Economic burden and cost drivers of interstitial lung disease: a systematic review

Abstract

Background

Interstitial lung disease (ILD) consists of many subtypes, with idiopathic pulmonary fibrosis (IPF) and subtypes that manifest as progressive pulmonary fibrosis (PPF) exhibit faster progression and poorer prognosis. Recent updates in the definition of PPF and expanded approval of antifibrotic treatments warrant a comprehensive review to understand the economic impact and key cost drivers.

Objective

This systematic review aims to summarize published studies on the economic burden of ILD on patients, the healthcare system, and society, and to identify the prevailing cost drivers.

Method

A literature search was conducted using PubMed, Scopus and Web of Science to identify full-length publications in English on the ILD economic burden, cost of illness, or cost-effectiveness studies, from inception until January 2024. The review protocol was registered with PROSPERO (CRD42024504116) and followed Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guideline. This review compares the direct, indirect, and overall costs across ILD subtypes and countries.

Results

Forty studies fulfilled the inclusion criteria. Annual direct costs of ILD ranged from $1,593 (South Korea) to $177,526 (United States) per patient, while indirect costs ranged from $45 to $19,507. Hospitalisation costs ranged from $1,500 to $18,792 per admission, and end-of-life care costs were $52,614 annually per patient. Inpatient (40%-89%) and medication costs (36%-82%) were identified as the major cost drivers. The predictors of increasing costs include disease severity, presence of comorbidities, and antifibrotic use. Additionally, regional disparities and healthcare system differences influenced the economic burden.

Conclusion

Interstitial lung disease imposes a substantial economic impact. Early diagnosis, effective treatments targeting ILD and comorbidities, appropriate social support, and an optimized health delivery system may reduce the overall cost and improve health outcomes.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10198-025-01836-w.

Key points for decision-makers

• The economic burden of interstitial lung disease (ILD) varies greatly among countries. Therefore, it is essential to assess it within the local setting by taking into account the local healthcare structure and financing, to more accurately represent the real burden on a country’s healthcare system and guide policymakers in judiciously allocating the healthcare budget.

• Inpatient costs of ILD is the leading cost driver reported by most studies, contributing 40%–89% of the total costs.

• ILD presents both clinical and economic challenges, particularly in progressive subtypes such as idiopathic pulmonary fibrosis (IPF) and progressive pulmonary fibrosis (PPF). The rising costs are influenced by the disease’s clinical complexity, including comorbidities, exacerbation, and hospitalisation, as well as the differences in the healthcare system and regional disparities.

• A comprehensive approach is required to address the growing economic burden. This may include early diagnosis that limits the disease to early stages, and definitive treatments targeting ILD and its comorbidities to slow disease progression and prevent exacerbations. Apart from that, effective symptomatic treatment or job modifications may improve patients’ overall performance. Health system reforms and coordinated efforts to refine policy and clinical practice through research are essential to ensure sustainable healthcare and reduce the long-term economic impact of ILD. Collectively, these measures could potentially improve patients’ overall well-being and reduce the economic burden of ILD on our healthcare system.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10198-025-01836-w.

Introduction

Interstitial lung disease (ILD) constitutes a diverse group of disorders characterised by inflammation and fibrosis of lung tissues, resulting in reduced lung function, quality of life, and potential mortality [1]. The prevalence of ILD ranges from 6.3 to 71 per 100,000 people globally, varying between countries and different subtypes. For instance, in France, the prevalence is estimated to be 3.42–5.73/100,000 for systemic sclerosis-associated ILD (SSc-ILD), while the prevalence for progressive pulmonary fibrosis (PPF) is higher at 6.6–19.4/100,000 [2, 3]. The prevalence continues to rise with the advancement in ILD diagnostic capabilities.

There were no disease-specific treatments for ILD until the introduction of antifibrotic therapy, which reduces lung fibrosis and is proven to slow disease progression. However, antifibrotics are only indicated for certain ILD subtypes, namely idiopathic pulmonary fibrosis (IPF), SSc-ILD, and PPF, and are associated with substantial costs [4, 5]. Although their potential economic impact has not been widely studied to date, it is important to explore it due to their growing role in ILD management. Apart from medications, ILD patients, especially those with progressive disease, are prone to acute exacerbation, leading to frequent hospitalisations.

Studies in the United States have estimated that the direct cost per admission ranges from $10,438 to $16,042 [6–8]. Additionally, routine monitoring and palliative care contribute further to the overall management cost. Therefore, ILD poses a significant economic impact on the healthcare system. However, there is wide variation in the economic impact of ILD across countries, with the annual cost per IPF patient ranging from $31,655 in Australia, $108,224 in the USA, €26,997 in Spain, and $1,744 in Korea [9–12]. These discrepancies highlight the impact of the healthcare system, policy, and disease management strategies on the overall costs.

The concept of progressive pulmonary fibrosis has gained attention lately, and the latest 2022 ATS/ERS/JRS/ALAT guideline also outlines a clear definition of PPF [13]. PPF, also referred to as progressive fibrosing ILD (PF-ILD) in previous studies, exhibits similar characteristics, disease progression, and treatment needs to IPF. Thus, more recent studies have expanded their focus to include other ILD subtypes beyond IPF, which had previously been the primary subtype studied. A previous systematic review published in 2022 by Wong et al. provided an overview of 25 studies on the cost of ILD [14]. However, there has been a substantial growth in published evidence since then, particularly with the emergence of the PPF concept and expanded use of antifibrotic therapy, both of which may significantly impact disease-related costs. The current review expands on the previous cost of ILD evidence by including a broader geographical representation, newer therapies, and offers a more detailed synthesis of cost exploration. Furthermore, the available published economic studies differ in their study designs, patient cohorts, cost calculation methods, and presentations, which makes interpreting the overall economic burden challenging. Therefore, this review aims to consolidate the current evidence on the economic burden of ILD and identify the associated predictors of increasing costs.

Methods

This systematic review protocol was registered with the International Prospective Register of Systematic Reviews [PROSPERO (CRD42024504116)]. The review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines [15].

Search strategy

A systematic literature search was conducted using the following databases: PubMed (Medline), Web of Science, and Scopus. The literature search methodology adhered to the PRISMA guidelines and the PICOS (population, intervention, comparators, outcomes, study design) review system.

Search terms related to interstitial lung disease and economic burden were used, including “interstitial lung disease”, “ILD”, “cost”, combined using suitable Boolean operators. The search terms used are listed in the supplementary information (S1). Database search was performed by two independent investigators (S.Y. and M.M.). Then, the identified studies were screened for eligibility and cross-referenced. All relevant articles published from inception up to January 2024 were considered for inclusion.

Inclusion and exclusion criteria

Full publications in English reporting on the economic burden, cost of illness studies, or cost-effectiveness studies (if full costs were reported) for ILD were considered for inclusion. Studies are eligible if the participants were adult patients aged 18 years and above, diagnosed with ILD or any of its subtypes, and reported at least one cost outcome. The cost outcomes could be either direct costs, indirect costs, or both. Direct costs refer to costs associated with ILD patient care, including hospitalisation, emergency visits, outpatient consultation, laboratory monitoring, imaging, diagnostic tests, procedures, medications, pulmonary rehabilitation, end-of-life care, home care or transportation. Indirect costs include costs associated with the loss of work productivity, disability, or burden to the caregiver. Studies that were not original research, such as reviews, editorials, or commentaries, were excluded from this review.

Study selection

The screening process was performed using Mendeley software. Two independent reviewers (S.Y. and M.M.) conducted the screening, with discrepancies resolved by consensus or the involvement of a third reviewer (S.C.). Duplicates were removed using the built-in function of “check for duplicates” in Mendeley software. Then, the titles and abstracts were screened for eligibility. Full-text articles for studies included in the abstract phase were retrieved and reviewed based on inclusion and exclusion criteria. Finally, the two reviewers carried out data extraction and quality assessment. References of included studies were screened to identify additional publications that were not included during the initial search. The process of study selection is presented in Fig. 1.

