Impact of time to diagnosis in patients with progressive fibrotic ILD: a systematic review Authors: Zhongyi Deng, Stefan Karrasch, Peter Alter, Pilar Rivera-Ortega, Juergen Behr, Rudolf Jörres Published: 2026 ABSTRACT AbstractIntroductionProgressive fibrosing interstitial lung diseases (PF-ILDs), including idiopathic pulmonary fibrosis (IPF), are characterised by irreversible fibrosis and poor prognosis. Despite the importance of timely identification, diagnostic delays remain common. This review aimed to summarise studies reporting time to diagnosis in PF-ILDs to identify reasons for diagnostic delays and the impact on several clinical and economic outcomes.MethodsDatabase (Embase, MEDLINE, Cochrane Library, LILACS and grey literature) searches were conducted on 15 July 2024 and updated on 5 June 2026. Observational studies reporting time to diagnosis in PF-ILD published between 2002 and 2026 were identified.Results34 studies representing 36 621 patients from 41 countries met the inclusion criteria. Dyspnoea, cough and fatigue were the most common presenting symptoms. Median time from symptom onset to diagnosis ranged between 6.0 and 25.2 months; mean estimates varied from 9.6 to 39.3 months in IPF and exceeded 30.2 months in non-IPF PF-ILD. Between 30% and 77% of patients experienced delays >12.0 months. An increased diagnostic time was associated with more advanced disease, worse survival, poorer quality of life and increased hospitalisation rate. Across included studies, between 21% and 87% of patients were misdiagnosed (often with cardiovascular or other respiratory diseases) and received treatment for other conditions. Predictors of delayed diagnosis included older age, comorbidities, misdiagnosis, previous therapy and absence of early imaging or multidisciplinary evaluation.ConclusionsPatients with PF-ILD, including IPF, face a long time to diagnosis, associated with unfavourable clinical and healthcare resource outcomes, with limited evidence on cost outcomes. Direct economic burden resulting from delayed diagnosis is currently poorly understood and needs further research.Keywords: Interstitial Fibrosis, Idiopathic Pulmonary Fibrosis WHAT IS ALREADY KNOWN ON THIS TOPIC Progressive fibrosing interstitial lung diseases (PF-ILDs), including idiopathic pulmonary fibrosis, are characterised by irreversible progressive lung fibrosis and short life expectancy. Progressive fibrosing interstitial lung diseases (PF-ILDs), including idiopathic pulmonary fibrosis, are characterised by irreversible progressive lung fibrosis and short life expectancy. WHAT THIS STUDY ADDS Patients with PF-ILDs were found to face long diagnostic delays that worsen clinical outcomes and increase healthcare use. Patients with PF-ILDs were found to face long diagnostic delays that worsen clinical outcomes and increase healthcare use. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY The findings support earlier recognition and streamline referral to multidisciplinary ILD teams and call for improved diagnostic systems that are needed to reduce this burden. The findings support earlier recognition and streamline referral to multidisciplinary ILD teams and call for improved diagnostic systems that are needed to reduce this burden. Introduction Progressive fibrosing interstitial lung disease (PF-ILD) represents a group of chronic ILDs characterised by inflammation and irreversible progressive lung fibrosis.1 The overall prevalence of PF-ILD is estimated to be 16.7 per 100 000 persons,2 with idiopathic pulmonary fibrosis (IPF), the most frequent and severe PF-ILD, serving as prototype.3 It exhibits a prevalence ranging from 3.3 to 45.1 cases per 100 000, with a mean survival of 4 years after diagnosis and absence of antifibrotic therapy,4 5 underscoring the substantial health burden. Other common types of PF-ILD share pathophysiological mechanisms driving self-perpetuating progressive fibrosis,5–7 including those associated with connective tissue disease (CTD-ILD), fibrotic hypersensitivity pneumonitis (fHP), idiopathic non-specific interstitial pneumonia (iNSIP) and unclassifiable fibrotic ILD.8–10 It is estimated that up to 32% of these patients will develop progressive fibrosis,11 characterised by deterioration of lung function, increased risk of acute exacerbations and early mortality.1 9 Achieving timely diagnosis in PF-ILD is challenging. Early symptoms, such as dyspnoea and cough, are often non-specific and may be misattributed to more common conditions.12–14 Limited disease recognition among patients and primary care providers may explain delays in seeking medical care, specialist referral and diagnostic testing.15–19 Even for experienced clinicians, the heterogeneity of PF-ILDs and the complexity of distinguishing them from other ILDs are likely contributors to delays in achieving an accurate diagnosis,15 17 20 despite the use of established guidelines.5 21 Delays between symptom onset and accurate diagnosis burden both patients and healthcare systems. A study from the European MultiPartner IPF Registry showed that patients with IPF who experienced diagnostic delay had more progressive disease development.22 While the past decade has witnessed considerable developments in understanding the pathogenesis of PF-ILD and antifibrotic treatments, which have been shown to reduce lung function decline and prolong survival,23 24 cumulated fibrosis cannot be reversed.25 Consequently, patients diagnosed at later stages may have poorer prognoses and greater medical-resource needs. The diagnostic pathway, often involving numerous consultations and investigations,26 also imposes financial burdens. However, the available evidence on time to diagnosis varies widely.27 28 Whether delayed diagnosis may potentially impact disease progression and subsequent outcomes remains inconclusive.19 22 29 It is also unclear whether specific patient factors or healthcare providers influence the delay. To date, no systematic review has synthesised the evidence on this topic, leaving a knowledge gap that hinders the development of targeted strategies to improve early diagnosis and optimise patient outcomes. The aim of this systematic literature review was first to summarise the data on the reported timespan from onset of symptoms to accurate diagnosis in patients with PF-ILDs, and second, to assess the impact of longer diagnostic time on clinical and economic outcomes. The third objective was to investigate potential predictors of longer diagnostic intervals. We intended to evaluate epidemiological information, both peer-reviewed publications and abstracts presented at scientific conferences, quantifying the diagnostic timespan in patients and exploring the extent of heterogeneity. In summary, the