Severe volcanic SO₂ exposure and respiratory morbidities in the Icelandic population – a register study

Introduction

The Holuhraun volcanic eruption in the Barðarbunga central volcanic system in the fall and winter of 2014-2015 was the largest eruption in Iceland since the Laki eruption in 1783-1784. Some 12 million tons of sulphur dioxide, SO₂, was emitted from the eruption and the lava field,(Gíslason et al. 2015) and was dispersed widely over Iceland according to meteorological conditions,(Ilyinskaya et al. 2017) reaching the capital area, some 250 km from the eruption site where the 24-hour air quality guideline limit for SO₂, 125 µg/m³ (WHO 2007) was exceeded repeatedly during the fall of 2014.. A previous study of professionals working at the eruption site with very high exposures experienced no serious health effects after exposure, perhaps because they were wearing protective equipment such as masks.(Carlsen et al., 2019). Meanwhile,population-level health effects remain to be investigated, in particular in sensitive populations such as children and the elderly.

SO₂ exposure at concentrations over 500 µg/m³ is associated with irritation of the respiratory tract in susceptible individuals.(WHO 2007) At higher concentrations, SO₂ exposure can trigger respiratory symptoms such as acute bronchial asthma, pulmonary oedema, and respiratory distress – especially in individuals with hyper-reactivity syndrome.(Nowak et al. 1997; WHO 2007, EPA 2008) Exposure to SO₂ from active volcanoes is associated with increased rates chronic cough and phlegm and dry and sore throat.(Ishigami et al. 2008; Iwasawa et al. 2009, 2010; Kochi et al. 2017; Longo 2013; Longo et al. 2010, 2008; Longo and Yang 2008) Fatalities from very high SO₂-related exposure near volcanoes have also been reported.(IVHHN.ORG) While some of the previous studies report dose-response relationship between SO₂ and respiratory symptoms, their study designs and methods leave most of them prone to bias as symptoms were most often self-reported and participants were aware of their exposure status.(Hansell and Oppenheimer 2004) Moreover, most of the existing literature pertains to long-term area-wide exposure whereas SO₂ exposure in Iceland during the Holuhraun eruption was intermittent with few hours or days of high SO₂ concentrations followed by periods of low SO₂ concentrations when wind directions changed.(Ilyinskaya et al. 2017)

With population-based registers on medicine dispensing and health care utilization as well as vigorous monitoring of air pollution in the capital area, Iceland provides an ideal setting for studying health effects of exposure to SO₂ from volcanic eruption. The objective of this study was therefore to study the acute health effects of exposure to volcanic SO₂ on respiratory health in the general population of Iceland. We hypothesized that respiratory health care utilization due to respiratory system illness would be increased in the general population following days with SO₂ concentrations above the air quality guideline value of 125µg/m³ per 24 hours. Furthermore, we investigated the associations of health outcomes with continuous exposure, the lag-times of the effects and effects in age groups and respiratory infections and obstructive respiratory disease.

The Holuhraun volcanic eruption in North-East central Iceland began 31 August 2014 and ended 27 February 2015. The study period was 1 January 2010 – 31 December 2014, and the time before the eruption was used a reference period. The Holuhraun eruption persisted for until end of February 2015, whereas our study period ends 31 December 2014 due to a change in the database recording of events. However, SO₂ never exceeded the air quality guideline limit during January and February 2015, although daily mean concentrations were still higher than before the eruption.

