{"paper_id":"1b5bffd7-3dbc-4b8f-bd8d-058153cd07ac","body_text":"All-cause and cardiac mortality in \nrelation to smoke exposure nine \nyears after the Hazelwood coal mine \nfire \n \nAuthors \nKaren Walker-Bone1,2, Caroline X. Gao2,3, Catherine L Smith2, Tingting Ye4, David \nBrown2, Natasha Kinsman1,2, Jillian F Ikin1,2, Matthew Carroll5, Michael J Abramson2, \nYuming Guo4, Tyler J Lane1,2 \nAffiliations \n \n1 Monash Centre for Occupational and Environmental Health School of Public Health \nand Preventive Medicine, Monash University, Melbourne, Victoria, Australia \n2 School of Public Health and Preventive Medicine, Monash University, Melbourne, \nVictoria, Australia \n3 Orygen, Centre for Youth Mental Health, The University of Melbourne, Parkville, \nVictoria, Australia \n4 Monash Climate, Air Quality Research (CARE) Unit, School of Public Health and \nPreventive Medicine, Monash University, Melbourne, Victoria, Australia \n5 Monash Rural Health Churchill, Monash University, Churchill, Victoria, Australia \n \nKey words: All-cause mortality; cardiac mortality; coal mine fire; air pollution; \nparticulate matter and PM2.5   \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \nNOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.\n\n 2\nAbstract \nBackground \nIn 2014, an open-cut coal mine caught fire in a rural area of Australia. Local \nresidents were affected by dense black smoke over several weeks.  Little evidence \nwas available to guide health policy-makers about risks to health, vulnerable \npopulations or strategies for harm mitigation. A long-term health study was \ncommissioned, which showed excess cardiac-related deaths in the 6 months \nfollowing the fire. \nMethods \nAn adult cohort of 4,056 people was recruited in 2016-17 using the electoral roll, \nsampling from the town with the highest exposure to coal mine fire smoke and \nanother similar town without significant exposure. To estimate individual-level smoke \nexposure, participant’s time-location diaries were blended with high resolution \nestimates of hourly fire-related PM\n2.5 based on meteorological and chemical \ntransport models. All-cause mortality data were obtained up to mid-2023 (six to \nseven years after the cohort was established, and nine years after the fire) by \nlinkage with the National Death Index. Cardiac-related deaths were identifiable up to \nDecember 2021 and predicted thereafter to mid-2023. Differences in all-cause \nmortality were evaluated with Cox proportional-hazards models and differences in \ncardiac mortality with a competing risk survival model.  \nResults \nIn total, 2,725 (67%) cohort members consented to linkage. Residents of both towns \nwere similar except those in the minimally exposed town were more socio-\neconomically advantaged. Over six to seven years of follow-up, 303 (11%) of the \nsample died. While all-cause mortality was not increased, mortality from cardiac \ncauses was increased in association with PM\n2.5 exposure, robust to adjustment for \nknown confounders. \nDiscussion \nIn a community where we have previously shown excess cardiac deaths in the 6 \nmonths following the fire, we now find additional excess cardiac mortality between 2 \nand 9 years after the event, associated with mine fire PM\n2.5 exposure. This finding \nconcords with a growing body of evidence after acute exposure to wildfires and \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 3\nchronic exposures from household or ambient air pollution. Measures are needed to \nprotect individuals from exposure to smoke during and after large fire events. \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 4\n1 Introduction \nDuring severe fire weather in mid-Summer 2014, embers from a grass fire in regional \nVictoria, Australia ignited the open-cut brown coal mine adjacent to the Hazelwood \npower station. The resultant coal mine fire burned for six weeks, enveloping nearby \ncommunities in dense smoke and ash. More than 7000 firefighters were deployed, \n500 or so at a time, to suppress the fire and many firefighters were treated for smoke \ninhalation. At the same time, there were complaints from people living nearby of \nnumerous symptoms, including headaches, blurred vision, cough, shortness of \nbreath, epistaxis, fatigue, gastrointestinal symptoms (including nausea, vomiting and \ndiarrhoea), and chest pain. Such were the levels of concern in the local community \nthat there were two judicial Inquiries and calls for a research study. Ultimately it was \ndecided that a longitudinal health study lasting at least 20 years should be performed \nto determine the long-term health effects of smoke. Accordingly, the Victorian \nDepartment of Health issued a request for tender for what would become the \nHazelwood Health Study, which started later the same year \n(1) . \nCoal mine fires can occur wherever coal is found and many are currently burning \nglobally, particularly where there are rich subterranean coal deposits, such as in \nChina, India and the USA (2–4). Coal mine fires release polluting emissions \nincluding high concentrations of toxic gases, volatile organic compounds and trace \nelements including carbon monoxide, polycyclic aromatic hydrocarbons, benzene \nand particulate matter (PM).  