A total of 14 of the 19 publications met the criteria for meta-analysis (Table 1). With the available data, pooled effects were possible to estimate for asthma (Fig. 1) and wheezing outcomes (Fig. 2), as well as a combination of asthma and asthma-related symptoms, including wheezing and exercise-induced wheezing, cough and night cough, and dyspnoea (Fig. 3). Heterogeneity between studies seems moderate for all measured outcomes when accounting for the different parameter subsets, with the association between PM10 and wheezing being the only subgroup resulting in an I2 > 50%.

Fig. 1.

Forest plot representing the pooled effects for the association between indoor air pollutants and asthma, resultant from the meta-analysis of the selected studies in Portugal.

(0.47MB).

Fig. 2.

Forest plot representing the pooled effects for the association between indoor air pollutants and wheezing, resultant from the meta-analysis of the selected studies in Portugal.

(0.4MB).

Fig. 3.

Forest plot representing the pooled effects for the association between indoor air pollutants and asthma or asthma related symptoms, including wheezing and exercise-induced wheezing, cough and night cough, and dyspnoea, resultant from the meta-analysis of the selected studies in Portugal.

(0.45MB).

The pooled estimate (and 95% CI) for the effect of indoor air quality on asthma was 1.13 [0.98 : 1.30], with significant associations found for CO2 (1.09 [1.00 to 1.19], based on 4 studies), ETS (2.90 [1.90 to 4.42], single study) and PM10 (1.24 [1.00 to 1.55], 2 studies). Regarding wheezing, the pooled effect estimate was 1.04 [0.99 to 1.10), with significant associations found for CO2 (1.04 [1.01 to 1.07], 4 studies), ETS (4.20 [1.13 to 15.68], single study), formaldehyde (0.69 [0.50 to 0.96], single study) and TVOC (1.05 [1.00 to 1.11], 4 studies). Finally, the pooled effect estimates showed a significant association between IAQ parameters and asthma or asthma-related symptoms (1.08 [1.03 to 1.13]), with significant contributions from CO2 (1.08 [1.02 to 1.14], 13 inputs from 4 studies), ETS (3.00 [2.01 to 4.48], 2 studies), fungi (2.05 [1.13 to 3.72], 3 inputs from 2 studies) and TVOC (1.04 [1.01 to 1.07], 9 inputs from 4 studies). On the other hand, the only study performed in Portugal that explored the associations between indoor formaldehyde concentrations and respiratory health surprisingly reported a significant protective effect (0.68 [0.51 to 0.91], 2 inputs from a single study).

Four reviewed studies focused on ETS, which is recognised as one of, if not the primary source of IAQ pollutants with highly documented evidence of causing respiratory morbidity. This has been reinforced by one of the reviewed studies showing that female individuals with asthma were more likely to be exposed to ETS when compared with those without the disease (OR [95% CI] = 2.90 [1.90 to 4.41]).30 It used data collected in 2010 from the Inquérito Nacional sobre Asma (The Portuguese National Asthma Survey), a cross-sectional study conducted through telephone interviews based on the GA2LEN questionnaire,45 targeting a randomised sample of individuals living in Portugal.46

Interestingly, a different study found converse results when focusing exclusively on children, showing no significant differences in secondhand exposure to ETS in indoor environments between asthmatic and non-asthmatic children.27 This survey also started in 2010 and was conducted by administering a self-reported questionnaire to 4th grade students attending schools in nine regions of Portugal.

The two remaining studies were centred on the effects of the smoke-free law that prohibited smoking indoors in public settings. The first was published in 2009 and refers to the Spanish smoke-free law, implemented in 2006, roughly two years before the law being applied in Portugal.28 It consisted of a longitudinal study that followed Spanish hospitality workers for 12 months after implementation of the regulation mentioned above. However, the article still provides a relevant perspective since it used Portuguese workers as controls in a timeframe when the law was still to be implemented in the country. Results demonstrated that, 12 months after smoking was banned, hospitality workers in Spain showed a significant decrease in the prevalence of respiratory symptoms. Although the prevalence also decreased in Portuguese workers, results were not statistically significant.28

In 2013, a study monitored IAQ in 10 Portuguese restaurants pre- and post-smoke ban (2 years since law implementation), as well as changes in respiratory symptoms and comfort parameters reported by 28 individuals working in those restaurants.29 Results showed that banning indoor smoke in these restaurants significantly reduced PM concentrations, total VOC (TVOC), benzene and CO. In addition, workers experienced fewer irritative symptoms after the ban, but there was no significant change in dyspnea.