The mean population of Iceland during the study period was 320 000 inhabitants. The capital area, Reykjavík and surrounding municipalities, had 205 282 residents at the beginning of the study period, and 215 965 residents at the end.(Statistics Iceland 2017) The analysis was restricted to the capital area (residential postcodes 101-171, 200-225, and 270) where adequate information about SO₂ exposure was available for the whole study period. The Icelandic health care system is state-centred, mainly publicly funded system with universal coverage. (Sigurgeirsdóttir et al. 2014) We obtained data on respiratory health from 1)) the National Medicines Register; 2) Primary care centres (that function as first point of contact) and 3) Landspitali, the national university hospital, the country’s centre of clinical excellence (Sigurgeirsdóttir et al. 2014). All three registers are held by the Icelandic Directorate of Health and extraction is subject to approval from the Icelandic Bioethical Committee. From the National Medicines Register we extracted data on dispensing (pharmacy sales to individuals) of prescription anti-asthma medication (AMD) classified by Anatomical Therapeutic Chemical code R03.(WHO 2015b) This class of drugs are prescribed for relieving symptoms of asthma and chronic obstructive pulmonary disease, COPD, and in some cases to relieve breathing difficulties in individuals with respiratory infections. Asthma medication dispensing is a proxy for respiratory health in a population and mild disease episodes (Menichini and Mudu, 2010; Conti et al., 2015; Furu et al., 2010). The data were extracted along with individual data on residence (postcode), age, sex and an anonymous personal identification number. From the primary care centers (PCC) and hospital emergency department (HED) databases at the Directorate of Health we extracted data on individuals diagnosed with respiratory illnesses (International Classification of Disease, ICD code J (WHO 2015a)) along with data on residence (postcode), age, sex, and an anonymous personal identification number. As the same bout of illness is likely to result in recurring health care contacts we included only the first instance of an individual registration within a 14-day period for the same diagnosis category to avoid exposure misclassification with respect to the timing of the outcome. For primary care, we analysed the number of MD visits (PCMD) in the main analysis (the total number of contacts including phone calls and consultations were analysed in a sensitivity analysis). For HED, all visits were included regardless whether the individual was admitted to hospital. For each outcome, we constructed daily time series starting 1 January 2010 to 31 December 2014 for the following age groups; children (under 18 years of age), adults (18-64 years), and elderly (age 65 years and above).

We obtained SO₂, PM₁₀, and NO₂, data along with meteorological data from the Icelandic Environment Agency’s stationary air pollution monitor located in Reykjavík (Figure 1) for the study period and constructed a time series of daily mean values.

• Download figure
• Open in new tab

Figure 1. The Icelandic capital region with eruption site indicated on the inset map.

Results

The daily mean SO₂ concentrations in Iceland’s capital area were low or moderate, until the Holuhraun eruption began 30 August 2014 (Table 1, Figure 2, Figure S5). After the eruption began, SO₂ concentrations surpassed the air quality guideline limit (125 µg/m³) on ten days, and both mean and median concentrations were significantly higher than before the eruption began.

• Download figure
• Open in new tab

Figure 2.

Respiratory health outcomes during the exposed period July to December 2014 (solid lines) and the unexposed previous years 2010-2013 (broken lines) for Capital area anti-asthma medication dispensing (top), primary care contacts (middle) and hospital emergency department visits (bottom) for respiratory diseases as 7-day running means. SO₂ concentration (3-day running means) indicated in grey, broken vertical line mark the beginning of the eruption, grey vertical line denote first nation-wide SO₂ warning. Grey shaded areas indicate MD strike in PCC and HED.

Before the eruption, an average of 129.4 individuals per day were registered purchasing (=were dispensed) AMD. The individuals, were mostly female (56.3%) and the mean age was 35.7 (SD 22.9) years. After the eruption began, there were an average of 158.4 individuals per day. The mean age was 44.5 (SD 26.3) years. Before the eruption, 106 528 individuals, attended primary care s for respiratory disease, the mean age was 32.8 (SD 22.5) years and 60.9% were women. After the eruption began, 15 806 individuals attended primary care for respiratory diseaseand the mean age was 35.0 years (SD 26.6), 60.4% were women. The most common form of contact was GP visits (61%) and phone calls (33%) (See flowchart in supplement for detail of the selection process). In HED, 19 541 individuals were registered seeking care for respiratory health outcomes before the eruption. The mean age was 36.3 years (SD 29.4) and 49.8% were women. After the eruption began, there were 1 932 individuals visited HED with a mean age of 39.4 years (SD 30.5) and 51.8% women (Table 1, Figure 2).

