{"paper_id":"4e80683f-6f8b-4dd1-a605-aa52dce9b6c0","body_text":"1\nOriginal Research Article 1 \n 2 \nThe diagnostic accuracy of chest Xray screening for silicosis: A systematic 3 \nreview, meta-analysis and modelling study 4 \n 5 \nAuthor list and affiliations:  6 \n1. *Patrick Howlett MBChB BSc MSc 1 – p.howlett@imperial.ac.uk 7 \n2. *Ashwin Durairaj BSc 1 – ashwin.durairaj20@imperial.ac.uk  8 \n3. Maia Lesosky BSc MSc PhD 1 – m.lesosky@imperial.ac.uk  9 \n4. Johanna Feary BM BS MSc FRCP PhD 1,2 – j.feary@imperial.ac.uk   10 \n 11 \n1. National Heart & Lung Institute, Imperial College London, Guy Scadding 12 \nBuilding, Cale Street, London, SW3 6LY.  13 \n2. Occupational lung disease department, Royal Brompton Hospital, Guys and 14 \nSt Thomas’ Hospital, London 15 \n* Joint first authors 16 \n 17 \nCorresponding author: Patrick Howlett (p.howlett@imperial.ac.uk). National Heart 18 \n& Lung Institute, Imperial College London, Guy Scadding Building, Cale Street, 19 \nLondon, SW3 6LY. Tel +44 7793198119.  20 \n 21 \nContributorship statement 22 \nPH – Guarantor: full access to all of the data in the study and takes responsibility for 23 \nthe integrity of the data and the accuracy of the data analysis, including and 24 \nespecially any adverse effects, Writing – Original draft preparation, 25 \nConceptualisation, Methodology, Project Administration, Data collection, Data 26 \nCuration, Visualisation, Formal Analysis 27 \nAD – Original draft preparation, Conceptualisation, Methodology, Project 28 \nAdministration, Data collection, Data Curation, Visualisation, Formal Analysis 29 \nML - Conceptualisation, Formal Analysis, Visualisation, Writing – Review & Editing 30 \nJF – Conceptualisation, Project Administration, Supervision, Writing – Review & 31 \nEditing 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 (which was not certified by peer review)\nThe copyright holder for this preprint this version posted May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \nNOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.\n\n 2\n 1 \nKeywords: Silicosis, Pneumoconiosis, Silicon Dioxide, Quartz, Mining, 2 \nIndustry, Occupational Disease, Occupational Exposure, Respiratory, 3 \nComputed Tomography, Radiograph, X-ray 4 \n 5 \nWord counts 6 \nAbstract: words: 250 words 7 \nManuscript:  3182 words  8 \n 9 \nReferences: 37 10 \n  11 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 3\nAbstract 1 \n 2 \nObjectives: Chest Xray (CXR) is widely used for silicosis diagnosis, despite concerns 3 \nregarding sensitivity. We investigated the diagnostic accuracy of CXR for silicosis 4 \nscreening compared to computed tomography (CT), high-resolution CT (HRCT) and 5 \nautopsy, and modelled the relationship between CXR sensitivity and disease 6 \nseverity. 7 \n 8 \nMethods: Medline, Embase, Scopus, and Web of Science databases were searched 9 \non 2nd July 2024 (Prospero registration: CRD42024513830). Meta-analyses were 10 \nperformed by reference standard and at increasing reference test severity cut-offs. 11 \nThe Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool 12 \nassessed risk of bias. In scenarios of fixed and relative sensitivity, according to 13 \ndisease severity, we estimated missed silicosis cases and the number needed to 14 \nscreen (NNS) in hypothetical populations of low (5%), medium (15%) and high (30%) 15 \nsilicosis prevalence.  16 \n 17 \nResults: Twenty studies included 2156 participants and 1105 silicosis cases. CXR 18 \nhad moderate sensitivity (0.76; 95% confidence interval (CI): 0.63-0.86, I2=84%) and 19 \nhigh specificity (0.89, 95% CI: 0.77-0.95, I2=57%) compared to HRCT in 12 studies, 20 \nand low sensitivity (0.50, 95% CI: 0.45-0.55, I2=0%) and high specificity (0.91, 95% 21 \nCI: 0.87-0.93, I2=20%) compared to autopsy in two studies. CXR sensitivity 22 \nincreased with higher reference test severity cut-offs. Clinically relevant numbers of 23 \ncases were missed in fixed and relative sensitivity scenarios; increased prevalence 24 \nand less severe disease resulted in more missed cases and a lower NNS.  25 \n 26 \nConclusions: Silicosis severity and reference test type both plausibly influence CXR 27 \nsensitivity. Assuming either fixed or relative sensitivity results in missed silicosis 28 \ncases. Judicious HRCT screening is likely to improve case detection.  29 \n 30 \n 31 \n  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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 4\nWhat is already known on this topic 1 \nIt is widely understood that Chest Xray (CXR) underdiagnoses silicosis compared to 2 \nmore accurate methods, such as high resolution computed tomography (HRCT) and 3 \nautopsy.  4 \n 5 \nWhat this study adds 6 \nOur systematic review and meta-analysis demonstrated that the sensitivity of CXR 7 \nwas lowest when compared to autopsy (50%), followed by HRCT (76%). This 8 \ndifference may be explained by the increased accuracy of autopsy as a reference 9 \ntest. Another potential explanation for differences between study results could be 10 \nthat – because severe silicosis is more easily diagnosed by CXR – studies with a 11 \nhigher proportion of severe disease recorded higher sensitivity results. Importantly, 12 \nregardless of whether differences between studies are explained by different 13 \nreference test modalities or the proportion of severe disease, when modelled among 14 \na population of silica-exposed workers, many silicosis cases are missed.  