According to the World Health Organization (WHO), 13% of deaths in the European Union (EU) in 2012 (last update) were linked to environmental factors [1]. This represents 630,000 deaths each year, of which 90% are due to noncommunicable diseases, the main ones being cancer, ischemic heart disease, chronic obstructive pulmonary disease (COPD), and stroke. In addition to deaths, WHO estimates that 20 million healthy life-years are lost worldwide each year due to environmental factors [1]. Since populations are generally exposed to multiple pollutants at once, the European Environment Agency underlines the importance of studying multiple-pollutant exposures and their impacts on health [2]. The 2 main environmental factors negatively impacting human health are air pollution and noise [2].

The main air pollutants are particulate matter (PM), ozone (O₃), and nitrogen dioxide (NO₂). Among these, PM_2.5 (aerodynamic diameter<2.5 µm) is the largest contributor, with 379,000 of the 400,000 premature deaths attributed to air pollution in the EU-28 (European Union with 28 member states) in 2018 [2]. This is despite a steady decline in PM concentrations in Europe since 1990 [2]. Between 2016 and 2018, more than 74% of Europe’s urban population was exposed to PM_2.5 levels exceeding the WHO threshold. PM affects the human body by causing respiratory problems, cardiovascular diseases, COPD, and even reproductive issues. Concerning indoor air quality, PM is also the first factor responsible for indoor pollutant–related diseases [2].

After air pollution, noise is the second-most important contributor to the environmental burden of disease in Europe [2]. Noise can affect human health either by damaging the auditory system (eg, hearing loss) or by inducing physiological stress (eg, sleep disturbance, impaired concentration with impacts on cognitive development, and metabolic effects potentially leading to cardiovascular diseases) [3,4]. In Europe, the main source of environmental noise is road traffic. Hänninen et al [5] estimated that 400-1500 disability-adjusted life-years (DALYs = years of life with disability + years of life lost) per million people are attributable to traffic-related noise. For example, Basner et al [4] reported that an increase of 10 dB (in LAeq, a weighted equivalent continuous sound pressure level) leads to a 7% increase in hypertension, a 15% increase in strokes, and a 17% increase in myocardial infections.

Extremely low-frequency magnetic fields (ELF-MFs) are fields produced by the movement of electrically charged particles. The “extremely low” range includes 50-60 Hz fields, which are the frequency generated by electricity distribution around the world. These fields, at chronic exposures of 0.3-0.4 µT, have in some studies been associated with an increased risk of childhood leukemia [6-9] and have therefore been classified as “possibly carcinogenic to humans” by the International Agency for Research on Cancer (IARC) [10]. To date, no causal relationship has been concluded, no action mechanism has been identified nor explained, and the Scientific Committee on Health, Environmental and Emerging Risks (SCHEER) 2024 report [11] concludes that evidence of this potential association is weak. ELF-MFs are also suspected to be linked to the development of neurodegenerative diseases, such as Alzheimer’s disease or amyotrophic lateral sclerosis (ALS) [12-16], but results are too mitigated and the evidence too weak to conclude a causal association [11,17]. Despite the absence of causal pathways, exposure to ELF-MFs and their potential health effects remain relevant due to an ongoing societal debate on perceived health risks.

Urban citizens are continuously and simultaneously exposed to a wide variety of environmental factors, such as noise, air pollution, and magnetic fields (MFs), that originate from different sources and can enter the human organism in different ways. Little is known about how simultaneous coexposure to multiple types of pollutants leads to interactive, or cocktail, effects. It is hypothesized that the interaction of different exposures may lead to antagonism, synergy, or the cumulation of effects [18]. For example, it has been observed that combined exposure to air pollution and high temperatures leads to higher levels of morbidity and mortality [2].

Urban populations are particularly affected by simultaneous exposures to noise and air pollution. Indeed, road traffic, denser in urban areas, is the main source of both poor air quality and high noise levels [2]. Moreover, both air quality and noise affect similar systems in the human body, such as the cardiovascular and endocrine systems [2].

In 2015, the Scientific Committee on Emerging and Newly Identified Health Risks (SCENIHR) underlined the possible role that ELF-MFs may play in interaction with other exposures [19]. Indeed, in the presence of 50 Hz MFs from high-voltage power lines, PM gets electrically charged and becomes “corona ions” potentially more dangerous to human health than nonelectrically charged PM [20]. Moreover, PM is suspected to be associated with similar kinds of diseases than ELF-MFs are, such as neurodegenerative diseases [21,22] or childhood leukemia [23,24].