After the eruption began, SO₂ was significantly higher than before; 35.7 µg/m³ (SD 71.2) vs 1.4 µg/m³ (SD 1.1. The daily number of number of individuals with AMD was significantly increased compared with the reference period (129.4 vs 158.4, p<0.001). Neither total nor age categories of HED and PCMD MD visits were significantly higher or lower during the eruption period as a whole compared to the period before (Table 2, Figure 2).

Regression analysis results

In the adjusted regression analyses of the whole study period 2010-2014 we found that after adjusting for other pollutants and weather, exposure to SO₂ at short lags (lag 0-2, figure S2) was associated with an estimated increase in the number of AMDs to individuals of all ages by 1.05% (95% CI 0.48 – 1.62%) per 10 µg/m³ SO₂. For children 0-17 years of age, the estimate was higher, 1.51% (95% CI 0.63 – 2.40%) and for individuals 65 years and older, the estimate was lower, 0.77% (95% CI 0.05 – 1.51). SO₂ concentrations exceeding the air quality guideline of 125µg/m³, were associated with a statistically significant increase in dispensing of AMD at lag 0–2 by 13.5% (95% CI 5.3-21.7%), the increase was 20.4% (95% CI 8.0-32.9%) in individuals under 18 years of age (Table 3). In the age-stratified analysis, the association effect estimates were highest in children under 18 by 20.4% (95% CI 8.0 - 32.9%) and with a lower estimate for elderly, 8.8% (95%CI -1.9 – 19.5%) which did not reach statistical significance.

For PCMD and HED, including an indicator variable for days where MDs were on strike improved the model fit and was an effect modifier which increased the effect estimates, so this variable was added to the PCMD and HED models. PCMD visits increased by 1.52% (95% CI 1.04 – 2.00%) per 10 µg/m³ SO₂ at lag 0–2 for all ages, and stratifying by age, the increase was highest in adults.

For PCMD, short term exposure to SO₂ over the air quality guideline value was associated with increases by 21.6% (95% CI 15.1 – 28.6%). The increase was highest in adults by 24.1% (95% CI 16.8 – 31.3%).

For HED visits, SO₂ exposure was associated with increased risk for individuals older than 64 at lag 0– 2 (Figure S1) by 1.54% (95% CI 0.02 – 3.07%) per 10 µg/m³ (Table 3). For exposure over the air quality guideline limit, the effect estimate significantly increased only in elderly by 26.3% (95% CI 5.56 – 47.0%).

Stratifying by diagnosis category, SO₂ exposure was associated with increased PCMD visits for respiratory infections at lag 0–2 by 1.35% (95% CI 0.80 – 1.89%) per 10 µg/m³. PCC contacts for obstructive disease were increased by 2.48% (95% CI 1.57 – 3.39%). Following air quality guideline value exceedances, the estimated increase in PCC MD visits were 19.0 (95% CI 10.5 – 28.1%) and 34.3% (95% CI 18.5 – 52.2%) for respiratory infections and obstructive disease, respectively. For HED visits, SO2 exposure was not associated with significant increases in visits due to respiratory infections or asthma or COPD although all estimates were positive (Table 4).

In the sensitivity analysis, the unadjusted associations between SO₂ exposure and health outcomes was only marginally different (<10 percentage point change) from the adjusted results (Table S1) with the exception of the estimates for HED visits which decreased from 1.12% (95% CI 0.19 – 11.2%) to 0.84% (95% CI -0.08-1.76%). However, the estimated association with HED visits in elderly increased from 1.54% (95% CI 0.02 – 3.70%) in the adjusted analysis to 1.70% (0.26 -3.16%) in the unadjusted analysis.

