – max] °C

Average RH [min – max]

%

Type of measure-ment

Average °C [min-

max] µg/m3

Rela-tive dev-iation

Class-room A

29/06/2015 16.00

06/07/2015 8.15

6.7 27.6

[25.0; 31.0] 51.2

[39; 60]


[0; 58.4]

Class-room B

29/06/2015 16.00

06/07/2015 8.15

6.7 29.1

[26.5; 31] 45.6

[37; 55]


[0; 41.0]

Class-room C

29/06/2015 16.00

06/07/2015 8.15

6.7 27.9

[25.5; 31.5] 50.0

[36; 58]

Radiello 27.7 - 5.4% NEMo 26.2

[0; 71.9]

The average relative deviation over 7 days compared with the Radiello reference measurement is 24.4% in school no.1 and 9.8% in school no.2. In this case, the diffusive samplers retranscribe lower concentrations than the continuous analyser on average. It appears that when the concentrations measured using diffusive samplers exceed 50 µg/m3 on average, the maximum concentrations measured over a 2-hour time interval systematically exceed 120 µg/m3. On the other hand, the study shows that average concentrations of less than 30 µg/m3 can be associated with maximum concentrations over 2-hour time intervals of less than 60 µg/m3.

4.4.3 Comparison of 3 sampling methods


Based on the measurements performed in the context of the IMPACTAIR study conducted in La Rochelle in 2016, it has been possible to compare the continuous sampling and passive sampling results over a 1-week time interval (4.5 days), as well as the continuous sampling and active sampling results.

The comparison tests for the two schools have shown a close correlation between the passive sampling results and continuous analyser data, albeit with a +20% (Grandes Varennes) to +25% (Lavoisier) deviation over the 4.5-day period in favour of passive sampling (Cormerais, B., 2017).

Comparison of measurements by passive samplers and continuous

measurements at Lavoisier School

Comparison of measurements by passive samplers and continuous

measurements – Grandes Varennes School

The formaldehyde concentrations obtained with the analyser and by active sampling were also compared.

Lavoisier School – Inter-comparison of the formaldehyde concentrations obtained with

the analyser and by active sampling

Grandes Varennes School – Inter-comparison of the formaldehyde concentrations obtained with

the analyser and by active sampling

The comparison tests for the two schools have shown a close correlation between the results of the two methods with the continuous analyser and by active sampling, with a 3% (Grandes Varennes) to 10% (Lavoisier) deviation – DNPH cartridge sampling slightly overestimated the results (Trocquet et al., 2016).



of 71

5. Existing indoor air quality guidelines and benchmarks

An indoor air quality guideline (IAQB) corresponds to an airborne concentration of a chemical substance below which no health effects are expected for the general population.

Generally speaking, the indoor air quality benchmark (IAQB) is the concentration below which no specific corrective action is required. In terms of management, this can be considered as a provisionally acceptable maximum level for the pollutant in question, in light of the distribution of values observed in enclosed spaces and under the conditions of their long-term regular occupation. That being so, compliance with this IAQB may be associated with health consequences – not least for the most vulnerable individuals – when it exceeds the IAQG, and the aim should ultimately be for this level to draw nearer to the IAQG.

5.1 At national level

5.1.1 ANSES' indoor air quality guidelines for formaldehyde

In 2007, the Agency had recommended IAQGs of 50 μg/m3 for short-term exposure and 10 μg/m3 for long-term exposure. This proposal was based on the choice of ATSDR-defined TRVs.

Since then, new scientific data has been published, bearing on the toxicokinetics, irritating effects (studies of controlled exposure in humans), respiratory sensitisation, the association between indoor air pollution and respiratory effects (epidemiological studies), genotoxicity (studies conducted in humans) and the carcinogenic effects of formaldehyde (nasopharyngeal cancer and leukaemia). Following an analysis of this data, in 2018 ANSES updated the reference values for formaldehyde, the IAQGs in particular.

Given the knowledge available at the time its last report was being written and the international practices for establishing TRVs, ANSES decided on the presence of a threshold for acute and chronic irritating effects. Taking eye irritation as the critical effect, it concludes that keeping formaldehyde concentrations below values that bring on eye irritation can help to protect against the onset of nasopharyngeal cancer.

Accordingly, with respect to updates to the IAQGs for formaldehyde, ANSES proposes a single IAQG for short-term exposure in view of protecting the general population, as regards acute and chronic effects alike.



of 71

The reasons justifying this proposal are as follows. Eye irritation has been selected to establish acute and chronic TRVs. This effect is the first key event and a precursor of more severe irreversible effects such as nasopharyngeal carcinogenic effects of formaldehyde. Considering the threshold mode of action for nasopharyngeal cancer, compliance with the acute value, characterised by a high confidence level, would protect against the occurrence of long-term effects.

Given a TRV of 123 μg/m3 and for consistency's sake with the indoor air quality guideline proposed by WHO in 2010 (100 μg/m3), ANSES proposes a short-term IAQG of 100 μg/m3.

For this to happen, as WHO underlined in 2010, the proposed value should be complied with for repeated and continuous short-term exposure over an entire day.

General comments concerning ANSES' guidelines

The IAQGs aim at protecting the population from any harmful effect associated with exposure to a substance. Compliance with them does not guarantee the absolute absence of effect at concentrations below the guidelines proposed, however – especially where particularly vulnerable individuals are concerned. By the same token, a health effect is not necessarily expected for all individuals if IAQGs are exceeded.

Lastly, it should be borne in mind that, since IAQGs are established for substances that are assessed individually, it cannot be ruled out that effects might arise, owing to possible synergies, at lower levels, as a result of simultaneous exposure to several pollutants or of multiple types of exposure to the same pollutant (skin and/or oral).

5.1.2 HCSP's indoor air quality benchmarks

In 2009, the HCSP proposed, for the long-term, adopting four values for formaldehyde: a target value to be achieved eventually, an objective quality benchmark, an information and recommendations value and lastly a rapid action value.

The target value corresponded to the long-term IAQG defined by AFSSET in 2007: 10 μg/m3.

• The air quality benchmark defined in 2009 was 30 μg/m3.