Most studies on exposures to environmental pollutants (air and noise) use data monitored from fixed stations. They provide average estimates of exposure at the levels of geographical areas (eg, neighborhoods) through extrapolation of measurements at these locations. Although this makes it possible to study potential links between health and exposures via ecological studies, the modeled data reduce the real heterogeneity of individual exposures in each area. Several authors [18,25-27] have pointed out that this results in reduced statistical power and increased measurement error.

Collecting exposure data at the level of individual citizens has several advantages. It provides information about exposure levels, detailed to whether the participant is indoors or outdoors, indoors being where people spend up to 90% of the time [2,28]. It also provides insights into exposure at the level of microenvironments (eg, kitchens, living rooms, cars, parks), as has been conducted for air pollution [25,28-30], MFs [31], and noise levels [32]. This allows us to relate exposures to the personal profile of activities of the exposed individuals. As observed by Ma et al [32], the means of exposure (in their case to noise) can be different between citizens from the same neighborhood. Buonanno et al [29] found that women in their sample were more exposed to ultrafine particles than men because they spent more time in kitchens, whereas men were most exposed while in public transport. In addition, this level of detail helps identify potential factors associated with exposure. Sivanantham et al [33] related the multiple-pollutant exposure patterns of different schools to different characteristics, such as the age of the buildings, cleaning habits, or ventilation habits. Finally, individual exposure measurements usually have a higher frequency of measurement (eg, every minute or second), whereas fixed monitors usually measure only once an hour [34].

A literature review on exposure to environmental factors during travel by Poom et al [34] found that 60% of the studies included in their review were published after 2015 and that 65% of those studies used portable sensors to collect data. This shows a growing interest in environmental exposures and in collecting data at the individual level using portable sensors.

However, there are several limitations to individual data that need to be considered. The main challenge is that collecting samples that are representative of the population is nearly impossible. Indeed, participation is voluntary, and as pointed out by Nieuwenhuijsen et al [35], only highly motivated people will agree to wear devices for several hours or days. Canha et al [28] also remind us of the importance of using reliable devices validated by comparisons with reference measurement tools. In the end, despite their imperfections, these methods allow us to improve decision-making, while acknowledging inaccuracies and uncertainties [36].

This data collection at the individual and microenvironmental levels will be insightful for further studies on the impacts of multiple-pollutant exposures on human health, especially given the growing interest in studying exposures to pollution as they occur in real life, with multiple-pollutant exposures that occur simultaneously.

ExpoHealth-1 is an observational study collecting data on exposure to multiple pollutants (PM, noise, ELF-MFs) at the individual level for 490 inhabitants of the Brussels Capital Region. To the best of our knowledge, this is the first study of its kind in the region. Our objectives are to (1) assess daily exposures to PM, noise, and ELF-MFs at the individual level for a sample of inhabitants of the Brussels Capital Region, (2) assess the levels of exposure to these 3 pollutants in the main microenvironments encountered on a normal weekday and evaluate the contribution of each microenvironment in the trends of exposure measured, and (3) assess the potential role of real exposure in the reporting of perceived exposure. To this end, we collected (1) individual-level exposure data for PM, noise, and ELF-MFs over 24 hours distributed over 2 normal weekdays, (2) a microenvironment diary for each participant during the 24-hour measurement period, and (3) participant information, such as gender, age, socioeconomic status, perceived health, and perceived exposure via a self-administered questionnaire.

Because of the exploratory purpose of this study, the sample size estimated for the data collection was first based on feasibility in terms of the number of devices available, the timing, and the size of the field team. This led to a target of 600 participants. At the end of the data collection, we had enrolled 490 participants.

Data were collected with 3 main instruments: a questionnaire, a set of 3 pollution exposure measurement devices, and a microenvironments diary.

Data analysis will be mainly descriptive using R software version 4.2.3 (R Foundation for Statistical Computing). The final database will gather the following data for each participant: (1) exposure levels for noise, PM_2.5, and ELF-MFs in means per minute over 24 hours, (2) the microenvironment in which each exposure was measured, and (3) answers to the questionnaire, which combines personal information, perceived health, perceived exposure, and several life and home habits.