Fig. 1.

PRISMA flowchart

Data extraction

Two independent reviewers (S.Y. and M.M.) carried out the data extraction process, and disagreements were resolved by consensus or the involvement of a third reviewer (S.C.). The outcome variables were compiled using a Microsoft Excel spreadsheet. The data extracted from the included studies are listed in Table 1, including the first author’s name, publication year, and the study characteristics (study country, population, study period, study size, cost component measured, epidemiological approach, method of resource valuation, perspective, study design, mean age of study cohort and ILD diagnosis). Regarding ILD diagnosis, we note that the term progressive pulmonary fibrosis (PPF) was introduced in the 2022 ATS/ERS/JRS/ALAT guideline. However, earlier studies often used the term progressive fibrosing ILD (PF-ILD). In this review, we retain the terminology used by each original study to describe their population, but interpret the findings in the context of the current PPF definition. The study outcomes extracted are reported in Tables 2, 3, 4 and 5 and include currency and year of cost data, cost component, and the reported annual cost per patient. For cases requiring clarification, the corresponding authors were contacted. If the study characteristics were not clearly stated in a study, decisions were made by consensus between the two reviewers.

Table 1.

Summary of study characteristics

Study	Country	Population	Study period	Study size	Cost component	Epidemiological approach	Method of resource valuation	Perspective	Study design	Age, mean (SD) (years)	Method of ILD diagnosis
Fischer, 2018 [16]	USA	SSc-ILD	2003–2014	219	Direct	Incidence a	Bottom-up a	Third-party payer and patient a	Retrospective	57	ICD-9
Cottin, 2017 [17]	France	IPF	2006–2013	6476	Direct (Hospitalisation)	Incidence a	Top-down a	Third-party payer	Retrospective	77 (69.0–83.0) c	ICD-10
Collard, 2012 [18]	USA	IPF	2001–2008	9286	Direct	Prevalence a	Bottom-up a	Third-party payer a	Retrospective	74 (9.2)	ICD-9
Singer, 2022 [19]	USA	PF-ILD	2015–2019	11,025	Direct	Prevalence a	Bottom-up a	Third-party payer a	Retrospective	72.7 (11.2)	ICD-9 &10
Tarride, 2018 [20]	Canada	IPF	2006–2011	5152	Direct	Incidence	Bottom-up a	Provider*	Retrospective	76.8 (12)	ICD-9 &10
Raimundo, 2016 [21]	USA	IPF	2011	1136	Direct	Prevalence	Bottom-up a	Third-party payer a	Retrospective	NA	ICD-9
Algamdi, 2020 [22]	Canada	CTD-ILD	2015–2017	375	Productivity loss	Prevalence a	Bottom-up a	Employer a	Prospective	60 (12)	CARE-PF cohort
Algamdi, 2019 [23]	Canada	Fibrotic ILD	2015–2017	650	Productivity loss	Prevalence a	Bottom-up a	Employer a	Prospective	68.6 (10.1)	CARE-PF cohort
Knarborg, 2022 [24]	Denmark	SSc-ILD	2002–2015	275	Direct and indirect	Incidence a	Econo-metric	Societal a	Retrospective	NA	ICD-10
Zheng, 2020 [25]	China	IPF	2012–2015	219	Direct (Hospitalisation)	Prevalence a	Bottom-up a	Provider a	Retrospective	65	ATS/JRS
Davidsen, 2021 [26]	Europe (8 countries)	SSc-ILD	2019	805	Direct	Prevalence a	Bottom-up a	Provider a	Expert consensus	NA	NA
Rodríguez-Nieto, 2023 [12]	Spain	IPF	2017–2018	180	Direct and indirect	Prevalence a	Bottom-up a	Societal	Prospective	70.8 (7.6)	ATS/ERS/JRS/ALAT
Zhou, 2019 [27]	USA	SSc-ILD	2004–2016	479	Direct and indirect	Prevalence a	Bottom-up a	Societal a	Retrospective	60.2 (12.5)	ICD-9
Frank, 2019 [28]	Germany	IPF	2009–2014	14,453	Direct	Prevalence a	Bottom-up a	Third-party payer a	Retrospective	71.6 (11.4)	ICD-10
Frank, 2019 [28]	Germany	Sarcoidosis	2009–2014	9106	Direct	Prevalence a	Bottom-up a	Third-party payer a	Retrospective	55.1 (15.5)	ICD-10
Lee, 2022 [29]	USA	Sjögren syndrome-ILD	2006–2015	815	Direct	Prevalence a	Bottom-up a	Third-party payer	Retrospective	58.47 (11.43)	ICD-9
Kreuter, 2022 [30]	Germany	IPF	2014–2019	1737	Direct	Incidence	Bottom-up a	Third-party payer a	Retrospective	72.1 (10.4)	ICD-10
Cottin, 2021 [2]	France	SSc-ILD	2010–2017	3333	Direct	Incidence a	Bottom-up a	Third-party payer	Retrospective	60.6 (14.4)	ICD-10
Nasser, 2021 [3]	France	PF-ILD	2010–2017	14,413	Direct	Incidence a	Mixed a	Third-party payer	Retrospective	68.4 (15.0)	ICD-10
Lassenius, 2020 [31]	Finland	IPF	2005–2017	266	Direct	Incidence a	Bottom-up a	Provider a	Retrospective	74.3 (8.45)	ICD-10
Kim, 2017 [11]	Korea	IPF	2009–2013	18,006	Direct	Prevalence a	Bottom-up a	Third-party payer a	Retrospective	67.3	ICD-10
Olson, 2020 [32]	USA	PF-ILD	2014–2016	373	Direct	Prevalence a	Bottom-up a	Third-party payer a	Retrospective	NA	ICD-9 &10
Olson, 2022 [33]	USA	PF-ILD	2011–2015	5815	Direct	Incidence	Bottom-up a	Third-party payer a	Retrospective	68.7 (13.7)	ICD-9
Corral, 2020 [34]	USA	IPF	2014–2015	3546	Direct	Prevalence a	Bottom-up a	Third-party payer a	Retrospective	75.9 (5.7)	ICD-10
Mooney, 2017 [7]	USA	IPF	2009–2011	22,350 b	Direct (Hospitalisation)	Prevalence a	Bottom-up a	Healthcare system a	Retrospective	70.0 (0.32)	ICD-9
Fan, 2020 [35]	USA	IPF	2014–2016	300	Direct (Hospitalisation)	Prevalence a	Bottom-up a	Healthcare system a	Retrospective	69 (8.0)	IPF-PRO Registry
Jeganathan, 2024 [6]	USA	ILD	2008–2018	345,063 b	Direct (Hospitalisation)	Prevalence a	Top-down a	Healthcare system a	Retrospective	NA	DRG code
Yu, 2016 [8]	USA	IPF	2006–2011	1735	Direct (hospitalisation)	Prevalence a	Bottom-up a	Third-party payer	Retrospective	71.5 (12.7)	ICD-9 code
Raimundo, 2019 [36]	USA	RA-ILD	2004–2013	750	Direct	Incidence	Bottom-up a	Third-party payer a	Retrospective	NA	ICD-9 code
Wuyts, 2020 [37]	Europe (8 countries)	PF-ILD	2019	1674	Direct	Prevalence a	Bottom-up a	Provider a	Expert consensus	NA	INBUILD criteria
Morrisroe, 2020 [38]	Australia	SSc-ILD	2008–2015	335	Direct	Incidence a	Bottom-up a	Provider a	Retrospective	55.1 (12.8)	ASCS cohort
Cox, 2023 [10]	Australia	IPF	2018–2019	162	Direct	Prevalence	Bottom-up	Partial societal	Retrospective	73.8 (7.6)	AIPFR cohort
Corral, 2020 [9]	USA	IPF	2014–2018	614	Direct	Prevalence a	Bottom-up a	Third-party payer a	Retrospective	70.9 (9.5)	ICD-9 & 10 code
Morell, 2016 [39]	Spain	IPF	2012–2013	NA	Direct	Prevalence	Bottom-up a	Provider	Expert consensus	NA	ATS/ERS/JRS/ALAT guideline
Collard, 2015 [40]	USA	IPF	2000–2011	7855	Direct	Incidence	Bottom-up a	Provider a	Retrospective	78.5 (6.9)	ICD-9
Kalluri, 2020 [41]	Canada	IPF	2012–2018	78	Direct (end-of-life care)	Prevalence a	Bottom-up a	Provider	Retrospective	73 (67–83) c	ICD-10
Hilberg, 2018 [42]	Denmark	IPF	2003–2009	120	Direct	Prevalence a	Bottom-up a	Third-party payer a	Retrospective	66.5	ICD-10
Gayle, 2020 [43]	UK	SSc-ILD	2005–2016	127	Direct	Prevalence a	Bottom-up a	Provider a	Retrospective	71 (13)	ICD-10
Pedraza-Serrano, 2017 [44]	Spain	IPF	2004–2013	22,214	Direct (hospitalisation)	Incidence a	Top-down a	Provider a	Retrospective	73.11 (12.28)	ICD-9
Cottin, 2023 [45]	France	IPF	2015–2017	1313	Direct	Prevalence a	Bottom-up a	Provider	Retrospective	72.8–73.2	ICD-10
Wilson, 2014 [46]	UK	IPF	2011–2012	72	Direct and indirect	Prevalence a	Bottom-up a	Societal	Prospective	NA	NA