This is the value below which no corrective action is required based on available data in 2009. Efforts to bring this down towards the target value were linear over the years, with the following milestones: 20 μg/m3 after 5 years (2014) and 10 μg/m3 after 10 years (2019).

• The information and recommendation value in 2009 was 50 μg/m3.

• The rapid action value was 100 μg/m3.



of 71

5.2 Values set by international organisations or applicable in other countries

For formaldehyde, a number of countries have outlined an "indoor air" guideline. In addition to the international guidelines set by the supranational bodies, national values have been proposed in Europe (Netherlands), Canada and the United States (State of California). These are non-binding values generally speaking. In most cases, these guidelines are provided for "long-term" and "short-term" exposure scenarios. They have been listed in ANSES' 2018 paper on formaldehyde, with the main data presented briefly in the paragraphs below.

5.2.1 Values set by supranational bodies

At European level, the INDEX project was aimed at establishing an initial list of priority chemical pollutants in indoor environments likely to be regulated in the future, based on clearly defined criteria, a bibliographic review (until September 2004) and reference values gathered from among international bodies. Based on the information available and after examining the existing data, the INDEX project steering committee defined a priority list of 14 substances out of the 41 compounds initially proposed – and this included formaldehyde.

In indoor air, the final report of the INDEX project: "Critical Appraisal of the setting and implementation of indoor exposure limits in the EU”, 2005, drawn up by the Institute for Health and Consumer Protection established, for formaldehyde, a concentration above which its presence is a concern over the long term of 1 µg/m3 (short-term and long-term IAQG). This value is based on the TRV proposed by OEHHA (1999) of 3 μg/m3, itself based on a NOAEL in the occupational environment of 90 μg/m3, and to which a further reduction of 3 has been added to factor in the specific sensitivity of children. However, since the INDEX group considered that median indoor air concentrations in Europe were generally higher (the median is 26+/-6 µg/m3, and the 90th percentile 59 +/- 7 µg/m3) it also set forth a target of 30 µg/m3 (INDEX, 2005).

In 2010, WHO (WHO guidelines for indoor air quality: selected pollutants, 2010) published an indoor air quality guideline of 100 µg/m3, applicable for 30-minute exposure, to protect from sensory irritation. WHO considers that this threshold should not be exceeded at any 30-minute interval during a day (WHO, 2010).

The lowest concentration associated with sensory eye irritation is 380 μg/m3 for 4 hours. Increases in eye blink frequency and conjunctival redness appear at 600 μg/m3, considered equal to the no observed adverse effect level (NOAEL) by WHO. In light of this NOAEL, WHO applied an assessment factor of 5 to take account of interindividual variability of sensory irritation thresholds. The final value obtained, 120 μg/m3, was rounded down to 100 μg/m3.



of 71

5.2.2 Values set by national bodies

Health Canada has issued a Residential Indoor Air Quality Guideline for Formaldehyde. This establishes maximum acceptable formaldehyde levels for two types of exposure:

The short-term exposure limit protects against health problems that may arise on account of exposure to high levels over a short time interval (e.g.: one hour). This type of exposure could occur, for example, when applying paint or varnish containing formaldehyde. In order to avoid possible eye, nose and throat irritation due to short-term exposure, the level of formaldehyde in indoor air, according to Health Canada, should not exceed 123 μg/m3 (100 ppb). This level is considered to be less than the level of formaldehyde responsible for irritations, according to scientific studies. Health Canada adopts the lowest value to provide better health protection, since sensitivity to formaldehyde varies from individual to individual.

The long-term exposure limit protects against health problems that may arise on account of repeated exposure to low levels of formaldehyde over a long period of time (days, weeks, months, etc.). In order to prevent respiratory problems owing to long-term exposure, i.e. over a period of days, months or years, the indoor air level recommended by Health Canada is 50 μg/m3 (or 40 ppb). Health Canada considers that when formaldehyde levels exceed this value, the risk of respiratory problems or allergic sensitisation also increases, particularly in children.

For short-term exposure, Japan (MHLW, 2002), the UK (COMEAP, 2004) and Norway (NIPH, 1999) propose a guideline value of 100 µg/m3 over a 30-minute time interval.

California (OEHHA, 2004) proposed specific guidelines for formaldehyde that were revised in August 2004 following the re-classification by the IARC. These are particularly based on the TRVs proposed by OEHHA. As such, a guideline of 95 µg/m3 is recommended for exposure over 1 hour and 34 μg/m3 for 8 hours of exposure or more, to protect from irritation effects.

Texas has, since 2002, proposed a guideline of 50 μg/m3, protecting against irritation effects over the long term. It is proposed for public buildings.

China (2001) has defined long-term guidelines for numerous categories of buildings divided into two groups: housing, hospitals, retirement homes, childcare centres, schools (group I) and offices, shops, hotels, leisure centres, museums, public transport and restaurants (group II). For these two groups of premises, the formaldehyde limit is set respectively at 80 and 120 μg/m3. On 1 January 2002, the "Health standard for indoor air quality" also came into force, promulgated by the Ministry of Health with a specific guideline of 80 μg/m3 (hourly average).

Poland has set admissible indoor concentrations for around 30 chemical compounds. For formaldehyde, the guidelines are equal to 50 μg/m3 in



of 71

buildings which people might occupy all day and 100 μg/m3 in premises where we spend 8 to 10 hours a day.

In Germany, the value of 124 μg/m3 proposed by BFR (German Federal Institute for Risk Assessment) was calculated on the basis of a model factoring in two mechanisms of action underlying the toxicity of formaldehyde (cytotoxicity, cell proliferation and genotoxicity). According to this information, it appears that the guideline proposed, with no indication of time, thus protects against local carcinogenic effects incurred by formaldehyde (nasopharyngeal cancer).

In Switzerland, the Federal Office of Public Health recommends not exceeding the concentration of 125 μg/m3 in housing and places of hospitality, so as to avoid adverse health effects.

Finland, since 2000, has proposed different concentration levels for formaldehyde. These are divided across three indoor environment categories:

• category S1: indoor air quality is very good and the thermal environment is as comfortable in summer as it is in winter (formaldehyde concentration of less than 30 μg/m3) ;

• category S2: indoor air quality is good. Temperatures exceed the comfortable temperature threshold on the hottest days in the summer (formaldehyde concentration of between 30 and 50 μg/m3);

• category S3: the indoor air quality and thermal environment meet the minimum requirements set by the building codes. The indoor air quality may occasionally feel stuffy and odours may appear. Temperatures can exceed the comfortable temperature threshold during hot summer days (formaldehyde concentration of between 50 and 100 μg/m3).