present review provides a comprehensive overview of the current evidence, identifies potential areas for future research and underscores the critical need for initiatives aimed at reducing delays in this vulnerable patient population. Methods Protocol and registration The systematic review was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines30 and was registered on the International Prospective Register of Systematic Reviews (PROSPERO) in 2024 (CRD42024582276). The completed PRISMA 2020 checklist is provided in online supplemental table S1. Search strategy and selection criteria The search strategy was constructed based on the predefined inclusion criteria using keywords (free-text terms) and appropriate subject headings (MeSH/Emtree terms) (online supplemental tables S2,S3). The strategy was peer reviewed by an information specialist using the Peer Review of Electronic Search Strategies checklist (online supplemental table S3).31 The databases were originally searched on 15 July 2024, and an updated search was performed on 5 June 2026. We searched for studies published as full texts or abstracts from 1 January 2002 onwards, based on the introduction of international guidelines for the diagnosis and management of IPF in 2002,32 including MEDLINE (Ovid), Embase (Ovid), LILACS, Cochrane Central Register of Controlled Trials (CENTRAL) and ClinicalTrials.gov, WorldCat and NTIS, without language or regional restriction. Hand searches based on reference lists were executed. We included observational studies investigating adult patients diagnosed with PF-ILD. Eligible disease subtypes included IPF and other non-IPF PF-ILDs (fHP, CTD-ILD (including rheumatoid arthritis-associated ILD (RA-ILD) and systemic sclerosis-associated ILD (SSc-ILD))), iNSIP and other unclassifiable ILD with a strictly fibrotic, progressive phenotype. Studies were required to report either: (1) a quantitative measure of the time from symptom onset to diagnosis, or (2) the impact of diagnostic delay on subsequent outcomes. Studies focused exclusively on non-progressive or non-fibrotic forms of ILD were excluded. Study outcomes The primary outcome was time to diagnosis, reported as mean, median, minimum or maximum duration in months. Secondary outcomes included the impact of delayed diagnosis on outcomes in patients with PF-ILDs, determinants of prolonged diagnostic time and pathway intervals. To assess the impact of delayed diagnosis, we defined the following outcomes: disease severity, mortality, lung transplantation rate, quality of life (QoL), healthcare utilisation rate and direct economic outcomes. Data extraction and quality assessment The titles and abstracts resulting from the search, after duplicate articles were removed, were evaluated by two reviewers (ZD and AKR) using the CADIMA tool.33 Studies relevant for further eligibility assessment were retrieved in full text. Any discrepancy between the two reviewers was resolved by a third, independent reviewer. The complete reading of the articles was performed by the two reviewers (ZD and AKR), with reasons for study exclusion recorded. Data were extracted by the ZD and independently validated MK and ZM. Extracted variables included study characteristics (year of study, country, design, data source (questionnaire or electronic records), setting, population size, study duration), patients’ demographic characteristics (age, sex ratio), clinical characteristics (PF-ILDs subtype, disease stage at diagnosis, treatment patterns), patients’ diagnostic journey (time in diagnosis interval and subintervals, diagnosis procedures, attributes to longer diagnosis time) and the impact of delayed diagnosis on outcomes (survival, mortality, QoL, disease stages, healthcare resource utilisation (HCRU)). Any discrepancies between the extractor and validators were resolved through discussion to reach a consensus. We extracted the adjusted HR, with 95% CIs, for each of the events of interest. If these were unavailable, we extracted raw data where possible. For studies identified in the updated search, eligibility assessment, data extraction and methodological quality appraisal were performed by ZD and validated by RJ, with discrepancies resolved by discussion. Quality rating was conducted using the Joanna Briggs Institute critical appraisal tool checklist.34 The majority of the studies were graded as medium quality (online supplemental table S4). Data synthesis approach Though protocol specified that quantitative synthesis would be undertaken where feasible, formal meta-analysis was not performed because of substantial heterogeneity across studies in study populations, PF-ILD subtypes, definitions of diagnostic delay, data sources and reported outcome measures. Findings were synthesised in accordance with the Synthesis Without Meta-analysis (SWiM) reporting guideline.35 The SWiM checklist is displayed in online supplemental table S5. Studies were grouped by outcome domain. Within these outcomes of diagnostic delay, findings were compared descriptively by disease phenotype, geographical region and subinterval of diagnostic pathway. To ensure comparability across studies, mean time intervals reported in discrete phases were summed to calculate the total time to diagnosis. All duration times were standardised to months. These intervals were further categorised into patient-level and healthcare-level stages to synthesise the primary sources of systemic delay. The interpretability of synthesis findings was considered narratively, informed by study-level methodological quality, consistency of findings, directness to the review question, availability of adjusted estimates and completeness of reporting. Results Study selection The initial search identified 1342 records published from 1 January 2002 to 15 July 2024. After removing 103 duplicates, 1239 records were screened. Of these, 1079 were excluded during title and abstract screening, leaving 160 articles assessed for eligibility. During assessment, a further 138 reports were excluded with given reasons. Citation searching identified five additional eligible studies. During revision, an updated search identified seven newly published studies from 15 July 2024 to 5 June 2026. Finally, 34 studies reporting diagnostic waiting times and/or delayed diagnosis related to outcomes in patients with PF-ILD were included (online supplemental figure S1). Characteristics of the studies This systematic review synthesised data from 34 studies encompassing 36 621 patients across 41 countries. A total of 17 studies were from Europe, 9 from North America, 1 from Latin America, 6 from Asia and 1 published from multiple nations across Europe, Asia and North America. In total, 33 studies described time from