AIPFR Australian Idiopathic Pulmonary Fibrosis Registry, ASCS Australian Scleroderma Cohort Study, ATS/ERS/JRS/ALAT American Thoracic Society/European Respiratory Society/Japanese Respiratory Society/Asociación Latinoamericana de Tórax, CARE-PF Canadian Registry for Pulmonary Fibrosis, CTD connective tissue disease, ICD International statistical classification of diseases and related health problems, ILD interstitial lung disease, INBUILD Efficacy and Safety of Nintedanib in Patients With Progressive Fibrosing Interstitial Lung Disease (PF-ILD) trial, IPF Idiopathic pulmonary fibrosis, NA not available, PF-ILD progressive fibrosing interstitial lung disease, RA rheumatoid arthritis, SSc systemic sclerosis, UK United Kingdom, USA United States of America

a Not clearly stated in the study, consensus achieved by discussion between reviewers

^b Number of hospitalisation

^c Median (Interquartile range)

Table 2.

Summary of studies reported both total direct and indirect costs from a societal perspective and the cost estimates

Study	Country	Study population	Year of cost data	Reported annual cost per patient (local currency of study year)	Mean annual cost per patient (2024 USD)	GDP per capita (2024)	Annual cost (% of GDP per capita) e
Knarbog, 2022 [24]	Denmark	SSc-ILD	2020 EURO	€29,725 d	33,004.42	69,273.05	47.64
Rodríguez-Nieto, 2023 [12]	Spain	IPF	2023 EURO	€26,997	28,915.15	35,788.73	80.79
Zhou, 2019 [27]	USA	SSc-ILD	2018 USD	$51,031	62,457.36	86,601.28	72.12
Wilson, 2014 [46]	UK	IPF	2012 GBP	£13,271	22,752.20	52,423.29	43.40

GBP Great British Pound,GDP Gross domestic product, ILD interstitial lung disease, IPF Idiopathic pulmonary fibrosis, SSc systemic sclerosis, UK United Kingdom, USA United States of America, USD United States Dollar

^d Measured as total excess cost for the SSc-ILD cohort as compared to non-ILD SSc cohort

^e Annual cost (% of GDP per capita) was calculated as the percentage of the cost for a particular country against its respective GDP per capita in the year 2024

Table 3.

Summary of studies reported direct costs of ILD and the cost estimates

Study	Country	Study population	Year of cost data	Cost component	Reported annual cost per patient (local currency of study year)	Mean annual cost per patient (2024 USD)	GDP per capita (2024)	Annual cost (% of GDP per capita)
Fischer, 2018 [16]	USA	SSc-ILD	2014 USD	Total direct	$38,221.4	49,627.45	86,601.28	57.31
Collard, 2012 [18]	USA	IPF	2008 USD	Total direct	$26,378	37,521.66	86,601.28	43.33
Singer, 2022 [19]	USA	PF-ILD	2019 USD	Total direct	$52,584	63,313.30	86,601.28	73.11
Tarride, 2018 [20]	Canada	IPF	2016 CAD	Total direct	Can$18,639	16,822.20	53,834.48	31.25
Raimundo, 2016 [21]	USA	IPF	2011 USD	Total direct	$59,379	81,262.27	86,601.28	93.83
Knarborg, 2022 [24]	Denmark	SSc-ILD	2020 EURO	Total direct	€20,275	22,511.79	69,273.05	32.50
Davidsen, 2020 [26]	Europe (8 countries)	SSc-ILD	2019 EURO	Total direct	€6191.34- €25,354.25 i (Average: €13,494.85)	7,519.79–29,307.72 i (Average:16,226.17)	24,341.51–90,433.67	44.19-105.96 j
Rodríguez-Nieto, 2023 [12]	Spain	IPF	2023 EURO	Total direct	€26,955	28,870.16	35,788.73	80.67
Zhou, 2019 [27]	USA	SSc-ILD	2018 USD	Total direct	$37,505	45,902.75	86,601.28	53.00
Frank, 2019 [28]	Germany	IPF	2014 EUROg	Total direct	€12,111.00	16,623.39	55,521.35	29.94
		Sarcoidosis		Total direct	€8,793	12,069.15	55,521.35	21.74
Lee, 2022 [29]	USA	Sjögren syndrome-ILD	2020 USD	Total direct	$46,384	55,115.70	86,601.28	63.64
Kreuter, 2022 [30]	Germany	IPF	EURO 2019+	Total direct	$15,721.00	19,875.39	55,521.35	35.80
Cottin, 2021 [2]	France	SSc-ILD	EURO 2017+	Total direct	€25,752.80	31,679.53	48,011.83	65.98
Nasser, 2021 [3]	France	PF-ILD	EURO 2020	Total direct	€18,362 h	21,395.10	48,011.83	44.56
Lassenius, 2019 [31]	Finland	IPF	2017 EURO	Total direct	€5,827.00	7,364.43	54,773.99	13.45
Kim, 2017 [11]	Korea	IPF	2016 USD f	Total direct	$1,744	1,592.60	36,131.85	4.41
Olson, 2020 [32]	USA	PF-ILD	2019 USD f	Total direct	$77,666.00	93,513.06	86,601.28	107.98
Olson, 2022 [33]	USA	PF-ILD	2018 USD	Total direct	$54,215	66,354.29	86,601.28	76.62
Corral, 2020 [34]	USA	IPF	2015 USD	Total direct	$137,994	177,526.40	86,601.28	204.99
Raimundo, 2019 [36]	USA	RA-ILD	2014 USD	Total direct	$34,681	45,030.52	86,601.28	52.00
Wuyts, 2020 [37]	Europe (8 countries)	PF-ILD	2019 EURO	Total direct	€15,648.71- €69,021.27 i (Average: €34,530.08)	19,006.38–79,783.71 i (Average:41,488.89)	24,431.51–90,433.67	20.43-51.23j
Morrisroe, 2019 [38]	Australia	SSc-ILD	AUD 2019 f	Total direct	AUD$9,322	7,189.48	65,965.62	10.90
Cox, 2023 [10]	Australia	IPF	AUD 2021	Total direct	AUD$31,655	22,810.32	65,965.62	34.58
Corral, 2020 [9]	USA	IPF	2017 USD	Total direct	$108,224.00	135,491.04	86,601.28	156.45
Morell, 2016 [39]	Spain	IPF	EURO 2013	Total direct	€26,435.00	34,188.27	35,788.73	95.53
Collard, 2015 [40]	USA	IPF	2012 USD	Total direct	$20,887	28,062.20	86,601.28	32.40
Hilberg, 2018 [42]	Denmark	IPF	2010 EURO	Total direct	€17,495	21,765.92	69,273.05	31.42
Gayle, 2020 [43]	UK	SSc-ILD	2016 GBP	Total direct	£6,375 h	10,293.02	52,423.29	19.63
Cottin, 2023 [45]	France	IPF	2022 EUROf	Total direct	€32,212-€34,523	35,922.22–38,499.40	48,011.83	74.82–80.19
Wilson, 2013 [46]	UK	IPF	2012 GBP	Total direct	£1,893	3,245.42	52,423.29	6.19
Specific Direct Cost
Study	Country	Study Population	Year of cost data	Cost component	Reported cost per admission (local currency of study year)	Mean cost per admission (2024 USD)	GDP per capita (2024)	Annual cost (% of GDP per capita)
Cottin, 2017 [17]	France	IPF	2016 EURO	Hospitalisation	€4,510 h	5,568.97 h	48,011.83	11.60
Zheng, 2020 [25]	China	IPF	2015 CNY	Hospitalisation	CNY9378.30 h	1,499.66 h	12,968.57	11.56
Mooney, 2017 [7]	USA	IPF	2011 USD	Hospitalisation	$16,042	21,954.05	86,601.28	25.35
Fan, 2020 [35]	USA	IPF	2016 USD	Hospitalisation	$13,975 l	17,809.27 l	86,601.28	20.56
Jeganathan, 2024 [6]	USA	ILD	2023 USDk	Hospitalisation	$10,438	10,687.35	86,601.28	12.34
Yu,2016 [8]	USA	IPF	2012 USD	Hospitalisation	$13,987	18,791.88	86,601.28	21.70
Pedraza-Serrano, 2017 [44]	Spain	IPF	2016 EUROk	Hospitalisation	€5,249.35	6,741.65	35,788.73	18.87
Kalluri, 2020 [41]	Canada	IPF	2017 CAD	End-of-life care	Can$59,795 l	52,614.19 l	53,834.48	97.73