These are chronic exposure values that only factor in the building's emissions (excluding human activities). And yet they are not based solely on health criteria (and are not associated with the onset of specific effects in the event the values are exceeded either).

In Flanders, a Flemish Government Decree stipulates that "A home where the intervention value, is exceeded (…) is deemed uninhabitable". The intervention value for formaldehyde is set at 100 μg/m3 for 30 minutes.



of 71

6. Regulatory provisions European Directive 98/8/EC concerning the placing of biocidal products on the market, transposed into French law in Articles L.522-1 et seq. of the French Environment Code, is aimed at harmonising the regulations of EU Member States – hitherto very unequal – and at ensuring market unity. The primary objective of this Directive is to ensure a high level of protection for humans, animals and the environment by ensuring that only effective biocidal products posing acceptable risks are placed on the market and by encouraging the marketing of active substances posing ever fewer risks for humans and the environment. The measures are particularly aimed at preventing long-term effects: carcinogenic or toxic for reproduction as well as the effects of bioaccumulative toxic substances that do not readily degrade.

Accordingly, biocidal products containing formaldehyde can only be placed on the market for product-types (PTs) listed in the review programme on biocidal active substances, namely PTs 2, 3 and 22:

PT 2: disinfectants and algaecides not intended for direct application on humans or animals,

PT 3: products used for veterinary hygiene purposes,

PT 22: embalming and taxidermist fluids.

The evaluation is performed by reporting Member States and the evaluation report is validated by ECHA's Biocidal Products Committee.

The regulations governing indoor air quality are based both on public health prevention associated with certain pollutants as well as on the following Grenelle Environment Forum pledges:

Indoor air quality monitoring in establishments open to the public.

Introduction of labelling of materials that can emit pollutants into indoor air;

Decree no. 2015-1000 of 17 August 2015 lays down the procedure for monitoring indoor air quality in establishments open to the public, including establishments accommodating children. It particularly stipulates that the ventilation methods must be assessed and certain pollutants, formaldehyde among them, shall have to be measured.

1 January 2018 is the deadline by which all buildings welcoming children under 6 years of age (preschools, day nurseries and infant schools) and junior schools must have implemented the indoor air monitoring procedure



of 71

for the first time; 1 January 2020 for leisure centres and secondary schools and 2023 for the other establishments.

Measurements of pollutants should particularly be compared against the indoor air quality guidelines and limits activating additional investigations.

The limits are defined in Decree No. 2015-1926 of 30 December 2015 amending Decree No. 2012-14 of 5 January 2012 on the assessment of ventilation methods and measurement of pollutants carried out in the context of indoor air quality monitoring in certain establishments open to the public. The limit for long-term exposure to formaldehyde is set at 100 μg/m3.

In terms of regulations, "indoor air quality guideline" refers to a level of a pollutant's airborne concentration indoors, set for a given enclosed space, in a bid to avoid or prevent, or reduce, the harmful effects on human health, and to be met, as far as possible within a given time-limit. To date, indoor air quality guidelines have been defined for formaldehyde: the guideline for formaldehyde set for long-term exposure is 30 μg/m3, applicable since 1 January 2015, and this is expected to be lowered to 10 μg/m³ on 1 January 2023.

Where a measuring campaign is not carried out, the establishment must set up an action plan based on an assessment performed pursuant to the practical guide for better air quality inside buildings receiving children.

In the context of the Grenelle Environment Forum, compulsory labelling of building and furnishing products as well as wall coverings and flooring, paint and varnish which emit substances into the surrounding air was proposed and enshrined in the Environment Code (Article L221-10). The regulations in France concern building and decoration products as well as scented products for burning with compulsory labelling. The terms governing this labelling are grounded in the characterisation of ten substances and the "total volatile organic compounds" (TVOC) parameter regarding emission, including formaldehyde.

Considering the multiple sources of exposure to formaldehyde and the possibility of co-exposure to other substances in the adehyde family, the measures stipulated by the Environment Code in terms of compulsory labelling of building and decoration products are particularly based upon compliance with a maximum emission threshold of 10 µg/m3 for the highest-performing category (A+). ANSES considers these measures to be consistent in that they reflect the operator's strong commitment to limiting – or removing – any situation exposing individuals to formaldehyde.



of 71

7. Proposals of benchmarks for formaldehyde In 2009 the HCSP had recommended four values for indoor air pollution by formaldehyde in enclosed spaces: a target value to be achieved eventually, an objective quality benchmark, an information and recommendations value and lastly a rapid action value, based on the scientific data available at the time. Given the more recent scientific data now available, these recommendations are due an update.

7.1 Benchmarks and management values for indoor air quality

7.1.1 Indoor Air Quality Benchmark (IAQB)

In accordance with the methodological approach set out in the January 2019 report on benchmarks for managing indoor air quality, and in light of the fact that ANSES has defined an IAQG over a short time interval for all formaldehyde-related effects (whether they arise over the short- or long-term), the HCSP recommends an Indoor Air Quality Benchmark (IAQB), equal to ANSES' IAQG, of 100 µg/m3 for air inside residential buildings or establishments open to the public. This benchmark has been adopted with eye irritation taken as the critical effect, since such symptoms appear early on following airborne exposure; compliance with this maximum limit helps to protect against the onset over time of nasopharyngeal cancer.

The IAQB should be measured over a 1- to 4-hour time interval.

There is little data currently available concerning concentrations measured over short time intervals (1 to 4 hours). Nevertheless, the values that are available, through the exploratory studies presented in Chapter 4.2, indicate concentrations that are much lower than 100 µg/m3 overall. The HCSP has not therefore deemed it necessary to define a higher IAQB than the IAQG with a timeline mapped out for improving the value. The HCSP recommends that this IAQB become immediately applicable and complied with in all buildings, with, in the event it is exceeded, a maximum time-limit for taking corrective action of 6 months.

