Artificial stone and silicosis: the Korean experience Authors: Jun-Pyo Myong Published: 2026 ABSTRACT AbstractBackgroundEngineered stone (ES), also termed artificial stone, is a quartz-rich composite slab (typically 85–95% crystalline silica) used for kitchen and bathroom surfaces, and has driven a global resurgence of accelerated silicosis. Korea presents a distinctive context in which individual workers accumulated multi-era exposure across cementitious terrazzo (Dokkidashi), natural stone, polymer solid surface and high-quartz ES. We combine a narrative review of the Korean ES industry, respirable crystalline silica (RCS) exposure and clinical phenotype with a single-center case series.MethodsThis study reviewed Korean trade, employment, exposure-measurement, and adjudicated-case data, and conducted a retrospective case series at the Pneumoconiosis Clinic, Seoul St. Mary’s Hospital (index window 1–31 March 2026). Rapid-progression cases required radiographic pneumoconiosis plus progressive massive fibrosis (PMF)/complicated disease, a forced vital capacity (FVC) decline ≥ 150 mL/year, or accelerated disease at presentation, with ≥1 year of follow-up and serial spirometry.ResultsThe only ES-specific Korean exposure dataset (Institute of Occupation and Environment, C company, 2018) showed personal quartz of 0.045–0.179 mg/m3 (up to 3.6 times over the Korean occupational exposure limit) and cristobalite of 0.026 mg/m3 even under nominally wet processing; slab X-ray diffraction returned quartz contents of 24%, 65%, 70% and 100%, the 24%-quartz product additionally containing 28% cristobalite. Five rapid-progression cases (all men; median age 67 years) had 34–50 years of total stone work but only 3–7 years of ES exposure; all five had PMF, and the median annual FVC decline was −0.14 L/year (range −0.47 to −0.11).ConclusionsES exposure was superimposed on decades of prior mixed silica exposure—a difference of degree rather than of kind from series reported elsewhere, possibly reflecting the shorter latency available to younger ES-only workers. Surveillance, exposure standards and compensation criteria should address multi-era exposure rather than ES alone.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12995-026-00528-2.Keywords: Engineered stone, Artificial stone, Silicosis, Respirable crystalline silica Introduction Silicosis is among the oldest recognized occupational diseases, yet it is being reinvented in the twenty-first century by a contemporary product: engineered stone (ES). ES is a composite slab manufactured by binding crushed quartz aggregate (typically 85–95% by weight) with unsaturated polyester or acrylic resin and pigments. It became globally popular for kitchen and bathroom surfaces because it can be cast in uniform colors, is non-porous, and is competitively priced relative to natural granite. The very property that makes it desirable, however, also makes it dangerous: the silica fraction of ES is several-fold higher than that of granite (typically 10–45%) or marble (typically < 10%) [1, 2]. The same material is also called artificial stone, reconstituted or agglomerated stone, quartz agglomerate, silica conglomerate or, in the countertop trade, simply “quartz”; we use engineered stone and artificial stone interchangeably throughout. The first published cluster of ES-related silicosis appeared in Spain in 2010, followed shortly by reports from Israel, Italy, Australia, and subsequently the United States, the United Kingdom, China, and Belgium [1–4]. A consistent message has emerged across those reports: workers are young, the latency from first ES exposure to diagnosis is short (often 7–19 years, sometimes < 5 years), and a substantial minority progress to PMF, end-stage fibrosis, lung transplantation, or death [3–6]. The scale is now considerable. Spain illustrates the trajectory: the annual average of notified occupational silicosis cases rose from 28.4 in 1990–1994 to 404.2 in 2015–2019, and among notifications with detailed industry coding 55.9% arose in stone cutting, shaping and finishing, the category that encompasses quartz-agglomerate and natural-stone processing [7]. The Californian epidemic has been documented in successive reports [4, 5, 8, 9]. The most recent surveillance series identified 592 cases, including 31 deaths and 65 lung transplantations, between January 2019 and June 2026 [10]. Australia took the unprecedented regulatory step of prohibiting the manufacture, supply, processing and installation of all engineered-stone benchtops, panels and slabs with effect from 1 July 2024, extending the prohibition to importation from 1 January 2025 [11]. Critically, the hazard of crystalline silica extends beyond fibrosis: the International Agency for Research on Cancer classifies inhaled crystalline silica in the form of quartz or cristobalite from occupational sources as a Group 1 human carcinogen, with sufficient evidence for lung cancer [12]. Korea is a relevant national context for at least three reasons. First, Korean stone-surface work should be interpreted as two product streams rather than as a single linear substitution. In the floor, stair, and public-building surface stream, on-site cast cementitious terrazzo (Dokkidashi) (commonly Dokkidashi; Portland cement with marble, limestone, or mixed aggregate, wet-ground in situ) was progressively displaced by precast terrazzo tiles and other factory-finished floor products. In the countertop and bathroom interior stream, natural granite was followed by resin-based solid-surface panels and subsequently by high-quartz ES. These streams are technically distinct, but many Korean workers moved across both during long careers, so compensation and clinical attribution must handle multi-era mixed exposure. Second, traditional pneumoconiosis surveillance in Korea has historically focused on coal mining and metal smelting rather than on small stone-fabrication shops. Third, no national registry captures workers ever exposed to ES; affected workers are generally identified only when they present to occupational pulmonary clinics or when detailed exposure reconstruction is performed during a workers’ compensation evaluation. This article combines a narrative mini-review of the ES industry, respirable crystalline silica (RCS) exposure measurement and clinical phenotype with a single-center case series. Our aim is to provide a Korean-context synthesis useful to clinicians, industrial hygienists and regulators, and to characterize the clinical course of these patients against both classical Korean pneumoconiosis and the international ES literature. Methods This study has two components: a single-center retrospective clinical case series, which is the original contribution, and a narrative review of the Korean engineered-stone industry, RCS exposure and adjudicated occupational disease, which provides the context required to interpret it. Material definitions. Engineered stone denotes the quartz-rich resin composite, and artificial stone is used synonymously; no other material is so termed. Cementitious terrazzo denotes the Portland-cement product, including the on-site cast variant known as Dokkidashi. Polymer solid surface denotes acrylic or unsaturated polyester resin with aluminum-trihydrate filler, marketed in Korea as artificial marble and internationally exemplified by Corian®. Natural stone denotes quarried granite and marble. Each exposure period was assigned to one of these four categories; periods that could not be assigned with confidence were classed as mixed (Supplementary Table S1). Study setting and ascertainment. We conducted a single-center retrospective case series at the Pneumoconiosis Clinic, Department of Occupational and Environmental Medicine, Seoul St. Mary’s Hospital, The Catholic University of Korea. During the index period of 1–31 March 2026, 405 patients attended. The clinic evaluates the full range of occupational lung disease and its attendances therefore do not constitute a silicosis cohort: of the 405, 241 had a documented history of occupational silica exposure and 141 carried a confirmed discharge diagnosis of silicosis (ICD-10 J62.8). Patients with documented engineered-stone exposure were identified from this pool for screening. At each visit an occupational-medicine physician takes a structured face-to-face occupational history, and patients undergo pre- and post-bronchodilator spirometry and chest imaging. Inclusion and case definition. Inclusion required all of: (1) confirmed occupational engineered-stone exposure (cutting, fabrication or installation) from the structured occupational history; (2) chest imaging consistent with pneumoconiosis by the ILO Classification (2011)—small-opacity profusion ≥ 1/0 on radiograph read by an experienced radiologist and Airpneumo program-trained occupational medicine physician; and (3) ≥1 year of follow-up with ≥2 technically acceptable spirometry tests. Classification as a severe or rapidly progressive case additionally required