AUD Australian dollar, CAD Canadian dollar, CNY Chinese Yuan renminbi, GBP Great British Pound, GDP Gross domestic product, ILD interstitial lung disease, IPF Idiopathic pulmonary fibrosis, PF-ILD progressive fibrosing interstitial lung disease, RA rheumatoid arthritis, SSc systemic sclerosis, UK United Kingdom, USA United States of America, USD United States Dollar

^f Cost year was not reported, assumed a year before publication

^g Cost was reported as full amounts documented in the year of study (2009–2014), thus inflated based on the last year of study

^h Cost reported in median

ⁱ The study involved multiple countries, cost reported as a range and average

^j The study involved multiple countries, thus the annual cost (% of GDP per capita) was calculated as the percentage of the cost for each country against its respective GDP per capita in the year 2024 and reported as a range

^k Cost year was not reported, assumed a year before publication

^l Cost reported as cost per patient per year

Table 4.

Studies reported the total indirect cost of ILD and its cost estimates

Study	Country	Study population	Year of cost data	Cost component	Reported annual cost per patient (local currency of study year)	Mean annual cost per patient (2024 USD)	GDP per capita (2024)	Annual cost (% of GDP per capita)
Algamdi, 2020 [22]	Canada	CTD-ILD	2017 CAD	Productivity loss	Can$13,593	11,960.61	53,834.48	22.22
Algamdi, 2019 [23]	Canada	Fibrotic-ILD	2017 CAD	Productivity loss	Can$11,610	10,215.75	53,834.48	18.98
Knarborg, 2022 [24]	Denmark	SSc-ILD	2020 EURO	Difference in earnings	€13,636	15,140.36	69,273.05	21.86
Rodríguez-Nieto, 2023 [12]	Spain	IPF	2023 EURO	Productivity loss	€42	44.98	35,788.73	0.13
Zhou, 2019 [27]	USA	SSc-ILD	2018 USD	Productivity loss	$13,526	16,554.61	86,601.28	19.12
Wilson, 2014 [46]	UK	IPF	2012 GBP	Productivity loss	£11,378	19,506.79	52,423.29	37.21

CAD Canadian dollar, CTD connective tissue disease, GBP Great British Pound, GDP Gross domestic product, ILD interstitial lung disease, IPF Idiopathic pulmonary fibrosis, SSc systemic sclerosis, UK United Kingdom, USA United States of America, USD United States Dollar

Table 5.

Summary of studies that investigate the predictors of increasing costs

Study	Cost components	Predictors	Estimates (95% CI) or Odds ratio (95% CI)	Standard error	P value
Zheng, 2020 [25]	Direct cost (hospitalisation) in IPF	• Length of hospital stay • Intensive care • Pulmonary infection • Respiratory failure	B = 0.057 B = 1.260 B = 0.186 B = 0.213	-	< 0.001 < 0.001 0.002 0.003
Rodríguez-Nieto, 2023 [12]	Total direct and indirect costs in IPF	• Antifibrotic treatment • Patients with FVC% predicted 50–80% at baseline vs. FVC predicted < 50% • Patients with FVC% predicted > 80% at baseline vs. FVC predicted < 50%	€25,483.09 (CI 17,824.17; 33,142.51) −€14,224.85 (− 22,235.47; −6213.97) −€13,580.18 (− 23,403.98; −3,755.36)	3878.26 4056.28 4974.58	< 0.0001 0.0006 0.007
Frank, 2019 [28]	Direct cost of ILD (IPF and sarcoidosis)	• Comorbidity (lung cancer) • Comorbidity (pulmonary hypertension) • Comorbidity (diabetes mellitus)	Surcharge factor 1.979 (1.966–1.992) Surcharge factor 1.71 (1.697–1.723) Surcharge factor 1.318 (1.305–1.331)	-	< 0.05 < 0.05 < 0.05
Mooney, 2017 [7]	Direct cost (hospitalisation) in IPF	• Age • Primary diagnosis of IPF • Bacterial pneumonia • Non-invasive pneumonia • Mechanical ventilation	-$148 (−215 - −81) $2,099 (1,042–3,155) $2,978 (1595–4362) $5,500 (2373–8628) $36,911 (32253–41568)	-	< 0.001 < 0.001 < 0.001 < 0.001 < 0.001
Morrisroe, 2020 [38]	Direct cost in SSc-ILD	• Age at ILD onset • ILD severity - Mild - Moderate - Severe • Presence of pulmonary arterial hypertension	OR 1.03 (1.0, 1.1) baseline OR 1.96 (1.0, 3.7) OR 5.09 (2.1, 12.4) OR 2.58 (1.2, 5.4)	-	0.01 0.04 < 0.001 0.01
Cox, 2023 [10]	Direct cost in IPF	• Comorbidities • Antifibrotics • FVC% (10-unit increase) • GAP total (1-unit increase) • CPI index (10-unit increase)	OR 1.38 (1.00-1.87) OR 2.41 (1.85–3.13) OR 0.93 (0.88–0.99) OR 1.15 (1.04–1.28) OR 1.22 (1.09–1.35)	-	< 0.05 < 0.05 < 0.05 < 0.05 < 0.05
Algamdi, 2019 [23]	Productivity loss in fibrotic ILD	• Dyspnea score (UCSD SOBQ) • Cough severity (VAS)	OR 0.96 (0.93–0.99) OR 0.97 (0.94–0.99)	-	0.01 0.02

CI 95% Confidence interval, CPI composite physiological index, FVC forced vital capacity, GAP gender, age, physiology assessment, ILD interstitial lung disease, IPF Idiopathic pulmonary fibrosis, SSc systemic sclerosis, OR Odds ratio, UCSD SOBQ the University of California San Diego Shortness of Breath Questionnaire, which is a 24-item patient-reported questionnaire that assesses dyspnea associated with activities of daily living (total score= 0-120), VAS 100-mm visual analog scale (VAS) used to measure cough severity

Standardization of cost

The costs were presented as the mean annual cost per patient. For studies that reported costs over more than one year, the average annual cost was calculated. In studies reporting monthly or 6-monthly costs, the costs were converted into annual costs by multiplying the relevant costs into 12-month costs.