These include PM with a median aerodynamic diameter \nless than 10 \nμ m (PM10) and fine PM less than 2.5 μ m (PM2.5), which are fine enough \nto reach the alveoli.  Some health effects of these emissions have received attention \nthrough studies of ambient urban background air pollution by PM\n2.5 and after \nexposure to humans during forest and/or peat fires or through studies after domestic \ncoal use.  However the effects of coal mine fire exposure on long-term health \noutcomes have received considerably less attention (2). One particular concern is \nmortality, given that several constituents of coal fire smoke have been found to be \nassociated with increased mortality, including PM\n2.5 (all-cause mortality), carbon \nmonoxide, benzene, formaldehyde, phenols and naphthalene (cancer), and some \ntrace elements. Moreover, increased mortality rates have been shown after forest \nfire exposure by some (5–11), but not all (12,13) studies.  \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 5\nOur prior analysis of mortality found that coal mine fire smoke PM2.5 exposure was \nassociated with an increase in injury-related deaths during the mine fire and cardiac-\nrelated deaths in the six months after the fire (14) . However, this was an ecological \ntime series study, with mine fire smoke exposure estimated based on residential \nareas recorded in the mortality database. These analyses could not account for \nindividual-level smoke exposure as well as relevant and important individual-level \nconfounding factors such as pre-fire health status, tobacco use and socio-economic \ndeprivation. \nIn this paper, we build on our previous analyses of the mine fire effects on \npopulation-level mortality (14) using individual-level data collected from participants \nin the Hazelwood Health Study Adult Cohort \n(15) . Survey data from the cohort were \nlinked to the Australian National Death Index (NDI), which we analysed to answer \nthe following research questions: 1. Was there any long-term impact of the mine fire \non all-cause mortality or cardiac mortality after 6-7 years of follow-up? 2. Were \nindividual levels of exposure to fire-related PM\n2.5 associated with all-cause mortality \nand/or cardiac mortality?  \n2 Methods \n2.1 Study cohort \nThe Hazelwood Health Study Adult Cohort comprised people who completed the \nAdult Survey between May 2016 and February 2017. The electoral roll maintained by \nthe Victorian Electoral Commission was chosen to sample people who were \nresidents in two towns in eastern Victoria (electoral registration is compulsory for \nAustralian citizens aged \n≥ 18 years). The first town was Morwell, adjacent to the mine \nfire and most severely affected by smoke exposure (exposed group). A comparison \n(minimally or not exposed) group was recruited from people who, during the mine \nfire, were living in 16 selected statistical areas (SA1s) of Sale, a town approximately \n65km away. These SA1s were chosen to reflect areas of Sale with characteristics as \nsimilar as possible to those of Morwell in terms of median age, household size, \nsocioeconomic factors and population stability. Data from air pollution modelling \nshowed that Sale had minimal smoke exposure during the mine fire (16). Residents \nidentified as living in these locations were sent postal invitations to participate along \nwith information leaflets about the study. In addition to completing the Adult Survey, \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 6\nparticipants were invited to provide written informed consent to link their data with \nNDI. More information about the Adult Cohort is available in our cohort profile (15) .  \n2.2 Exposures \nAir quality monitoring in Morwell and its surrounds during the mine fire were \ninadequate, particularly in the earliest days when smoke levels were at their most \nintense (1). Consequently, high resolution spatial and temporal estimates of hourly \nfire-related PM2.5 distribution were generated through emission, chemical transport \nand meteorological models (17). To determine Individual-level smoke exposure, \nAdult Survey participants provided time-location diaries for the mine fire period in 12-\nhour periods (15), and the diary data were then combined with the modelling output \n(16), see previous analyses of this cohort for more detail (15,17). \n2.3 Outcome \nMortality was measured in survival days by linkage to the National Death Index (NDI) \n(18) . The NDI captures all deaths occurring in Australia from 1980 onwards and is \nhoused by the Australian Institute for Health and Welfare. Subject to ethical review, \nprocesses are in place for researchers to obtain access to the NDI for research \npurposes via record linkage for all deaths. Cohort data were linked to NDI-identified \ndeaths up to June 2023. The underlying cause of death was available from the NDI \nup to, and including, deaths until December 2021 and could be used to identify \ncardiac-related mortality (ICD-10: I00-I99, G45, G46).  \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 7\n2.4 Risk Factors \nThe following risk factors were explored in the analyses (see Statistical Analysis \nbelow), all measured at the time of the survey: age, sex, educational attainment \n(“secondary up to year 10”, “secondary years 11-12”, “certificate/diploma/tertiary \ndegree”), study site (Morwell or Sale), tobacco use (smoker status: “current”, “former \nwith at least 100 lifetime cigarettes”, and “never” as well as tobacco pack-years), and \nself-reported pre-fire comorbidities (either self-reported angina, myocardial infarction, \nheart failure, stroke, COPD, cancer or diabetes). To control for socioeconomic \ndifferences between participants, we also included the Index of Relative \nSocioeconomic Advantage and Disadvantage (IRSAD) score for each participant’s \nStatistical Area Level 1 (SA1) based on 2016 census data \n(19) .  \n2.5 Statistical analysis \nTo identify potential cardiac-related mortality in 2022-23 when the cause of death \nwas not available, a predictive model was developed using a nested cross-validated \n(outer Loop: 5-fold and inner loop: 10-fold) XGBoost algorithm (20). This prediction \nmodel took advantage of the data-linkage study design and obtained potential \npredictors from the Adult Survey (e.g., demographics, smoking, alcohol \nconsumption, socioeconomic status, psychological distress, and self-reported doctor-\ndiagnosed conditions). The model included predictors extracted from three linked \nhealthcare records: the Victorian Admitted Episodes Dataset (VAED) for hospital \nadmissions (21), the Victorian Emergency Minimum Dataset for emergency \npresentations (22), and Ambulance Victoria electronic patient care records for \nambulance attendances up to 30 June 2022 (23). These predictors included the total \nnumber of healthcare services utilised, as well as service utilisation for specific \ncondition groups (classified by ICD-10 chapters and corresponding ambulance \ndiagnostic groups), within the 5 years preceding the mortality event. Due to the high \ndimensionality of the data, we first ran the prediction model to identify the top 20 \npredictors and re-trained it with selected predictors to improve model accuracy.  \nDescriptive statistics were used to characterise the cohort. Differences in all-cause \nbased on fire-related PM\n2.5 exposure were evaluated with Cox proportional-hazards \nmodels (24), while differences in cardiac-related mortality based on fire-related PM2.5 \nexposure were evaluated with a competing risk survival model, which accounted for \ndeaths due to other causes (25).  \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 8\nSurvival was measured from the date of the Adult Survey until the end of follow-up \nlinkage with NDI data (27 June 2023). Analyses were adjusted for risk factors that \nwere added to an initial crude model in the following order: age, sex, educational \nattainment, IRSAD score, study site, tobacco use and comorbidities. To determine \nwhether any risk factors influenced the relationship between exposure and outcome, \nwe performed a series of moderator analyses, incorporating interactions between \nexposure and the moderator, using covariates from the fully-adjusted model. To \nenhance interpretability, continuous moderators were standardised into z-scores and \nthe daily mean fire-related PM\n2.5 across the mine fire period was expressed in units \nof 10 µg/m³ and centred at 10 µg/m³. \nMissing data were addressed with random forest multiple imputations (26)  and \npooled according to Rubin’s rules (27). The number of imputations was equivalent to \nthe proportion of cases with missing data (10 imputed datasets). All analyses were \nconducted in R \n(28)  with RStudio (29) . Cleaning and analytical code are available on \nour public repository (30). \n2.6 Ethics \nThis study was approved by the Monash University Human Research Ethics \nCommittee as part of the Hazelwood Adult Survey and Health Record Linkage Study \n(Project ID: 25680; previously CF15/872 - 2015000389 and 6066).  Participants \nprovided written informed consent to linkage.  \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 9\n3 Results \n3.1 Description \nFrom 4,056 Adult Survey participants, 2,115 from Morwell (68%) and 610 (64%) from \nSale consented to the linkage of their Survey responses with NDI data. Participants \nfrom both towns had similar age and sex distributions and similar levels of cigarette \nsmoking exposure (pack years) at the time of the Survey. However, Sale participants \nhad slightly higher levels of educational attainment and were more socio-\neconomically advantaged than those from Morwell (Table 1). As expected, Morwell \nresidents had much higher estimated PM\n2.5\n exposure during the mine fire. In total, \nduring the 6-7 years of follow-up, 234 deaths occurred among residents of Morwell \nand 69 among residents of Sale, equating to 11% of each group. While the \nproportion of cardiac-related deaths was higher in Morwell, it was not statistically \nsignificantly so. \nTable 1. Descriptive statistics by study site \n Morwell \n n = 2,115 \nSale \n n = 610 \n \np-\nvalue \nAll-cause mortality 234 (11%) 69 (11%) 0.884 \nCause of death* \nCardiac \nActual \nPredicted \nOther  \nActual \nPredicted \nSurvived \n75 (3.5%) \n55 (2.6%) \n20 (0.9%) \n158 (7.5%) \n106 (5.0%) \n53 (2.5%) \n1,881 (89%) \n14 (2.3%) \n11 (1.8%) \n3 (0.5%) \n56 (9.2%) \n42 (6.9%) \n13 (2.1%) \n541 (89%) \n0.160 \nDaily