15 \n 16 \nHow this study might affect research, practice or policy 17 \nThis study suggests the careful implementation of HRCT screening for silicosis 18 \nwould improve case detection.   19 \n 20 \n 21 \n 22 \n  23 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 5\nIntroduction 1 \n 2 \nSilicosis is a preventable yet incurable occupational fibrotic lung disease caused by 3 \nthe inhalation of respirable crystalline silica (RCS) dust. Recent reports of the high 4 \nmorbidity and mortality of artificial stone silicosis are reminders of the potentially 5 \ndevastating effects of this disease[1,2]. The likely underdiagnosis of silicosis and the 6 \nlack of clarity around modelling methods mean the global burden of disease is 7 \nunknown and likely higher than currently reported estimates[3,4].  8 \n 9 \nSilicosis diagnosis relies on positive Chest Xray (CXR) and/or Computed 10 \nTomography (CT) findings in conjunction with a compatible exposure history. The 11 \nInternational Labour Organisation (ILO) classification system[5] is one system that 12 \nprovides a standardised approach to screening for possible silicosis cases in at-risk 13 \npopulations. The 12-point, stepwise classification describes the severity and 14 \nprofusion of small-opacities, along with other relevant findings. In UK and US 15 \npractice, a profusion cut off of > 1/0 often qualifies as a positive screening result[6,7]; 16 \nCT is normally reserved for cases with an abnormal CXR. Due to availability and 17 \ncost, for most workers globally, CXR is the only test performed to diagnose silicosis. 18 \nIn practice, therefore, a positive screening CXR in an exposed worker will be 19 \nclassified as a case of silicosis. In many settings, the expertise required to apply the 20 \nILO classification precludes its use, relying instead on a simple radiographic 21 \ndiagnosis.  22 \n 23 \nThe concern that some silicosis may not be visible on CXR, sometimes termed 24 \n“subradiological silicosis”, dates back to autopsy studies of South African miners, 25 \nwhich demonstrated a sensitivity of 31% at a CXR >= 1/1 ILO cut-off[8,9]. More 26 \nrecently, among artificial stone workers in Australia, a CXR sensitivity of 48% was 27 \nreported compared to HRCT[10]. These enduring concerns are changing practice. A 28 \n2019 Royal Australian and New Zealand College of Radiologists Position Statement 29 \nstrongly recommended CT scanning as the “primary imaging modality for screening 30 \nexposed workers”. However, the potential limitations of CXR have not been 31 \nsystematically reviewed or quantified[11]. In practice, and for the purposes of this 32 \npaper, HRCT and reconstructed multidetector CT (MDCT) images are collectively 33 \nreferred to as HRCT. 34 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 6\n 1 \nIt is intuitive that CXR sensitivity is related – at least in part – to severity of disease, 2 \nsuch that CXRs with a greater severity of disease have a higher sensitivity. In South 3 \nAfrican miners, CXR sensitivity increased from 38% in the subgroup with “slight” 4 \ndisease at autopsy, to 83% in those with “marked” disease[9]. The counterfactual 5 \nassumes that sensitivity does not vary with silicosis severity. These two relationships 6 \nare illustrated by directed acyclic diagrams in Figure 1A and 1B.  7 \n 8 \nQuantifying the sensitivity of CXR compared to high-resolution computed 9 \ntomography (HRCT) informs the debate on the role of HRCT in silicosis 10 \nscreening[12]. Understanding whether silicosis severity may modify CXR sensitivity 11 \nis helpful clinically and programmatically. Understanding the impact of the 12 \ncounterfactual scenarios of fixed and relative sensitivity, according to silicosis 13 \nseverity, on the number missed cases may be helpful in identifying priority 14 \npopulations for HRCT screening.  15 \n 16 \nWe therefore performed a systematic review and meta-analysis of the sensitivity and 17 \nspecificity of CXR compared to computed tomography (CT), HRCT and autopsy 18 \nreference standards. We then modelled how sensitivity and specificity of CXR 19 \nchanges according to severity of underlying disease. We used this model to estimate 20 \nthe number of cases missed and number needed to screen (NNS) in example 21 \npopulations of silica-exposed individuals.  22 \n 23 \n  24 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 7\nMethods 1 \n 2 \nOne reviewer (AD) searched Medline, Embase, Scopus and Web of Science 3 \ndatabases from inception up to 2nd July 2024 (search strategy in online 4 \nsupplementary materials 1) and reviewed bibliographies of included studies 5 \n(Prospero ID: CRD42024513830).  6 \n 7 \nStudy inclusion criteria were: (1) cohort, cross-sectional, or case-control study 8 \ndesign; (2) Population >18 years-old with occupational RCS exposure; (3) CXR used 9 \nas the index test for assessing silicosis, using the ILO classification or equivalent; (4) 10 \nCT, HRCT, or autopsy used as the reference standard for assessing silicosis; (5) 11 \nsufficient information to construct 2x2 tables. In the case of a high risk of duplicated 12 \ndata, the most comprehensive study was retained.  13 \n 14 \nTitle and abstract screening and full-text review was performed in parallel by two 15 \nreviewers (AD, PH) using Covidence. Data was extracted by a single reviewer (AD) 16 \nand then checked (PH). Disagreements were resolved through consensus. 