onset of symptoms to diagnosis, while 8 examined its association with clinical outcomes and 6 investigated its determinants. All included studies were observational. Most studies (n=19) were based on electronic health records data from hospitals or registries, eight retrieved data from surveys, three used mixed sources, two used quantitative information from qualitative studies, one used claims data and one was based on analysis of referral letters. Twenty-three studies included only patients with IPF, eight studies included mixed IPF and other ILDs, and two studies specifically targeted patients with non-IPF PF-ILD (table 1). Participants had a mean age ranging from 59.028 to 73.6 years,36 and most cohorts had a male predominance (sex ratio range: 41.337–87.5%38). High proportions of people who smoked previously/people who smoke currently were identified. The common clinical sign of disease was crackles, observed in up to 97.2% of patients during lung auscultation.39–44 Table 1Characteristics of the studies and patients includedAuthorsStudy designCountryNILD typeAge, yearsMaleMean (SD)Median (IQR)n (%)IPF studies Martinez et al46Retrospective cohortUSA54IPF64.7 (8.2)NR30 (55.6) Shankar et al36Retrospective cohortUK2359IPF73.6 (8.03)NR1875 (79.5) Patel et al47Retrospective cohortUSA33IPFNRNRNR Caro et al40Retrospective cohort14 Latin American countries*761IPF71.9 (8.3)NR569 (74.7) Hoffman et al29Retrospective cohortNetherlands409IPFNR68.2313 (77.0) Hoyer et al19Prospective cohortDenmark264IPF73.26 (7.64)NR199 (75.4) Peña Miguel et al48Cross-sectional surveySpain69IPF66.5 (7.6)NRNR Lancaster et al27Cross-sectional surveyFrance, Germany, Japan, USA1249IPF65.6 (10.61)†NR849 (68.0)Bordas-Martínez et al84Retrospective cohortSpain130IPF69 (7.8)NR105 (80.8) Lyu et al28QualitativeChina16IPF59NR13 (81.2) Snyder et al55Ambispective cohortUSA347IPFNR70 (64–75)264 (76.1) Sköld 2019‡, 59Retrospective cohortSweden2826IPFNRNR1751 (62.0) Sköld 2019§, 59Retrospective cohortSweden677IPFNRNR416 (61.4) Hoyer et al44Prospective cohortDenmark204IPF73.7 (7.8)NR158 (77.5) Purokivi et al41Retrospective cohortFinland91IPF55NR52 (57.1) Vašáková et al22Retrospective cohortEMPIRE registry1170IPFNRNRNR Ghoshal et al62Cross-sectional surveyIndia1217IPFNRNRNR Schoenheit et al52QualitativeGermany, France, Italy, Spain, UK45IPFNR6722 (49.0) Guenther et al45Retrospective cohort8 European countries¶525IPF65.2 (11.6) at diagnosis; 68.1 (11.1) at enrolmentNR387 (73.7) Lamas et al25Prospective cohortUSA129IPF63 (8)NR98 (76.0) Alhamad et al49Retrospective cohortSaudi Arabia212IPF66.4 (11.7)NR150 (70.8) Natarajan and Subramanian37Retrospective cohortIndia46IPFMale: 68.36 female: 66.25NR19 (41.0) La Blasca et al38Observational case–controlItaly24IPF69.8 (8.2) for the MDT group; 76.9 (7.7) for the control groupNR21 (87.5) Phillips et al70Retrospective cohortUSA679IPF70.4 (9.3)NR373 (54.9) Pimple et al53Retrospective cohortUSA19 946IPFNR74 (66–81)11 818 (59.3)IPF and mixed ILD studies Grant-Orser 2024**, 54Observational studyCanada97Mixed††65.5 (16)NR54 (55.6) Grant-Orser 2024‡‡, 54Mixed-methods observationalCanada70Mixed††69.8 (8.1)NR35 (50.0) Van der Sar 202112Cross-sectional survey13 European countries§§273IPF (78%)¶¶NRNRNR Cosgrove et al18Cross-sectional surveyUSA600Mixed***NRMale: 69; female: 62300 (50.0) Hyldgaard et al39Retrospective cohortDenmark431Mixed†††61 (14.1)NR236 (54.7) Chai et al58Retrospective cohortSingapore97IPF, CTD-ILD67 (11)NR50 (51.5) Kaneko et al51Cross-sectional surveyJapan382IPF, PPFIPF: 71.9 (7.3); CTD-ILD: 64.1 (12.6); other PPF 70.6 (10.8)NR166 (43.5) Krauss et al57SurveyGermany, Austria, Switzerland, Finland69IPF (64.7%)NRNRNR Suvorova et al60Retrospective cohortRussia153IPF, HPNR70 (65–73)59 (38.6)Non-IPF PF-ILD studies Wijsenbeek et al11Observational studyUSA patients; physicians from seven countriesNANon-IPF PPFNRNRNR Chaudhuri et al50Cross-sectional surveyFrance, Germany, Italy, Spain, UK1335non-IPF PPF60.4 (11.6)NR559 (41.9)*Argentina, Bolivia, Brazil, Chile, Colombia, Ecuador, Guatemala, Mexico, Panama, Paraguay, Peru, Dominican Republic, Uruguay, Venezuela.†Mean age stratified by countries: France 65.9 (11.50) Germany 62.2 (9.16) Japan 70.2 (9.33) USA 65.0 (10.65).‡Patients from cohort 1.§Patients from cohort 2.¶Germany, France, Italy, Austria, Spain, Czech Republic, Hungary, UK.**Patient group from chart review.††IPF, CTD-ILD, fHP, occupational ILD, unclassifiable ILD.‡‡Patient group from patient survey.§§Spain, Belgium, UK, Italy, Germany, Netherlands, Bulgaria, France, Poland, Austria, Ireland, Norway, Romania.¶¶Sarcoidosis 10%, other types pf pulmonary fibrosis 12%.***IPF (46.5%), non-IPF ILD (autoimmune or CTD 15.7%; NSIP 15.5%).†††IPF, CTD-ILD, fHP, NSIP.CTD-ILD, connective tissue disease-associated interstitial lung disease; EMPIRE, European MultiPartner IPF Registry; fHP, fibrotic hypersensitivity pneumonitis; ILD, interstitial lung disease; iNSIP, idiopathic non-specific interstitial pneumonia; IPF, idiopathic pulmonary fibrosis; MDT, multidisciplinary team; NA, not applicable; NR, not reported; NSIP, non-specific interstitial pneumonia; PPF, progressive pulmonary fibrosis; RA-ILD, rheumatoid arthritis-associated interstitial lung disease; SSc-ILD, systemic sclerosis-associated interstitial lung disease. Characteristics of the studies and patients included Argentina, Bolivia, Brazil, Chile, Colombia, Ecuador, Guatemala, Mexico, Panama, Paraguay, Peru, Dominican Republic, Uruguay, Venezuela. Mean age stratified by countries: France 65.9 (11.50) Germany 62.2 (9.16) Japan 70.2 (9.33) USA 65.0 (10.65). Patients from cohort 1. Patients from cohort 2. Germany, France, Italy, Austria, Spain, Czech Republic, Hungary, UK. Patient group from chart review. IPF, CTD-ILD, fHP, occupational ILD, unclassifiable ILD. Patient group from patient survey. Spain, Belgium, UK, Italy, Germany, Netherlands, Bulgaria, France, Poland, Austria, Ireland, Norway, Romania. Sarcoidosis 10%, other types pf pulmonary fibrosis 12%. IPF (46.5%), non-IPF ILD (autoimmune or CTD 15.7%; NSIP 15.5%). IPF, CTD-ILD, fHP, NSIP. CTD-ILD, connective tissue disease-associated interstitial lung disease; EMPIRE, European MultiPartner IPF Registry; fHP, fibrotic hypersensitivity pneumonitis; ILD, interstitial lung disease; iNSIP, idiopathic non-specific interstitial pneumonia; IPF, idiopathic pulmonary fibrosis; MDT, multidisciplinary team; NA, not applicable; NR, not reported; NSIP, non-specific interstitial pneumonia; PPF, progressive pulmonary fibrosis; RA-ILD, rheumatoid arthritis-associated interstitial lung disease; SSc-ILD, systemic sclerosis-associated interstitial lung disease. Extent of diagnostic delay Diagnostic time was summarised as means, medians, ranges and proportions