The study protocol was approved by the Erasme-ULB Hospital Ethics Committee (approval number P2018/454; approval date November 6, 2018). Consent was obtained from all participants, who signed the information and consent forms. Participants had the option to opt out at any time during data collection. Each participant’s data were associated with a pseudonym. A separate table linking each participant’s personal information (eg, address and phone number) to their pseudonym was created on a fixed computer and made accessible to the researchers’ team only. No compensation was offered to participants.

The objective of this study is to assess personal and simultaneous exposures to multiple pollutants (noise, PM_2.5, and ELF-MFs) over a normal weekday, with detailed exposure levels for the main microenvironments encountered (home, means of transportation, workplace).

In Brussels, daily averages of exposure to noise and air pollutants are freely accessible from the Brussels environmental agency (Bruxelles Environnement/Brussel Leefmilieu) [62] and the interregional environmental unit IRCELINE [63] websites, respectively. These exposures are measured by several fixed-site monitoring stations distributed in the region and then modeled to estimate levels of exposure in the entire region.

To the best of our knowledge, this study is the first assessment of Brussels’ inhabitants’ personal exposure to multiple pollutants. The results will provide estimates of personal as well as microenvironmental trends of exposure to noise, PM_2.5, and ELF-MFs for inhabitants of the areas studied. This is a first step toward a detailed picture of citizens’ profiles of multiple-pollutant exposures and toward identifying the main microenvironments affected by these 3 pollutants in Brussels.

The study results will be transmitted to the relevant public institutions: the 3 municipalities in which the data collection took place, the Brussels environmental agency (Bruxelles Environnement/Brussel Leefmilieu), and the federal agency for public health Sciensano.

Every participant in the study has already received their personal results in the shape of averages per 24 hours of measurements. Many of them have shown a strong interest in knowing the levels to which they are exposed in their daily lives. This shows the importance of citizen science.

The main strength of this study is that the exposure levels for several pollutants were collected at an individual level, which provides a much higher level of detail than extrapolated data. This, together with the microenvironment diary, the time of the day, and the profile of the participating population, will allow us to identify factors that can influence the levels of exposure, and therefore to highlight concrete targets to reduce exposures. As the measurements were taken during typical days of the week, they reflect the real-life exposures of the participants. Finally, the study focuses on 3 types of daily exposures measured simultaneously. This responds to the needs underlined by the European Environmental Agency to focus on multiple-pollutant exposures [2].

The main limitation of the study is the lack of representativeness of the participants due to the nonprobabilistic sampling method. As participation in the data collection was voluntary and without any form of reward, the participating individuals appeared to show similar socioeconomic levels, especially in the level of education received. This makes it impossible to test for differences in socioeconomic status. Moreover, no causal relationship will emerge from this study because of its cross-sectional design.

Second, data collection took place only once, during a 24-hour period, for each participant. This will prevent us from drawing any conclusion about long-term exposure.

Moreover, to gather these many data on multiple-pollutant exposures at the individual level, the best option is often low-cost measuring devices. Although the noise and ELF-MF levels measured will be corrected with a simple correction factor, the levels of PM_2.5 measured with Airbeam 2 will need a corrective equation. Indeed, these devices show systematic underestimations of PM_2.5 concentrations. A corrective equation adapted to the equipment used in the study is being developed from linear models after colocations of the Airbeam 2 devices and a reference instrument (DustTrak 8534 TSI).

In our data collection for exposure to PM, we did not include a real-time measurement of physical activity, as has been done by other authors [25,64]. This would have allowed us to assess the inhalation rate and therefore the true exposure to air pollutants.

During this data collection, a new model of Airbeam was launched. Airbeam 3 seems to be significantly more efficient and accurate than Airbeam 2. This newer version would have potentially improved our data quality.

Finally, this study focuses on measured exposures and perceived exposures. A cohort design would be more relevant to explore potential links between multiple-pollutant exposures and health impacts.

This paper aims to describe, with a high level of detail, the methodology and protocol for the assessment of individual exposure to noise, PM, and ELF-MFs for 24 hours on a normal weekday, applied to 490 inhabitants of the Brussels Capital Region. We hope that this detailed description of the protocol will make it easily reproducible, foster further improvements to multiple-pollutant exposure methodologies, and permit a wider use of the database produced. The database created will be shared after the publication of the first results. Characterizing the multiple-pollutant exposures faced by individuals in different microenvironments in Brussels could lead to improved understanding of the sources of exposure and contribute to creating better policies to face multiple-pollutant exposures.