17 \nMethodological quality was assessed using the Quality Assessment of Diagnostic 18 \nAccuracy Studies-2 (QUADAS-2) tool[13] and entered into RevMan. Extraction tools 19 \nare presented in online supplementary materials 2 and description of our application 20 \nof the QUADAS-2 tool in supplementary materials 3.  21 \n 22 \nMeta-analysis 23 \n 24 \nOur pre-specified CXR meta-analysis of sensitivity and specificity of CXR for silicosis 25 \nincluded studies using the ILO CXR cut-off of ≥ 1/0 or equivalent for the index 26 \ntest[14]. The HRCT group included reconstructed MDCT scans, all of which were 27 \nlabelled as HRCT in their respective studies. Post-hoc sensitivity analyses included 28 \nstudies with other CXR cut-offs and restriction to studies without a high/unclear risk 29 \nof patient selection or index test bias. 30 \n 31 \nTo demonstrate whether CXR diagnostic accuracy varied according to silicosis 32 \nseverity, we repeated our meta-analyses at increasing reference test severity cut-33 \noffs. We separated studies by radiology (HRCT and CT) and autopsy reference 34 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 8\nstandards. Although recent consensus classifications are available[15], no single 1 \nHRCT/CT or autopsy classification was consistently applied. We therefore 2 \nconsidered four of seven studies that used a 4-level scale (similar to the ILO CXR 3 \nsystem) as equivalent and mapped the remainder to that scale (details in online 4 \nsupplement materials 4).  5 \n 6 \nA bivariate generalized linear mixed model (GLMM) with logit transformation was 7 \nused to calculate pooled sensitivity and specificity estimates using the “meta” 8 \npackage in R (Version 4.3.2). The bivariate GLMM approach accounts for the 9 \ncorrelation between sensitivity and specificity and is preferred when 2x2 cell counts 10 \nare low[16]. Heterogeneity was assessed using the I2 statistic.  11 \n 12 \nMeasures of impact 13 \n 14 \nWe performed a mixed-effects linear meta-regression with the exposure of the ratio 15 \nof ILO category ≥ 2/1 relative to ≥ 1/0 silicosis and the outcome of CXR ≥ 1/0 16 \nsensitivity compared to HRCT. ILO category ≥ 2/1 was used as it represents major 17 \ncategory 2 disease and provides a robust numerator. Using this model, and to 18 \ndemonstrate the impact of both fixed and relative sensitivity scenarios, we estimated 19 \nthe sensitivity of CXR for silicosis among hypothetical populations of 1000 silica-20 \nexposed workers with low (5%), medium (15%) and high (30%) prevalences of CXR 21 \nILO ≥ 1/0 category silicosis. We also assumed two fixed proportions of less severe 22 \ndisease (20% CXR ILO ≥ 2/1 relative to ILO >1/0 silicosis) and more severe disease 23 \n(40% CXR ILO ≥ 2/1 relative to ILO >1/0 silicosis). In each population, we then 24 \ncalculated the number of missed cases that would be identified by HRCT and the 25 \nnumber needed to screen using HRCT to detect a single extra case of silicosis. A 26 \nsensitivity analysis used the prevalence of CXR (non-reference confirmed) CXR ILO 27 \ncategory ≥ 2/1 silicosis as the meta-regression exposure. 28 \n 29 \nAll code and data are publicly available at 30 \nhttps://github.com/pjhowlett/da_silic_cxr/tree/main.  31 \n 32 \nHuman Research Ethics Committee approval was not required as only previously 33 \npublished data was used.   34 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 9\nResults 1 \n 2 \nFrom a total of 825 title and abstracts, 47 studies underwent full-text screening and 3 \n20 were included (Figure 2). Fifteen studies used HRCT, three used CT and two 4 \nused autopsy as a reference standard. A total of 2156 participants (HRCT 1207, CT 5 \n151, Autopsy 798) were included, of whom 1105 (HRCT 587, CT 112, Autopsy 406) 6 \nwere identified by the reference test to have silicosis. There was at least moderate 7 \nagreement for title and abstract and full-text reviews; Cohen’s Kappa 0.65 and 0.45, 8 \nrespectively. 9 \n 10 \nAll studies represented mining and non-mining industries in middle or high-income 11 \neconomies. The median prevalence of reference test silicosis was 63% (range 16-12 \n99%) [17–24,15,25–32,8,33,34] (online supplementary table 1). Study size ranged 13 \nfrom 11 to 557 participants. Exposures ranged from 7.6 to 31 years; dental 14 \ntechnician, sandblaster and artificial stone worker studies had median exposures of 15 \n<15 years[25,28–30]. Autopsy and CT studies were either performed or published 16 \nbefore HRCT in most cases, although HRCT publication dates ranged from 1991 to 17 \n2024. 18 \n 19 \nAmong 15 studies with HRCT as the reference test, CXR sensitivity ranged from 20 \n0.18 (95% CI 0.02, 0.52) to 0.95 (95% CI 0.76, 1.00) while specificity ranged from 21 \n0.29 (95% CI 0.04, 0.71) to 1.00 (95% CI 0.99, 1.00) (Figure 3A). For two studies 22 \nwith autopsy as the reference test, CXR sensitivity was 0.50 (95% CI 0.39, 0.61) and 23 \n0.50 (95%CI 0.44, 0.56), while specificity was 0.89 (95% CI 0.84, 0.93) and 0.93 24 \n(95% CI 0.87, 0.96) (Figure 3B). For three studies with CT as the reference test, 25 \nCXR sensitivity ranged from 0.94 (95% CI 0.73, 1.00) to 1.00 (95% CI 0.92, 1.00) 26 \nwhile specificity ranged from 0.50 (95% CI 0.21, 0.79) to 1.00 (95% CI 0.48, 1.00) 27 \n(supplementary figure 1). 