7.1.2 Provisional management value

To ensure compliance in practice with the IAQB, measured over a short time interval, and given the kinetics of formaldehyde concentrations in indoor air, it would be necessary to carry out continuous measurements or active measurements repeated over the space of one week.

In order to limit the number and, consequently, the cost of measurements to be taken, the HCSP recommends a provisional management value of 30 µg/m3 measured over a 1-week period, based on monitoring study findings and



of 71

the various study results presented above. This provisional management value may eventually be abolished once new continuous measuring instruments have been standardised and become available at a reasonable price.

Below this value, there is no objective reason for taking specific corrective action, but there would be merit in undertaking renovations or in changing pieces of furniture to opt for less polluting materials and so foster progress over time towards lower and lower levels, in keeping with the ALARA principle.

For premises where the levels measured range between 30 and 100 μg/m3, action to reduce emissions is encouraged. However, at these concentration levels, given the ubiquitous nature of formaldehyde, it is often the case that the emissions come from multiple diffuse sources; it is therefore often preferable to start by upgrading the building's ventilation system, so as to bring the average levels down below 30 μg/m3. The time-limit for checking (via new measurements) that this action regarding ventilation does indeed have this effect, should be no more than a year.

This type of situation corresponds to the range of values at the top end of the distribution of values in housing – between the 80th and 95th percentiles of the distribution of values in the OQAI-led housing campaign conducted between 2003 and 2005. The Report on indoor air quality monitoring in establishments open to the public indicates that around 16% of measurements come within this range (INERIS, 2018).

On the other hand, where a level is measured exceeding 100 μg/m3 over a week's time interval (4.5 to 7 days), this must be considered a value requiring corrective action within a set time-limit of 6 months. Pending the end of works, ventilation of the room in question shall be ramped up to help remove the presence of formaldehyde. The goal will be to swiftly bring the average levels in the surrounding air down below 30 μg/m3 during measurements carried out over a week's time interval (4.5 to 7 days).

Remember that this 100 μg/m3 value was not recorded during the OQAI-led nationwide housing campaign, but that it has already been measured occasionally in educational buildings.

7.2 Rapid action value (RAV)

The IAQB recommended by the HCSP is equal to the IAQG proposed by ANSES – the purpose of which is to prevent the onset, over the long-term and owing to possibly short-term, repeated exposures, of severe toxic effects (nasopharyngeal cancer). As such, the HCSP does not consider it necessary to recommend a short-term RAV.



of 71

7.3 Proposal of a sampling and measuring strategy

The IAQB, which equals the IAQG established by ANSES in 2018 and the guideline proposed by the World Health Organisation in 2010, should be complied with for repeated and continuous short-term exposure over an entire day. This means that changes in formaldehyde concentrations over time must be measured to ensure that there are no exposure peaks exceeding 100 µg/m3.

It thus strikes as important, in the context of an indoor air quality survey, to try to pinpoint the sources and conditions conducive to peaks in formaldehyde concentrations. This new approach to characterising formaldehyde exposure entails the implementation of suitable measuring strategies for making sure that exposure peaks do not occur. Such strategies might involve successive measurements over 1- to 4-hour time intervals – at least during the periods of occupation of the buildings in question. Otherwise, measurement over the space of one week (4.5 to 7 days) via passive sampling could also be organised.

7.3.1 Sampling strategy

In the various studies carried out over recent years in France (OQAI's nationwide housing campaign in particular, local measurements in preschools and schools), what comes across is that the formaldehyde levels measured in hot weather are much higher than those measured in cooler weather.

This is why the monitoring campaigns initiated by the Ministry of the Environment provide for a winter measurement and a summer measurement to obtain an estimate of the annual average value that more accurately reflects reality than a single value, which could lead to an over- or under-estimation of the measurements depending on the season.

This strategy corresponds to the protocol developed by the Central Laboratory for Air Quality Monitoring (LCSQA) for the measuring campaigns scheduled across 300 preschools and schools between 2009 and 2011. It goes without saying that, where values exceed the indoor air quality benchmark, this strategy should not postpone communication to occupants or put off a decision regarding necessary intervention.

Moreover, measurements aimed at assessing exposure among individuals, the rooms to be equipped as a priority are those where the occupants spend the most time (so for housing, the living room or bedroom should be given precedence).

During compliance checks, the sampling device should be placed at least 1 m or 2 m away from a wall and at a height of 1.5 m or 1 m to 1.2 m in the case of offices, schools or preschools where occupants spend most of the time in the seated position. Under these circumstances, one device per room is normally sufficient for sampling purposes. It is recommended that places in direct sunlight, located near heating systems, in draughty areas or near ventilation ducts be avoided as these may skew the measuring results.



of 71

The samples should be taken under normal conditions which characterise the indoor air in these rooms. That said, such conditions should remain within the zone of comfort. Measurements may be taken when occupants are present.

7.3.2 Measuring methods

Indoor air quality measurement for the purposes of comparison with the indoor air quality benchmark (IAQB) and the provisional management value should be carried out via a standard-approved, renowned method.

The main standardised measuring techniques currently available are described below.

Recommendations for comparison with the benchmarks:

1- to 4-hourly time intervals:

Monitoring of the IAQB may be carried out via 1- to 4-hourly sampling, using a 2,4 dinitrophenylhydrazine-coated cartridge containing silica gel, with acetonitrile desorption, and high-performance liquid chromatography with UV detection.

According to the protocol, the sampling rate recommended is between 0.1 and 2 L/min, without exceeding 1.5 L/min owing to the high pressure drop of the cartridge. A flow rate of 2 L/min is viable for certain cartridges with lower pressure drops. The typical flow rate is 1 L/min (0.8 L/min for two standard cartridges). Checking the stability of flow in very dusty environments is also to be recommended.

Systematic use of an ozone scrubber (usually containing potassium iodide, KI) is required, since ozone is known to interfere negatively with the method. Checks should be carried out beforehand, though, to ensure that the ozone scrubber used does not reduce the formaldehyde levels collected by the cartridge.

For indoor environments, this method relies particularly on standard NF ISO16000-3 (2011), which is partly based on US EPA method TO-11A (1999) for the measurement of airborne formaldehyde, as well as US EPA method IP-6A (1990) and US EPA method 0100 (1996) – the latter is itself associated with US EPA method 8315A (1996).