exceeding predefined delay cut-offs. The median diagnostic time in patients with IPF ranged from 6 months in a group of eight European countries (Germany, France, Italy, Austria, Spain, Czech Republic, Hungary and the UK) to 25.2 months in Denmark44 45 while mean times ranged from 9.6 to 39.3 months27 28 41 45–49 (figure 1, online supplemental table S6). When comparing across continents, no consistent pattern emerged. In patients with non-IPF PF-ILDs,11 Wijsenbeek et al reported mean time to diagnosis was 37.7, 31.2, 36.1, 30.2, 32.6 and 30.9 months in patients with RA-ILD, SSc-ILD, other CTD-ILDs, iNSIP, hypersensitivity pneumonitis and other unclassifiable ILDs, respectively. Chaudhuri et al50 reported a mean time of 15.5 months in a cohort of patients with non-IPF progressive pulmonary fibrosis, whereas Kaneko et al51 observed mean times of 14.3 months in CTD-ILD and 12.8 months in other ILDs. Median and mean time to diagnosis of patients with PF-ILD (in months) by country/region. *5 European countries (Germany, France, Italy, Spain and UK); 8 European countries (Germany, France, Italy, Austria, Spain, Czech Republic, Hungary and UK); 4 European countries (Germany, Austria, Switzerland and Finland); 14 Latin American countries (Argentina, Bolivia, Brazil, Chile, Colombia, Ecuador, Guatemala, Mexico, Panama, Paraguay, Peru, Dominican Republic, Uruguay and Venezuela). †For, Wijsenbeek et al11 2019 subtype-specific analyses estimates (RA-ILD, SSc-ILD, other CTD-ILDs, iNSIP, hypersensitivity pneumonitis, unclassifiable ILD) are displayed respectively; for, Kaneko et al51 2025 subtype-specific analyses estimates (CTD-ILDs, other ILDs) are displayed, respectively. Bordas-Martinez 2021a-c, Sköld 2019a,b and La Blasca 2025a,b refer to different subpopulations derived from their respective studies. CTD, connective tissue disease; ILD, interstitial lung disease; IPF, idiopathic pulmonary fibrosis; iNSIP, idiopathic non-specific interstitial pneumonia; PF, progressive fibrosis; RA, rheumatoid arthritis; SSc, systemic sclerosis. When subgrouping patients by cut-offs, we found that 56.2–88.6% of patients had a diagnostic time of more than 6 months,12 36 52 53 29.6–77.4% more than 12 months12 18 19 22 36 48 52 54–56 and 17.9–51.6% more than 24 months.29 36 44 48 55 57 Several studies evaluated median waiting times with subintervals broken down within the symptoms to diagnosis time in patients with IPF as illustrated in online supplemental figure S2. Specifically, the median waiting time from symptoms onset to first medical visit was 1.2–6.0 months.12 18 27 44 Time from symptoms onset to first specialist visit was 5.0–12.0 months,29 39 58 and from referral to first specialist visit was found to range from 0.9 to 2.4 months.44 47 Median time from the first specialist visit to diagnosis was 1.2–10.0 months.12 29 44 47 Full subintervals time information is presented in online supplemental table S7. The definition of diagnostic delay for IPF patients has not reached a consensus. Shankar et al and Natarajan et al used >6 months from symptom onset to diagnosis as diagnostic delay to assess its influence on disease prognosis36 37 53; >12 months was used as a cut-off time in four studies19 22 52 55 and >24 months by two other studies29 44 (table 2). Table 2Definition of diagnosis delay and association with adverse impact on patients with IPFAuthorsDisease typeSample sizeDefinition of delay according to article/control groupOutcome measuresPatient outcomesAdjusted coefficientShankar et al36IPF1380<6 months vs 6–12 months vs 12–24 months vs >24 monthsOverall survivalPatients with delay of 6–12 months had significantly decreased overall survival than those with delay of <6 months (HR (95% CI) 1.34 (1.02 to 1.77), p=0.034)Age, smoking status, GAP index, time to referral and travel distance to hospitalHoffman et al29IPF409>24 months vs <24 monthscontinuousMedian survivalNo significant impact of diagnosis delay on survival time or survival rateMedian survival time 1012 days vs 1322 days, p=0.05Time to diagnosis to survival rate (categorical) (HR (95% CI) 1.051 (0.800 to 1.380), p=0.72)Time to diagnosis to survival rate (continuous) (HR (95% CI) 1.000 (0.997 to 1.004), p=0.97)DLCO, forced vital capacity, sex and ageHoyer et al19IPF264>12 months vs <12 monthsProgression-free survivalPatients with delay of >12 months had significantly decreased progression-free survival in entire cohort and in patients with mild disease (FVC >80% predicted) groupEntire cohort (HR (95% CI) 1.70 (1.18 to 2.46), p=0.004)Mild disease (FVC >80% predicted) (HR (95% CI) 2.43 (1.45 to 4.01), p<0.001)Age, sex, FVC% predicted at baseline, DLCO% predicted at baseline and antifibrotic treatment received during follow-upQoL scoresPatients with delay of >12 months had worse QoL scores (SD).Mean total SGRQ score: 49.33 (17.65) vs 40.91 (14.63), p=0.03Mean baseline derived SGRQ-I: 45.41 (22.26) vs 34.37 (19.97), p=0.02Age, sex, FVC% predicted at baseline, DLCO% predicted at baseline and antifibrotic treatment received during follow-upMean hospitalisation ratesPatients with delay of >12 months had significantly higher mean hospitalisation rates during first year after diagnosis and entire follow-up periodFirst year after diagnosis:All-cause admissions: 0.71 vs 0.38 per person-year (IRR (95% CI) 3.28 (1.35 to 8.55), p=0.01)Respiratory admissions: 0.25 vs 0.13 per person-year (IRR (95% CI) 5.80 (1.19 to 42.12), p=0.04)Entire follow-up period:All-cause admissions: 0.68 vs 0.57 per person-year, (IRR (95% CI) 1.74 (1.01 to 3.02), p=0.04)Respiratory admissions: 0.35 vs 0.27 per person-year, (IRR (95% CI) 1.70 (0.81 to 3.65), p=0.16)Age, sex, FVC% predicted at baseline, DLCO% predicted at baseline and antifibrotic treatment received during follow-up173Mean FVCPatients with delayed diagnosis had worse disease progression by mean FVC (% predicted) in mild disease group: 98.19% (12.82) vs 104.43% (16.46), p<0.01Age, sex, FVC% predicted at baseline, DLCO% predicted at baseline and antifibrotic treatment received during follow-upSnyder et al55IPF347>12 months vs <12 monthsCombined endpoint of death or lung transplantNo significant association with death or lung transplant. Unadjusted HR (95% CI) 0.83 (0.53 to 1.30), p=0.41; adjusted HR (95% CI) 0.76 (0.48 to 1.21), p=0.25Age, use of supplemental oxygen at rest or with activity, FVC% predicted and DLCO% predictedVašáková et al22IPF1170>12 months vs <12 monthsMean vital capacityPatients with delay of >12 months had worse disease progression by mean vital capacity at time of diagnosis: 2.49 L vs 2.72 L, p=0.002NRMedian survival timePatients with delay of >12 months had significantly decreased mean survival time: 50.9 vs 74.4 months, p=0.001NRLamas et al25IPF129<12 months vs 