28 \n 29 \nQuality Assessment  30 \n 31 \nEight studies had unclear bias and three had high risk of bias regarding patient 32 \nselection (online supplementary figure 2). Unclear or high risk of bias in the index (6 33 \nstudies) or reference (4 studies) test was most commonly due to poor detailing of 34 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 10\ntest blinding. Six studies had unclear bias and three had high risk of bias for flow and 1 \ntiming, due to flow concerns (2 studies) or unknown or inappropriate intervals 2 \nbetween the index and reference test (7 studies). Five studies had high or unclear 3 \napplicability, mainly related to different ILO CXR cut-offs. 4 \n 5 \nMeta-analysis 6 \n 7 \nPooled sensitivity for CXR with a HRCT reference standard was 0.76 (95% CI 0.63, 8 \n0.86; n = 12), with a pooled specificity of 0.89 (95% CI 0.77, 0.95; n = 12) (Figure 9 \n3A). Heterogeneity was substantial for sensitivity (I2 = 84%) and moderate for 10 \nspecificity (I2 = 57%). Compared to an autopsy reference standard, pooled sensitivity 11 \nand specificity for CXR were 0.50 (95% CI 0.45, 0.55; n = 2) and 0.91 (95% CI 0.87, 12 \n0.93; n = 2), with low heterogeneity (I2 = 0% and I2 = 20%, respectively) (Figure 3B). 13 \nCompared to a CT reference standard, pooled sensitivity and specificity for CXR 14 \nwere 0.97 (95% CI 0.86, 0.99; n = 2) and 0.69 (95% CI 0.20, 0.95; n = 2), with low (I2 15 \n= 0%) and moderate (I2 = 57%) heterogeneity, respectively (supplementary figure 1). 16 \nThe sensitivity analyses, which included different ILO cut-offs or the removal of 17 \nstudies with a high/unclear risk of patient selection bias or index test bias, did not 18 \nsignificantly alter our results (online supplementary figures 3 and 4).  19 \n 20 \nCXR demonstrated increased sensitivity at higher grade cut-offs of silicosis in the 21 \nreference test (online supplementary figures 5 and 6). Sensitivity was 0.86 (95% CI 22 \n0.72, 0.93; I2 = 47%; n = 4) compared to combined HRCT and CT test category 1 23 \ndisease or greater. All participants bar one with reference level 2 or higher disease 24 \non CT or HRCT were identified correctly by CXR. For autopsy reference standards, 25 \nsensitivity increased from 0.72 (95% CI 0.27, 0.95; I2 = 89%; n = 3) at the lowest cut-26 \noff to 0.83 (95% CI 0.70, 0.91; I2 = 0%; n = 3) at the highest. Correspondingly, 27 \nspecificity showed a decreasing trend across both radiological and autopsy 28 \nreference standards with increasing grade of cut off.  29 \n 30 \nMissed cases and number needed to screen 31 \n 32 \nOur meta-regression estimated a 1.15% (95% CI 0.44, 1.86) increase in sensitivity 33 \nfor each 1% increase in the ratio of ILO category ≥ 2/1 relative to ≥ 1/0 silicosis. When 34 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 11\nthere were no ILO category ≥ 2/1 silicosis cases (i.e. the ratio was zero) the 1 \nsensitivity of CXR was 16%. The R2 value was 63% and heterogeneity substantial (I2 2 \n= 86%) (Figure 4).  3 \n 4 \nIn a low prevalence scenario (5% or 50 silicosis cases per 1000 exposed) (Table 1 5 \nand supplementary figure 7A) and under the assumption of fixed CXR sensitivity, the 6 \nnumber of missed cases per 1000 exposed was 16 cases and the NNS was 62 7 \npersons. Under the assumption of relative sensitivity, the number of cases missed 8 \nwas higher and the NNS lower; when we assumed a lower proportion of severe 9 \ndisease (20% cases > ILO 2/1 relative to >1/0 silicosis), 79 cases per 1000 exposed 10 \nwere missed with NNS of 13 persons. When we assumed severe disease was more 11 \ncommon (40% cases > ILO 2/1 relative to >1/0 silicosis) 31 cases per 1000 person 12 \nwere missed with a NNS of 32 persons. Both increasing prevalence of silicosis and a 13 \nhigher proportion of severe disease led to a higher number of missed cases and, 14 \nhence, a lower NNS. Our sensitivity analysis, using the prevalence of CXR ILO >2/1, 15 \nresulted in similar estimates of missed cases and NNS (supplementary table 2, 16 \nsupplementary figures 7B and 8). 17 \n  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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 12\nTable 1. Missed cases and number needed to screen in a silica-exposed population. 1 \nCalculated using the metaregression model with exposure of the ratio ILO >2/1 2 \nrelative to >1/0 silicosis.  The number of missed cases per 1000 exposed persons 3 \nand number needed to screen to identify 1 extra case of silicosis in three scenarios 4 \nof low silicosis prevalence (5% or 50 cases per 1000), as defined by CXR ILO >1/0, 5 \nmedium silicosis prevalence (15% or 150 cases per 1000) and high silicosis 6 \nprevalence (30% or 300 cases per 1000), and under the assumptions of fixed and 7 \nrelative sensitivity of CXR, compared to HRCT. For the scenario of relative 8 \nsensitivity, we have further assumed either less common severe disease (20% of 9 \ncases are > ILO 2/1 relative to >1/0 silicosis) and more common severe disease 10 \n(40% of cases are > ILO 2/1 relative to >1/0 silicosis).  11 \n 12 \n \nLow prevalence (5%) Medium prevalence \n(15%) High  prevalence (30%) \nMissed \ncases per \n1000 \npersons  \nNNS to \ndetect 1 \nextra case \nMissed \ncases per \n1000 \npersons \nNNS to \ndetect 1 \nextra case \nMissed \ncases per \n1000 \npersons \nNNS to \ndetect 1 \nextra case \nFixed CXR sensitivity 16 62 47 21 95 11 \nRelative \nCXR \nsensitivity \nLess severe \ndisease  \n(20% > ILO \n2/1) \n79 13 236 4 471 2 \nMore severe \ndisease  \n(40% > ILO \n2/1) \n31 32 93 11 185 5 \n 13 \n 14 \n 15 \n 16 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 13\nDiscussion 1 \n 2 \nDespite concerns regarding the diagnostic accuracy of CXR for silicosis, no 3 \nsystematic review or meta-analysis has been performed to date. We found that, for 4 \nthe diagnosis of ≥ 1/0 silicosis, CXR had a moderate pooled sensitivity compared to 5 \nHRCT (0.76; 95% CI 0.63, 0.86) and poor sensitivity compared to autopsy (0.50; 6 \n95% CI 0.45, 0.55). Specificity was high for both autopsy and HRCT. Heterogeneity 7 \nwas high for HRCT (I2 = 84%) while two large autopsy studies reported identical 8 \nsensitivity values. 