Cartridges should be stored in a cool, dark place for no more than 90 days prior to sampling. After sampling, the cartridge may be stored in a cool, dark place for no more than 15 days.

1-week time intervals

Monitoring of the provisional management value may be carried out via diffusive sampling on a badge, filter or silica gel cartridge coated in a reagent, 2,4-dinitrophenylhydrazine (DNPH). Analysis is then performed by



of 71

high-performance liquid chromatography (HPLC) to separate the different derivatives, with UV detection performed at a wavelength of 360 to 370 nm.

For indoor environments, this method particularly relies on standard NF ISO 16000-4 (2012) and US EPA method IP-6C (1990).

In recent years, several direct-read, continuous measuring instruments have been developed, and marketed in some cases, in a bid to provide information on changes in indoor air concentrations. Using this type of instrument is particularly worthwhile for dynamic measurements aimed at identifying higher exposure peaks over a given period.

The technical characteristics and performances of these instruments have not been validated or adequately documented to date, however.

7.4 Overview of values (benchmark and provisional management value)

The HCSP proposes adopting two long-term values for formaldehyde: a quality benchmark and a provisional management value.

The benchmark is ANSES' IAQG: 100 μg/m3. This value is immediately applicable. It is validated by successive 1- to 4-hourly measurements over the course of a day – under occupied conditions.

The provisional management value is 30 μg/m3. This is a value for which the IAQB is deemed to be complied with in cases where the technical or financial criteria have resulted in measurement over a 1-week time interval being chosen (i.e. 4.5 to 7 days) – and the value below which no corrective action is required.

7.5 Case of new buildings

Regarding new buildings delivered and equipped from 2020, the concentrations measured must be as low as possible and, in all cases, less than the IAQB or, failing that, the provisional management value. The same applies for buildings where extensive renovations are being carried out. To that end, architects and construction managers shall endeavour particularly to act on indoor emission sources, interior design and building materials.



of 71

8. Occupational exposure limits (OELs) The IAQB and IAQB are defined to protect the general population against the harmful health effects of pollutants present more specifically in indoor environments (homes, schools or offices for example, excluding premises exposed to specific pollution). In the context of occupational exposure, OELs (occupational exposure limits) should be borne in mind.

The average occupational exposure limit (8h-aOEL) established in France is 0.5 ppm (610 µg/m3) The STEL (short-term exposure limit) over 15 minutes is 1 ppm (1,230 µg/m3). These values, purely indicative, were adopted by a Ministry of Labour circular in 1993 (DGT 93-18) (Toxicological datasheet 22 of the INRS, 2011) and do not therefore have any regulatory status in their own right.

Since then, assessment of the health effects of occupational exposure to formaldehyde has been subject to a change in method within the EU (exclusive consideration of health criteria) and new data, including the data published by the Scientific Committee on Occupational Exposure Limits (SCOEL) in 2008. In this regard, the SCOEL held the critical effect to be pro-inflammatory effects on the nasal mucosa following exposure to formaldehyde. That said, owing to insufficient data for establishing a NOAEL for this critical effect, the SCOEL selected irritation of the eye mucous membranes, deemed more sensitive, as the indicator of the aforementioned critical effect.

In 2008, a short-term exposure limit (15 min-STEL) of 500 μg/m3 (0.4 ppm) was recommended by ANSES' OEL Expert Committee, based on the study by Lang et al. (2008) protecting against the irritant effects of formaldehyde. An 8-hour occupational exposure limit (8h-OEL) of 250 μg/m3 (0.2 ppm) was also recommended. The critical effects selected were sensory irritation and eye irritation. For that value, the studies by Paustenbach et al. (1997) for eye irritation, and Arts et al. (2006) for sensory irritation were selected as key studies.

The values that ANSES initially selected by critical analysis of the health effects associated with occupational exposure to formaldehyde were 0.3 ppm and 0.5 ppm over 8 hours and 15 minutes respectively. However, for the sake of harmonisation, ANSES eventually opted for the very similar values previously published by the SCOEL.

In 2017, ANSES proposed the following OEL:

8h-OEL = 350 μg/m3 (rounded), based on the study by Lang et al. (2008) and study by Mueller (2012), which was not published in ANSES' 2008 paper;

15 min-STEL = 700 μg/m3 (rounded), based on the study by Lang et al. (2008) by taking eye irritation as the critical effect.



of 71

Overall, a certain uniformity can be observed in the values proposed by the different organisations at international level:

ACGIH1

• TLV-STEL (Threshold limit value – Short term exposure limit) = 0.3 ppm = 0.37 mg/m3

Germany: MAK2:

• aOEL (8 hours) = 0.3 ppm = 0.37 mg/m3

Consistency between ANSES' IAQG and the OEL

The OELs defined by ANSES are based on the same key studies as those selected for establishing the TRVs and IAQG.

The critical effect selected is the same, the only difference lying in the application of safety factors: there is only a factor 3 between the 8h-OEL and the IAQG proposed by ANSES.

1 American Conference of Governmental Industrial Hygienists

2 Maximale arbeitsplatz-konzentration (maximum workplace concentration)



of 71

Bibliography Ademe. 2016. Projet MERMAID : mesures expérimentales représentatives et modélisation

air intérieur détaillée. May 2016.

AFSSET, 2008. Evaluation des risques sanitaires liés à la présence de formaldéhyde dans les environnements intérieurs et extérieurs. Toxicité du formaldéhyde. Etat des connaissances sur la caractérisation des dangers et choix des valeurs toxicologiques de référence.

AFSSET, 2009. Risques sanitaires liés à la présence de formaldéhyde. Étude de filières, risques professionnels, relation entre composition et émission - April 2009.

Andersen I. and Molhave L., 1983. Controlled human studies with formaldehyde. Formaldehyde Toxicity. Washington, Hemisphere Publishing Corporation. J. E. Gibson, pp.154-165

Annesi-Maesano I, Hulin M, Lavaud F, Raherison C, Kopferschmitt C, de Blay F, Charpin DA, Caillaud D. 2012. Poor air quality in classrooms related to asthma and rhinitis in primary schoolchildren of the French 6 Cities Study. Thorax.