Forty five day care centres from Lisbon and Porto, the two largest cities in Portugal, were included in a study aiming to evaluate the associations between indoor CO2 concentrations and symptomatology in preschool children.26 Results showed that CO2 was significantly associated with a higher prevalence of wheezing in children (1.04 [1.01 to 1.07]). Authors could also inversely correlate indoor CO2 in daycare centres with good ventilation practices, such as opening the windows.26 This result emphasises the need for instructing school and day care staff on ventilation behaviours to maintain adequate IAQ.

In a second phase of the same study, part of the ENVIRH project, the research group performed a thorough IAQ in 19 of the 45 day care centres with the worst ventilation practices.25 Authors found CO2 to be significantly associated with a diagnosis of asthma (1.10 [1.00 to 1.20]), while TVOC and concentrations of Der p1 in dust were associated with wheezing in the previous 12 months (1.06 [1.01 to 1.11] and 1.06 [0.99 to 1.12], respectively).

The first study to measure the associations between IAQ and respiratory health in adolescents in Portugal was published in 2008 and was part of the EPITeen project.41 Capitalizing on the previously validated Portuguese version of the International Study of Asthma and Allergies in Childhood (ISAAC) questionnaire, and after measuring temperature, relative humidity, CO2 and TVOC in 9 middle schools (7th, 8th and 9th grade, mean age 14.0) in the city of Porto, authors found significant associations between high levels of CO2 in classrooms and exercise-induced wheeze (1.86 [1.20 to 2.89]), hence corroborating the previously mentioned results of the ENVIRH project, in younger children.26 On the other hand, there were no significant associations between TVOC and respiratory symptoms.41

In another study, CO2 levels were assessed in 81 classrooms in 51 primary schools in Coimbra.40 Again, CO2 was above the reference level (1625 ppm for naturally ventilated buildings, according to Portuguese law) in most classrooms during Autumn and Winter, and was significantly correlated with lack of concentration (p = 0.002). No associations between respiratory health and CO2 were reported.

The first studies reporting the clinical outcomes of the Portuguese ARIA project were progressively published in 2017.38,42,43 The project aimed to study the associations between IAQ at schools in Porto and the development of allergic diseases and asthma in children. The project was distinguished for merging a complete evaluation of indoor air parameters in 71 classrooms from 20 primary schools with a comprehensive clinical evaluation of 858 children attending those classrooms, including skin-prick tests, spirometry with bronchodilation, exhaled nitric oxide (eNO), pupillometry, exhaled breath condensate (EBC) collection and anthropometric measurements. The children's legal guardians were also asked to fill out an ISAAC-based self-reported questionnaire. By analysing data from the ARIA project, significant associations were identified between high fungal diversity in classrooms and a lower risk of allergic sensitisation in children (0.60 [0.40 to 0.92]).38 However, no associations between microbial diversity and asthma were found.

Specific VOC were also measured in the ARIA project, with researchers being able to significantly associate a pattern of VOC – characterised mainly by high levels of toluene, o-xylene, m/p-xylene, benzene, ethylbenzene and styrene – with respiratory symptoms and asthma in children, particularly those with the obese asthma phenotype (1.43 [1.01 to 1.98]).42,47

Another publication on the same study population showed that children exposed to higher concentrations of particulate matter (both PM2.5 and PM10) were at a significantly higher risk of developing asthma, but this association was mediated by dietary practices (interaction effect = 1.44 [1.01 to 2.21] and 1.29 [1.03 to 1.68], for PM2.5 and PM10, respectively).39 In other words, children with a pro-inflammatory diet – according to the dietary inflammatory index – were less protected from the effects of PM2.5 and PM10 as asthma determinants, when compared with those under an anti-inflammatory diet.