To ease comparability across the included studies, all costs were reported in 2024 United States dollars (USD) using the following conversion method. Costs reported in the studies were first inflated using the Gross Domestic Product (GDP) implicit price deflator to reflect local currency values in 2024. Subsequently, the values were converted into 2024 USD using 2024 exchange rates. Data on the GDP implicit price deflator and exchange rate were obtained from the International Monetary Fund database [47, 48].

Statistical analysis

To facilitate comparison across studies, subgroup analyses were conducted on total annual direct costs based on geographical region, ILD subtypes, and whether antifibrotic therapy costs were included. Due to substantial heterogeneity and limited data availability, a meta-analysis was not feasible. Instead, a generalised linear model with a gamma distribution and log link was used to identify the potential predictors of direct costs associated with ILD. All statistical analyses were conducted using SPSS version 29, with a significance level of p-value less than 0.05.

Quality assessment

Quality assessment was conducted by the two independent reviewers (S.Y. and M.M.), with disagreements resolved by consensus or the involvement of a third reviewer (S.C.). We adapted three complementary tools: (i) the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) checklist to assess reporting quality [49], (ii) the consensus-based checklist for the critical appraisal of cost-of-illness (COI) studies by Schnitzler et al. [50] to evaluate methodological rigor, and (iii) the ECOBIAS checklist to assess potential source of bias [51, 52]. The rationale for choosing these tools was to ensure a comprehensive appraisal from different aspects, allowing us to evaluate how well studies were reported, the robustness of methods and the potential risk of bias that could influence the cost estimates.

CHEERS 2022 is a 28-item checklist that focuses on the reporting quality of health economic studies to enhance their interpretability and reproducibility. Meanwhile, the Schnitzler et al. 17-item checklist emphasizes the methodological rigor of COI studies, including key aspects such as epidemiological and costing approaches, data collection processes, identification, measurement, and valuation of cost components.

ECOBIAS is a 22-item checklist that focuses on methodological biases rather than reporting standards. Since the included studies were non-model-based, we applied only Part A, which evaluates general bias in economic evaluation and covers 11 domains.

For the quality assessment in this review, items 16 and 17 from the CHEERS 2022 checklist were excluded because they are intended for economic modelling studies, which were not applicable. Furthermore, items 7, 11, 12, and 13 were deemed not applicable to the cost-of-illness studies, which only examine costs and cost offsets. The overall reporting score was calculated as the percentage of applicable items fulfilled. On the other hand, the results of the appraisal using the Schnitzler et al. checklist were summarized in a table indicating whether each study fulfilled each item. Item-level findings were reported to address the common strengths and methodological gaps across the included studies. ECOBIAS results were summarised separately in a table to provide an overview of biases in each study.

Results

Search results

The databases search yielded a total of 4967 articles, of which 40 articles were finally included in this review. The PRISMA flowchart is depicted in Fig. 1.

Characteristics of the included studies

Study characteristics are summarised in Table 1. Among the 40 included studies, the majority focused on the ILD subtype of IPF (23 studies), followed by SSc-ILD (7 studies) and PPF (5 studies). Twenty seven studies reported on total direct costs, while eight studies focused on the direct costs of a specific component, of which seven studied the cost of hospitalisation [6–8, 17, 25, 35, 44], and one studied the cost of end-of-life care [41]. There were four studies that reported both direct and indirect costs [12, 24, 27, 46], while two studies reported only indirect costs [22, 23].

North American and European populations formed the basis of the vast majority of studies (16 from the United States, 4 from Canada, 14 European countries, and 2 from the United Kingdom), whereas only two studies were conducted in Australia and another two studies among Asian populations.

Majority of the studies did not specify the approaches adopted in their cost-of-illness studies. Therefore, consensus was achieved between reviewers to determine the methodology employed. Retrospective studies using administrative or insurance claims databases were the most common method used among included studies. Only one study, by Rodríguez-Nieto et al., used a prospective multicentre design to evaluate the total costs of IPF patients in Spain [12], while Wilson et al. conducted an economic evaluation using a randomized controlled trial [46]. Most studies were conducted from the perspective of third-party payers (n = 17), followed by healthcare providers (n = 14), and utilized a prevalence-based (n = 26), bottom-up (n = 35) costing approach. The primary data sources and cost components measured in the cost analysis of the included studies are presented in Table S2.

Table S3 compares the characteristics of studies included in the previous review by Wong et al. [14] and those in the current review. The current review included an expanded number of countries, more ILD subtypes beyond IPF, and a greater number of studies that included antifibrotic therapy in their cost analysis. Notably, the annual cost per patient remained high and showed an increasing trend over time. The costs associated with PPF have been studied more frequently in recent years, highlighting the growing recognition of their economic impact. Further results are discussed below.

Total cost

Only four studies evaluated both the direct and indirect costs from a societal perspective and reported the total annual cost per patient of $62,457 in the United States [27], $28,915 in Spain [12], and $22,752 in the United Kingdom [46]. (Table 2) Across these studies, direct medical costs accounted for approximately 14.3–99.8% of the total annual costs, whereas indirect costs contributed 0.2–85.7%. Another study by Knarborg reported a total excess cost of $33,004 in SSc-ILD patients compared to the non-SSc cohort [24]. The total annual cost reported in the United States was significantly higher than in the other countries. Interestingly, the proportion of direct and indirect costs varied among studies. Rodríguez-Nieto et al. reported an annual indirect cost of only $45 per patient [12]. In contrast, Wilson’s study found that a significant proportion of the total cost (86%), which is $19,506, was contributed by indirect costs [46]. The total annual cost reported in Zhou’s study in the US was contributed mainly by direct costs (74%) [27].

Direct costs

The summary of direct cost estimates is presented in Table 3. There was wide variability in the reported total annual direct costs among the included studies, ranging from $1,593 in Korea [11] to $177,526 in the United States [7] per patient.

Direct medical costs included in most of the studies were inpatient care, outpatient services, emergency visits, medications, diagnostic and laboratory tests, and procedures. Direct non-medical costs were reported in six studies, and items categorised under this cost included home care, transportation, informal care, and community and social services [10, 12, 20, 24, 45, 46].

Figure 2(A), (B), and (C) depict the cost trends over time, stratified by geographical region and ILD subtypes. Studies from North America (the United States and Canada) consistently reported significantly higher mean annual costs than other regions. The mean annual cost per patient among studies in this region was $68,888 (±$44,772.40), compared to $21,658 (±$11,237.60) in Europe and Central Asia, and $10,531 (±$10,996.40) in the East Asia and Pacific region (Table S4, p < 0.001) We observed different trends in annual direct costs over time, varied by region. In East Asia and Pacific, the annual cost per patient steadily increased from $7,189 in 2015 to $22,810 in 2019, showing a growing economic burden. In Europe, costs remained relatively stable, ranging from $21,766 to $41,488, with a slight upward trend in recent years. In the US, costs varied widely across studies, with the most recent study reporting lower costs than earlier years. These trends showed regional differences in the economic burden of ILD, with North America consistently reporting the highest cost compared to other regions.

Fig. 2.