mean exposure to fire-related PM2.5 \n(µg/m³) 11 (7-19) 0 (0-0)  \nAge at survey 60 [IQR: 48-70] 60 [IQR: 47-72] 0.543 \nMale 980 (46%) 265 (43%) 0.213 \nEducational attainment   \nSecondary to year 10 \nSecondary to year 12 \nCertificate/diploma/tertiary degree \nMissing \n \n671 (32%) \n421 (20%) \n1,011 (48%) \n12 \n \n139 (23%) \n99 (16%) \n369 (61%) \n3 \n<0.001 \nIRSAD scores \nMissing \n833 [IQR: 782-\n889] \n57 \n898 [IQR: 844-952] \n0 \n<0.001 \nSmoking status \nNon-smoker \nFormer smoker \nCurrent smoker \nMissing \n \n974 (46%) \n763 (36%) \n365 (17%) \n13 \n \n318 (53%) \n205 (34%) \n82 (14%) \n5 \n0.013 \nCigarette pack years (current/former \nsmokers) \nMissing \n15 [IQR: 5-30) \n21 \n15 [IQR: 6-30] \n4 \n0.455 \nAny pre-fire comorbidity 629 (30%) 170 (28%) 0.391 \nNotes: Continuous data presented by median and inter-quartile range (compared with Kruskal-\nWallis tests); dichotomous/categorical data presented by number and percent (compared with \nFisher’s exact tests); * cause of death predicted for n = 89 who had died in the most recent 18 \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 10\nmonths for which death notification was available but cause was not; p-value is based on \ncomparison of combined predicted and actual causes of death. \n \n3.2 Prediction model  \nThe prediction model with a total of 225 observations, including 67 known cardiac \ndeaths as training data, had a cross-validation Area Under the Curve of 0.76.  With a \nprobability cut-off of 0.45, the algorithm had a sensitivity of 0.98 and specificity of \n0.80 on the training data. The model predicted an additional 23 cardiac deaths \namong 93 new mortality records that did not specify a cause of death. The variable \nimportance from the final prediction model is provided in Figure S1, which indicates \nmine fire exposure is one of the key predictors differentiating between cardiac and \nnon-cardiac deaths. \n3.3 Main results \nFigure 1 and Table S1 summarise the results of the Cox proportional hazards (all-\ncause mortality) and competing risks (cardiac mortality) survival models, providing \nestimated Hazard Ratios (HRs) and 95% confidence intervals for each model. Model \n1 for each outcome shows the crude association/effect of PM\n2.5 with overall mortality \nand cardiac mortality respectively. Models 2 to 6 show the effect of PM2.5 after \nadjusting for confounders which are added incrementally into the model. In the plots \nbelow, Model 6, each confounder was assessed for effect modification by adding the \ninteraction between the potential effect modifier and PM\n2.5 to Model 6, with only the \nresulting interaction and their main effects shown in the plot.  \nIn the main effects models 1-6, which do not include interaction terms, PM2.5 \nexposure had no detectable effect on all-cause mortality except in the Model 1, \nwhich only evaluated the crude (unadjusted) association. However, main effects \nunadjusted and adjusted models of cardiac-related mortality found that increasing \nPM\n2.5 exposure was associated with increasing mortality in a dose-response \nrelationship. The association attenuated with each additional adjustment in models \nwithout interaction terms, with the smallest effect being an HR of 1.17 (95%CI 1.01-\n1.36) per 10µg/m\n3 of fire-related PM2.5 (fully adjusted model 6).  \nNone of the tested potential moderators emerged as either protective against or an \nexacerbator of PM\n2.5 effects.   \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 11\nFigure 1. Forest plot of PM2.5 (per 10µg/m3 increase) effects on mortality, from crude to \nadjusted models, and moderated effects in fully-adjusted models \n \n  \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 12\n4 Discussion \nNine years after the Hazelwood coal mine fire of 2014, this study explored the rates \nof mortality from all-causes and cardiac causes amongst people living in the vicinity \nat the time.  Some were from Morwell, the town that was covered with dense smoke \nfrom the fire, and the remainder from Sale, a similar town minimally affected by \nsmoke. In these analyses, rates of mortality from all-causes were not found to be \nelevated, except in the crude model.  This meant that, with this study size and \ndesign, mine fire smoke exposure had no detectable association with all-cause \nmortality in exposed cohort members. However, cardiac mortality rates were found to \nbe increased in a dose response relationship with PM\n2.5 exposure at the time of the \ncoal mine fire. As expected, cardiac mortality was also associated with increasing \nage, cigarette smoking and comorbidities. However, the effects of PM\n2.5 exposure on \ncardiac mortality were robust to sequential adjustment for demographic factors, \nsocioeconomic status, study site, tobacco use, and comorbidities, although \nattenuated somewhat with each adjustment. None of these factors showed evidence \nof moderating the impact of PM\n2.5 exposure on cardiac mortality. \nWith regard to chronic exposure to the effects of air pollution, data from the Global \nBurden of Disease Study suggested that an estimated 6.7 million deaths in 2019 \ncould be attributed to the effects of air pollution (31). The authors found that after \ntobacco, elevated