ANSES. 2011. CLH report Proposal for Harmonised Classification and Labelling Based on Regulation (EC) No 1272/2008 (CLP Regulation), Annex VI, Part 2. Substance Name: FORMALDEHYDE.

ANSES. 2017. Valeurs toxicologiques de référence (VTR). Élaboration de VTR par inhalation pour le formaldéhyde (CAS n°50-00-0).

ANSES. Mise à jour de valeurs guides de qualité d’air intérieur. Formaldehyde. Avis de l’Anses. Rapport d’expertise collective. February, 2018.

ASPA, 2005. Campagne de mesure du formaldéhyde dans les établissements scolaires et d’accueil de petite enfance de la ville de Strasbourg : bilan des niveaux mesurés - June 2005.

ATMO Rhône-Alpes, 2009. Mesure du formaldéhyde dans l'air intérieur des écoles maternelles et des crèches en Rhône-Alpes. Rapport de synthèse, January 2009.

ATSDR, 1999. Toxicological profile for formaldehyde. Agency for Toxic Substances and Disease Registry, Atlanta, GA: U.S department of Health and Human Services, Public Health Services. https://www.atsdr.cdc.gov/ToxProfiles/tp.asp?id=220&tid=39 .

Banerjee S, Annesi-Maesano I (2012) Spatial variability of indoor air pollutants in schools. A multilevel approach. Atmos Environ 61:558-561. doi:10.1016/j.atmosenv.2012.08.007

Barker S, Weinfeld M, Murray D. 2005. DNA-protein crosslinks: their induction, repair, and biological consequences: Mutat Res, v. 589, no. 2:111-135.

Bender J.R., Mullin L.S., Graepel G.J. and Wilson W.E., 1983. Eye irritation response of humans to formaldehyde. Am Ind Hyg Assoc J, 44, 6, 463-465.

https://www.atsdr.cdc.gov/ToxProfiles/tp.asp?id=220&tid=39



of 71

BfR. 2006. Toxicological Assessment of Formaldehyde. Opinion of BfR. No. 023/2006 of 30 March 2006.

Blair A., Saracci R., Stewart P.A., Hayes R.B. and Shy C. 1990. Epidemiologic evidence on the relationship between formaldehyde exposure and cancer. Scand J Work Environ Health,16, 6, 381-393.

Brown T, Dassonville C, Derbez M, Ramalho O, Kirchner S, Crump D, Mandin C, 2015. Relationships between socioeconomic and lifestyle factors and indoor air quality in French dwellings. Environ Res 140:385-396. doi:10.1016/j.envres.2015.04.012

Canha N, Mandin C, Ramalho O, Wyart G, Riberon J, Dassonville C, Hanninen O, Almeida SM, Derbez M, 2016. Assessment of ventilation and indoor air pollutants in nursery and elementary schools in France. Indoor Air 26 (3):350-365. doi:10.1111/ina.12222

Collins J.J., Caporossi J.F. and Esmen N.A. 1997. An updated meta-analysis of formaldehyde exposure and upper respiratory tract cancers. J Occup Environ Med, 39, 639- 651.

Collins JJ, Ness R, Tyl RW, Krivanek N, Esmen NA, Hall TA. 2001. A Review of AdversePregnancy Outcomes and Formaldehyde Exposure in Human and Animal Studies. Regul Toxicol Pharmacol; 34:17–34.

Cormerais B., Blondeau P., 2017. INCIT’AIR et IMPACT’AIR : deux projets au coeur des établissements scolaires de La Rochelle.

Dallongeville A, Costet N, Zmirou-Navier D, Le Bot B, Chevrier C, Deguen S, Annesi-Maesano I, Blanchard O. 2015. Volatile and semi-volatile organic compounds of respiratory health relevance in French dwellings. Indoor Air 26: 426–438.

Dassonville C, Demattei C, Laurent AM, Le Moullec Y, Seta N, Momas I., 2009. Assessment and predictor determination of indoor aldehyde levels in Paris newborn babies' homes. Indoor Air. 2009 Aug;19(4):314-23. Epub 2009 Jan 19.

Day J.H., Lees R.E., Clark R.H. and Pattee P., 1984. Respiratory response to formaldehyde and off-gas of urea formaldehyde foam insulation. Can Med Assoc J, 131, 9, 1061-1065.

Doty RL, Cometto-Muniz JE, Jalowayski AA, Dalton P, Kendal-Reed M, Hodgson M. 2004. Assessment of upper respiratory tract and ocular irritative effects of volatile chemicals in humans. Critic Rev Toxicol, 34:85-142.

Duboudin C (2009). Pollution inside the home: Descriptive analyses. Part I: Analysis of the statistical correlations between pollutants inside homes. Environnement, Risques et Santé 8 (6), 485-496.

Duong A, Steinmaus C, McHale CM, Vaughan CP, Zhang L. 2011. Reproductive and developmental toxicity of formaldehyde: A systematic review. Mutat Res, 728:118-138.

European Commission. Joint Research Centre (JRC), 2005. Final Report. Critical Appraisal of the Setting and Implementation of Indoor Exposure Limits in the EU: The Index project. Institute for Health and Consumer Protection, Physical and Chemical Exposure Unit. January, 2005. 337 pages. I-21020 Ispra (VA), Italy.



of 71

European Commission. Joint Research Centre (JRC), 2005b. HEXPOC: Human Exposure Characterisation of Chemical substances, quantification of exposure routes. Physical and Chemical Exposure Unit, EU 21501 EN. 126 pages. I-21020 Ispra (VA), Italy. 2005.

Fable S et al., 2009. Air intérieur, mesure du formaldéhyde ; December 2009 – 39 pages.

François R. et al.2016. Développement d’un capteur de mesure en continu du taux de formaldéhyde dans l’air intérieur : application au contrôle de la qualité de l’air intérieur dans les lieux accueillant des enfants, Pollution atmosphérique [Online], No.231 - 232, December 2016. Updated on: 09/02/2017, http://lodel.irevues.inist.fr/pollution- atmospherique/index.php?id=6016

Franks SJ, 2005. A mathematical model for the absorption and metabolism of formaldehyde vapour by humans. Toxicol Appl Pharmacol. 2005; 206 : 309-320.