9 \n 10 \nThe sensitivity of CXR in detecting silicosis increased across autopsy, HRCT and 11 \nCT, reflecting the improved resolution and hence diagnostic accuracy of autopsy 12 \nover HRCT, and HRCT over CT. That the sensitivity of CXR compared to CT 13 \napproached 1 suggests that plain CT, as applied at the time of the included studies 14 \n(1986-1993), had little benefit for the screening detection of silicosis beyond 15 \nstandard CXR. Older plain CT technology has been superseded by HRCT or 16 \nreconstructed MDCT images, thus limiting the generalisability of our CT results to 17 \ncurrent CT imaging. Only one, small study from 1991 directly compared CT and 18 \nHRCT; although this reported equal sensitivity, the authors commented on improved 19 \nclarity and firmer interpretation with HRCT compared to CT [17]. Technological 20 \nimprovements over time in both CXR and reference tests may influence results. 21 \nOlder HRCT studies tended to have reduced sensitivity compared to more recent 22 \nstudies (Figure 3). This suggests greater improvement in CXR accuracy over time 23 \nrelative to HRCT, although these studies were also often smaller. 24 \n 25 \nWe observed increasing CXR sensitivity at higher cut-off grades of reference test 26 \nsilicosis. This supports the intuitive and previously reported understanding that 27 \nsevere silicosis is more readily diagnosed on CXR[8,9]. We then found that CXR 28 \nsensitivity increased by 1.15% (95% CI 0.44, 1.86) for every 1% increase in the ratio 29 \nof CXR ILO >2/1 relative to ≥ 1/0 silicosis, although heterogeneity was considerable 30 \n(I2 = 86%).  31 \n 32 \nBased on our findings, it is plausible that both silicosis severity and reference test 33 \ntype influence the diagnostic accuracy of CXR for silicosis. In the case of HRCT, it 34 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 14\nmay therefore be considered that the true sensitivity and specificity of CXR lies 1 \nbetween the assumptions of fixed and relative sensitivity. It is therefore important to 2 \nnote that, regardless of assumption, a clinically meaningful number of cases were 3 \nmissed (Table 1 and supplementary figure 7). Furthermore, higher silicosis 4 \nprevalence and greater proportions of severe disease led to higher numbers of 5 \nmissed cases and a lower NNS. This suggests HRCT may have the greatest impact 6 \namong high prevalence populations with more severe disease. Conversely, although 7 \nthe absolute impact of HRCT may be reduced among lower prevalence groups, if the 8 \nrelative sensitivity scenario is considered probable, the proportional benefit from the 9 \nearlier diagnosis of silicosis is greater, particularly in the plausible case of less 10 \nfrequent severe disease. 11 \n 12 \nFrom a pragmatic perspective, our results would suggest a low threshold for HRCT 13 \nscreening among groups or individuals considered to be at a higher risk of silicosis. 14 \nFrom the patient and physician perspective, although no established treatment for 15 \nsilicosis exists, the use of HRCT may allow earlier diagnosis of silicosis, leading to 16 \nmore informed work choices. For employers, industrial hygienists and regulators, 17 \nmore accurate, earlier-stage silicosis prevalence data may prompt pro-active 18 \nexposure testing, adjustments to work practices and legislation.  19 \n 20 \nGiven the dose-response relationship between silica and lung cancer, and often high 21 \nrates of smoking among silica-exposed workers[3,35], our analysis represents the 22 \nminimum potential benefit of HRCT screening. Practical considerations of HRCT 23 \nscreening include accessibility, cost-effectiveness, radiation dose and risks of 24 \nsubsequent investigations, such as lung biopsy of high-risk nodules. Correctly risk 25 \nstratifying nodules among silica-exposed populations, particularly if TB is common, 26 \npresents an unexplored challenge. Low-dose CT screening with high resolution 27 \nreconstruction, such as implemented for lung cancer screening and silicosis 28 \nscreening in Western Australia, is possible[11]. An alternative approach may involve 29 \na combination of computer-aided detection for CXR, trained with HRCT as a 30 \nreference standard. If successful, this approach would have global impacts for the 31 \ndiagnosis and classification of silicosis; our results can help design studies.   32 \n 33 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 15\nIndividual study methods may influence sensitivity or specificity. Recruitment from 1 \nlargely pre-screened silicosis populations, in whom prevalences of silicosis are 2 \nnotably high (range 16-99%), may falsely raise sensitivity and reduce specificity. 3 \nDifferences in test reader experience or reading and consensus methods may 4 \nincrease between study variance. However, restriction to studies without high or 5 \nunclear risk of bias in these domains did not alter our findings. Future studies should 6 \naim for an unselected population and a standardised reading and consensus 7 \napproach. Comorbidities including emphysema or tuberculosis did not appear to 8 \nreduce sensitivity or specificity[32,33]. 