at least one of: (a) PMF or complicated pneumoconiosis on HRCT (ILO large opacity A–C, conglomerate masses ≥ 1 cm, or whole-lung distortion); (b) mean annual decline in absolute FVC ≥ 150 mL/year; or (c) accelerated disease at first presentation, defined as PMF or ILO profusion category 3 on the diagnostic imaging. The 150 mL/year threshold approximates the ~157 mL/year distinguishing the most rapidly declining quartile of a large engineered-stone silicosis cohort [13]. For context, it lies well above the ≥60 mL/year rapid-decline threshold used in the older coal-miner literature [14]. Exposure history. Histories were obtained by an occupational-medicine physician using a structured occupational questionnaire, in interviews lasting approximately 30 minutes, and were reconstructed on a calendar basis with each material transition anchored to a datable employment event rather than to unaided recall. For every period we recorded employer and setting, materials worked, tasks (cutting, edging, polishing, drilling, hole-cutting, installation, clean-up), wet or dry method, working hours, and use of local exhaust ventilation and respiratory protective equipment (including type and consistency of use). Accounts were corroborated wherever documentary sources existed—workers’-compensation case files, employer and subcontractor records, and health-examination records—and where the recalled and the documented history diverged, the documented record was used and the discrepancy noted. Data source and availability Occupational and clinical information was entered into the electronic medical record during routine care and accessed retrospectively under the institution’s standard for medical-record-based case report research, with the analysis variables abstracted into a structured dataset. Because combinations of occupational history, dates and clinical variables could permit re-identification in a series of five patients, neither the source records nor the abstracted dataset are published; the de-identified dataset can be provided to the editors and reviewers on request and is available from the corresponding author on reasonable request, subject to Institutional Review Board approval. Measurements. For each patient we extracted demographics, smoking history in pack-years, age at first stone and first engineered-stone work, exposure years by material, working hours, task composition, and use of wet methods, local exhaust ventilation and respiratory protection. Spirometry was performed before and after bronchodilator on a calibrated flow-volume spirometer and accepted only when it met ATS/ERS 2019 criteria [15], with predicted values from Korean reference equations [16]. Chest CT images (1–3 mm slice thickness) and chest radiographs were independently reviewed by an ILO-trained occupational medicine physician. For each patient, annual change in pre-bronchodilator FVC and FEV1 was calculated as the difference between the first and the last acceptable measurement divided by the follow-up interval (L/year); this endpoint-based value is the measure reported throughout. Korean HS 6810.99 trade (2016–2025): import and export value and quantity. Source: Korea Customs Service trade statistics Service (tradedata.Go.kr) and KITA K-stat (stat.Kita.net), retrieved May 2026 [18] HS 6810.99 is a heterogeneous heading (articles of cement, concrete or artificial stone) and should be read as an upper bound for engineered-stone trade Narrative review and data sources. We conducted a narrative (non-systematic) review of the international and Korean literature on engineered-stone silicosis, the composition and toxicology of stone-trade materials, and RCS exposure, searching PubMed and Google Scholar together with relevant Korean-language sources (RISS [Research Information Sharing Service] and KCI [Korean Citation Index]), and hand-searching reference lists and relevant government and agency reports. Searches combined material terms (engineered stone, artificial stone, agglomerated stone, reconstituted stone, quartz conglomerate, quartz surfacing, solid surface, terrazzo, artificial marble; in Korean 인조석, 인조대리석, 테라조, 도끼다시), disease terms (silicosis, accelerated silicosis, pneumoconiosis, progressive massive fibrosis; 규폐증, 진폐증) and exposure terms (respirable crystalline silica, quartz, cristobalite, respirable dust, occupational exposure limit, work-environment measurement; 결정형 유리규산, 호흡성 분진, 작업환경측정), each with and without the qualifier Korea. No protocol was registered and no formal screening algorithm was applied; sources were selected for relevance to the Korean material eras, so the review provides interpretive context rather than an exhaustive synthesis. Korean industry and workforce context was characterized from official statistics: international trade in HS 6810.99 from the Korea Customs Service Trade Statistics Service and Korea International Trade Association, K-stat (2016–2025), and worker-population proxies from Statistics Korea (KOSIS; Mining and Manufacturing Survey and National Census on Establishments, classified by KSIC code) [17, 18]. Domestic exposure benchmarks were drawn from published Korean field studies of stone grinding, building-stone installation, and cement work, and from the only ES-specific Korean measurement available—the Institute of Occupation and Environment (IOE) work-environment evaluation of C company (13 December 2018). The Korean adjudicated-case landscape was summarized from IOE work-relatedness expert-assessment case files (2000–2021) [19–22]. Source selection for the evidence tables. Exposure sources were selected by an explicit hierarchy: direct Korean measurement of engineered-stone tasks where available, otherwise direct Korean measurement of the closest analogous stone task, otherwise an indirect domestic proxy, labeled as such and not treated as an engineered-stone measurement. Material-composition data (Supplementary Table S2) were restricted to Korean mineralogical analyses and to bulk analyses of products in actual use. Ethics and human research approval. The study was approved by the Institutional Review Board of Seoul St. Mary’s Hospital, The Catholic University of Korea (approval no. KC26WISI0342). The analysis used de-identified clinical data collected during routine occupational lung-disease care, and the requirement for individual informed consent was waived by the Board. Use of the de-identified IOE expert-assessment material was covered by the same approval. Search of the IOE public assessment reports. Work-relatedness expert-assessment reports issued by the IOE, Korea Workers’ Compensation & Welfare Service, between 2000 and 2021 are published on the Institute’s public notice board; the addresses and retrieval procedure are given in the footnote to Supplementary Table S3. The published series was screened using the terms 인조석, 인조대리석 and engineered stone with their English equivalents, and each retrieved report was then read in full by an occupational-medicine specialist to confirm documented quartz-resin engineered-stone exposure as a primary or significant substrate rather than an incidental mention. Of 32 stone-trade cases identified, four met that standard (인조석 (artificial stone), 인조대리석 (artificial marble) and engineered stone, with their English equivalents; Supplementary Table S3). Results The Korean stone-surface industry: material streams, exposure context and workforce Five stone-trade workers met the case definition; all had progressive massive fibrosis and lost forced vital capacity at a median of −0.14 L/year despite engineered-stone exposure of only 3–7 years within careers of 34–50 years. The exposure evidence is presented first because it establishes the material and measurement context in which those careers have to be read; the case findings follow in Section 4. Parallel material trajectories and worker transitions The Korean stone-surface material history is best organized by end-use and exposure mechanism (Table 1). The historical Dokkidashi/terrazzo stream is primarily a floor, stair, corridor, and institutional-surface technology, whereas ES is primarily an indoor countertop and bathroom-surface technology. Separating these streams avoids the misleading implication that ES simply replaced terrazzo. The occupationally important transition is that many individual workers, not products, moved across these streams: from on-site cementitious terrazzo/Dokkidashi grinding and natural-stone work into solid-surface and then ES fabrication. Table 1Korean stone-surface materials by end-use stream and period of dominancePeriod of dominanceMaterialPrincipal applicationShared substrate—worked in both streams throughoutLong established; mechanized era 1980s–2000s, continuingNatural stone (granite, marble)Finishing, cladding, flooring, some countertopsFloor and building-surface stream~1950s–1980s; residual repair thereafterOn-site cast cementitious terrazzo (Dokkidashi)Floors, stairs, corridors, public buildingsLate 1980s–presentPrecast terrazzo