Prevalence of drug-induced interstitial lung disease in Saudi Arabia: a real-world data study

Abstract

Drug-induced interstitial lung disease (DI-ILD) is a serious but often under-recognized complication of various medications. Despite its potential reversibility if detected early, estimating its true prevalence is difficult due to clinical overlap with other interstitial lung disease (ILD) subtypes and the absence of specific biomarkers. We conducted a retrospective observational study using the Saudi Real-world Evidence Network between 2016 and 2023. ILD cases were identified using ICD-10 code J84.XX, and DI-ILD was defined either by diagnostic codes (J70.2–J70.4) or as a potential DI-ILD group based on recent exposure to drugs previously associated with pulmonary toxicity within six months prior to diagnosis. Demographics, comorbidities, and prescribing patterns were analyzed, and prevalence was calculated per 100,000 individuals. Among 2,858,141 individuals, 1,907 ILD cases were identified (66.7 per 100,000). Coded DI-ILD accounted for 1.26 per 100,000, while potential DI-ILD based on drug exposure was 28.6 per 100,000. Patients with potential DI-ILD were more often female (62.5%), older (mean age 61.1 ± 15.4 years), and had a higher comorbidity burden (mean Charlson Comorbidity Index 2.0 ± 2.1). Among ILD patients, 42.9% had recent exposure to at least one drug associated with pulmonary toxicity, most commonly antiarrhythmics (18.7%), including amiodarone (6.5%). These findings highlight the diagnostic complexity of DI-ILD in routine clinical practice and underscore the importance of systematic medication history review when evaluating patients with ILD. Exposure-based prevalence estimates should be interpreted cautiously and may serve as indicators of potential risk rather than confirmed diagnoses, supporting the need for enhanced diagnostic awareness and strengthened pharmacovigilance approaches.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-43586-6.

Introduction

Interstitial lung disease (ILD) encompasses a heterogeneous group of pulmonary disorders characterized by inflammation and fibrosis of the lung parenchyma¹. Drug-induced interstitial lung disease (DI-ILD) is a clinically significant and potentially reversible subset of ILD caused by a wide range of drugs including chemotherapeutics (e.g., bleomycin), immunosuppressants (e.g., methotrexate), and targeted therapies (e.g., immune checkpoint inhibitors)^2,3. The underlying pathogenesis of DI-ILD is complex and multifactorial, involving direct cytotoxicity, immune-mediated injury, oxidative stress, and idiosyncratic reactions, which contribute to alveolar damage and fibrosis^3,4. The clinical presentation of DI-ILD can vary significantly, ranging from asymptomatic radiographic findings to severe respiratory failure, which makes diagnosis and management particularly challenging^3,5. Early detection and drug discontinuation are critical for favorable outcomes, as timely intervention can prevent irreversible lung damage².

Diagnosing DI-ILD remains challenging due to the absence of definitive clinical, laboratory, or radiologic markers⁶. While high-resolution computed tomography (HRCT) and pulmonary function tests (PFTs) are useful for detecting interstitial lung abnormalities, they do not differentiate drug-induced ILD from other types⁶. DI-ILD diagnosis often relies on clinical judgment, exclusion of other possible causes, and establishing a temporal relationship between drug exposure and the onset of respiratory symptoms—typically within days to several months of initiating drug⁷. A comprehensive evaluation includes a detailed medical history, capturing both prescription and over-the-counter drug use, as well as any relevant occupational or environmental exposures⁸. Further complicating the diagnostic process is the frequent use of multiple drugs, particularly in patients with cancer or autoimmune diseases, making it difficult to attribute lung injury to a single causative drug^9,10.

The prevalence of DI-ILD remains poorly defined, partly due to underreporting and diagnostic uncertainty¹. The frequency varies due to differences in prescribing practices, diagnostic criteria, and environmental and genetic susceptibility³. A systematic review found that cancer therapies are the most common cause (23–51%), followed by disease-modifying antirheumatic drugs (6–72%), antibiotics (6–26%), and antiarrhythmic drugs (0–9%)^3,11.

Despite the clinical importance of DI-ILD, there is a limited data available from Middle Eastern countries, particularly Saudi Arabia¹². Regional differences in prescribing patterns, genetic background, and environmental exposures may affect both the prevalence and risk profile of DI-ILD³. Furthermore, local epidemiological data are essential for tailoring pharmacovigilance strategies and drug safety regulations^12,13. Previous research has largely been limited to case reports and small case series, which limit the ability to generate prevalence estimates of the disease¹⁰. This highlights the necessity for studies to assess the prevalence of DI-ILD and associated drug exposures in larger population.