systolic blood pressure and dietary risk, pollution was the fourth \nmost important risk factor for mortality globally, more important than obesity, \ncholesterol, physical inactivity or alcohol excess. These effects could be attributed to \nboth indoor air pollution (predominantly from use of biomass for cooking or kerosene \nlamps for lighting) (32) and outdoor air pollution (33). Of the excess deaths \nattributable to pollution, an estimated 50% were due to cardiovascular causes (31), \naccounting for almost 20% of all global deaths from cardiovascular disease. Whilst \nthese researchers found evidence of an encouraging reduction in the exposure to \nhousehold air pollution through socio-economic development, they found evidence of \n“concerning” increases in exposure to ambient particulate matter pollution in excess \nof 0.5% annually (31).   \nAir pollution not only exacerbates the course of cardiovascular diseases, but also \nincreases the onset through combinations of mechanisms which may include: local \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 13\nlung inflammatory responses triggered by particulate matter; penetration of \nparticulate matter deep into the lungs and therefore the bloodstream causing direct \nactivation of systemic inflammation; oxidative stress; elevated stress hormones and \ninsulin resistance; imbalance of autonomic and sympathetic nervous system \nfunction; impacts on the microbiome of the gut which may increase inflammation, \natherosclerosis and metabolic syndrome; and direct damage to respiratory mucosa \nwhich leads to increased permeability for cardiotropic micro-organisms (34). \nAlthough air pollution is a complex, dynamic mixture of gases and particles from a \ndiversity of sources, three common pollutants are most commonly the focus of \nresearch, measurement and communication: ozone, nitrogen dioxide and particulate \nmatter (35). Of these, the most consistent evidence about effects on cardiovascular \ndisease has been found to date for particulate matter, which has established impacts \non ischemic heart disease and stroke. Not only have these impacts been shown on \nshort-term mortality (36) but also longer-term impacts on incidence and mortality of \ncardiovascular disease have been found (37).  \nUntil now, there has been limited research about the mortality caused by pollution \nfrom coal mine fires (2). However, Liu and colleagues undertook a systematic review \nof the health effects of a different types of acute exposure to outdoor air pollution, \nnamely that caused by forest fires, wildfires and peat fires (38). Including studies \npublished between 1986-2014, the authors identified 13 studies of mortality after \nwildfire smoke exposure, among which nine showed increased rates of all-cause \nmortality. However, this review highlighted that 10 of 13 studies included accidental \ndeaths related to the wildfires in their all-cause mortality rates (5,6,13); one of which \n(13) found no increased all-cause mortality rate after short-term exposure to \nwildfires.  \nUnfortunately, the reviewers found no studies of cardiac mortality specifically, but \nidentified 14 studies which considered cardiac morbidity following acute exposure to \nwildfire smoke, assessed by presentations or admissions with cardiac arrest or \nsymptoms of a cardiovascular disease after the exposure occurred (38). \nInterestingly, they reported some geographic variation such that five out of six \nstudies examining exposure to wildfire smoke in USA were associated with \nincreased hospital admissions for cardiovascular diseases (including cardiac arrest \nand chest pain). In contrast, seven studies from Australia and Canada found no \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 14\nimpact on cardiovascular morbidity, measured similarly. Only one other study in the \ncity of Porto, Portugal, found a significant increase in admissions for: hypertensive \ndisease; ischaemic heart disease; and other cardiac diseases, including heart failure \nover a 3-month spate of summer forest fires in 2005 (39). The reviewers did not put \nforward a hypothesis to explain the apparent geographic variation although they \npointed out that USA had much higher rates of cardiovascular diseases at the time \nthan many of the other areas studied (38).   \nAlthough our findings are concordant with the growing body of literature suggesting a \nlong-term impact of exposure to airborne particulate matter in polluted air, they were \nunexpected given that the participants in this cohort study could only be recruited 2 \nyears after the coal mine fire. We have already shown that there was an excess of \ndeaths from cardiovascular disease associated with PM\n2.5 exposure within 6 months \nof the mine fire (14). Consequently, the individuals who died during the fire or \nimmediately afterwards could not be recruited into this cohort study. This has likely \nintroduced survivor bias into the cohort where these early deaths, which could also \nbe attributed to the mine fire, have acted to deplete the most susceptible members of \nthe community prior to recruitment (40). For this reason, we suggest that the \nestimated excess cardiac mortality reported here is likely to be under-estimated. \nThis study has some strengths: the analyses were based on an established cohort \nfor whom we had collected considerable information about individual risk factors, \nsupplemented with diaries about their movements during the mine fire period.  