tileInstitutional flooring, stairs, pavingCountertop and interior-surface streamLate 1990s–2000sPolymer solid surface (artificial marble)Kitchen countertops, bathroom vanities2000s–presentQuartz engineered stoneCountertops, sink tops, bathroom and built-in surfacesFabrication and installation setting—where the same tasks are performed, not a further material generationMid-2010s–present; partial adoptionOff-site CNC/waterjet dimensioningShop fabrication of slabs of any materialLong-standing; continues alongsideOn-site placement and fittingTrimming, hole cutting and edge adjustment at installationA skeleton of the framework only. The work process, exposure profile and Korean exposure/adjudication relevance of every row are given in Supplementary Table S4. End-use stream refers to the application a material was principally made for, not to an industry or an employer. Period of dominance is the era in which a material was the usual choice in Korean practice within its stream, not the date of first or last use; earlier materials remain in service and are still encountered in repair, demolition and refurbishment work. Natural stone heads the table because it was the substrate worked in both streams throughout rather than a stage in either sequence. The final block is a different dimension from the material rows above it: off-site dimensioning and on-site fitting are alternative locations for the same fabrication tasks, apply to any of the materials listed, and are not sequential. Section 1.5 develops this point Korean stone-surface materials by end-use stream and period of dominance A skeleton of the framework only. The work process, exposure profile and Korean exposure/adjudication relevance of every row are given in Supplementary Table S4. End-use stream refers to the application a material was principally made for, not to an industry or an employer. Period of dominance is the era in which a material was the usual choice in Korean practice within its stream, not the date of first or last use; earlier materials remain in service and are still encountered in repair, demolition and refurbishment work. Natural stone heads the table because it was the substrate worked in both streams throughout rather than a stage in either sequence. The final block is a different dimension from the material rows above it: off-site dimensioning and on-site fitting are alternative locations for the same fabrication tasks, apply to any of the materials listed, and are not sequential. Section 1.5 develops this point This framework has two implications for the present review. First, the historical floor-surface pathway (on-site cementitious terrazzo/Dokkidashi → precast terrazzo tile) and the kitchen countertop pathway (natural granite (plus marble for interior/exterior finishing) → acrylic or polyester solid surface → quartz engineered stone) should not be conflated. Engineered stone is not a successor to terrazzo flooring; it is a high-silica successor to granite and to low-silica solid-surface products in the countertop niche. Second, the Korean worker is often the linking unit across these streams. A man who began with Dokkidashi or natural-stone installation in the 1970s–1980s may have accumulated decades of mineral-dust exposure differing in composition and of variable, at times considerable, intensity before a shorter engineered-stone period of far greater silica intensity. Compositional characteristics and toxicological distinctions The respiratory risk associated with these materials differs both quantitatively and qualitatively according to substrate composition and task. For occupational attribution the important question is not whether a material is called “artificial stone”, but whether the task generated respirable crystalline silica, alkaline cement dust, organic vapor, or a mixture of these hazards. Cementitious Dokkidashi /terrazzo - cement-dust dominant, but aggregate-dependent: Historical on-site cementitious terrazzo/Dokkidashi work should not be described as silica-free. The common marble- or limestone-chip formulation is low in quartz, and its occupational-hygiene profile is characterized more by cementitious and irritant dust than by crystalline silica; however, granite-chip terrazzo, adjacent natural-stone cutting, or repair grinding can add meaningful RCS. Because no direct Korean RCS measurement study of historical Dokkidashi/terrazzo was identified, exposure reconstruction should state the likely aggregate type, whether grinding was performed in situ, whether dry cutting occurred, and whether the worker also performed granite/marble fabrication. Acrylic/polyester solid-surface panels—low RCS comparator: ATH-filled MMA solid-surface products contain negligible crystalline silica in dominant formulations, so their main industrial-hygiene concerns are sanding dust, adhesives, and VOCs rather than silicosis. In a multi-era Korean career, this interval can be treated as a relative step-down in RCS compared with granite, followed by a step-change upward when the same worker transitions to quartz ES. Engineered stone-high-intensity RCS hazard: ES contains a high proportion of crystalline silica, and Korean site-specific IOE data confirm that commercially used slabs may reach the upper end of the international range, with one product containing both quartz and cristobalite [19, 20]. Cutting, edge-grinding, polishing and hole-cutting generate freshly fractured respirable silica particles capable of driving macrophage injury, oxidative and inflammatory responses, and progressive fibrosis, with lung cancer as an additional endpoint because occupationally inhaled crystalline silica is carcinogenic [12, 23–25]. The C company data show that even nominal wet processing may exceed the Korean OEL when dry edge-finishing or air-blowing is embedded in the workflow [19, 20]. Material composition across these material categories is summarized in Supplementary Table S2, and the Korean exposure benchmarks in Table 2. Table 2Korean exposure benchmarks relevant to cementitious terrazzo/Dokkidashi, natural-stone fabrication and installation, and engineered-stone fabricationSettingTaskSample typeSampling duration (min)Total dust (mg/m3)Respirable dust (mg/m3)RCS—quartz (mg/m3)RCS—cristobalite (mg/m3)C company (indoor fabrication)Wet cutting/polishingaPersonal (Worker A)3241.9180.2440.045n.d.C company (indoor fabrication)Wet cutting/polishingaArea3100.4320.1820.0200.004C company (indoor fabrication)Wet cutting/polishingPersonal (Worker B)3171.1780.1870.1220.006C company (indoor fabrication)Wet cutting/polishingArea3040.3140.1470.0310.004C company (indoor fabrication)Hole-cutting / edge-finishingPersonal (Worker C)3283.3680.4480.1790.026C company (indoor fabrication)Hole-cutting / edge-finishingArea3131.9130.2240.0840.005Outdoor installation [26]bCutting, polishing, drilling - stone attachmentPersonal (14 workers, 4 sites)8-h TWA-0.489 personal, 0.118 area0.043 (mean)-Korean OEL-----0.050.025 (ACGIH TLV)Cement manufacturing/handling [27] (cement-dust proxy)Cement manufacturing/handlingTD n = 53; RD n = 42-GM 0.10; production 0.207; handling 0.085GM 0.08; production 0.362; handling 0.054not measured-Stone grinding [28]Stone grindingPersonal (n = 15)240–3901.372 (GM)0.170 (GM); 93.3% >Korean OEL-a: Worker A was stationed adjacent to the hole-cutting cell (interior wall, door open); exposure reflects partial bystander contribution rather than pure cutting/polishing. n.d., not detected;—, not measured or not applicable; GM, geometric mean; TD, total dust; RD, respirable dust; C company values are reproduced from Institute of Occupation and Environment; work-environment evaluation, 13 December 2018 [19, 20]. Yi et al. [26] and Kim et al. [28] provide direct Korean benchmarks for stone installation and stone grinding, respectively. Bae et al. [27] is included only as an indirect Portland-cement dust benchmark for cementitious terrazzo/Dokkidashi; it is not a terrazzo-specific RCS measurement. Korean OEL for crystalline silica: 0.05 mg/m3 (8-h TWA); ACGIH TLV for cristobalite: 0.025 mg/m3 Korean exposure benchmarks relevant to cementitious terrazzo/Dokkidashi, natural-stone fabrication and installation, and engineered-stone fabrication a: Worker A was stationed adjacent to the hole-cutting cell (interior wall, door open); exposure reflects partial bystander contribution rather than pure cutting/polishing. n.d., not detected;—, not measured or not applicable; GM, geometric mean; TD, total dust; RD, respirable dust; C company values are reproduced from Institute of Occupation and Environment; work-environment evaluation, 13 December 2018 [19, 20]. Yi et al. [26] and Kim et al. [28] provide direct Korean benchmarks for stone installation and stone grinding, respectively. Bae et al. [27] is included only as an indirect Portland-cement dust benchmark for cementitious terrazzo/Dokkidashi; it is not a terrazzo-specific RCS measurement. Korean OEL for crystalline silica: 0.05 mg/m3 (8-h TWA); ACGIH TLV for cristobalite: 