This study aims to describe the prevalence of DI-ILD and explore recent exposure to drugs previously associated with pulmonary toxicity among ILD cases in a large patient cohort in Saudi Arabia. By characterizing patterns of drug exposure within this population, the findings may support future efforts by the Saudi Food and Drug Authority (SFDA) to strengthen pharmacovigilance strategies. These may include issuing safety communications, refining drug labeling, or prioritizing risk mitigation studies.

Methods

Study design, data source and study population

This study was a retrospective observational design to estimate the prevalence of DI-ILD and explore recent exposure to drugs previously associated with pulmonary toxicity among ILD cases. Data were extracted from the Saudi Real-world Evidence Network (SRWEN), covering the period from 2016 to 2023.

SRWEN is a national initiative led by the Saudi Food and Drug Authority (SFDA), developed to standardize and integrate electronic health records from various healthcare institutions across Saudi Arabia¹⁴. The database contains de-identified patient-level data, including demographics, diagnostic codes, drug use, and comorbidities.

The study population was all patients (of all ages) in the SRWEN database (n = 2,858,141). Patients without complete demographic or drug exposure data were excluded.

Case definition

ILD and DI-ILD cases were identified using diagnostic codes from the International Classification of Diseases, Tenth Revision (ICD-10). The following codes were used: J70.2–4 (acute drug-induced ILD), J70.3 (chronic drug-induced ILD), and J70.4 (unspecified drug-induced ILD), or ILD using ICD-10: J84.XX (J84-J84.9) were included in the analysis¹⁵. DI-ILD cases were defined using two approaches:

• Coded definition: patients with specific DI-ILD diagnostic codes (J70.2–J70.4).

• Exposure-based definition: ILD patients (J84.XX) who received at least one drug previously associated with DI-ILD (based on literature) within six months prior to diagnosis. These cases are referred to as potential DI-ILD (exposure-based), used for exploratory purposes.

To reduce misclassification, patients with documented risk factors for ILD unrelated to drug exposure—such as tobacco use, occupational/environmental exposures, or toxic agents—were excluded from DI-ILD groups using the following ICD-10 codes: F17, Z57, Z71, Z72, Z77, T65.

Patients were categorized into three groups:

• ILD group: All patients diagnosed with ILD (ICD-10: J84.xx), including those with DI-ILD-specific codes.

• DI-ILD (coded): Patients with drug-induced ILD diagnostic codes (J70.2–J70.4).

• Potential DI-ILD (exposure-based): ILD patients (J84.xx) with recent exposure (within 6 months) to at least one drug previously associated with DI-ILD.

Data collection and variables

Data were collected from the SRWEN database and included two main categories: drug-related exposures and patient demographic and clinical characteristics.

Drug characteristics

Medication history was reviewed for all ILD patients within a six-month window prior to diagnosis. This timeframe reflects the typical latency of drug-induced lung injury^3,4. Drugs with known or suspected pulmonary toxicity were categorized into the following therapeutic groups:

• Antineoplastics (methotrexate, cyclophosphamide, bleomycin, vinblastine, carmustine, busulfan, doxorubicin, erlotinib, chlorambucil),

• Immunosuppressants (azathioprine, sirolimus, tacrolimus, mycophenolate),

• Antiarrhythmics/antihypertensive (amiodarone, hydralazine),

• DMARDs (rituximab, etanercept, infliximab, sulfasalazine, penicillamine),

• Antimicrobials (nitrofurantoin, isoniazid).

These classifications were used to analyze prior drug exposure among patients diagnosed with ILD. These therapeutic categories were selected based on literature reports of pulmonary toxicity cases or studies (Supplementary Table 1).

Patient demographic and clinical characteristics

Demographic variables included age, sex, and nationality. Age was calculated at the time of ILD diagnosis. Healthcare setting (inpatient, outpatient or ER) at the time of ILD diagnosis was collected for analysis.