Distribution of total annual direct costs stratified by: (A) Region and study year, (B) ILD subtypes and study year, (C) Region and ILD subtype, (D) Antifibrotic therapy and study year, (E) Cost per hospitalisation event. Abbreviation: AU-ILD: autoimmune-associated ILD; ILD: interstitial lung disease; IPF: Idiopathic pulmonary fibrosis; PPF: progressive pulmonary fibrosis; RA: rheumatoid arthritis; SSc: systemic sclerosis; SjS: Sjögren’s syndrome; USD: United States Dollar Poly. indicates a polynomial trendline used to visualise the trend over time

When stratified the annual direct costs by ILD subtypes, studies focusing on the IPF and PPF subtypes reported higher annual direct costs compared to other ILD subtypes, ranging from $1,593 to $177,526 for IPF and $21,395 to $93,513 for PPF, respectively (Fig. 2B and C). The average cost among PPF studies was the highest, with a mean of $57,213 (±$27,242.60) and a median of $63,313 (IQR $48,491.70), followed by IPF with a mean of $41,890 (±$48,989.30) (Table S4). In contrast, autoimmune-associated ILD, such as Rheumatoid arthritis associated ILD (RA-ILD), SSc-ILD, and Sjögren syndrome-ILD, had a mean cost of $31,508 (±$18,137.20), while Sarcoidosis had comparatively lower costs at $12,069. A similar trend was observed when costs were stratified by region and subtype, with North America incurring higher costs, regardless of the subtype. One study by Tarride et al. investigated the cost patterns among IPF patients in Canada, both pre- and post-diagnosis, over a 5-year period [20]. The average annual cost per patient in the year preceding IPF diagnosis was $11,664, then increased to $17,398 and $12,978 per patient 1 year and 2 years post-diagnosis, respectively, indicating a higher economic burden of IPF.

PPF is a relatively new concept but has recently gained wide attention due to its progressive nature and early mortality comparable to IPF. Five studies investigated patients with PPF [3, 19, 32, 33, 37], with three studies reporting hospitalisation costs as the leading cost, ranging from 39–88% [3, 19, 32]. When examining the treatment received in these studies, corticosteroids were identified as the primary therapy [19, 33]. However, none of these studies assessed the costs of antifibrotic therapy, indicating a gap in knowledge regarding antifibrotic use in this patient group.

Based on the studies that provided cost breakdowns, the majority (13 out of 25) identified inpatient costs as the leading cost driver of the economic burden of ILD, contributing 40–89% of the total costs, with an average of 47.2%. This was followed by medication costs (reported in 6 out of 25 studies), which accounted for 36–82% of total costs, averaging 25%. The cost breakdown is presented in Fig. 3. Direct non-medical costs were specified in only six studies, ranging from $340 to $3,542 [10, 12, 20, 24, 45, 46]. When examining cost breakdowns by ILD subtype (Fig. 4A), inpatient care remained the primary cost driver across all subtypes. In studies focusing on PPF, outpatient costs became a more significant contributor to the overall cost compared to the other subtypes. The proportion of medication costs was higher in IPF studies, but with a wide range, from 7–83% of total direct costs. This variation is likely attributable to whether antifibrotic therapies were included, as studies that included antifibrotic therapy reported a higher proportion of medication costs (29–82%) than those that did not (7–25%). The impact of antifibrotic therapy is further explored and discussed below. Compared to the previous review, the current review observed a wider range in the proportion of medication costs, and a relatively lower proportion of inpatient costs (Fig. 4B). This highlights the impact of the availability of newer treatment options on overall cost.

Fig. 3.

Distribution of healthcare costs by study and cost category

Fig. 4.

Proportion of annual-per-patient direct cost by cost components (inpatient, emergency, outpatient, medication, and other services): (A) by ILD Subtype, and (B) by Study Type (new vs. previous review). Abbreviation: AU-ILD: autoimmune-associated ILD; ILD: interstitial lung disease; IPF: Idiopathic pulmonary fibrosis; PPF: progressive pulmonary fibrosis; Autoimmune-associated ILD includes SSc-ILD, RA-ILD and Sjögren syndrome ILD

Recent studies incorporating antifibrotic therapy reported higher costs. Figure 2D compares the annual cost reported in 11 studies that included antifibrotic therapies to those that did not. Although not statistically significant, studies that accounted for antifibrotic use reported higher mean annual costs of $50,718 (±$53960.1), versus $34,705 (±$26777.80) in others (Table S4), highlighting antifibrotic therapy as an important cost driver. The difference in costs is particularly notable in patients with IPF. For instance, Corral et al. reported a monthly cost of $14,794 (or an annual cost of $ 177,526) in IPF patients treated with antifibrotics, with treatment costs accounting for the largest share (79%) [34]. In a subsequent study, he reported that 67% of the annual cost of $135,491 per patient was attributed to antifibrotic drugs [9]. Similarly, studies in Australia and Spain reported 61% and 83% of total costs were spent on antifibrotic medications [10, 12]. These findings emphasize the high financial burden of disease-specific treatments across regions. Nonetheless, most of these studies were on IPF subtype, indicating a gap in evidence regarding antifibrotic costs in PPF and SSc-ILD.

Hospitalisation remains a major cost contributor in ILD. Seven studies investigated the direct cost of hospitalisation for IPF, and another study examined the cost of end-of-life care (Table 3). Hospitalisation costs ranged from $1,500 to $21,954 per admission; higher costs were reported in North America compared to other regions [6–8, 17, 25, 35, 44]. Interestingly, the North America region observed a slight decline in the cost per hospitalisation event from 2011 to 2018 (Fig. 2E). Most of the studies excluded the cost of lung transplants from hospitalisation costs. Only one study by Cottin reported a median cost per stay of $95,337 per lung transplantation admission, indicating the high cost of treatment at the end stage of the disease [17]. This is consistent with the findings from Kalluri et al. that highlighted the high burden of end-of-life care, accounting for 98% of Canada’s GDP per capita ($52,614), driven mainly by increasing inpatient costs as the patient’s condition deteriorated [41]. Yu et al. assessed the economic burden of hospitalisation and exacerbation and reported the cost of $19,791 per IPF exacerbation event [8]

Indirect costs

The components and methods involved in the calculation of indirect costs varied among studies. However, the total indirect costs reported did not vary widely among subtypes and countries, ranging from $10,216 to $19,507, except for one study that reported only $45 [12]. Table 4 summarises indirect costs reported by included studies; the cost breakdown of indirect costs is presented in supplementary information (Table S5). Most studies (n = 4) employed the human capital method, calculating the cost of productivity loss due to absenteeism, which refers to the days or time off from work due to health-related issues. The reported cost of absenteeism ranged from $0 to $5,717 [12, 22, 23, 27].

Two studies by Algamdi included the cost due to both absenteeism and presenteeism (impaired functionality at work despite physical presence) using Work Productivity and Activity Impairment (WPAI) questionnaires. The cost of presenteeism was higher than absenteeism, $6,930 vs. $5,030 for connective tissue disease-associated ILD (CTD-ILD) patients and $7,244 vs. $2,971 among fibrotic-ILD patients [22, 23].

Rogriguez-Nieto et al. used the opportunity cost method to calculate informal care costs but reported only $45 per patient annually, and zero missed workdays over 12 months [12]. The disability cost was reported by Zhou et al. as $11,008 per patient by using data from disability claims, which was roughly double the absenteeism cost ($5,547) [27]. Wilson et al. did not itemize the cost but only reported the total cost of patient and carer time off work and informal care as $19,507 [46].

Lastly, Knarborg et al. used an econometric approach to estimate the indirect costs based on the difference in earnings and social transfer payments between the SSC-ILD patients and controls, reported a difference of $15,140 [24].

Predictors of cost

A few studies analysed the determinants of the economic burden of ILD by using regression analysis (Table 5). The common predictors for increasing total cost reported in the studies were disease severity, the presence of comorbidities, and the usage of antifibrotic therapy. Morrisroe et al. classified SSc-ILD severity based on the extent of abnormal lung involvement shown on HRCT, and found that patients with severe SSC-ILD (> 30% involvement) are five times more likely to have higher median annual total healthcare costs compared to patients with mild SSc-ILD (< 10% involvement) (OR: 5.09, p < 0.001) [38].