12–24 months vs 24–48 months vs >48 monthsSurvivalPatients with longer delay had significantly increased risk of death: HR (95% CI) 1.3 (1.03 to 1.6), p=0.03Age, sex, FVC% predicted, private insurance and college educationLung transplantationNot associated with a lower likelihood of undergoing lung transplantation HR 1.0, p=0.68Age, sex, FVC% predicted, private insurance and college educationNatarajan and Subramanian37IPF46≥6 months and administered with anti-tuberculosis medications vs those who were notSurvivalPatients with delay ≥6 months had significantly decreased survival: p=0.05NRPimple et al53IPF19 946>6 months vs <6 monthsInitiation of antifibrotic treatmentNo association with antifibrotic treatment initiation (HR (95% CI) 0.94 (0.85 to 1.03)Baseline demographic, clinical characteristicsDLCO, diffusing capacity of the lung for carbon monoxide; FVC, forced vital capacity; GAP, Gender, Age, Physiology; IPF, idiopathic pulmonary fibrosis; IRR, incidence rate ratio; NR, not reported; QoL, quality of life; SGRQ, St. George’s Respiratory Questionnaire; SGRQ-I, St. George’s Respiratory Questionnaire–IPF-specific version. Definition of diagnosis delay and association with adverse impact on patients with IPF DLCO, diffusing capacity of the lung for carbon monoxide; FVC, forced vital capacity; GAP, Gender, Age, Physiology; IPF, idiopathic pulmonary fibrosis; IRR, incidence rate ratio; NR, not reported; QoL, quality of life; SGRQ, St. George’s Respiratory Questionnaire; SGRQ-I, St. George’s Respiratory Questionnaire–IPF-specific version. Diagnostic pathway The definitions for both the disease and the diagnostic delay varied across literature. In studies listing the initial symptoms for patients with PF-ILD, we found commonly reported respiratory symptoms were dyspnoea12 18 19 25 27–29 37 41 45 50–57 59 60 and cough,12 18 19 27–29 37 41 45 50–57 59 60 while systemic symptoms included fatigue12 18 19 27 45 52 54 55 59 and weight loss12 18 19 52 55 (online supplemental table S6, figure S3). Symptom onset was most commonly ascertained from patient-reported symptom history or survey responses. Thirteen studies defined IPF in accordance with the international American Thoracic Society and European Respiratory Society guidelines,3 21 61 while six studies used International Classification of Diseases (ICD) codes.11 46 47 53 56 59 The remaining studies defined with seven self-reported diagnoses,12 18 27 48 50 51 54 six physician-reported diagnoses28 38 52 57 60 62 and one through a registry record.63 One study used physician-reported diagnoses to assess progressive fibrosis phenotypes in non-IPF PF-ILDs after classifying ILD subtypes with ICD codes.11 Only nine studies mentioned that diagnoses were confirmed via multidisciplinary team (MDT) discussion.19 29 36 38 45 49 55 58 60 During the process, patients were found to undergo several diagnostic investigations. Imaging studies (eg, X-ray, CT, high-resolution CT (HRCT)) were most frequently reported, while biopsy was used in fewer studies (online supplemental table S8). Notably, as shown in table 3, 21–87.5% of patients underwent misdiagnoses, including other respiratory diseases like chronic obstructive pulmonary disease (COPD),27 28 44 50 52 54 asthma,18 28 44 50 52 54 bronchitis18 44 54 and pneumonia,18 28 52 as well as cardiovascular disease.41 52 54 These patients were found to receive treatments or management procedures targeting other conditions18 28 29 37 54 58 and experienced longer time to diagnosis,18 29 37 with the median time to correct diagnosis being 11 months, as described by Cosgrove et al.18 Table 3Summary of misdiagnoses experienced by patients with PF-ILD prior to diagnosisStudyDisease type% of misdiagnosisMisdiagnosed asReported outcomesHoffman et al29IPF16Other ILD diagnosesLonger time to correct diagnosis (47.0 months vs 19.0 months; p<0.001). More frequent use of prednisolone or other immunosuppressive medications (69% vs 17%; p<0.001)Lancaster et al27IPF45COPDNot specified in detailLyu et al28IPF87.5Other respiratory diseases (asthma, COPD, pneumonia)Symptom-relieving medications used; delayed seeking further medical advice until condition significantly deterioratedHoyer et al44IPF41Heart disease, asthma, chronic bronchitis, COPD, emphysema, othersNot specified in detailGhoshal et al62IPF59Pulmonary tuberculosisPulmonary tuberculosis remains a differential diagnosis for 59% of HCPsSchoenheit et al52IPFNRAsthma, COPD, pneumonia, pertussis, pleurisy and pericarditisThere was a tendency among these patients to perceive the initial diagnosis not as a working hypothesis but rather as an erroneous or missed diagnosisNatarajan and Subramanian37IPF21TuberculosisThis patient group had delay in diagnosis (6 months to 2 years). Lower survival rate for the latter group (20% vs 50%, p=0.05)Hyldgaard et al39IPF, CTD-ILD, fHP, NSIP32.2Unclassified ILDNot specified in detailChai et al58IPF, CTD-ILD50.5ILD with uncertain classification, CTD-ILD, iNSIPChanges in management after correct diagnosis, including initiation or augmentation of appropriate therapiesGrant-Orser et al54IPF, CTD-ILD, fHP, occupational ILD, unclassifiable ILD45COPD, asthma, bronchitis, heart disease, othersTreated for other conditions, leading to inappropriate management of actual ILDCosgrove et al18IPF (46.5%), non-IPF ILD (autoimmune or CTD 15.7%; NSIP 15.5%)55Asthma, pneumonia, bronchitisMedian time to correct diagnosis was 11 months, significant delays and exposure to ineffective or harmful therapiesChaudhuri et al50Non-IPF PPF17.4Asthma, COPDNot specified in detailKaneko et al51IPF, PPF6.8Asthma and COPDNot specified in detailCOPD, chronic obstructive pulmonary disease; CTD, connective tissue disease; fHP, fibrotic hypersensitivity pneumonitis; HCP, healthcare professional; ILD, interstitial lung disease; iNSIP, idiopathic non‑specific interstitial pneumonia; IPF, idiopathic pulmonary fibrosis; NR, not reported; PF, progressive fibrosing; PPF, progressive pulmonary fibrosis. Summary of misdiagnoses experienced by patients with PF-ILD prior to diagnosis COPD, chronic obstructive pulmonary disease; CTD, connective tissue disease; fHP, fibrotic hypersensitivity pneumonitis; HCP, healthcare professional; ILD, interstitial lung disease; iNSIP, idiopathic non‑specific interstitial pneumonia; IPF, idiopathic pulmonary fibrosis; NR, not reported; PF, progressive fibrosing; PPF, progressive pulmonary fibrosis. Factors associated with diagnostic delay Six studies identified determinants associated with longer diagnostic time in IPF (table 4). Patient characteristics associated with a longer delay included older age,44 59 sex differences,44 56 certain comorbidities,25 lower income56 and symptom profiles.29 59 A family history of