0.025 mg/m3 Product classification and Korean market dynamics Engineered-stone products are captured within HS 6810.99 (“Articles of cement, concrete or artificial stone, n.e.s.”) in Korean customs data—a heading that also includes cement and concrete articles and should therefore be read as an upper-bound proxy for ES trade rather than a clean ES-only series [18, 29]. Year-by-year Korean import/export data for HS 6810.99 from 2016 to 2025 are illustrated in Fig. 1; full annual trade values and partner breakdown are provided in Supplementary Table S5. Three patterns are epidemiologically relevant. Korea is a substantial net importer over the decade (cumulative imports ≈778 M USD vs. exports ≈477 M USD; net −301 M USD), consistent with a substantial downstream fabrication and installation market. Export volume to the United States surged 2019–2021 (peak 86.7 M USD/153,678 t in 2020) and declined thereafter. The principal Korean manufacturers, LX Hausys and Hyundai L&C, are export-oriented, and their plant workers constitute an additional at-risk population that downstream fabrication statistics do not capture. The apparent transition from solid-surface products toward quartz engineered stone in the countertop market should be interpreted cautiously. Plausible market and technical factors include stone-like appearance, heat and scratch resistance, remodelling demand, increasing availability of quartz slabs, and compatibility with established stone-fabrication practices. Fire-safety changes following the 2010 Woosin Golden Suite exterior-cladding fire, and Article 61 of the Enforcement Decree of the Building Act, are relevant to the broader Korean market for non-combustible and semi-combustible finishing materials [30, 31], but they should not be presented as direct evidence that quartz engineered stone replaced solid surface in indoor countertops in the absence of product-specific market data. For occupational health the essential point is substrate toxicity: the same cutting, polishing and hole-cutting tasks deliver far greater RCS potential on quartz engineered stone than on ATH-filled solid surface. Worker population and surveillance gap Three occupational groups are relevant. Manufacturing-plant workers mix resin, quartz and pigments and operate vibro-compaction and curing lines; their dust exposures depend on whether quartz is handled wet or dry. Fabrication-shop workers (countertop “stoneworkers”) cut, edge, polish, drill and dry-grind ES slabs to dimension; this group has shown the highest reported RCS exposures internationally and is the dominant reservoir of severe disease [1, 2, 13]. Installation workers perform on-site cutting and polishing during fitting; this is often dry, dusty work in confined indoor spaces [3, 4]. Korean worker-population proxies (KOSIS): No Korean Standard Industrial Classification (KSIC) code is specific to ES fabrication, so two proxies of different scope are used. The narrow manufacturing proxy, KSIC 23,999 (other non-metallic mineral product manufacturing, n.e.c.), comprised approximately 2,000 workers in the 2016–2019 Mining and Manufacturing Survey and 3,206 workers in the 2024 National Census on Establishments; because these figures come from different statistical instruments, they indicate approximate scale rather than a directly comparable increase over time. The broader construction-trade proxy, which adds the installation specialty trades (KSIC 423, 424 and 4259, all enterprises, 2024), gives 32,457 workers in 2024. Neither proxy is engineered-stone-specific: the narrow one omits installation work and includes unrelated manufacturing, and the broader one includes stone-trade workers who never handle engineered stone. Both series are shown in Fig. 2 and given in full in Supplementary Table S6 [17]. Korean stone-trade worker population proxies, 2016–2024 (KOSIS). Statistics Korea (KOSIS). Two proxies of different scope, neither specific to engineered stone: KSIC 23,999 (other non-metallic mineral product manufacturing, n.e.c.), the narrowest available manufacturing code; and construction specialty trades (KSIC 423, 424, 4259), which include stone workers who never handle engineered stone. 2016–2019, Mining and manufacturing survey (enterprises ≥ 10 workers); 2023–2024, national Census on Establishments (all enterprises). The change of source produces the break in the series; the segments are not directly comparable, and the figure shows approximate scale rather than a count of exposed workers Absence of an official at-risk denominator: No Korean administrative register provides a reliable count of workers currently or formerly exposed to ES, and the size of the ES-exposed workforce cannot be derived from any single registry. In practice, engineered-stone exposure histories are currently identified mainly through occupational pulmonary clinics and through workers’ compensation evaluations, because no population-based surveillance system or exposure registry provides quantitative ascertainment of engineered-stone-related silicosis in Korea. Fabrication setting and implications for exposure control Two historically distinct shifts moved dust-generating processing away from the point of installation, although they occurred in different material streams and for different reasons. The first was the transition from in-situ cast cementitious terrazzo to factory-produced precast tile from the late 1980s—a change of material and construction system as well as of location—which transferred casting, grinding, polishing and the associated effluent from the construction site into a manufacturing plant and left adhesive setting and limited cutting on site. This period coincided with the extension of environmental controls to construction sites: construction became a fugitive-dust generating activity subject to prior notification and prescribed suppression under the Clean Air Conservation Act, and alkaline cement slurry from in-situ wet grinding could no longer be discharged to storm-water or sewer lines under the Water Quality Conservation Act (renamed the Water Environment Conservation Act in 2018) and the Sewerage Act, requiring settling and neutralization on site. These obligations bore on site-based wet grinding and may have favored factory production, alongside shortening construction schedules and the declining availability of the skilled trade required for multi-stage in-situ grinding. We identified no Korean study establishing their relative contribution and do not present the regulatory changes as a demonstrated cause; the drivers are compared in Supplementary Table S7. The second was the increasing transfer of countertop fabrication to off-site workshops from the mid-2010s, a change principally of fabrication location and workflow. Quartz composite requires stone-cutting equipment and controls rather than the woodworking-type tooling commonly used for solid-surface fabrication, and sink cut-outs, mitred edges and seam tolerances are achieved by digital templating followed by CNC or waterjet cutting, with a mis-cut wasting an entire slab. Wet cutting is in any case impractical in the occupied apartments where much Korean countertop replacement takes place, and the principal manufacturers supply through authorized fabrication shops, which separates fabrication from installation. Off-site dimensioning removes the highest-dust tasks from the installation site. KCS 41 35 04 provides the construction-standard framework for terrazzo works [32]. Uptake by small fabrication shops is incomplete; final trimming, hole cutting and edge adjustment continue at the point of installation, often dry and in enclosed rooms; and no Korean measurement program has systematically compared off-site fabrication with on-site fitting, the available C company data characterizing only a single small indoor fabrication shop. Exposure assessment: respirable crystalline silica International ES-specific measurements International measurements of RCS in ES fabrication typically exceed regulatory limits. Exposure surveys in US fabrication shops have documented 8-hour time-weighted-average RCS in the range of tens to a few hundred µg/m3, with a substantial fraction of personal samples exceeding the OSHA permissible exposure limit of 50 µg/m3 [6, 33]. Enforcement in the United States has since been directed at this industry through a focused inspection initiative [34]. Laboratory characterization of dust generated by ES cutting demonstrates an unusually high respirable fraction (<1 µm) and a high quartz-plus-cristobalite content of the airborne dust, consistent with the slab composition itself [23, 25]. Korean exposure data: a literature-based context For the present manuscript, the Korean evidence base relevant to stone-surface work can be organized into three exposure settings: cementitious terrazzo/Dokkidashi, natural-stone fabrication and installation, and engineered-stone fabrication. No Korean study was identified that directly measured respirable crystalline