Rodríguez-Nieto et al. reported that patients with impaired lung function and the usage of antifibrotic therapy were the primary determinants of cost in IPF patients. For patients with more impaired lung function categorised by forced vital capacity (FVC) predicted of less than 50% at baseline, the annual costs were €14,224.85 and €13,580.18 higher compared to patients with FVC predicted 50–80%, and those with more than 80%, respectively. The higher costs were mainly driven by hospitalisation (€9,293 vs. €992 and €814), including intensive care unit (ICU) stay (€3,363 vs. €216 and 0) and non-pharmacological treatment (€2,668 vs. €270 and €26) reported in this group of patients (p < 0.001). The same study by Rodríguez-Nieto et al. also identified antifibrotics usage as a cost predictor, associated with a €25,483.09 increase in total cost [12].

Common comorbidities of ILD, such as pulmonary hypertension, lung cancer, and diabetes, are among the cost predictors reported in a few studies [10, 28, 38]. Frank reported that healthcare costs for individuals with comorbidities of lung cancer and pulmonary hypertension are approximately 2 times and 1.7 times higher compared to individuals without comorbidities [28]. The presence of pulmonary arterial hypertension was also identified as a cost driver in Morrisroe’s study, increasing the odds of higher total cost by 158% [38].

The direct cost of hospitalisation was increased by hospitalisation-related interventions and patients’ conditions. Mooney reported an increase of $36,911 among patients requiring mechanical ventilation, and $5,500 among those requiring non-invasive ventilation, indicating an association between higher cost and patients with respiratory failure [7]. Zheng et al. also reported that intensive care, pulmonary infection, and respiratory failure led to an increase in the cost of hospitalisation [25].

For predictors of indirect costs, Algamdi et al. reported that for every point increase in dyspnea score and 1-mm increase in cough severity, the odds of reporting productivity loss increased by 4% and 3% respectively [23].

To synthesise cost predictors across studies, we conducted a generalised linear model analysis with gamma distribution and a log link to examine the pooled effect of cost predictors on total direct costs, based on data retrieved from 31 studies reporting total direct costs (Table 6). Two significant cost predictors were identified: geographical region and inclusion of antifibrotic therapy in cost analysis. Compared to patients from North America, patients from Europe and Central Asia had significantly lower total costs, by approximately 87.5% (Exp(B) = 0.125, p < 0.001). Similarly, East Asia and Pacific region had an estimated 90% lower costs compared to North America (Exp(B) = 0.099, p < 0.001). Antifibrotic treatment is significantly associated with increased total annual cost (Exp(B) = 3.465, p < 0.001), indicating a 3.5-fold increase in costs compared to those studies without antifibrotic treatment.

Table 6.

Regression analysis: predictors of total direct costs in ILD patients

Variables	Exp (B) (Cost Ratio)	95% CI for Exp (B)	p-value
Demographics
Age (mean)	1.024	0.964–1.087	0.445
Percentage of male patients	1.010	0.974–1.048	0.585
Region
North America	reference
East Asia and Pacific	0.099	0.060–0.164	< 0.001*
Europe and Central Asia	0.125	0.064–0.244	< 0.001*
Subtype
IPF	reference
PPF	1.423	0.773–2.618	0.257
Autoimmune-associated ILDm	2.447	0.446–13.420	0.303
Percentage of hospitalised patients	1.003	0.992–1.013	0.633
Antifibrotic therapy included in cost analysis (yes versus no)	3.465	1.884–6.372	< 0.001*

CI 95% Confidence interval ILD interstitial lung disease, IPF Idiopathic pulmonary fibrosis, PPF progressive pulmonary fibrosis, SSc systemic sclerosis

^m Autoimmune-associated ILD includes SSc-ILD, RA-ILD and Sjögren syndrome ILD

^* Statistical significance at p < 0.05

Quality assessment

The quality assessments of the methodology and the potential risk of bias of included studies are presented in supplementary information (Table S6 – S8). The results showed that the overall quality of the included studies was relatively high: 27 studies (68%) met at least 80% of the criteria in the CHEERS checklist, nine studies (23%) fulfilled between 70 and 79%, and only four studies met 50–70% of the CHEERS criteria (Table S6). Many studies did not clearly mention the perspective of the cost analysis, report the year of the currency and conversion, nor specify the discount rate. About half of the studies did not describe the methods used to characterize uncertainty, nor report the results.

Quality assessment using the consensus-based checklist by Schnitzler et al. (Table S7) yielded similar results, showing that study perspective, the epidemiological and costing approach, and sensitivity analysis are parameters often overlooked by many studies.

ECOBIAS assessment (Table S8) showed that most studies adequately reported cost components, cost measurements, and disclosed sponsorship information. However, several potential biases were commonly observed, particularly narrow perspective bias, invalid valuation bias, limited sensitivity analysis bias, and reporting and dissemination bias.

Discussion

This systematic review highlights the significant economic impact of ILD globally. The mean and median direct costs per patient were $40,385 (± $38,541.18) and $28,870 (IQR $33,004.06), respectively, with a range from $1,593 to $177,526, equivalent to 4–205% of the respective country’s GDP per capita. On the other hand, indirect costs ranged from $45 to $19,507 (0.1%−37% of the respective country’s GDP per capita). This reflects a significant impact of ILD on the healthcare system and society, warranting improvement in the health system and strategies for cost containment.

The higher economic burden of ILD, especially compared to other chronic lung diseases such as asthma ($679 to $3,970 per patient per year) [53] and chronic obstructive pulmonary disease (COPD) ($1,544 to $11,787 per patient per year) [54], is influenced by both clinical and structural factors. Clinical challenges in ILD management include delayed diagnosis, the complexity of the disease itself with comorbidities and complications, and more frequent hospitalisations. For instance, the need for frequent admissions, due to exacerbation and ILD complications, particularly in progressive subtypes like IPF and PPF, contributes to a higher economic burden. This review identified that hospitalisation alone accounted for 39–88% of total direct costs. Several studies also reported high rates of exacerbation (4%−72%) and hospitalisation (25%−95%) among IPF and PPF patients, leading to a high in-hospital mortality rate (36.5%) [3, 8, 11, 17, 37, 55, 56]. Furthermore, the lack of definitive and effective treatment in certain ILD subtypes leads to higher healthcare resource utilization, thereby increasing costs.

Beyond clinical characteristics of the disease, differences in healthcare systems and financing structures also significantly influenced the magnitude of costs. In countries that practice a fragmented and privatized healthcare system, such as the US, the absence of a centralized expenditure regulation and limited price competition contribute to higher costs. Additionally, the extensive utilization of advanced technologies in diagnostics and treatment, along with higher drug costs, further increases the economic burden [57, 58]. In contrast, in countries with a centralised healthcare system, such as those in Europe, total expenditure is controlled by practicing a stricter drug price regulation system and public health insurance coverage. For example, the cost of diagnostic tests ranged from $9,142 to $21,862 in the US [29, 33, 36] but were significantly lower in France ($181) [2], Spain ($904) [12], and Australia ($1,866) [10]. Reducing trend in hospitalisation costs observed in the US, as supported by Jeganathan et al., who reported a 23.1% reduction in hospitalisation rate from 2008 to 2018 [6], and shorter mean length of stay from 9 days to 5.5 days [6, 8]. The reduction was likely due to improvements in outpatient ILD management, such as the initiation of antifibrotic therapy, which reduces the risk of acute exacerbation and hospitalisation [59, 60].

South Korea presents a different scenario with a lower reported economic burden of ILD. This was due to the omission of outpatient costs, possible underreporting of medication costs as effective treatment was unavailable during the study period, and the implementation of a national policy that subsidizes 90% of the total costs for rare diseases [11]. Given these discrepancies among countries, it is crucial to understand the structural and policy challenges within each healthcare system and to emphasize the importance of standardized reporting methods that enable cross-country cost comparison.