Gaylor DW, Lutz WK, Conolly RB. 2004. Statistical Analysis of Nonmonotonic Dose-Response Relationships: Research Design and Analysis of Nasal Cell Proliferation in Rats Exposed to Formaldehyde. Toxicol Sci 77, 158–164.

HCSP, 2009. Valeurs repères d’aide à la gestion dans l’air des espaces clos. Le Formaldéhyde. October 2009.

HCSP, 2019. Valeurs repères d’aide à la gestion de la qualité de l’air intérieur. Présentation de la démarche méthodologique. January 2019.

Ineris, 2010. Fiche de données toxicologiques et environnementales des substances chimiques. DRC-10-109974-00925A. February 2010.

Ineris, 2015. Utilisation de produits ménagers et qualité de l’air intérieur : enjeux sanitaires, substances d’intérêt, bonnes pratiques. DRC-14-144018-04822B. May 2015.

Ineris, 2018. - Opération "1000 kits pour les écoles et crèches" Déploiement / Retour d’expérience DRC-17-149968-10992B. October 2018.

INRS, 2011. Fiche de données toxicologiques n°7. Aldéhyde formique et solutions aqueuses (N. Bonnard, M. Falcy, E. Pasquier, J.-C. Protois). Institut National de Recherche et de Sécurité. http://www.inrs.fr/publications/bdd/fichetox/fiche.html?refINRS=FICHETOX_7

IARC, 2006. Monographs on the evaluation of carcinogenic risks to humans. Formaldehyde, 2- Butoxyethanol and 1-tert-Butoxypropan-2-ol. Volume 88

IPCS (International Program on Chemical Safety). 1989. Environmental Health Criteria EHC n°89. Formaldehyde. WHO, UNEP, ILO.

Kulle T.J. 1993. Acute odor and irritation response in health nonsmokers with formaldehyde exposure. Inhal Toxicol, 5, 323-332.

Lang I, Bruckner T, Triebig G., 2008. Formaldehyde and chemosensory irritation in humans: A controlled human exposure study. Regul Toxicol Pharmacol, 50:23–36.

Marchand C, Bulliot B, Le Calvé S, Mirabel P., 2006. Aldehyde measurements in indoor environments in Strasbourg (France). Atmos Environ 40 (7):1336-1345. doi:10.1016/j.atmosenv.2005.10.027

http://lodel.irevues.inist.fr/pollution-atmospherique/index.php?id=6016

http://lodel.irevues.inist.fr/pollution-atmospherique/index.php?id=6016

http://www.inrs.fr/publications/bdd/fichetox/fiche.html?refINRS=FICHETOX_7



of 71

Marchand C, Le Calvé S, Mirabel P, Glasser N, Casset A, Schneider N, de Blay F., 2008. Concentrations and determinants of gaseous aldehydes in 162 homes in Strasbourg (France). Atmos Environ 42 (3):505-516. doi:10.1016/j.atmosenv.2007.09.054

Michelot N, Mandin C, Ramalhc O, Riberon J, Marchand C, Malherbe L, Ramel M, Personnaz MB, Delmas V, Urban S, Carrega M, 2011. Monitoring indoor air quality in French schools and day-care centres Results from the first phase of the pilot survey. Pollut Atmos (211):267-279

Monticello TM, Swenberg JA, Gross EA, Leininger JR, Kimbell JS, Seilkop S, Starr TB, Gibson JE, Morgan KT., 1996. Correlation of regional and nonlinear formaldehyde-induced nasal cancer with proliferating populations of cells: Cancer Res., v. 56, no. 5:1012-1022.

Mueller JU, Bruckner T., 2013. Exposure study to examine chemosensory effects of formaldehyde on hyposensitive and hypersensitive males, International Archives of Occupational and Environmental Health V. 86, Issue 1:107-117.

Nicolas M., Chiappini L., D’Anna B., 2013. Activités domestiques et qualité de l’air intérieur : émissions, réactivité et produits secondaires. Final report. April 2013.

WHO (World Health Organisation). 2010. WHO Guidelines for indoor air quality. Selected pollutants. WHO European Centre for Environment and Health, Bonn Office. WHO Regional Office for Europe. 484 p.

OQAI. 2004. Campagne pilote dans 90 logements et 9 écoles. Rapport synthétique - July 2004.

OQAI, 2006. Campagne nationale Logements : État de la qualité de l'air dans les logements français, Rapport final référencé DDD/SB- 2006-57, Kirchner S., Arenes J-F., Cochet C. et al. ; November 2006 – 165 pages

OQAI, 2011. Qualité d’air intérieur, qualité de vie. 10 ans de recherche pour mieux respirer. Ouvrage collectif. Kirchner S., Mandin C., Derbez M., Ramalho O., Ribéron J., Dassonville C., Lucas J-P., Ouattara M. ; August 2011 – 208 pages.

OQAI, 2016. Base de référence nationale sur la qualité de l’air intérieur et le confort des occupants de bâtiments performants en énergie : OQAI-BPE. Deuxième état descriptif de la qualité de l’airintérieur et du confort de bâtiments d’habitation performants en énergie. Mickaël DERBEZ,

Partanen T., 1993. Formaldehyde exposure and respiratory cancer - A meta-analysis of the epidemiologic evidence. Scand J Work Environ Health, 19, 8-15.

Paustenbach D, Alarie Y, Kulle T, Smith R, Swenberg J, Witschi H, Horowitz S. 1997. A recommended occupational exposure limit for formaldehyde based on irritation. J Toxicol Environ Health, 50:217-263.

Ramalho O, Wyart G, Mandin C, Blondeau P, Cabanes PA, Leclerc N, Mullot JU, Boulanger G, Redaelli M (2015) Association of carbon dioxide with indoor air pollutants and exceedance of health guideline values. Build Environ 93 (P1):115-124. doi:10.1016/j.buildenv.2015.03.018



of 71

Ritchie IM, Lehnen RG. 1987. Formaldehyde-related health complaints of residents living in mobile and conventional homes. Am J Public Health; 77(3):323-8.