silica during historical on-site terrazzo or Dokkidashi work; this exposure era should therefore not be quantified using unrelated non-stone construction data. It is best interpreted as a mixed cementitious-dust exposure whose RCS potential depends strongly on aggregate type: marble- or limestone-chip terrazzo would generate little quartz, whereas granite-chip terrazzo could generate RCS approaching that of natural granite. As an indirect domestic benchmark for the cementitious component, Bae et al. reported geometric mean total and respirable dust concentrations of 0.10 and 0.08 mg/m3, respectively, in Korean cement manufacturing and handling workplaces; respirable dust was higher in cement-production tasks than in handling tasks. This study should be cited only as a proxy for cement-dust exposure, not as a direct terrazzo RCS measurement [27]. For natural-stone fabrication, the most directly relevant Korean study is the stone-grinding survey by Kim et al., in which the geometric mean quartz concentration in respirable dust was 0.170 mg/m3 and 93.3% of samples exceeded the Korean occupational exposure limit [28]. This provides a domestic quantitative basis for reconstructing RCS exposure during granite and other natural-stone cutting, grinding and polishing before the widespread adoption of engineered stone. An earlier Korean manufacturing-industry study also reported an average personal respirable-dust concentration of 0.24 mg/m3 in stone workplaces, supporting the presence of measurable respirable-dust exposure in the pre-ES stone-manufacturing sector [35]. For installation work, the most applicable Korean benchmark is Yi’s 2011–2012 field survey of building-stone installation work, in which 14 workers at four construction sites performed cutting, polishing and drilling. The mean respirable-dust concentration was 0.489 (personal) and 0.118 (area), and the mean RCS concentration was 0.043 mg/m3, close to the Korean OEL of 0.05 mg/m3 [26]. This value is directly relevant to stone-setting and installation histories and can be used as a domestic benchmark for outdoor or semi-outdoor stone-installation exposure. The only ES-specific Korean exposure dataset currently available is a targeted workplace measurement performed by the Institute of Occupation and Environment (IOE), Korea Workers’ Compensation & Welfare Service, on 13 December 2018 at a sink-countertop fabrication facility (C company) as part of an occupational-disease epidemiologic investigation [19, 20]. C company is a small (4–5 worker) Korean enterprise that began operating in 2002 and has, since approximately 2004, predominantly handled engineered stone with quartz content of 90% or more for residential sink and benchtop manufacture. The plant uses a continuous wet workflow of slab intake → wet cutting → wet edge polishing → hole-cutting for sink basins and cooktops → finished-product dispatch, with cells partitioned by full-height interior walls; nominal hole-cutting was wet, but dry edge-grinding and air-blowing were also performed at hole margins on the day of measurement. Bulk X-ray diffraction (NIOSH 7500; Bruker D8) on the four engineered-stone products in active use at the plant showed quartz contents of 65, 70 and 100% in three products; the fourth contained 24% quartz, 28% cristobalite and 13% resin. The shop inventory therefore included products at the upper end of the international engineered-stone silica range. Personal and area air samples collected the same day, sized to the respirable fraction, returned the 8-hour TWA exposures (NIOSH 0500/0600/7500 methods) summarized in Table 2 [19, 20]. Four observations follow from these measurements. First, even with a wet workflow, the personal quartz exposure of the cutting/polishing worker at C company reached 0.122 mg/m3—about 2.4 times higher than the Korean OEL of 0.05 mg/m3. Second, the hole-cutting/edge-finishing personal sample reached 0.179 mg/m3 quartz (3.6 times the Korean OEL) and 0.026 mg/m3 cristobalite, slightly above the ACGIH TLV, with dry edge touch-up and air-blowing embedded within an otherwise wet workflow. Third, Korean natural-stone grinding and building-stone installation provide the most relevant domestic quantitative benchmarks for pre-ES stone work: stone grinding clearly exceeded the quartz OEL [28], while outdoor installation averaged 0.043 mg/m3 RCS [26]. Fourth, the Dokkidashi/terrazzo era differs qualitatively from natural-stone and ES work: the available Korean cement data support cementitious dust exposure [27], but no direct Korean RCS measurement for historical terrazzo/Dokkidashi was identified. To our knowledge, the C company dataset has not previously been incorporated into a Korean systematic exposure review of the engineered-stone industry. It should therefore be described as a site-specific workers’ compensation investigation rather than a representative national survey. Its value is nonetheless high: the shop was small but used partitioned, nominally wet processing, so the exceedances were recorded under conditions more controlled than unpartitioned dry fabrication or on-site fitting. Material composition—Korean natural stone versus engineered stone The clinical magnitude of the C company numbers can only be fully appreciated alongside the crystalline-silica content of the slab materials being cut. Korean adjudicated case files [19, 20] uniformly cite the mineralogical analysis of Choi et al. [36] for Korean granite quartz content: Iksan granite 32.3–34.2%, Pocheon granite 24.6–49.8%, and Geochang granite 20.4–42.8%, with imported Chinese granite at 28.3–42.7%. Korean carbonate marble contains essentially negligible crystalline silica (<5%). Polymer “solid surface” products (MMA / unsaturated polyester resin with aluminum trihydrate filler) contain essentially negligible crystalline silica (<1%) in ATH-filled formulations, which predominate among Korean brands. Cement-based terrazzo has highly variable crystalline silica content depending on aggregate type: marble-chip or limestone-chip terrazzo carries essentially negligible quartz (<1%), while granite-chip terrazzo may approach the granite range (~20–40%) [36]. Engineered stone generally contains approximately 65–95% crystalline silica [24, 25]. In the C company inventory, three products contained 65, 70 and 100% quartz, while the fourth contained 24% quartz together with 28% cristobalite [19]. These values are consolidated in Supplementary Table S2. The comparisons that carry the argument are that Korean granite contains roughly 20–50% quartz whereas most of the engineered-stone products tested at C company contained substantially more, and that one of the four products analyzed at C company contained 28% cristobalite, which is more biologically reactive than α-quartz. The C company and building-installation datasets are complementary rather than comparable and together bracket the range of Korean stone-trade exposure. The C company measurements characterize engineered-stone fabrication indoors in a single small enclosed workshop, whereas the 2011–2012 survey of four building sites characterized the installation of natural stone (granite and marble cladding) outdoors or semi-outdoors, with mean respirable dust of 0.489 (personal) and 0.118 (area) and mean RCS of 0.043 mg/m3 [26]. Outdoor natural-stone installation therefore approaches the Korean OEL even outside enclosed shops, while indoor engineered-stone fabrication exceeds it severalfold. Adjudicated occupational disease cases with documented engineered-stone exposure Occupational disease recognition data cannot be read as a disease-incidence registry, but neither are they a pure exposure record. As the historiography of occupational disease has shown, and of silicosis in particular, recognition data simultaneously reflect exposure, disease and the categories and evidentiary thresholds of the medico-legal system through which claims pass [37]. What these files therefore demonstrate is which combinations of exposure and disease were recognizable within the Korean compensation system at a given time. Of 32 stone-trade cases reviewed for the period 2000–2021, four had engineered stone documented as a primary or significant exposure substrate.These four are tabulated case by case in Supplementary Table S3; a detailed analysis of the Korean compensation experience is beyond the scope of the present report and is being prepared separately. Two features of this recognized stream are notable. First, it is dominated by COPD and lung cancer rather than silicosis: one worker with about 40 years of stone installation was recognized with severe COPD, one with a mixed foundry and stone history with IPF, and two with about 9–13 years of engineered-stone cutting and polishing with lung cancer at 41 and 48 years of age. Second, engineered stone was documented in only four of the 32 cases and only from 2019 onward. Exposure intensity in these files varied by method and substrate: open-air installation averaged 0.489 (personal) and 