Discussion

In this study, a comparison of health care utilisation in Iceland’s capital are before and after the eruption began showed that only AMD increased significantly during the Holuhraun eruption, but primary care and hospital care was not higher during the eruption than before. However, in adjusted time series regression analysis, SO₂ concentrations were associated with increased AMD, MD visits in primary care for respiratory causes and hospital emergency department visitsat lag 0-2. Using a cut-off or continuous measure of SO₂ resulted in largely similar estimates (Table 3). The effect estimates for specific disease categories indicate that individuals with obstructive lung disease were particularly affected (Table 4).

In the regression, our choice of lags were informed - exploratory analysis (Figures S1). For AMD and PCC visits the observed increase occurred during the same day and up to two days after a peak in SO₂ concentration. HED visits increased one to three days after a peak in SO₂ which could indicate that primary care is the first point of contact for a majority of individuals and hospital care is sought only after other care options have been exhausted. In the compliance analysis, the effect of initial warnings of volcanic gas exposure appeared to be limited, as excluding the first day or week of high SO₂ yielded similar estimates for AMD and PCC visits, although there was some loss of statistical power. This indicates that the increase reflected actual increased respiratory morbidities rather than only compliance as it persisted after first exposure (Supplemental Table S2). We present both unadjusted results (Table S1) and results adjusted for other air pollutants (Tables 3 and 4) as well as plotted an exposure-response function with a spline. Comparing the results from the continuous analysis and the cut-off analysis, we found little indication of significant non-linear effects, which is also observed in the spline plot of the exposure-response function (Figure S2). Our results are mostly consistent with those from previous studies of SO₂ exposure in urban settings; e.g. respiratory mortality rates were estimated to increase by 2.4% per 27 µg/m³ SO₂ (Li et al. 2017), corresponding to a point estimate of 0.88% per 10 µg/m³ SO₂, which is lower than the estimates for HED and PCMD estimate of respiratory morbidity associated with SO₂ found in this study; 1.12% (95% CI 0.19-2.05%) and 1.52% (95% CI 1.04 – 2.00%) per 10 µg/m³ SO₂, respectively. In previous studies of SO₂ exposure during volcanic eruptions, the SO₂ concentration on Miyakejima Island increased after Mount Oyama erupted in 2000. After evacuating the island at the time of the eruption, residents and aid workers returned from 2005 and onwards. Aid volunteers (n=611) reported respiratory irritation and the rates were associated with SO₂ exposure - particularly among women and non-smokers.(Ishigami et al. 2008) During the same eruption, children, who were exposed to daily mean concentration of 125 µg/m³ SO₂, had increased rates of wheezing.(Iwasawa et al. 2010) In a follow-up study 2006-2011, permanent residents of Miyakejima Island (n=168) who lived in areas with SO₂ reported increased rates of cough and wheeze few effects.(Kochi et al. 2017) None of the studies of people exposed at Miyakejima Island found adverse effects on lung function.(Ishigami et al. 2008; Iwasawa et al. 2009, 2010; Kochi et al. 2017) The Kilauea Volcano in Hawaii has been erupting continuously since 1983 with periods of increased eruption activity from 2008. SO₂ concentrations spiked from a daily mean concentration of 72 µg/m³ to 215 µg/m³ in a populated area downwind during this period and the rates of cough, and acute pharyngitis diagnosed at a clinic in the exposed community increased subsequently.(Longo et al. 2010) A cross sectional study found that those living in areas exposed to SO₂ had increased risk of cough, phlegm, and hay fever. A qualitative survey of subjects from the same study population revealed that participants reported that symptoms disappeared 7-10 days after leaving the exposed area, but reappeared one week after returning.(Longo et al. 2008) In a study of individuals exposed at the Holuhraun eruption site we found that while some reported respiratory symptoms during their stay at the eruption site, most symptoms had resolved and lung function was normal at a clinical examination one to six days later.(Carlsen et al. 2019) Regarding individual susceptibility, we observed the highest effect estimates for PCMD visits for asthma and COPD (Table 4). The effect estimate for HED visits were also increased but did not reach statistical significance. In controlled trials, hyper-responsiveness to SO₂ has previously been reported as common (20-25%) in individuals with positive asthma test (methacoline test).(Nowak et al. 1997), indicating that they be particularly vulnerable to severe SO₂ exposure. Similarly, respiratory infections were increased in PCMD visits and tended to be increased for HED visits, although the increase did not reach statistical significance. Finally, an association between SO₂ and hospital visits for upper respiratory tract infections has previously been reported(Li et al. 2017).