pulmonary fibrosis was associated with a shorter diagnostic time.29 As for healthcare providers, absence of chest X-ray,29 absence of MDT evaluation,38 previous other ILD diagnoses29 and other treatments prior to final diagnosis were linked to longer delay.29 44 Table 4Risk factors and their association with diagnostic delay in patients with IPFAuthorsDisease typeCharacteristicsHoffman et al29IPF Have no family member with pulmonary fibrosis vs presence (median time to diagnosis 24.5 vs 19.0 months; p=0.03).Symptoms that did not start directly after a respiratory tract infection vs start directly (median time to diagnosis 25.0 vs 9.5 months; p<0.001).Absence of chest X-ray prior to the start of symptoms vs presence (median time to diagnosis 26.0 vs 17.0 months; p=0.01).Received steroids prior to diagnosis vs none (median time to diagnosis 37.0 vs 18.5 months; p<0.001), received other immunosuppressant prior to diagnosis vs none (median time to diagnosis 48.0 vs 23.0 months; p=0.002).Prior diagnosis of another ILD (median time to diagnosis 47.0 vs 19.0 months; p<0.001). Sköld et al59IPF Older age on index date (being diagnosed within 1 year, HR (95% CI) 0.988 (0.981 to 0.996)).Early symptom type: cough vs dyspnoea (being diagnosed within 1 year, HR (95% CI) 0.81 (0.74 to 0.89)); fatigue vs dyspnoea (being diagnosed within 1 year, HR (95% CI) 0.78 (0.69 to 0.89)). Hoyer et al44IPF Sex, men vs women (patient delay, IRR (95% CI) 3.84 (1.17 to 11.36), p=0.006).Received previous inhalation therapy vs none (patient delay, IRR (95% CI) 4.68 (1.77 to 13.37), p=0.0004).Older age (total delay, IRR (95% CI) 0.03 (1.01 to 1.06), p=0.004). Lamas et al25IPFComorbidities: participants with diabetes mellitus vs others (median delay 3.5 vs 2.1 years, p=0.049), with gastro-oesophageal reflux disease vs others (median delay 2.8 vs 1.8 years, p=0.01).La Blasca et al38IPFAbsence of MDT evaluation: The interval from the first outpatient visit to a definite diagnosis was 3±2.3 months with MDT vs 12.8±9.4 months without MDT (p=0.02).Phillips et al70IPF Sex: Males were more likely to receive a faster diagnosis than females (p<0.001).Income: Higher-income groups are associated with faster diagnosis (p=0.010). ILD, interstitial lung disease; IPF, idiopathic pulmonary fibrosis; IRR, incidence rate ratio; MDT, multidisciplinary team. Risk factors and their association with diagnostic delay in patients with IPF Have no family member with pulmonary fibrosis vs presence (median time to diagnosis 24.5 vs 19.0 months; p=0.03). Have no family member with pulmonary fibrosis vs presence (median time to diagnosis 24.5 vs 19.0 months; p=0.03). Symptoms that did not start directly after a respiratory tract infection vs start directly (median time to diagnosis 25.0 vs 9.5 months; p<0.001). Symptoms that did not start directly after a respiratory tract infection vs start directly (median time to diagnosis 25.0 vs 9.5 months; p<0.001). Absence of chest X-ray prior to the start of symptoms vs presence (median time to diagnosis 26.0 vs 17.0 months; p=0.01). Absence of chest X-ray prior to the start of symptoms vs presence (median time to diagnosis 26.0 vs 17.0 months; p=0.01). Received steroids prior to diagnosis vs none (median time to diagnosis 37.0 vs 18.5 months; p<0.001), received other immunosuppressant prior to diagnosis vs none (median time to diagnosis 48.0 vs 23.0 months; p=0.002). Received steroids prior to diagnosis vs none (median time to diagnosis 37.0 vs 18.5 months; p<0.001), received other immunosuppressant prior to diagnosis vs none (median time to diagnosis 48.0 vs 23.0 months; p=0.002). Prior diagnosis of another ILD (median time to diagnosis 47.0 vs 19.0 months; p<0.001). Prior diagnosis of another ILD (median time to diagnosis 47.0 vs 19.0 months; p<0.001). Older age on index date (being diagnosed within 1 year, HR (95% CI) 0.988 (0.981 to 0.996)). Older age on index date (being diagnosed within 1 year, HR (95% CI) 0.988 (0.981 to 0.996)). Early symptom type: cough vs dyspnoea (being diagnosed within 1 year, HR (95% CI) 0.81 (0.74 to 0.89)); fatigue vs dyspnoea (being diagnosed within 1 year, HR (95% CI) 0.78 (0.69 to 0.89)). Early symptom type: cough vs dyspnoea (being diagnosed within 1 year, HR (95% CI) 0.81 (0.74 to 0.89)); fatigue vs dyspnoea (being diagnosed within 1 year, HR (95% CI) 0.78 (0.69 to 0.89)). Sex, men vs women (patient delay, IRR (95% CI) 3.84 (1.17 to 11.36), p=0.006). Sex, men vs women (patient delay, IRR (95% CI) 3.84 (1.17 to 11.36), p=0.006). Received previous inhalation therapy vs none (patient delay, IRR (95% CI) 4.68 (1.77 to 13.37), p=0.0004). Received previous inhalation therapy vs none (patient delay, IRR (95% CI) 4.68 (1.77 to 13.37), p=0.0004). Older age (total delay, IRR (95% CI) 0.03 (1.01 to 1.06), p=0.004). Older age (total delay, IRR (95% CI) 0.03 (1.01 to 1.06), p=0.004). Sex: Males were more likely to receive a faster diagnosis than females (p<0.001). Sex: Males were more likely to receive a faster diagnosis than females (p<0.001). Income: Higher-income groups are associated with faster diagnosis (p=0.010). Income: Higher-income groups are associated with faster diagnosis (p=0.010). ILD, interstitial lung disease; IPF, idiopathic pulmonary fibrosis; IRR, incidence rate ratio; MDT, multidisciplinary team. Impact of diagnostic waiting time on adverse outcomes Seven studies explored the association between increased diagnostic time and subsequent prognosis in patients with IPF (table 2). Five studies reported a significant association between longer diagnostic time and worse survival outcomes,19 22 25 36 37 whereas two studies found no significance with either a composite endpoint of death or lung transplantation29 or survival rate.55 Importantly, only a subset of studies adjusted for baseline disease severity. Among these adjusted analyses, significant associations were reported by Shankar et al,36 Hoyer et al19 and Lamas et al,25 whereas Snyder et al55 and Hoffman et al29 found no significant association. In addition, two studies found that longer diagnostic time was associated with greater disease severity at diagnosis, reflected by worse mean forced vital capacity19 and vital capacity.63 Hoyer et al also found the delay was associated with a poorer QoL, with results that were consistent across three different QoL assessment tools.19 No included studies assessed delay-related costs. Evidence was limited to HCRU and treatment-related care pathway indicators. Hoyer et al reported an association between longer delay and increased hospitalisation rates, whereas Pimple et al found no significant association with initiation of antifibrotic treatment.19 53 Discussion This is the first systematic review synthesising evidence on time to