This \nenabled individual-level estimation of their PM\n2.5 exposure. Furthermore, the data \nabout mortality have been extracted up to nine years after the mine fire from \nadministrative databases, providing objective and comprehensive data that are \nknown to be accurate and take account of deaths occurring outside the State of \nVictoria. \nHowever, the findings must be considered alongside some limitations. Firstly, \nalthough the total follow-up period has been nine years following the mine fire, the \nnumber of deaths was still relatively small, limiting the statistical power of the current \nanalyses. For this reason, more follow-up is still needed to know for sure whether \ndeaths from all-causes have increased, and to precisely estimate the additional \ncardiac deaths caused by exposure to the coal mine fire. Secondly, the follow-up of \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 15\nthis particular cohort for mortality took place after a number of deaths occurred \nduring the mine fire and during the following 6 months. This has created a bias in \nthose eligible to be recruited, so that the risk ratios provided are likely to be under-\nestimated.  \nIn summary, this longitudinal analysis of mortality up to nine years after the \nHazelwood Mine fire found no detectable increase in all-cause mortality, but an \nexcess risk of deaths from cardiac causes, which appears to be attributable to the \neffects of individual-level PM exposure. \n  \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 16\nSupplementary materials \nTable S1. Associations between fire-related PM2.5 exposure and mortality in the \nHazelwood Health Study Adult Cohort \n All-cause mortality Cardiac mortality \nHR (95% CI) p-value HR (95% CI) p-value \n \nBaseline models (no interaction) \nModel 1 :  crude  fire-r e l a ted P M 2.5  \nex po s u r e  1.0 9  (1 .00 - 1.19 )  0 . 048  \n1.31 (1.14 - \n1.50) \n< 0.00 1 \nModel 2 :  m od e l 1  plus d e mogra p h ics  1.04 ( 0.96 - 1 . 13)  0. 31 3  \n1.24 (1.10 - \n1.41) \n0.001 \nModel 3 :  m od e l 2  plus SE S  1.01 ( 0.93 - 1 . 10)  0. 83 7  \n1.23 (1.08 - \n1.39) \n0.002 \nModel 4 :  m od e l 3  plus st ud y  si te  1.04 ( 0.95 - 1 . 15)  0. 37 5  \n1.21 (1.05 - \n1.40) \n0.009 \nModel 5 :  m od e l 4  plus toba cco  use  1.05 ( 0.95 - 1 . 15)  0. 33 4  \n1.20 (1.03 - \n1.39) \n0.016 \nModel 6 :  m od e l 5  plus c omor b i d i ties  1.03 ( 0.94 - 1 . 14) 0 . 5 0 3 \n1.17 (1.01 - \n1.36) \n0.033 \n \nAge m o de l  \nF i r e- r el a t ed  P M 2.5  e xpo sure 1.06 ( 0.89 - 1 . 25) 0 . 5 0 9 1.07  ( 0.77 - 1 . 47) 0.70 1 \nPM 2.5  *  age  a t S urve y  (z - s co re) 0.98 ( 0.86 - 1 . 12)  0. 77 6  1.09  ( 0.84 - 1 . 39)  0.52 1  \nA g e at  Su r ve y ( z - s co r e) 4.5 8  (3 .75 - 5.60 )  <0.001 4.55  ( 2.92 - 7 . 08)  <0 . 00 1 \n \nSex model\n \nF i r e- r el a t ed  P M 2.5  e xpo sure 1.00 ( 0.86 - 1 . 15) 0 . 9 5 3 1.04  ( 0.85 - 1 . 28) 0.67 6 \nPM 2.5  *  male  1.06 ( 0.90 - 1 . 26)  0. 47 6  1.22  ( 0.94 - 1 . 59) 0.13 3 \nMale 1.4 0  (1 .09 - 1.81 )  0 . 009 1.03  ( 0.63 - 1 . 69) 0.89 6 \n \nCo m o rbidit ie s  m o d e l  \nF i r e- r el a t ed  P M 2.5  e xpo sure 1.07 ( 0.92 - 1 . 24) 0 . 3 8 5 1.21  ( 0.93 - 1 . 57) 0.16 5 \nPM 2.5 *  c o m orb i d it i e s  0.95 ( 0.80 - 1 . 13)  0. 57 6  0.96  ( 0.71 - 1 . 31)  0.81 5  \nAn y  c o m o rb idit i e s †  1.84 ( 1.42 - 2 . 38) <0.001  3.20  ( 1.84 - 5 . 55) <0 . 00 1  \n \nSocioe co no m i c  a re a  (Ind e x  of  R el a tive S o cio e c on o m i c  Adva n tage and  D isadva n t a ge 2 0 1 6)\n \nF i r e- r el a t ed  P M 2.5  e xpo sure 0.99 ( 0.88 - 1 . 11) 0 . 8 7 5 1.14  ( 0.95 - 1 . 35) 0.15 5 \nPM 2.5  *  IRSAD  score (z-score) 0.91 ( 0.79 - 1 . 05)  0. 20 0  0.93  ( 0.76 - 1 . 14) 0.50 1 \nIR S AD score  (z-sc o re ) 0.8 7  (0 .77 - 0.99 )  0 . 036 1.06  ( 0.85 - 1 . 33) 0.61 9 \n \nEducationa l a ttain ment (r ef :  s econdary  up  to y ear 1 0 )  \nF i r e- r el a t ed  P M 2.5  e xpo sure 1.03 ( 0.89 - 1 . 18)  0. 69 7  1.2 6 (1 .02  - 1.5 5)  0.0 29 \nPM 2.5  *  certificate /di p loma /te rti a ry  1.05 ( 0.84 - 1 . 33)  0. 65 2  0.83  ( 0.53 - 1 . 31) 0.42 8 \nPM 2.5  *  Sec o n d a ry ye ar 1 1-12  0.99 ( 0.82 - 1 . 19)  0. 91 5  0.89  ( 0.67 - 1 . 17) 0.40 6 \nCertifi c ate /d ip l oma /tertia ry 1.14 ( 0.80 - 1 . 61)  0. 46 4  1.27  ( 0.67 - 2 . 41) 0.46 6 \nSe con dary  y ea r 11 -1 2 0.93 ( 0.71 - 1 . 20)  0. 56 4  0.83  ( 0.50 - 1 . 38) 0.47 8 \n \nSm o k e r sta tu s  a t su rv e y  \nF i r e- r el a t ed  P M 2.5  e xpo sure 1.07 ( 0.93 - 1 . 24) 0 . 3 4 0 1.28  ( 1.04 - 1 . 59)  0.02 2  \nPM 2.5  *  f orme r smoker  0.94 ( 0.79 - 1 . 13)  0. 52 6  0.91  ( 0.67 - 1 . 22) 0.51 4 \nPM 2.5  *  cu rre n t  sm o k er  0.93 ( 0.73 - 1 . 19)  0. 58 5  0.79  ( 0.55 - 1 . 13) 0.19 3 \nForme r smok e r  1.67 ( 1.28 - 2 . 18)  <0.001 1.04  ( 0.61 - 1 . 77)  0.88 2  \nCu rre n t sm o k er  2.92 ( 1.99 - 4 . 29)  <0.001 3.76  ( 1.95 - 7 . 23)  < 0.0 01  \n \nCi g arette pack -years a t surv e y  \nF i r e- r el a t ed  P M 2.5  e xpo sure 1.05 ( 0.95 - 1 . 