0.118 (area) mg/m3 respirable dust (RCS 0.043 mg/m3), whereas engineered-stone fabrication exceeded the OEL ~2.4-fold under wet cutting and polishing and ~3.6-fold during hole-boring. Both features are interpreted in the Discussion. Single-center case series: severe or rapidly progressive engineered-stone-associated silicosis Case ascertainment During the index window 405 patients attended the clinic. This is a general occupational-lung-disease population rather than a silicosis cohort: it comprised patients with coal workers’ pneumoconiosis, silicosis, asbestosis and humidifier-disinfectant lung injury, patients with lung cancer or COPD, and patients attending for evaluation in support of a workers’-compensation claim. Of these, 241 had a documented history of occupational silica exposure, and 141 carried a confirmed discharge diagnosis of silicosis (ICD-10 J62.8). Seventeen workers with documented engineered-stone exposure were identified from this pool and screened. Three had no radiographic evidence of pneumoconiosis and were excluded from the progression analysis; three had insufficient longitudinal data (one presented with a primary lung cancer and two had only a single spirometry measurement). Of the remaining 11, five met the case definition and form the basis of the FVC-trajectory analysis (Table 3). Table 3Case series of severe or rapidly progressive silicosis with prior engineered-stone exposure (n = 5)Case noAge/Dx age (year)Smoking history(PY)Total stone exposure periods (yrs)Exposure years by material (natural stone / solid surface / engineered stone)Solid surface / engineered stone start yearILO classificationsPMF yes/nomMRCPre-BD FVC % predicted (Δ%)Cumulative ΔFVC (L)Annual Δ FVC (L/yr)Annual Δ FEV1 (L/yr)PFT follow-up (yrs)Clinical courseInclusion criterion167/59 (2017)104425/13/61997/20092/1, q-tyes499 → 64 (Δ−35%)−1.45−0.28−0.235.1Worsening symptoms; PMF size increase(a) + (b)266/55 (2015)04020/15/51995/20102/3, r-qyesRadiograph–HRCT discordance; progression371 → 62 (Δ−9%)−0.42−0.11−0.143.9Radiograph–HRCT discordance; progression(a)371/57 (2012)105019/28/31982/20092/3, r-qyes470 → 48 (Δ−22%)−1.09−0.14−0.117.9Progressive over follow-up period(a) + (c)467/58 (2017)14030/3/72007/20103/3, r-ryes473 → 51 (Δ−22%)−0.89−0.13−0.087.0Continued FVC decline(a) + (c)569/66 (2023)0347/21/61989/20102/2, r-ryes496 → 26 (Δ−70%)−2.60−0.47−0.275.6Most rapid decline; non-smoker(a) + (b) + (c)PY, pack-years; ES, engineered stone; ILO, International Labor Organization small-opacity classification (major / minor reading; shape codes per ILO 2011); PMF, progressive massive fibrosis; mMRC, modified Medical Research Council dyspnea scale; PFT, pulmonary function test. Pre-BD = pre-bronchodilator. Annual ΔFVC and Annual ΔFEV1 were calculated by dividing the change between the first and last pre-bronchodilator measurement by the PFT follow-up duration, as specified in the Methods. Start years = estimated year of first solid-surface and first engineered-stone work based on structured occupational interview. Qualifying criteria for the severe/rapidly progressive definition: (a) PMF or complicated pneumoconiosis on HRCT (ILO large opacity A–C, conglomerate masses ≥ 1 cm, or whole-lung distortion); (b) mean annual decline in absolute FVC ≥ 150 mL/year; (c) documented PMF or profusion 3 pneumoconiosis at first presentation. Case series of severe or rapidly progressive silicosis with prior engineered-stone exposure (n = 5) yes Radiograph–HRCT discordance; progression PY, pack-years; ES, engineered stone; ILO, International Labor Organization small-opacity classification (major / minor reading; shape codes per ILO 2011); PMF, progressive massive fibrosis; mMRC, modified Medical Research Council dyspnea scale; PFT, pulmonary function test. Pre-BD = pre-bronchodilator. Annual ΔFVC and Annual ΔFEV1 were calculated by dividing the change between the first and last pre-bronchodilator measurement by the PFT follow-up duration, as specified in the Methods. Start years = estimated year of first solid-surface and first engineered-stone work based on structured occupational interview. Qualifying criteria for the severe/rapidly progressive definition: (a) PMF or complicated pneumoconiosis on HRCT (ILO large opacity A–C, conglomerate masses ≥ 1 cm, or whole-lung distortion); (b) mean annual decline in absolute FVC ≥ 150 mL/year; (c) documented PMF or profusion 3 pneumoconiosis at first presentation. Demographic and exposure summary All five were Korean-born males, with a median age of 67 years (range 66–71). Total stone work ranged from 34 to 50 years (median 40), whereas ES-specific exposure ranged from 3 to 7 years (median 6). Two were never-smokers; the remaining three had ≤ 10 pack-years. None reported consistent use of respiratory protective equipment. Lung-function trajectory and radiologic findings Median annual FVC change was −0.14 L/year (range −0.47 to −0.11 L/year) and median cumulative FVC change was −1.09 L (range −0.42 to −2.60 L) over a median follow-up of 5.6 years (range 3.9–7.9), across a median of six spirometries per patient (range 5–9). Pre-bronchodilator FVC at the start of the observation window ranged from 70% to 99% predicted; by the last available measurement every case had fallen by at least 9 percentage points of predicted, and three had fallen by more than 20. Annual ΔFEV1 tracked closely with annual ΔFVC (range −0.27 to −0.08 L/year), confirming a predominantly restrictive and progressive pattern rather than a variable obstructive one. Two cases (Cases 1 and 5) declined by ≥150 mL/year and therefore met qualifying criterion (b) in addition to criterion (a). The fastest-declining case (Case 5, −0.47 L/year) was a never-smoker; the slowest (Case 2, −0.11 L/year) showed bilateral multifocal conglomerate masses on HRCT despite low radiographic profusion. All five cases had PMF or complicated pneumoconiosis on HRCT. Individual FVC trajectories are shown in Fig. 3, plotting every available pre-bronchodilator measurement against time since each patient’s index spirometry, in liters (panel A) and as % predicted (panel B). Case 1 is notable for beginning within the normal range (4.00 L, 99% predicted) and falling to 2.55 L (64%) within 5.1 years. Case 5 shows a biphasic course, declining from 3.64 L (96%) to 1.04 L (26%)—the steepest trajectory in the series at −12.5 percentage points of predicted per year. Cases 3 and 4 decline steadily from an already reduced baseline over the longest follow-up, and although their absolute rates are almost identical (−0.14 and −0.13 L/year), Case 4 loses more relative to predicted (−3.2 versus −2.8 percentage points per year). Case 2, the slowest, shows the visit-to-visit variability that makes endpoint-based estimates less stable in that patient. Serial pre-bronchodilator FVC in the five cases, plotted against time since each patient’s index spirometry. (A) Measured FVC in liters, with the baseline and final values labeled and the annualized change shown in the key as reported in Table 3. (B) The same measurements expressed as % predicted, with baseline and final values labeled and the annualized rate of change shown in the key. Every available measurement is plotted; Korean reference equations were used for predicted values Selected case details Case 1—A 67-year-old man (diagnosed at 59; 10 pack-years) had worked 44 years in the stone trades: 25 years of natural-stone masonry, then 13 years fabricating solid-surface (artificial marble) sinks and kitchen countertops, and finally 6 years fabricating quartz engineered stone. Chest imaging (Fig. 4) showed PMF with ILO profusion 2/1 (q-t). Across five spirometries over 5.1 years, FVC fell from 99% to 64% predicted (pre-bronchodilator), with an annual ΔFVC of −0.28 L/year and a cumulative decline of −1.45 L, accompanied by progressive enlargement of the PMF lesion and worsening dyspnea (mMRC 4). Case 1—progressive massive fibrosis (PMF) in a 67-year-old fabricator (44 yr stone work; final 6 yr engineered stone). (A) June 2017 and (B) June 2020 chest CT show interval enlargement of the conglomerate (PMF) lesion Case 2—A 66-year-old man (diagnosed at 55; never-smoker) began stone-cutting and installation at age 26 and accumulated 40 years in the stone trades: 20 years of natural stone, 15 years of solid surface, and a final 5 years of engineered stone. His primary tasks were slab cutting and on-site installation rather than bench fabrication; he worked in enclosed small shops for more than 10 hours per day with no consistent use of respiratory protective equipment. Plain chest radiography in 2016 suggested only early, mild pneumoconiosis, whereas HRCT subsequently demonstrated bilateral multifocal conglomerate masses (ILO 2/3 r-q) (Fig. 5). Serial spirometry over 3.9 years showed the slowest decline in the series (annual ΔFVC −0.11 L/year; cumulative −0.42 L; FVC 71% → 62% predicted). Case 2—radiograph–HRCT discordance in a 66-year-old never-smoker. (A) July 2016 CT suggested only mild