9 \n 10 \nImportant limitations of our methods exist. Whilst we applied our risk of bias tool 11 \naccording to the guidance[13] and have attempted to do so transparently (including 12 \npublication of our raw data), we felt the QUADAS-2 may over-estimate risk of bias in 13 \nsome cases. For example, the preselection of a sample (such as in Hoy et al (29)) 14 \nresults in a penalisation of both Patient selection and Flow and Timing, possibly 15 \nover-stating the true risk of bias. Potentially exacerbated by our English-language 16 \nrestriction, our data is skewed to developed economies with relatively low 17 \nexposures[3] . When no ILO > 2/1 cases were present modelled sensitivity was 16%. 18 \nThis may represent inaccuracy outside our data range as our lowest ILO >2/1 19 \nrelative to >1/0 percentage was 38%. However, one study of only  ILO 0 and 0/1 20 \nCXRs in miners found 57/339 (17%) had silicosis on HRCT suggesting this low 21 \nsensitivity is plausible[36]. Wide confidence intervals in our meta-regression 22 \nrepresent data sparsity and heterogeneity and lend caution to our subsequent 23 \nmissed cases and NNS calculations. Whilst we maintain these estimates are useful, 24 \nthey should be viewed as illustrative estimates of the impact of changes in 25 \nparameters – such as proportion of severe disease – rather than exact figures. Our 26 \nsensitivity analysis (using coefficients from the prevalence of ILO >2/1 exposure 27 \nmodel) demonstrates similar values, particularly at higher prevalences of silicosis. At 28 \nlower ranges, the differences between the missed case and NNS estimates of our 29 \nprimary and sensitivity analysis suggests further caution regarding figures outside of 30 \nour data range. Whilst severity of disease is likely a major determinant of sensitivity, 31 \nother important factors that we did investigate that may determine CXR sensitivity 32 \ninclude the quality of images and experience of the reader and who hired them [37].  33 \n 34 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 16\nCompared to HRCT, CXR has moderate sensitivity and high specificity for silicosis 1 \ndiagnosis. Both the type of reference test and the proportion of severe disease 2 \nplausibly influence CXR sensitivity. Regardless, using CXR for silicosis screening 3 \nresults in missed cases of silicosis and – in most scenarios – a relatively low number 4 \nof participants are needed to be screened with HRCT to detect a single extra case of 5 \nsilicosis.  6 \n 7 \nConflict of interest statement: No conflicts of interest 8 \n 9 \nFunding support: PH - Medical Research Council (MR/W024861/1). ML is partially 10 \nsupported by the Academy of Medical Sciences Professorship (APR7\\1005).  No 11 \nother financial disclosures 12 \n 13 \nNotation of prior abstract publication/presentation: Accepted for presentation by 14 \nAshwin Durairaj at the British Thoracic Society Winter meeting, London, United 15 \nKingdom, 27-29th November 2024 16 \n 17 \nAcknowledgements:  18 \nPH – Guarantor: full access to all the data in the study and takes responsibility for 19 \nthe integrity of the data and the accuracy of the data analysis, including and 20 \nespecially any adverse effects, Writing – Original draft preparation, 21 \nConceptualisation, Methodology, Project Administration, Data collection, Data 22 \nCuration, Visualisation, Formal Analysis 23 \nAD –  Original draft preparation, Conceptualisation, Methodology, Project 24 \nAdministration, Data collection, Data Curation, Visualisation, Formal Analysis 25 \nML - Conceptualisation, Formal Analysis, Visualisation, Writing – Review & Editing 26 \nJF – Conceptualisation, Project Administration, Supervision, Writing – Review & 27 \nEditing 28 \n  29 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 17\nFigure 1.  Directed acyclic diagrams of the conceptual relationship between 1 \ncumulative silica exposure and silicosis. Two counterfactual relationships are 2 \nrepresented. A. The relationship between cumulative silica exposure and radiological 3 \nsilicosis is fixed – sensitivity does not vary according to disease severity. B. The 4 \nrelationship between cumulative silica exposure and identification on radiological test 5 \nis mediated and thus relative to the severity of silicosis – sensitivity varies relative to 6 \ndisease severity. 7 \n 8 \nFigure 2. PRISMA Diagram  9 \n 10 \nFigure 3. Forest plots of sensitivity and specificity from meta-analysis of chest X-ray 11 \n(CXR), at an ILO >1/0 cut-off, for the diagnosis of silicosis.  12 \n 13 \nA. Describes the sensitivity (left) and specificity (right) of CXR compared to high-14 \nresolution computed tomography (HRCT). 15 \n 16 \nB. Describes the sensitivity (left) and specificity (right) of CXR compared to autopsy 17 \n 18 \nRandom effects model uses a generalised linear mixed model. Heterogeneity is 19 \ndetermined through the I2 statistic and Chi-squared (χ ²) test. 20 \n 21 \nAbbreviations: C.I, confidence interval. 22 \n 23 \nFigure 4. A mixed-effects linear meta-regression model showing the association 24 \nbetween the sensitivity of CXR at ILO category ≥ 1 and the ratio ILO >2/1 relative to 25 \n>1/0 silicosis across 7 studies. Each circle represents a study, with the diameter 26 \nproportional to the study's size. The circle is coloured depending on reference 27 \nstandard (Red = autopsy, Blue = computed tomography and Green = high resolution 28 \ncomputed tomography). The solid line represents the predicted sensitivity based on 29 \nthe ratio ILO >2/1 relative to >1/0 silicosis. The model explains a moderate 30 \nproportion of the heterogeneity in CXR sensitivity between studies (R² = 63%, I2 = 31 \n86%). 