Roda C, Barral S, Ravelomanantsoa H, Dusséaux M, Tribout M, Le Moullec Y, Momas I (2011) Assessment of indoor environment in Paris child day care centers. Environ Res 111 (8):1010-1017.doi:10.1016/j.envres.2011.06.009

Roux A.L., Maupetit F., Nicolas M., 2011. Émissions par les meubles destinés à la petite enfance : impact sur la qualité de l’air des crèches et premières réflexions quant à la mise en place d’une procédure d’étiquetage

Health Canada. Residential Indoor Air Quality Guideline: Formaldehyde Cat.: H128-1/06-432-1F. 3 pages. April 2006.

Trocquet C., et al., 2016. Résultats complémentaires projet IMPACT’AIR. La Rochelle 26.

Verriele M, Schoemaecker C, Hanoune B, Leclerc N, Locoge N Do low energy public buildings (LEPB) comply with the recent IAQ regulations in france? What about unregulated VOC? In: 13th International Conference on Indoor Air Quality and Climate, Indoor Air 2014, 2014. International Society of Indoor Air Quality and Climate, pp 608-614

Wolkoff P., 2016. External eye symptoms in indoor environments. Indoor Air., 27: 246–260. doi:10.1111/ina.12322

Weschler CJ, Nazaroff WW (2014). Dermal uptake of organic vapors commonly found in indoor air. Environmental Science and Technology 48(2), 1230-1237

Wyart G. et al., 2016. Rapport CSTB-OQAI/2016-010



of 71

Appendices Appendix I: Referral



of 71



of 71

Appendix II: List of members of the Working Group set up to respond to this self-referral

Working Group of the Environmental Health

Expert Committee (CSRE)

Working group chair Mr Luc Ferrari Professor of toxicology – University of Lorraine

Member of the Environmental Health Expert Committee (CSRE)

Working group members: Mr Olivier Blanchard Research Professor at EHESP in exposure assessment

Mr Dany Chevalier Lecturer of toxicology, University of Lille

Mr Pierre Deroubaix Ventilation Engineer /IAQ and energy poverty –

Construction & Urban Planning Department, ADEME

Mr Didier Febvrel Chief Physician, Marseille's Department for Public Health and Disabled People, CSRE Member

Mr Bruno Fouillet Lecturer of toxicology, University of Claude

Bernard Lyon 1

Mr Guillaume Karr Design and Research Engineer in Health Risk Assessment, INERIS

Ms Juliette Larbre Director, Chemical Pollutants Laboratory, Paris City Hall

Mr Laurent Madec Lecturer of Health, Safety

& Environment, University Paris 13 and EHESP, CSRE Member

Mr Fabien Squinazi Clinical Biologist, Former Hospital Biologist,

Former Director, City of Paris Hygiene Laboratory and Head of the Health, Safety and Environment Bureau, Paris City Hall

CSRE Member Working group members

Ms Marie Verriele Research Professor in Atmospheric Physical Chemistry,

SAGE, IMT Lille Douai



of 71

Ms France Wallet Medical Expert, Environmental Health Risk Assessment, Medical Research Department, EDF

Mr Denis Zmirou-Navier Honorary Professor, University of Lorraine

(Nancy Faculty of Medicine). Former Director, Occupational & Environmental Health and Health Engineering Department, EHESP.

CSRE Chair

Experts interviewed Ms Catherine Boutet Deputy Head, Health & Environment

Unit, Normandy Regional Health Agency (ARS) - Calvados Département-level Delegation

Ms Cécile Canesse Health Engineer, Hauts-de-France ARS



of 71

Appendix III:


Conditions

Week 1 Week 2 Week 3 Week 4 Week 5

Empty room

Room +

furniture

Room + furniture +

usual ventilation guidelines

Room + furniture +

specific ventilation guidelines -

OQAI

Room + furniture +

light indicator

Min. value

over the 7 6.7 10.7 < DL 1.1 1.8 Lavoisier days

School Max. value

Formaldehyde over the 20.5 34.9 41.1 31.6 32.3 Analyser 7 days

µg/m3 Average

value 14.7 26.3 25.2 20.5 23.4 24/7 Min. value

Grandes over the 7

days < DL 9.6 6.5 3.9 1.6

Varennes School Max. value

Formaldehyde Analyser

over the 7 days

50.0 58.6 50.2 42.8 42.0

µg/m3 Average value

29.5

39.6

31.0

25.6

27.4

24/7

Formaldehyde concentrations continuously measured with the analyser (IMPACTAIR project) – La Rochelle, 2016



of 71

City of Grenoble – Ethera (2018)

Campaign

Measuring point

Sampling period

Formaldehyde concentrations measured in µg/m3

Average concentration

measured throughout the

period

Average concentration

measured only when children are present

Reception (2-4 year olds), period 1

27/04/18 to 04/05/18 14 5


09/05/18 to 16/05/18 46 25

Old school Reception & Year 1

(3-5 year olds), period 1

25/04/18 to 04/05/18 24 8

n = 1 Reception & Year 1 (3-5 year olds),

period 2

07/05/18 to 11/05/18 51 24

Reception & Year 1 (3-5 year olds),

period 3

11/05/18 to 18/05/18 47 16

1st, room 2, period 1 26/02/18 to 02/03/18 5 11 1st, room 2, period 3 12/03/18 to 19/03/18 7 16

Old school no. 2

1st, room 2, period 4 21/03/18 to 27/03/18 6 7

1st, room 2, period 5 27/03/18 to 05/04/18 8 11

GF, period 1 26/02/18 to 02/03/18 2 5

GF, period 2 06/03/18 to 12/03/18 4 5

GF, period 3 12/03/18 to 19/03/18 3 6

GF, period 4 19/03/18 to 27/03/18 4 4

GF, period 5 27/03/18 to 05/04/18 4 6

New school


25/04/18 to 04/05/18 10 4


07/05/18 to 14/05/18 25 25

Reception doctor 27/04/18 to 04/05/18 13 13

Primary school years 2 & 3 (7-9 year

olds)

28/05/18 to 05/06/18 10 7

Recent

school


olds), class 1

07/05/18 to 14/05/18 39 14


olds), class 2

25/04/18 to 04/05/18 17 4

Summary of the formaldehyde concentrations measured in Grenoble schools – Ethera (2018)

benchmarks for managing indoor air quality for formaldehyde

Documents