pneumoconiosis, whereas (B) January 2024 HRCT reveals bilateral multifocal conglomerate masses—despite the slowest functional decline in the series (−0.11 L/year) Case 3—A 71-year-old man (diagnosed at 57; 10 pack-years) had 19 years of natural-stone work followed by 28 years of solid-surface work and 3 years of engineered-stone fabrication, total 50 years. He had ILO 2/3 (r-q) PMF (Fig. 6) and mMRC 4 dyspnea. Nine spirometries over 7.9 years documented an FVC decrement from 70% to 48% predicted (pre-bronchodilator; cumulative −1.09 L; annual −0.14 L/year). He exemplifies a worker who began traditional stonework in his twenties and moved into engineered-stone fabrication late in his career. Case 3—disease progression over follow-up in a 71-year-old worker (50 yr stone work). (A) November 2012 and (B) August 2022 chest CT show progression of conglomerate fibrosis; FVC declined from 70% to 48% predicted (−0.14 L/year) Case 4—A 67-year-old man (diagnosed at 58; never-significant smoker, 1 pack-year) had 40 years of stone work: 30 years of natural-stone installation, 3 years of solid surface, and 7 years of engineered-stone fabrication of sinks and kitchen countertops including hole-cutting. He presented with complicated pneumoconiosis (numerous nodular infiltrations with bilateral large opacities, ILO 3/3 r-r) (Fig. 7), mMRC 4 dyspnea, and an FVC decline of −0.13 L/year over 7.0 years (pre-bronchodilator 73% → 51%; cumulative −0.89 L). Case 4—complicated pneumoconiosis at diagnosis (2019) in a 67-year-old worker. (A) July 2019 and (B) January 2021 chest CT show progression of conglomerate fibrosis Case 5—A 69-year-old never-smoker (diagnosed at 66) had spent 34 years in the stone trades, beginning with on-site cementitious terrazzo (Dokkidashi): 7 years of terrazzo and natural-stone work, 21 years of solid surface, and 6 years of engineered stone. HRCT (Fig. 8) showed clustered upper-lobe perilymphatic nodules with conglomeration, paracicatricial emphysema and early PMF (ILO 2/2 r-r; mMRC 4). Across six spirometries between August 2020 and April 2026 (5.6 years), FVC fell from 96% to 26% predicted—the most rapid decline in the series (annual ΔFVC −0.47 L/year; cumulative −2.60 L). The trajectory was biphasic: FVC fell by 0.76 L over the first 3.2 years (−0.24 L/year) and then by a further 1.84 L over the next 2.4 years (−0.77 L/year), indicating late acceleration rather than uniform decline. Case 5—rapidly progressive silicosis in a 69-year-old never-smoker. April 2024 HRCT shows clustered upper-lobe perilymphatic nodules with conglomeration, paracicatricial emphysema, and early PMF; the fastest decline in the series (−0.47 L/year; FVC 96% → 26% predicted) Discussion This single-center Korean series supports three observations, which we then use to frame four gaps in the Korean system for identifying and preventing engineered-stone disease. First, the clinical course in these patients was rapid rather than chronic-stable. All five had PMF or complicated pneumoconiosis, but this reflects the selection criteria—PMF was itself one of the qualifying features—and cannot be compared with the PMF prevalence at diagnosis reported from unselected engineered-stone cohorts. What is not determined by the selection criteria is the magnitude of functional loss. The median annual FVC decline of −0.14 L/year (range −0.47 to −0.11) lies well outside the range reported for healthy ageing adults: a systematic review of prospective cohorts found FVC decline estimates of 14.1–65.6 mL/year, the upper figure coming from a cohort of mean age 73 years [38]. Our median is more than twice that upper figure and Case 5 exceeds it sevenfold, and Cases 1 and 5 each lost more than 1.2 L of absolute FVC over the observation interval. Two of the five exceeded the 150 mL/year threshold, which was only one of three alternative qualifying criteria, and in Case 5 the loss accelerated from −0.24 to −0.77 L/year during follow-up. Second, residual confounding by smoking does not explain the trajectory. Two of five never smoked, and the fastest-declining case (Case 5, −0.47 L/year) had zero pack-years; the maximum burden among the three smokers was 10 pack-years. The gradient of FVC decline does not correlate with smoking intensity, and the three largest cumulative decrements (Cases 5, 1, and 3: −2.60, −1.45, and −1.09 L) involved workers with highly variable ES career fractions (18, 14, and 6%, respectively), indicating that cumulative prior silica burden—rather than ES fraction alone—determines disease severity. Third, the observed pattern is consistent with engineered stone acting as a late-career high-silica accelerator, although this interpretation remains hypothesis-generating. ES exposure was brief in all five cases (3–7 years; median 6) and occurred after decades of earlier stone-trade work (total career 34–50 years; median 40), including natural-stone and solid-surface work. Thus, unlike the predominantly younger, ES-focused workers described in several international outbreaks, the patients identified in this Korean series represent an older, multi-era exposure phenotype in which a relatively short period of high-intensity ES exposure was superimposed on a substantial prior occupational dust burden. The present study cannot determine whether ES independently accelerated disease progression, because it lacks a comparison group of similarly aged workers with comparable prior stone exposure but no ES exposure. Nevertheless, the hypothesis is supported by the marked increase in RCS exposure potential associated with ES compared with preceding material periods, the temporal proximity of ES exposure to subsequent severe disease in these cases, and the similarity of their functional decline to that reported in international ES-associated silicosis cohorts. The Korean pattern should therefore be interpreted as cumulative mixed exposure with a possible late-career high-silica accelerator, rather than as evidence of an established ES-specific Korean disease phenotype. The Korean experience in international context The engineered-stone silicosis literature has often described a young fabricator in a dedicated countertop workshop with short latency and little prior silica exposure [1, 2, 4, 13, 39, 40]. That description requires qualification in two directions. It is not uniformly accurate internationally: migrant status is prominent in Californian and Australian series but not in Spain, where the largest recognized national series involves a predominantly Spanish-origin workforce—a point of similarity with Korea, since differences between the two countries cannot be attributed to migration status. Nor is mixed mineral exposure unique to Korea: small fabrication workshops in Spain and elsewhere have continued to work marble and granite alongside engineered stone. The difference we observe is therefore one of degree and sequence rather than of kind. Our patients accumulated 28–47 years of natural-stone and solid-surface work before any engineered-stone exposure began, and engineered stone constituted only 6–18% of each career—a longer and earlier-starting prior burden than has typically been reported elsewhere. The cementitious terrazzo legacy adds an alkaline cement-dust layer consistent with the COPD-dominated diagnostic routing of older Korean adjudicated cases, while those who also cut natural granite accumulated meaningful pre-engineered-stone silica burden. Two non-mutually exclusive alternative explanations must also be considered. The first is ascertainment and latency. Engineered stone became widespread in Korea only during the 2000s, and the present cases were drawn from a tertiary pneumoconiosis clinic; this ascertainment pathway would be expected to identify older workers with long cumulative exposure earlier than younger workers whose careers have been dominated by engineered stone. If ascertainment and latency are major contributors, prospective surveillance should increasingly identify younger, engineered-stone-predominant Korean cases over the coming one to two decades, as latency accrues in the engineered-stone-exposed cohort. This is a testable prediction. The second is diagnostic routing. Among older workers with long cementitious and mixed-dust careers, chronic airflow obstruction may dominate the clinical presentation and lead to recognition or compensation under COPD rather than silicosis, even when the two conditions coexist. This interpretation is consistent with the Korean adjudicated-case series, in which COPD and lung cancer predominated over silicosis among recognized stone-trade disease. Clinically, occupational COPD in workers with long mixed stone careers should therefore prompt radiological assessment for coexisting silicosis rather than being assumed to represent an isolated airway disorder. Four gaps in the Korean system, and the measures they require The review and the case series together identify four gaps—in exposure assessment, in case identification and surveillance, in workers’ compensation, and in prevention. Each is stated below with the evidence bearing on it and the measure it requires. Gap 1: exposure assessment. Korea has one engineered-stone-specific workplace dataset, from a single day at a four-worker shop [19, 20]. It nonetheless recorded personal quartz of 0.122 mg/m3 during nominally wet cutting and polishing, about 2.4 times the Korean OEL, and 0.179 mg/m3 with detectable cristobalite during hole-cutting and edge finishing, because dry touch-up and air-blowing were embedded in an otherwise wet workflow. Nominally wet processing therefore does not necessarily provide adequate control, and off-site dimensioning should be treated as a hierarchy-of-controls measure that relocates rather than eliminates the hazard: surveillance and measurement must follow the high-exposure tasks wherever they are performed. The mixed material histories described in Table 1 should likewise be set out explicitly in both clinical review and workers’ compensation reconstruction. What is needed is an engineered-stone-specific program of 8-hour TWA personal and task-based sampling across fabrication and on-site installation, including natural-stone tasks in the same shops, so that mixed-exposure environments are characterized rather than averaged. Gap 2: case identification and surveillance. Engineered stone was documented in only four of 32 adjudicated stone-trade cases and only from 2019, and mixed exposure histories together with the diagnostic routing described above plausibly produce a structural undercount of silicosis among older workers. Active case finding should record material and task histories separately and extend beyond engineered-stone workers to all those with substantial exposure to any silicogenic substrate, with periodic spirometry and low dose CT or HRCT from five years of such exposure. Serological screening is not routine in Korean pneumoconiosis surveillance. Given the recognized association between silica and engineered-stone exposure and systemic autoimmune disease [9], this component of the disease burden is likely under-ascertained in Korea and warrants consideration in symptomatic workers. Gap 3: workers’ compensation. Single-product criteria poorly reflect the multi-era careers documented here, in which engineered stone represented well under a fifth of total stone work. Exposure reconstruction should integrate substrate and task across cementitious terrazzo, natural stone, solid surface and engineered stone, recording which periods involved cutting, grinding, drilling, polishing, dry touch-up and air-blowing, rather than attributing risk to a product label. Linking Korea Customs Service HS 6810.99 data [18, 29], KOSHA inspection records and business registration data would define the population at risk, for which no administrative source currently provides a denominator. Gap 4: prevention and regulation. Korean measurements show that RCS exposure can exceed the current OEL of 0.05 mg/m3 even during nominally wet engineered-stone processing, so the limit should be reviewed toward the ACGIH TLV of 0.025 mg/m3 or a more protective health-based target such as 0.02 mg/m3, and dry cutting, polishing and hole-cutting should be prohibited and paired with on-tool water suppression, local exhaust ventilation and enclosure. We offer this as an interim minimum, not as the Australian model: Australia prohibited the manufacture, supply, processing and installation of engineered stone outright from 1 July 2024, extending the ban to importation from 1 January 2025, and such a prohibition warrants serious consideration here; that decision has been analysed as an evidence-based and precautionary policy grounded in real rather than theoretical working conditions [41]. KCS 41 35 04 [32] should meanwhile be extended to engineered stone through explicit KOSHA guidance. Neither Korean change in fabrication setting was driven by worker respiratory health (Section 3), so dust reduction has been incidental and the residual dry work at installation has never been a control priority; a measure aimed directly at respirable crystalline silica would be the first of its kind in Korea. Strengths and limitations This study has several strengths. To our knowledge it is the first Korean synthesis to integrate the engineered-stone industry’s historical trajectory, domestic RCS exposure benchmarks, the adjudicated-case landscape and a clinical case series within a single multi-era framework. It incorporates the only ES-specific Korean workplace-measurement dataset alongside the most directly relevant domestic natural-stone and installation benchmarks, anchoring the analysis in Korean rather than imported exposure data. The case series applied a pre-specified and deliberately conservative progression definition, ILO-trained image reading, serial spirometry meeting ATS/ERS 2019 criteria, and Korean reference equations. Exposure histories were reconstructed on a calendar basis with task-level detail and documentary corroboration where available, which is rarely achievable in retrospective occupational series. Several limitations temper interpretation. The case series is single-center, small and retrospective, drawn from a tertiary pneumoconiosis clinic, so referral and survivor bias favor severe, compensation-seeking presentations and preclude prevalence estimates. The case definition itself selected for severity, so the proportion with progressive massive fibrosis and the rates of decline describe this selected group only and are not the frequency or natural history of engineered-stone silicosis in Korea. There was no comparison group without engineered-stone exposure, so its specific contribution cannot be isolated, and occupational histories remain subject to recall bias despite calendar anchoring and corroboration. Material composition was established analytically only for the four engineered-stone products at C company, whose measurements come from a single day at one small facility; for earlier periods it was inferred from work descriptions and product era, terrazzo binder type could not be determined retrospectively, and no individual cumulative-silica estimate could be derived, so misclassification is possible. At population level, HS 6810.99 trade data and KSIC codes are heterogeneous proxies, no register defines the at-risk denominator, and no direct Korean measurement exists for historical terrazzo work. The literature component is a narrative rather than a systematic review, so evidence selection was relevance-based and not reproducible. Connective-tissue-disease serology was obtained in only one patient, in whom it was negative, so we cannot comment on silica-associated systemic autoimmune disease in this series. Conclusion In this single-center Korean series of five severe or rapidly progressive cases, all five had PMF and the median annual FVC decline was −0.14 L/year (range −0.47 to −0.11), comparable to the most severe Spanish and Australian cohorts [1, 4, 13]; one never-smoker with 6 years of engineered-stone exposure lost 70 percentage points of predicted FVC in 5 years. Against personal RCS reaching 0.179 mg/m3 quartz with cristobalite under nominally wet processing, and the Group 1 carcinogenicity of crystalline silica [12], four measures follow. Exposure assessment requires an engineered-stone-specific, task-based measurement program. Case identification requires material-stream-specific active case finding, with spirometry and low dose CT or HRCT extended to self-employed fabricators and installation subcontractors, who fall outside employer-mandated examinations. Compensation requires criteria and a national register built around cumulative multi-era exposure. Prevention requires review of the 0.05 mg/m3 OEL toward the ACGIH TLV of 0.025 mg/m3 or a more protective health-based target and, at minimum, prohibition of dry processing of quartz engineered stone, with the comprehensive Australian prohibition warranting consideration. Whether the older multi-era profile proves characteristic of Korea, or is an early-phase feature that later cohorts will not share, is the question prospective surveillance should now answer. Electronic supplementary material Below is the link to the electronic supplementary material. Supplementary material 1 Author contributions J-PM conceived and designed the study, acquired and interpreted the data, and drafted and revised the manuscript. Funding This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of Seoul St. Mary’s Hospital, The Catholic University of Korea (approval no. KC26WISI0342). The Board waived the requirement for individual informed consent because the analysis used de-identified clinical data collected during routine occupational lung-disease care. Consent for publication Not applicable; no individually identifiable patient data or images are presented. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary material 1 Data Availability Statement No datasets were generated or analysed during the current study.