Temperature, relative humidity and PM2.5 were also associated with a decrease in EBC pH; another team study showed.43 As a more acidic EBC has been associated with asthma.,48 these IAQ parameters may eventually contribute to a particular mechanism in its pathophysiology.

In 2020, the most recent study comprising data from the European project SINPONIE was published.36 This project aimed to assess the IAQ and built environment in 115 European schools, including Portuguese institutions, and study possible associations with symptomatology in 7089 attending children.49 After dichotomizing the concentrations of pollutants by their median, results showed that classrooms with higher concentrations of PM2.5, benzene, limonene, ozone and radon in classrooms were associated with increased odds for upper and lower airways symptoms. A significantly higher risk of lower and upper airways symptoms was also observed for the classrooms within the highest tier of VOC concentrations.

The final study in this section is part of the INAIRCHILD project,50 which followed a similar strategy as the ARIA project, but also included pre-scholar infants (3 to 5 years old). Results showed that inhalable doses of NO2 were significantly and positively associated with current wheezing (1.27 [1.02 to 1.59]), and PM2.5 and O3 exposure was associated with reduced FEV1 in primary school children (1.90 [1.11 to 3.25] and 2.38 [1.23, 4.63], respectively), but there was no identifiable relationship between inhalable doses of any indoor air pollutant and asthma.37

A 2020 study evaluated associations between IAQ and wheezing in 131 families with newborns (0 to 36 months) from the Arco Ribeirinho region, south of Lisbon, across the Tejo river.31 Building characterisation was conducted, and the children's bedrooms were monitored for TVOC. Results showed no associations between indoor TVOC and wheezing in children (0.77 [0.20 to 3.02]). Unfortunately, no specific volatiles were measured during this study.

The only other identified study focusing on residential IAQ was part of the SMALL CELL project, which included one Portuguese hospital (Centro Hospitalar do Porto).33 Patients with small cell lung cancer were recruited, and 63 had their residences monitored for radon concentration. Authors found that radon concentrations in stage IV patients were significantly higher in individuals diagnosed at later ages (p = 0.032), suggestive of the long-term effect of radon exposure demonstrated in previous studies.51 These results were complemented in a later publication from the same group, where authors showed that individuals with concentrations higher than 147 bq/m3 were at a significantly higher risk of developing small cell lung cancer (2.08 [1.03 to 4.39]).32

In old-age, people are also highly susceptible to indoor air pollutants.52 A significant part of this population spends approximately 90% of their time in nursing homes, which reflects a pronounced opportunity for long-term exposure to indoor air pollutants. This is particularly important to the Portuguese perspective, since 19% of the population is aged over 65 years old. 53,54

A study which evaluated the influence of IAQ on the respiratory health of 668 residents from 21 long-term nursing homes showed PM10 to be significantly associated with allergic rhinitis (2.87 [1.14 to 7.24]) reported through questionnaire interview.35 However, no significant associations were observed for asthma-related symptoms, such as wheezing or coughing.

A later study with a similar objective was conducted in 33 nursing homes from Lisbon and 20 from Porto.34 In this study, a clinical evaluation of the participants was performed, including lung function measurement, EBC analysis and nitrite level quantification. Results showed that each 100 mg/m3 increment of TVOC was associated with a 5% average increase in the odds of respiratory infection in the previous three months (1.05 [1.00 to 1.09]). It was also shown that PM2.5 concentrations were inversely associated with EBC pH values (β [95% CI] = -0.04 [-0.06 to -0.01]), thus supporting the observations in the previously discussed study by Rama et al.43 In addition, total bacteria concentrations were significantly associated with an increase in FEV1/FVC (β [95% CI] = 0.22 [0.08 to 0.37]) and a decrease in FVC (β [95% CI] = -0.53 [-0.87 to -0.20]), when adjusted to respiratory disease.34 Nevertheless, it is necessary to states that each pollutant was fitted into models adjusted for different confounders or in combination with other covariates, with only significant effect estimates being shown in completely separated models. For this reason, caution is advised when interpreting these results.