16) 0 . 3 6 7 1.2 1 (1 .05  - 1.4 0)  0.0 08 \nPM 2.5  *  pa ck-y e a rs (squ a re roo t) 0.98 ( 0.92 - 1 . 05)  0. 59 5  0.94  ( 0.85 - 1 . 04) 0.20 9 \nPack - years (sq uare - ro ot) 1.37 ( 1.23 - 1 . 51)  <0.001 1.17  ( 0.95 - 1 . 43) 0.13 7 \n \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 17\nEducationa l a ttain ment (r ef :  s econdary  up  to y ear 1 0 )  \nF i r e- r el a t ed  P M 2.5  e xpo sure 1.03 ( 0.89 - 1 . 18) 0 . 6 9 7 1.2 6 (1 .02  - 1.5 5)  0.0 29 \nPM 2.5  *  certificate /di p loma /te rti a ry  1.05 ( 0.84 - 1 . 33) 0 . 6 5 2 0.83  ( 0.53 - 1 . 31) 0.42 8 \nPM 2.5  *  Sec o n d a ry ye ar 1 1-12  0.99 ( 0.82 - 1 . 19) 0 . 9 1 5 0.89  ( 0.67 - 1 . 17) 0.40 6 \nCertifi c ate /d ip l oma /tertia ry 1.14 ( 0.80 - 1 . 61) 0 . 4 6 4 1.27  ( 0.67 - 2 . 41) 0.46 6 \nSe con dary  y ea r 11 -1 2 0.93 ( 0.71 - 1 . 20) 0 . 5 6 4 0.83  ( 0.50 - 1 . 38) 0.47 8 \n† Self-reported pre-fire conditions including angina, myocardial infarction, heart \nfailure, stroke, COPD, cancer, and diabetes; SES includes socioeconomic area \n(IRSAD) and educational attainment; * indicates interactions (linear combination) \n  \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n   \nFigure S1. Variable importance plot of the top 20 factors in the prediction model for \ncardiac mortality among all mortality data.  \nNotes. Cover: relative quantity of observations that a feature splits; Frequency: the \nnumber of times a feature is used to split data across all trees in the model; Gain: the \nimprovement in accuracy (or reduction in loss) brought by a feature when it is used \nto split the data in the decision tree; ED: hospital emergency department; CVD: \ncardiovascular disease. \n \nFunding \nThis work was funded by the Victorian Department of Health. The paper presents the \nviews of the authors and does not represent the views of the Department. \n \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n5 References \n1. Teague B, Catford J, Petering S. Hazelwood Mine Fire Inquiry Report 2014 \n[Internet]. Government of Victoria; 2014. Available from: \nhttps://apo.org.au/node/41121 \n2. Melody SM, Johnston FH. Coal mine fires and human health: What do we know? \nInternational Journal of Coal Geology. 2015 Dec;152:1–14.  \n3. Stracher GB, Taylor TP. Coal fires burning out of control around the world: \nthermodynamic recipe for environmental catastrophe. International Journal of Coal \nGeology. 2004 Jul;59(1–2):7–17.  \n4. Stracher GB. Geology of coal fires: case studies from around the world. Vol. 18. \nBoulder, CO: Geological Society of America; 2007.  \n5. Analitis A, Georgiadis I, Katsouyanni K. Forest fires are associated with elevated \nmortality in a dense urban setting. Occup Environ Med. 2012 Mar;69(3):158–62.  \n6. Johnston F, Hanigan I, Henderson S, Morgan G, Bowman D. Extreme air pollution \nevents from bushfires and dust storms and their association with mortality in \nSydney, Australia 1994–2007. Environmental Research. 2011 Aug;111(6):811–6.  \n7. Morgan G, Sheppeard V, Khalaj B, Ayyar A, Lincoln D, Jalaludin B, et al. Effects of \nBushfire Smoke on Daily Mortality and Hospital Admissions in Sydney, Australia. \nEpidemiology. 2010 Jan;21(1):47–55.  \n8. Faustini A, Alessandrini ER, Pey J, Perez N, Samoli E, Querol X, et al. Short-term \neffects of particulate matter on mortality during forest fires in Southern Europe: \nresults of the MED-PARTICLES Project. Occup Environ Med. 2015 \nMay;72(5):323–9.  \n9. Nunes KVR, Ignotti E, Hacon SDS. Circulatory disease mortality rates in the \nelderly and exposure to PM2.5 generated by biomass burning in the Brazilian \nAmazon in 2005. Cad Saúde Pública. 2013 Mar;29(3):589–98.  \n10. Shaposhnikov D, Revich B, Bellander T, Bedada GB, Bottai M, Kharkova T, et \nal. Mortality Related to Air Pollution with the Moscow Heat Wave and Wildfire of \n2010: Epidemiology. 2014 May;25(3):359–64.  \n11. Sahani M, Zainon NA, Wan Mahiyuddin WR, Latif MT, Hod R, Khan MF, et al. \nA case-crossover analysis of forest fire haze events and mortality in Malaysia. \nAtmospheric Environment. 2014 Oct;96:257–65.  \n12. Hänninen OO, Salonen RO, Koistinen K, Lanki T, Barregard L, Jantunen M. \nPopulation exposure to fine particles and estimated excess mortality in Finland \nfrom an East European wildfire episode. J Expo Sci Environ Epidemiol. 2009 \nMay;19(4):414–22.  \n13. Vedal S, Dutton SJ. Wildfire air pollution and daily mortality in a large urban \narea. Environmental Research. 2006 Sep;102(1):29–35.  \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 20\n14. Dimitriadis C, Gao CX, Ikin JF, Wolfe R, Gabbe BJ, Sim MR, et al. Exposure \nto mine fire related particulate matter and mortality: A time series analysis from the \nHazelwood Health Study. Chemosphere. 2021 Dec;285:131351.  \n15. Ikin J, Carroll MTC, Walker J, Borg B, Brown D, Cope M, et al. Cohort profile: \nThe Hazelwood Health Study Adult Cohort. Int J Epidemiol. 2021 Jan \n23;49(6):1777–8.  \n16. Luhar AK, Emmerson KM, Reisen F, Williamson GJ, Cope ME. Modelling \nsmoke distribution in the vicinity of a large and prolonged fire from an open-cut \ncoal mine. Atmos Environ. 2020 May;229:117471.  \n17. Johnson AL, Gao CX, Dennekamp M, Williamson GJ, Brown D, Carroll MTC, \net al. Associations between respiratory health outcomes and coal mine fire PM2.5 \nsmoke exposure: A cross-sectional study. IJERPH. 2019 Nov 2;16(21):4262.  \n18. Australian Institute of Health and Welfare [Internet]. 2023 [cited 2023 Dec 13]. \nNational Death Index (NDI). Available from: https://www.aihw.gov.au/about-our-\ndata/our-data-collections/national-death-index \n19. Australian Bureau of Statistics. 2033.0.55.001 - Census of Population and \nHousing: Socio-Economic Indexes for Areas (SEIFA), Australia, 2016 [Internet]. \n2018 [cited 2019 May 13]. Available from: \nhttps://www.abs.gov.au/ausstats/abs@.nsf/mf/2033.0.55.001 \n20. Chen T, He T, Benesty M, Khotilovich V, Tang Y, Cho H, et al. xgboost: \nExtreme Gradient Boosting [Internet]. 2024. Available from: https://CRAN.R-\nproject.org/package=xgboost \n21. Department of Health (Victoria). Victorian Admitted Episodes Dataset \n[Internet]. State Government of Victoria, Australia; 2023 [cited 2023 May 17]. \nAvailable from: https://www.health.vic.gov.au/data-reporting/victorian-admitted-\nepisodes-dataset \n22. Department of Health (Victoria). Victorian Emergency Minimum Dataset \n(VEMD) [Internet]. State Government of Victoria, Australia; 2023 [cited 2023 May \n17]. Available from: https://www.health.vic.gov.au/data-reporting/victorian-\nemergency-minimum-dataset-vemd \n23. Cox S, Martin R, Somaia P, Smith K. The development of a data-matching \nalgorithm to define the ‘case patient’. Aust Health Review. 2013;37(1):54.  \n24. Therneau T. A Package for Survival Analysis in S [Internet]. 2015. Available \nfrom: https://cran.r-project.org/package=survival \n25. Sjoberg DD, Fei T. tidycmprsk: Competing Risks Estimation [Internet]. 2023. \nAvailable from: https://CRAN.R-project.org/package=tidycmprsk \n26. Mayer M. missRanger: Fast Imputation of Missing Values [Internet]. 2024. \nAvailable from: https://CRAN.R-project.org/package=missRanger \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 21\n27. van Buuren S, Groothuis-Oudshoorn K. ‘mice’: Multivariate Imputation by \nChained Equations in R. J Stat Softw. 2011;45(3):1–67.  \n28. R Core Team. R: A Language and Environment for Statistical Computing \n[Internet]. Vienna, Austria: R Foundation for Statistical Computing; 2024. Available \nfrom: https://www.r-project.org/ \n29. RStudio Team. RStudio: Integrated Development for R [Internet]. Boston, MA: \nRStudio, Inc.; 2024. Available from: https://www.rstudio.com \n30. Lane T. Bridges. Monash University; 2025 [cited 2025 Jan 15]. Cleaning and \nanalytical code: All-cause and cardiac mortality in relation to smoke exposure nine \nyears after the Hazelwood coal mine fire. Available from: \nhttps://doi.org/10.26180/28208279.v1 \n31. Murray CJL, Aravkin AY, Zheng P, Abbafati C, Abbas KM, Abbasi-Kangevari \nM, et al. Global burden of 87 risk factors in 204 countries and territories, 1990–\n2019: a systematic analysis for the Global Burden of Disease Study 2019. The \nLancet. 2020 Oct;396(10258):1223–49.  \n32. Lim S, Said B, Zurba L, Mosler G, Addo-Yobo E, Adeyeye OO, et al. \nCharacterising sources of PM\n2·5 exposure for school children with asthma: a \npersonal exposure study across six cities in sub-Saharan Africa. The Lancet Child \n& Adolescent Health. 2024 Jan;8(1):17–27.  \n33. Air pollution [Internet]. [cited 2024 Oct 2]. Available from: \nhttps://www.who.int/health-topics/air-pollution \n34. Konduracka E, Rostoff P. Links between chronic exposure to outdoor air \npollution and cardiovascular diseases: a review. Environ Chem Lett. 2022 \nOct;20(5):2971–88.  \n35. Brauer M, Casadei B, Harrington RA, Kovacs R, Sliwa K, the WHF Air \nPollution Expert Group. Taking a stand against air pollution—the impact on \ncardiovascular disease: a joint opinion from the world heart federation, american \ncollege of cardiology, american heart association, and the european society of \ncardiology. Circulation [Internet]. 2021 Apr 6 [cited 2024 Oct 2];143(14). Available \nfrom: https://www.ahajournals.org/doi/10.1161/CIRCULATIONAHA.120.052666 \n36. Liu C, Chen R, Sera F, Vicedo-Cabrera AM, Guo Y, Tong S, et al. Ambient \nParticulate Air Pollution and Daily Mortality in 652 Cities. N Engl J Med. 2019 Aug \n22;381(8):705–15.  \n37. Yusuf S, Joseph P, Rangarajan S, Islam S, Mente A, Hystad P, et al. \nModifiable risk factors, cardiovascular disease, and mortality in 155 722 \nindividuals from 21 high-income, middle-income, and low-income countries \n(PURE): a prospective cohort study. The Lancet. 2020 Mar;395(10226):795–808.  \n38. Liu JC, Pereira G, Uhl SA, Bravo MA, Bell ML. A systematic review of the \nphysical health impacts from non-occupational exposure to wildfire smoke. \nEnvironmental Research. 2015 Jan;136:120–32.  \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint \n\n 22\n39. Azevedo JM, Gonçalves FLT, De Fátima Andrade M. Long-range ozone \ntransport and its impact on respiratory and cardiovascular health in the north of \nPortugal. Int J Biometeorol. 2011 Mar;55(2):187–202.  \n40. Yadav K, Lewis RJ. Immortal time bias in observational studies. JAMA. 2021 \nFeb 16;325(7):686.  \n \n \n \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 15, 2025. ; https://doi.org/10.1101/2025.01.14.25320562doi: medRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}