Clinical variables included comorbidities such as hypertension, diabetes (with and without complications), malignancy, autoimmune disease, and chronic pulmonary conditions. All diagnoses were identified using the International Classification of Diseases Tenth Revision (ICD-10) codes (Supplementary Table 2). Comorbidities were quantified using the Charlson Comorbidity Index (CCI) and categorized into four levels: 0 (no comorbidities), 1–2 (low), 3–4 (moderate), and ≥ 5 (high)¹⁶.

Statistical analysis

Descriptive statistics were used to summarize the data. Categorical variables were presented as frequencies and percentages, and continuous variables were summarized using means and standard deviations. The primary analysis focused on estimating the prevalence of drug-induced interstitial lung disease (DI-ILD) within the overall database population. We calculated the prevalence based on the different case definitions, and included a subgroup analysis for potential DI-ILD accounting for autoimmune disease history. Prevalence estimates were expressed both as percentages and as rates per 100,000 individuals with 95% confidence intervals. All statistical analyses were conducted using RStudio (USA).

Ethical considerations

Ethical approval for this study was obtained from the SFDA Institutional Review Board (Ref. No. 2025_011). All methods were performed in accordance with the relevant guidelines and regulations, including the principles of the Declaration of Helsinki. Given the retrospective, de-identified nature of this study utilizing existing electronic health record data, the Institutional Review Board (IRB) of SFDA granted a waiver of informed consent. The study complied with ethical standards and was approved by the SFDA Institutional Review Board. Patient confidentiality was maintained through the use of de-identified data, and all analyses adhered to relevant data protection regulations.

Patient and public involvement

Patients were not involved in the design, conduct, reporting, or dissemination plans of this research, as it used secondary routinely collected health data.

Results

Prevalence of drug-induced interstitial lung disease (DI-ILD)

Among a total population of 2,858,141 individuals, 1,907 patients were diagnosed with interstitial lung disease (ILD), corresponding to a prevalence of 66.7 per 100,000 population (Table 1). Only 36 patients had a clinical diagnosis of drug-induced interstitial lung disease (DI-ILD) based on ICD-10 codes J70.2, J70.3, or J70.4, resulting in a prevalence of 1.26 per 100,000 population.

Table 1.

Prevalence of DI-ILD in the total population (n = 2,858,141).

Population Group	Subgroups	Cases	Prevalence (per 100,000)	95% CI (per 100,000)
ILD overall*		1,907	66.7	63.79–69.78
Coded DI-ILD **		36	1.26	0.91–1.74
Potential DI-ILD ***	All patients	819	28.65	26.76–30.69
	With autoimmune disease****	130	4.55	3.82–5.36
	Without autoimmune disease	689	24.1	22.34–25.98

When ILD patients with exposure to at least one drug associated with pulmonary toxicity within the six months preceding diagnosis were included, 819 patients met criteria for potential DI-ILD, corresponding to a prevalence of 28.65 per 100,000 population (0.0286%) (Table 1). Among these potential DI-ILD cases, 130 patients (15.9%) had a documented autoimmune or connective tissue disease, highlighting the potential contribution of underlying autoimmune conditions to observed drug exposure patterns. A subgroup analysis of the prevalence stratified by autoimmune disease status is presented in Table 1.

Demographic characteristics

The majority of ILD patients were female, comprising 53.6% of the overall ILD cohort and 62.5% of the potential DI-ILD subgroup. The mean age at the time of ILD diagnosis was similar between groups, with patients averaging 62.3 years (SD 17.1) in the ILD group and 61.1 years (SD 15.4) in the DI-ILD group. Most ILD diagnoses were made in outpatient settings (56% of ILD and 56.9% of DI-ILD), with fewer identified in emergency or inpatient settings. Table 2 presented a summary analysis for ILD and DI-ILD patients’ characteristics.

Table 2.

Characteristics of study population.