diagnosis and its consequences with PF-ILD. Across 34 studies encompassing over 36 000 patients from 41 countries, diagnostic delays was a relevant global issue. The time from symptom onset to diagnosis typically spans months to years, with median delays of 6–25.2 months and mean delays of 9.6–39.3 months. For non-IPF PF-ILDs, delays exceeded 30 months. Longer diagnostic time appears to be associated with adverse clinical and HCRU consequences. The updated search did not materially change the conclusions but reinforced the problem of prolonged diagnostic time. Taken together, the evidence identifies a complex ‘diagnostic odyssey’ driven by individual-level and healthcare-level factors, demonstrating the need for concerted strategies to expedite recognition, referral and MDT diagnosis, consistent with contemporary guidance for patients with PF-ILD.5 64 Interpretation of diagnostic delays and variation The reported time to diagnosis varied substantially across studies, likely reflecting differences in healthcare systems, study designs and definitions of diagnostic delay. Geographical patterns remain inconclusive. Some cohorts from well-resourced healthcare infrastructure, such as Denmark and the Netherlands, reported among the longest median times, suggesting that healthcare-system wealth alone does not ensure timely diagnosis.29 44 This may reflect ascertainment bias, wherein these robust longitudinal health records capture earlier, subtle symptoms or strict gatekeeping and resource constraints.27 54 For instance, the current national guidelines in Finland do not classify suspected parenchymal lung disease as grounds for urgent referral, mandating evaluation windows that inherently extend the diagnostic timeline.41 Conversely, variations within single countries suggest that access to advanced diagnostics (eg, HRCT, MDT) remains uneven, particularly affecting rural populations.28 36 54 Such observations call for caution regarding interpretations based on country income status alone. Subinterval analyses revealed that delays were not solely patient-driven. While the interval from symptom onset to first medical contact was significant, delays also occurred after the initial medical consultation, specifically between the first specialist visit and the final diagnosis (median between 1.2 and 10 months).12 29 44 47 The apparently short referral-to-specialist (0.9–2.4 months) should be interpreted cautiously, as the referral subinterval followed abnormal chest radiography and CT imaging or investigation at other community hospitals,44 47 suggesting it reflects a relatively late stage of work-up. These patterns suggest that parts of the diagnostic pathway may be affected by healthcare-system and specialist-level bottlenecks, including challenges in triage, serial imaging to confirm progression and the logistical complexity of MDT discussions. These requirements may be greater in non-IPF PF-ILD, where clinicians must document progression and resolve a broader differential, plausibly extending the pathway. Why delays occur: a multilevel view Our findings indicate that diagnostic delays stem from a multifactorial interplay of patient, clinician and healthcare system barriers. Patient-related factors associated with longer delays included older age, sex, type of symptom and comorbidities such as cardiovascular disease or diabetes, which may obscure the recognition of PF-ILD symptoms or lead clinicians to attribute them to ageing or more common chronic conditions.18 28 This is supported by a large US claims-based study demonstrating higher all-recorded comorbidity prevalence among patients with delayed diagnosis.53 Comorbidities are ubiquitous and continue to pose challenges in IPF,65 with current clinical reviews advocating routine and structured assessments for reflux disease, cardiovascular disease and obstructive sleep apnoea.66 Interestingly, a family history of pulmonary fibrosis was associated with shorter time to diagnosis,29 likely attributable to heightened patient awareness due to known familial risk and clinician suspicion. Misdiagnosis represents a major, modifiable determinant of delay. Between 21% and 87% of patients were initially misdiagnosed, most frequently with COPD, asthma or heart failure, and one included study showed a median of 11 months from initial incorrect diagnosis to the corrected one.18 This ‘diagnostic masquerade’ may not only prolong the time to antifibrotic therapy but may also expose patients to inappropriate and potentially harmful treatments, such as bronchodilator or systemic corticosteroid therapies.18 67 68 These clinical challenges are further compounded by structural barriers. While complex conditions like ILD require MDT discussion, structural hurdles often impede the necessary care coordination.12 38 69 Consequently, our observations mirror the ‘diagnostic odyssey’ common in other rare diseases, where fragmented care and the diffusion of clinical responsibility across multiple providers hinder definitive action.70 71 Diagnostic delay and outcomes The synthesis of outcome data underscores the clinical and resource urgency of timely diagnosis. Several studies demonstrated that longer diagnostic delays were associated with reduced survival, with a subset adjusted for baseline disease severity. Delay was also associated with reduced lung function at diagnosis, supporting the biological plausibility that unmanaged disease progresses and leads to irreversible function decline.72 73 At the same time, more advanced disease at diagnosis may explain part, though likely not all, of the observed relationship between diagnostic delay and poorer survival. While two studies did not detect a significant survival difference,25 55 these analyses were limited by either small sample sizes or the use of broad delay thresholds such as >24 months may introduce immortal time bias, where patients must survive a long delay to be classified as delayed, thereby selecting for a relatively healthier cohort.74 Timely diagnosis is clinically meaningful only if followed by timely management.5 75 Treatment therefore represents a potential mediator between diagnostic timing and downstream outcomes. No significant association between diagnostic delay and antifibrotic treatment initiation was found in review, suggesting that post-diagnostic factors, including treatment eligibility, access to specialist ILD care, reimbursement and patient preferences may also influence treatment initiation.64 76 The diagnostic delay also imposes a heavy burden on QoL and HCRU. Patients with delays exhibited significantly worse scores on QoL and higher rates of hospitalisation.19 No included studies directly quantified delay-related cost. However, literature indicates