To reduce the economic burden of ILD, it is important to identify possible predictors of increasing costs. This review identified geographical region, disease severity, presence of comorbidities, and the usage of antifibrotic therapy as cost predictors. Common comorbidities in ILD, such as chronic obstructive pulmonary disease, pulmonary hypertension, and gastro-oesophageal reflux, require additional treatment and are associated with a higher risk of complications, faster disease progression, and shorter survival time, leading to an increase in overall healthcare expenditure [61]. Wong et al. reported that males with obstructive sleep apnea (OSA) had a more rapid decline in FVC %-predicted compared to those without comorbidities [62]. Furthermore, as the disease deteriorates, more frequent exacerbations and hospitalisations are anticipated. Consequently, more aggressive treatments such as oxygen therapy and potentially lung transplant may be initiated, further increasing the healthcare costs [12, 38]. These findings highlight the potential role of early diagnosis and the timely initiation of effective treatment in reducing costs.

The availability of antifibrotic therapy, currently the only disease-specified treatment for certain ILD subtypes, had a significant cost-driving effect (estimated annual cost of $21,486 to $138,847 per patient) [9, 34, 45]. Antifibrotic therapy offers clinical benefits but adds significantly to total costs, particularly in regions with higher drug prices and lower treatment uptake. Despite the high cost, antifibrotic drugs have been proven to slow disease progression, improve the quality of life, and reduce exacerbations and hospitalisations among IPF and PPF patients, as shown in several randomized controlled trials [60, 63, 64]. Since healthcare resource utilization and costs tend to increase over time in IPF and PPF, early initiation of Nintedanib may hypothetically reduce the overall long-term economic burden. However, limited availability and high costs of antifibrotic therapies remain barriers to patient access. Given this promising role, future economic evaluations should study the long-term economic impact of the early initiation of antifibrotics. Studies exploring the cost prediction model based on treatment effectiveness could guide our policymakers on whether the reduction in morbidity, mortality, and long-term healthcare utilization offsets the high initial treatment costs.

Indirect costs provide insights into costs from a societal perspective. In this review, the cost of absenteeism ranged from 0 to $5,547. Zero cost reported by Rogriguez et al. was probably due to the low employment status (11.8%) among the study populations with a mean age of 70.8 years [12, 65, 66]. Notably, the cost of presenteeism, which was often neglected, was found to have a bigger contribution than absenteeism [22, 23]. This shows that patients’ disease status and symptoms, such as dyspnea and cough, can impair their work performance, even though they are physically present at work [23]. Therefore, apart from early definitive and symptomatic treatments, employers can play a role by adopting workplace interventions such as modifying job scopes to lighter duties and offering flexible working hours. This may not only improve their productivity and quality of life, but also significantly alleviate the economic burden on society.

Multiple approaches are required to reduce the economic burden of ILD, including early detection, effective disease management, and health system reforms. Early diagnosis can help identify and treat patients before they progress to severe disease states, allowing early intervention. Effective disease management, achieved through regular monitoring, early detection of exacerbations, and optimized use of antifibrotic therapies, helps maintain lung function and prevent acute events. Both help to prevent progression and improve survival, eventually reducing healthcare utilization and costly hospitalisation.

In addition, health technology assessment can be applied to justify the integration of cost-effective therapies at an early stage of disease by evaluating their ability to reduce the need for hospital care and improve quality-adjusted life years, making the initial investment worthwhile. Further cost-effectiveness studies on integrating digital health technologies such as home-based monitoring, telemedicine, or artificial intelligence-based tools in disease management, particularly for remote areas, could provide additional evidence on how these implementations can help reduce exacerbations and hospitalisations, and eventually contain costs. This information can guide the government to allocate the budget towards interventions with the greatest benefits efficiently.

Moreover, health system and policy reforms are necessary at the policymaker level. Establishing a comprehensive national healthcare system with coordinated and integrated care models can help avoid redundant services among specialties. Integrated care is likely to reduce costs and improve outcomes in a systematic review [67]. Implementing centralized cost control or an effective public-private partnership, where the government controls expenditure from the provider side while market competition maintains the quality and accessibility of supplies and services, may also help contain costs. Value-based healthcare is gaining attention to ensure optimal patient outcomes and costs through the adoption of value-based payment models. However, the effectiveness of the value-based payment model remains controversial, necessitating further research to determine how to enhance the system and integrate it into local healthcare settings [68].

This systematic review aligns with previous systematic reviews, indicating the substantial economic burden of ILD. However, the inclusion of newer studies and further analysis provides additional insights into cost trends, regional disparities, and differences within the healthcare system. It also identifies cost predictors that can be targeted to reduce the overall economic burden of ILD.

This review highlights the need for more research on progressive pulmonary fibrosis, a growing concern due to its significant economic burden and limited available data. Future studies should not only focus on the economic evaluations of cost-effective or cost-saving interventions such as antifibrotic therapies or telemedicine, but also explore the integration of digital health technology into disease management. This research could provide valuable information on strategies to reduce long-term healthcare costs. Furthermore, there is a need to standardise cost components, measurement, and reporting of cost analysis studies to improve the comprehensiveness and enable meaningful comparison among international studies. The consolidated health economic evaluation reporting standards (CHEERS) 2022 can be a useful tool when reporting health economic evaluations.

Limitations

Meta-analysis was not conducted due to the heterogeneity of the available data and differences in the methodological approaches across the included studies. Nonetheless, efforts were made to adjust and inflate all costs to a single base year, 2024, in US dollars, to allow for meaningful comparisons. Subgroup analyses and regression analyses were conducted on the available data to enable a more in-depth synthesis of results.

Not all studies reported key components of their methodology and cost analysis; thus, they were determined by consensus among reviewers. There is a lack of cost analysis from regions outside the United States and Europe; thus, the results may not be generalizable to populations from other continents.

Lastly, most of the included studies are based on insurance claims or administrative data, which are highly dependent on ICD coding. Given the complexity of the diagnosis process and the possibility of coding errors, there is a chance of an inaccurate ILD diagnosis. Most studies have attempted to address this limitation by applying specific inclusion criteria, such as the number of pulmonologist visits, relevant diagnostic tests (e.g., HRCT, lung biopsy), or prescription fills for antifibrotics, to increase accuracy in identifying ILD patients. However, the possibility of over- or under-diagnosis still exists.

Conclusion

This systematic review highlights that interstitial lung disease presents both clinical and economic challenges, particularly with the subtypes such as IPF and PPF. The rising costs are influenced by the disease’s clinical complexity, including comorbidities, exacerbation, and hospitalisation, which lead to high inpatient and medication costs. Additionally, differences in the healthcare system and regional disparities significantly impact the economic burden. Both direct and indirect costs highlight the substantial societal impact of ILD. A comprehensive approach is required to address the growing economic burden, encompassing early detection, effective disease management, support for workplace adaptations, and health system reforms. Coordinated efforts to refine policy and clinical practice through research are crucial to ensuring sustainable healthcare and reducing the long-term economic impact of ILD.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

The research aim was conceptualized by S.Y., M.M., and S.C. S.Y. and M.M. performed the literature search. Screening of articles, data extraction, and quality assessment were conducted by S.Y., M.M., and S.C. S.Y. wrote the first manuscript draft, and M.M. and S.C. commented on the draft. All authors read and approved the final manuscript.

Funding

Open access funding provided by The Ministry of Higher Education Malaysia and Universiti Sains Malaysia.

Data availability

This study is a systematic review; all data are extracted from published studies.

Code availability

Search terms used in the search strategy are available in the electronic supplementary information S1.

Declarations

Ethics approval

Not applicable.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Conflict of interest

The authors have no competing interests to declare that are relevant to the content of this article.