Previous studies of health effects of air pollution in Iceland have yielded lower effect estimates of daily air pollution on morbidity than the current study,(Carlsen et al. 2012b) other air pollution exposure types are thus not likely to bias the results. This includes H₂S, which has a low correlation (<0.1) with the exposure of interest during the study period (data not shown). During the reference period, there were two ash-rich volcanic eruptions, in Eyjafjallajökull 2010 and Grímsvötn 2011. While ash from Eyjafjallajökull had local respiratory health effects(Carlsen et al. 2012a) and there may have been adverse health effects in the capital area,(Carlsen et al. 2015) neither eruption had significant SO₂ emissions and sensitivity analysis excluding this period yielded similar results (Supplemental Table S2). Our study employs a time series design were individual level risk factors are not time-dependent and thus should not confound the association between short term exposure to air SO₂ and health outcomes, leaving bias due to unmeasured confounders, seasonal variation, and other intermittent exposures as main concerns. The exposed period October and November of 2014 did not coincide with any viral respiratory illnesses (e.g. influenza and RS-virus) epidemic,(Tilkynningarskyldir sjúkdómar 2018) but it did coincide with the MD labour conflict which resulted in lower PCC and HED attendance during those days, Hence, results from the analysis comparing the eruption period with the time before may have underestimated of true effect of the SO₂ from the eruption.

SO₂, PM₁₀ and NO₂ data was missing for a number of days which were excluded from the analysis. As the volcanic plume effectively changed the chemical composition of the atmosphere during the eruption period (Supplemental Tables S4a-b), the correlations of SO₂ with other air pollutants were altered after the eruption (we present unadjusted results in the supplement (Supplemental Table S1)). The results from these are nearly identical to the main analysis (Table 3), indicating that the effect of the very high concentrations of SO₂ was unambiguous.

It is a strength of the current study that health data were collected prospectively from population-wide registers, which minimizes the risk of information bias from individuals knowing their exposure status. Although the exposure would have been known to the public for at least part of the exposed period but we attempt to addresse this source of bias in the sensitivity analysis, showing only moderate changes to the results (Table S3). As a study outcome, we use dispensing of asthma-medication, a novel and more sensitive proxy than primary care attendance and hospital visits. It measures morbidity in individuals who in most cases already are in contact with the health care system. However, this outcome has been used in other studies and is known to be a proxy for asthma morbidity in a population (Furu et al,. 2010; Naureckas et al., 2005) and be associated with air pollution (Menichini and Mudu, 2010; Fattore et al., 2016) A limitation with register-data is that it is not collected for research purposes and diagnoses given in the health care system could be biased towards overestimation as medical professionals assume respiratory outcomes to be more likely during the eruption. However, as the estimated effect are similar across all respiratory outcomes we conclude that this sources of bias is not likely to fully explain our findings.. By basing exposure status on residential postcode we assume that people spend most of their time at home, however, outcomes in individuals not in the exposed area during the study period would bias the results toward the null. Reykjavík and the Icelandic capital area was exposed to SO₂ concentrations above 125µg/m³ during a total of ten days which occurred mostly during October and November 2014. This limits the statistical power of the study and our options for further analysis, as does the fact that our study period did not extend until after the eruption, meaning that we cannot fully assess the importance of population increase during the study period on our results.

In conclusion, this comprehensive study with prospectively collected data on volcanic air pollution exposure and respiratory outcome, is the first to firmly establish an association between spikes of high SO₂ concentrations and respiratory outcomes in the general population, particularly in individuals with prevalent respiratory disease. These findings emphasise the need for attention from authorities and parents to the health susceptible individuals during times of volcanic eruptions.