that IPF and fibrotic ILD may impose substantial overall economic burden after diagnosis, and modelling analyses suggest that earlier diagnosis may be cost-effective.77 78 Prolonged diagnostic pathways may also involve redundant consultations, unnecessary tests beyond recommendations and management of advanced disease.3 18 27 28 Patient surveys18 27 also suggested potential consequences for early retirement, disability claims and greater work impairment, indicating the potentially avoidable resources and costs by timely disease identification. Strength and limitation To our knowledge, this is the first systematic review comprehensively covering evidence on diagnostic delays in PF-ILDs, including IPF. Our methodology was rigorous, following PRISMA guidance with a pre-registered protocol and a comprehensive search strategy incorporating grey literature. However, there are several limitations. Most included studies were retrospective, with substantial heterogeneity regarding definitions, data sources and reporting standards. The lack of standardised definitions for ‘diagnostic delay’ and ‘symptom onset’ impeded direct comparability and precluded meta-analysis; therefore, findings were synthesised descriptively in accordance with SWiM guidance. Diagnostic confirmation was not uniformly based on MDT discussion, the gold standard recommended in major ILD guidelines, and reliance on physician or patient reported diagnosis introduces potential bias. Consistent with the risk-of-bias assessment, most studies were of moderate methodological quality, which supports the descriptive finding that diagnostic delay is common but limits causal interpretation. Evidence during the COVID-19 era remains limited. The pandemic likely created a mixed effect: healthcare system strain may have prolonged delays, while the surge in chest imaging and heightened awareness in the respiratory community may have accelerated time to recognition.79 80 Although some determinants of diagnostic delay were identified, the underlying reasons remain obscure due to the small number of studies addressing this question. Though economic outcomes were prespecified, the eligible literature did not provide direct empirical evidence for delay-attributable costs. Moreover, evidence on non-IPF PF-ILDs remains limited. Nevertheless, given the overlap between IPF and PF-ILDs regarding radiologic, prognostic and clinical attributes, extrapolation from IPF may be reasonable.5–7 81 Implications for practice and research Reducing delays in the diagnosis of PF-ILD requires a multipronged strategy.82 Education initiatives may encourage patients to seek timely medical attention. Enhancing ILD-specific training for general practitioners is critical to help raise early suspicion. This includes promoting the use of basic, readily available tools in primary care such as lung auscultation and spirometry. Practical cues such as persistent exertional dyspnoea, cough and ‘Velcro-like’ crackles should prompt early chest imaging and spirometry.14 75 Systemically, streamlining referral pathways, strengthening collaboration between primary care providers, pulmonary specialists and MDT ILD teams and expanding access to HRCT imaging are essential to resolving the bottlenecks identified. Moreover, implementation of comprehensive screening strategies for at-risk individuals may further support early detection.75 83 Future research should prioritise establishing a consensus-based definition for ‘timely diagnosis’ to facilitate benchmarking. Efforts are required to focus more on non-IPF PF-ILDs subtypes and access diagnostic pathways in countries across Oceania or Africa. Economic evaluations quantifying the impact of delayed diagnosis on direct utilisation costs, as well as indirect societal costs such as patient work disability and caregiver burden, will be critical to inform resource allocation and policy development. Finally, interventional studies are needed to test whether targeted screening of high-risk populations (eg, those with familial risk, occupational exposures or incidental interstitial lung abnormalities) can shorten diagnostic intervals and improve long-term prognosis. Conclusion Diagnostic delay in PF-ILDs including IPF is a prevalent, global phenomenon, contributing to adverse clinical and QoL outcomes and higher healthcare utilisation. This review also highlights that delays are not merely a result of patient hesitation but are compounded by systemic hurdles and misdiagnosis. Concerted efforts to improve early symptom recognition, standardise referral pathways and accelerate access to antifibrotic therapy are imperative to alter the trajectory of these devastating diseases. Supplementary material Acknowledgements The authors would like to thank Professor Jakob Linseisen, University of Augsburg, as a member of ZD’s PhD thesis advisory committee for his insightful suggestions. The authors thank Anna Katharina Renner, former employee of Boehringer Ingelheim GmbH, for her valuable contribution to the protocol conceptualisation, screening phases and feedback on early-stage drafts. We further thank Bettina Schmitt, Boehringer Ingelheim GmbH, for completing the PRESS peer review assessment and for her feedback on the search strategy. Masoumeh Kisomi, PhD, and Zoe Marjenberg, PhD, of Maverex Ltd. provided validation support for data extraction, which was contracted and funded by Boehringer Ingelheim. All authors read and approved the final manuscript. All named authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this article, take responsibility for the integrity of the work as a whole and have given their approval for this version to be published. The authors did not receive payment related to the development of the manuscript. Footnotes Funding: The study was funded by Boehringer Ingelheim International GmbH. The funder was given the opportunity to review the manuscript for medical and scientific accuracy as well as intellectual property considerations. Provenance and peer review: Not commissioned; externally peer reviewed. Patient consent for publication: Not applicable. Ethics approval: Not applicable. Data availability free text: The authors declare that the data supporting the findings of this study are available within the paper and its Supplementary Information files. Should any raw data files be needed in another format they are available from the corresponding author upon reasonable request. Data availability statement All data relevant to the study are included in the article or uploaded as supplementary information. References Associated Data Supplementary Materials Data Availability Statement All data relevant to the study are included in the article or uploaded as supplementary information.