32 \n  33 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 18\nBibliography 1 \n1.  Feary J, Devaraj A, Burton M, Chua F, Coker RK, Datta A, et al. Artificial stone 2 \nsilicosis: a UK case series. Thorax. 2024 Sep 18;79(10):979–81.  3 \n2.  Fazio JC, Gandhi SA, Flattery J, Heinzerling A, Kamangar N, Afif N, et al. 4 \nSilicosis Among Immigrant Engineered Stone (Quartz) Countertop Fabrication 5 \nWorkers in California. JAMA Intern Med. 2023 Sep 1;183(9):991–8.  6 \n3.  Howlett P, Mousa H, Said B, Mbuya A, Kon OM, Mpagama S, et al. Silicosis, 7 \ntuberculosis and silica exposure among artisanal and small-scale miners: A 8 \nsystematic review and modelling paper. PLOS Glob Public Health. 9 \n2023;3(9):e0002085.  10 \n4.  Chen S, Liu M, Xie F. Global and national burden and trends of mortality and 11 \ndisability-adjusted life years for silicosis, from 1990 to 2019: results from the 12 \nGlobal Burden of Disease study 2019. BMC Pulm Med. 2022 Jun 21;22(1):240.  13 \n5.  International Labour Office, editor. Guidelines for the use of the ILO international 14 \nclassification of radiographs of pneumoconioses. Revised edition 2011. 15 \nGeneva: International Labour Office; 2011. 48 p. (Occupational safety and 16 \nhealth series).  17 \n6.  OSHA. Occupational Exposure to Respirable Crystalline Silica. Final rule. 18 \n[Internet]. Federal Register (USA); 2016 p. 16285–890. Report No.: 81(58). 19 \nAvailable from: https://www.federalregister.gov/documents/2016/03/25/2016-20 \n04800/occupational-exposure-to-respirable-crystalline-silica 21 \n7.  Health and Safety Executive (UK). Health surveillance for those exposed to 22 \nrespirable crystalline silica [Internet]. [cited 2024 Aug 22]. Available from: 23 \nhttps://www.hse.gov.uk/pubns/priced/healthsurveillance.pdf 24 \n8.  Hnizdo E, Murray J, Sluis-Cremer GK, Thomas RG. Correlation between 25 \nradiological and pathological diagnosis of silicosis: An autopsy population based 26 \nstudy. Am J Ind Med [Internet]. 1993 Oct [cited 2022 Oct 25];24(4):427–45. 27 \nAvailable from: https://onlinelibrary.wiley.com/doi/10.1002/ajim.4700240408 28 \n9.  Ehrlich R, Murray J, Rees D. Subradiological silicosis. Am J Ind Med [Internet]. 29 \n2018 Nov [cited 2022 Oct 25];61(11):877–85. Available from: 30 \nhttps://onlinelibrary.wiley.com/doi/10.1002/ajim.22909 31 \n10.  Hoy RF, Jones C, Newbigin K, Abramson MJ, Barnes H, Dimitriadis C, et al. 32 \nChest x-ray has low sensitivity to detect silicosis in artificial stone benchtop 33 \nindustry workers. Respirology. 2024 Sep;29(9):785–94.  34 \n11.  Silicosis Position Statement | RANZCR [Internet]. [cited 2024 Oct 16]. Available 35 \nfrom: https://www.ranzcr.com/college/document-library/silicosis-position-36 \nstatement 37 \n12.  All-Party Parliamentary Group for Respiratory Health. APPG Report: Improving 38 \nSilicosis outcomes in the UK [Internet]. UK Government; 2023 Jan. Available 39 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 19\nfrom: https://www.appg-respiratory.co.uk/sites/appg/files/2023-01/APPG-1 \nSilicosis-Report-01-23.pdf 2 \n13.  Whiting PF, Rutjes AWS, Westwood ME, Mallett S, Deeks JJ, Reitsma JB, et al. 3 \nQUADAS-2: a revised tool for the quality assessment of diagnostic accuracy 4 \nstudies. Ann Intern Med. 2011 Oct 18;155(8):529–36.  5 \n14.  Hodous TK, Chen RA, Kinsley KB, Liu XT, McLaughlin JK, Chen JQ, et al. A 6 \ncomparison of pneumoconiosis interpretation between Chinese and American 7 \nreaders and classifications. J Tongji Med Univ. 1991;11(4):225–9.  8 \n15.  Tamura T, Suganuma N, Hering KG, Vehmas T, Itoh H, Akira M, et al. 9 \nRelationships (I) of International Classification of High-resolution Computed 10 \nTomography for Occupational and Environmental Respiratory Diseases with the 11 \nILO International Classification of Radiographs of Pneumoconioses for 12 \nparenchymal abnormalities. Ind Health. 2015;53(3):260–70.  13 \n16.  Harrer M, Cuijpers P, Furukawa TA, Ebert DD. Doing Meta-Analysis With R: A 14 \nHands-On Guide [Internet]. 1st ed. Boca Raton, FL and London: Chapman & 15 \nHall/CRC Press; 2021. Available from: https://www.routledge.com/Doing-Meta-16 \nAnalysis-with-R-A-Hands-On-Guide/Harrer-Cuijpers-Furukawa-17 \nEbert/p/book/9780367610074 18 \n17.  Bégin R, Ostiguy G, Fillion R, Colman N. Computed Tomography Scan in the 19 \nEarly Detection of Silicosis. Am Rev Respir Dis [Internet]. 1991 Sep [cited 2024 20 \nSep 3];144(3_pt_1):697–705. Available from: 21 \nhttp://www.atsjournals.org/doi/abs/10.1164/ajrccm/144.3_Pt_1.697 22 \n18.  Grenier P, Valeyre D, Cluzel P, Brauner MW, Lenoir S, Chastang C. Chronic 23 \ndiffuse interstitial lung disease: diagnostic value of chest radiography and high-24 \nresolution CT. Radiology. 1991 Apr;179(1):123–32.  25 \n19.  Talini D, Paggiaro PL, Falaschi F, Battolla L, Carrara M, Petrozzino M, et al. 26 \nChest radiography and high resolution computed tomography in the evaluation 27 \nof workers exposed to silica dust: relation with functional findings. Occup 28 \nEnviron Med. 1995 Apr;52(4):262–7.  29 \n20.  Antao VCDS, Pinheiro GA, Terra-Filho M, Kavakama J, Muller NL. High-30 \nResolution CT in Silicosis: Correlation With Radiographic Findings and 31 \nFunctional Impairment. Journal of Computer Assisted Tomography [Internet]. 32 \n2005 May [cited 2024 Sep 9];29(3):350–6. Available from: 33 \nhttp://journals.lww.com/00004728-200505000-00015 34 \n21.  Murgia N, Muzi G, Dell’Omo M, Sallese D, Ciccotosto C, Rossi M, et al. An old 35 \nthreat in a new setting: High prevalence of silicosis among jewelry workers. Am 36 \nJ Ind Med. 2007 Aug;50(8):577–83.  37 \n22.  Sun J, Weng D, Jin C, Yan B, Xu G, Jin B, et al. The Value of High Resolution 38 \nComputed Tomography in the Diagnostics of Small Opacities and 39 \nComplications of Silicosis in Mine Machinery Manufacturing Workers, 40 \nCompared to Radiography. Jrnl of Occup Health [Internet]. 2008 Sep [cited 41 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 20\n2023 Jul 11];50(5):400–5. Available from: 1 \nhttps://onlinelibrary.wiley.com/doi/abs/10.1539/joh.L8015 2 \n23.  Lopes AJ, Mogami R, Capone D, Tessarollo B, de Melo PL, Jansen JM. High-3 \nresolution computed tomography in silicosis: correlation with chest radiography 4 \nand pulmonary function tests. J Bras Pneumol. 2008 May;34(5):264–72.  5 \n24.  Meijer E, Tjoe Nij E, Kraus T, van der Zee JS, van Delden O, van Leeuwen M, 6 \net al. Pneumoconiosis and emphysema in construction workers: results of 7 \nHRCT and lung function findings. Occup Environ Med. 2011 Jul;68(7):542–6.  8 \n25.  Berk S, Dogan DO, Gumus C, Akkurt I. Relationship between radiological (X-9 \nray/HRCT), spirometric and clinical findings in dental technicians’ 10 \npneumoconiosis. Clin Respir J. 2016 Jan;10(1):67–73.  11 \n26.  Takahashi M, Nitta N, Kishimoto T, Ohtsuka Y, Honda S, Ashizawa K. 12 \nComputed tomography findings of arc-welders’ pneumoconiosis: Comparison 13 \nwith silicosis. Eur J Radiol. 2018 Oct;107:98–104.  14 \n27.  Ş ener MU, Ş imş ek C, Özkara Ş , Evran H, Bursali İ , Gökçek A. Comparison of 15 \nthe International Classification of High-resolution Computed Tomography for 16 \noccupational and environmental respiratory diseases with the International 17 \nLabor Organization International Classification of Radiographs of 18 \nPneumoconiosis. Industrial Health [Internet]. 2018 Oct 19 [cited 2024 Oct 19 \n17];57(4):495. Available from: 20 \nhttps://pmc.ncbi.nlm.nih.gov/articles/PMC6685792/ 21 \n28.  León-Jiménez A, Hidalgo-Molina A, Conde-Sánchez MÁ, Pérez-Alonso A, 22 \nMorales-Morales JM, García-Gámez EM, et al. Artificial Stone Silicosis: Rapid 23 \nProgression Following Exposure Cessation. Chest. 2020 Sep;158(3):1060–8.  24 \n29.  Hoy RF, Dimitriadis C, Abramson M, Glass DC, Gwini S, Hore-Lacy F, et al. 25 \nPrevalence and risk factors for silicosis among a large cohort of stone benchtop 26 \nindustry workers. Occup Environ Med [Internet]. 2023 Aug [cited 2024 Sep 27 \n3];80(8):439–46. Available from: 28 \nhttps://oem.bmj.com/lookup/doi/10.1136/oemed-2023-108892 29 \n30.  Crawford JA, Sanyal S, Burnett BR, Wiesenfeld SL, Abraham JL. Accelerated 30 \nsilicosis in sandblasters: Pathology, mineralogy, and clinical correlates. Am J 31 \nInd Med. 2024 Mar;67(3):179–99.  32 \n31.  Bergin C, Muller N, Vedal S, Chan-Yeung M. CT in silicosis: correlation with 33 \nplain films and pulmonary function tests. American Journal of Roentgenology 34 \n[Internet]. 1986 Mar 1 [cited 2024 Sep 3];146(3):477–83. Available from: 35 \nhttps://www.ajronline.org/doi/10.2214/ajr.146.3.477 36 \n32.  Cowie RL, Hay M, Thomas RG. Association of silicosis, lung dysfunction, and 37 \nemphysema in gold miners. Thorax [Internet]. 1993 Jul 1 [cited 2024 Sep 38 \n3];48(7):746–9. Available from: 39 \nhttps://thorax.bmj.com/lookup/doi/10.1136/thx.48.7.746 40 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 21\n33.  Corbett EL, Murray J, Churchyard GJ, Herselman PC, Clayton TC, De Cock 1 \nKM, et al. Use of Miniradiographs to Detect Silicosis: Comparison of 2 \nRadiological with Autopsy Findings. Am J Respir Crit Care Med [Internet]. 1999 3 \nDec 1 [cited 2023 Apr 14];160(6):2012–7. Available from: 4 \nhttps://www.atsjournals.org/doi/10.1164/ajrccm.160.6.9903040 5 \n34.  Bégin R, Bergeron D, Samson L, Boctor M, Cantin A. CT assessment of 6 \nsilicosis in exposed workers. AJR Am J Roentgenol. 1987 Mar;148(3):509–14.  7 \n35.  Wang D, Yang M, Liu Y, Ma J, Shi T, Chen W. Association of Silica Dust 8 \nExposure and Cigarette Smoking With Mortality Among Mine and Pottery 9 \nWorkers in China. JAMA Netw Open [Internet]. 2020 Apr 14 [cited 2023 Feb 10 \n25];3(4):e202787. Available from: 11 \nhttps://jamanetwork.com/journals/jamanetworkopen/fullarticle/2764378 12 \n36.  Carneiro APS, Da Silva LL, Silva FDCL, Hering KG, Algranti E. Volume-based 13 \ntomography for the diagnosis of incipient silicosis in former gold miners. Occup 14 \nEnviron Med [Internet]. 2022 Jun [cited 2024 Sep 12];79(6):427–32. Available 15 \nfrom: https://oem.bmj.com/lookup/doi/10.1136/oemed-2021-107922 16 \n37.  Akgün M, Ozmen I, Ozari Yildirim E, Tuzun B, Nur Toreyin Z, Kayinova A, et al. 17 \nPitfalls of using the ILO classification for silicosis compensation claims. 18 \nOccupational Medicine [Internet]. 2022 Jul 1 [cited 2025 Mar 19];72(6):372–7. 19 \nAvailable from: https://doi.org/10.1093/occmed/kqac010 20 \n21 \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \n\n 22\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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint \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 May 29, 2025. ; https://doi.org/10.1101/2025.05.28.25328086doi: medRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}