Variables	Level	ILD	Potential DI-ILD	Coded DI-ILD
n		1907	819	36
Patient Demographic
Sex	F	1023 (53.6%)	512 (62.5%)	23 (63.9%)
	M	884 (46.4%)	307 (37.5%)	13 (36.1%)
Age at ILD Diagnosis (mean ± SD)		62.33 (17.13)	61.1 (15.4)	50.69 (27.18)
Diagnosis Setting	ER	238 (12.5%)	89 (10.9%)	1 (2.8%)
	Inpatient	168 (8.8%)	75 (9.2%)	15 (41.66%)
	Outpatient	1067 (56%)	466 (56.9%)	20 (55.55%)
	NA	83 (4.4%)	10 (1.2%)	0
Death		129 (6.8%)	66 (7.9%)	1 (2.8%)
CCI Category (%)	0	629 (34.5%)	218 (26.6%)	4 (11.1%)
	1–2	701 (38.4%)	324 (39.6%)	19 (52.8%)
	3–4	342 (18.8%)	174 (21.2%)	5 (13.9%)
	≥ 5	152 (8.3%)	93 (11.4%)	8 (22.2%)
	NA	83 (4.4%)	10 (1.2%)	0
CCI Score (mean ± SD)		1.71 (1.99)	2 (2.1)	2.78 (2.36)
ILD Diagnosis
	DI-ILD Diagnosed	36 (1.9%)	36 (4.4%)	36 (100%)
	ILD Only Diagnosed	1907 (100%)	819 (100%)	36 (100%)

Abbreviations: CCI: Charlson Comorbidity Index, SD: standard deviation, ILD: Interstitial Lung Disease, DI-ILD: Drug-Induced Interstitial Lung Disease.

Clinical and comorbidity profiles

Comorbidities were frequent, with a slightly higher burden in the DI-ILD group (mean CCI: 2.0 vs. 1.71). A greater proportion of DI-ILD patients had a CCI ≥ 5 (11.4% vs. 8.3%). Mortality was also slightly higher among DI-ILD patients (7.9% vs. 6.8%) (Table 2).

History of drug exposure

Among patients diagnosed with interstitial lung disease (ILD), 42.9% had received at least one drug with known or suspected pulmonary toxicity within 6 months prior to diagnosis with a notable subset exposed to multiple agents (Table 3). Exposure was particularly common within the immunosuppressant class, including mycophenolate (14.1%), azathioprine (8.4%), and tacrolimus (3.1%). Notable exposure was also observed for antiarrhythmics such as hydralazine (12.2%) and amiodarone (6.5%). Other reported exposures included DMARDs (e.g., rituximab 3.2%, etanercept 1.8%) and antimicrobials (e.g., nitrofurantoin 5.7%, isoniazid 2.2%) (Supplementary Fig. 1). Data on drug exposure in patients with diagnosed DI-ILD and potential DIILD are presented in Supplementary Table 3.

Table 3.

Frequency of Exposure to DI-ILD–Associated Drugs among ILD and Potential DI-ILD Patients (within 6 months prior to diagnosis).

Associated Drug Category	ILD	Potential DI-ILD
Any Associated Drug	819 (42.9%)	819 (100%)
≥ 2 Associated Drugs	220 (11.5%)	220 (26.86%)
≥ 3 Associated Drugs	106 (5.6%)	106 (12.94%)

Abbreviations: ILD, interstitial lung disease; DI-ILD, drug-induced interstitial lung disease.

Because several commonly observed drugs—particularly immunosuppressants—are frequently prescribed for autoimmune diseases that can independently cause ILD, a subgroup analysis stratified by autoimmune disease status was performed (Supplementary Table 3). In patients without autoimmune disease, the most frequent exposures remained mycophenolate, hydralazine, azathioprine, amiodarone, and nitrofurantoin, whereas patients with autoimmune disease showed higher relative exposure to methotrexate, rituximab, etanercept, and azathioprine, reflecting treatment of the underlying condition.

Discussion

Prevalence of DI-ILD

This study estimates the prevalence of drug-induced interstitial lung disease (DI-ILD) in a large Saudi Arabian cohort using routinely collected real-world health data. A total of 1,907 patients were diagnosed with interstitial lung disease (ILD), corresponding to a prevalence of 66.7 per 100,000 population. Only 36 patients (1.9% of the ILD population) had a clinical diagnosis of DI-ILD, a proportion consistent with international estimates suggesting that DI-ILD accounts for approximately 3–7% of ILD cases^1,10.

When the definition was expanded to include ILD patients who had been prescribed at least one drug known to be associated with DI-ILD, the estimated prevalence increased to 28.6 per 100,000 population. This contrast underscores a discrepancy in the clinical recognition and coding of DI-ILD across healthcare settings¹¹. A systematic review attempted to estimate DI-ILD rate, reporting rates that varied between 4.1 and 12.4 cases per million per year³. This study highlights the challenges of identifying drug-induced disease using routine data alone. Notably, exposure-based prevalence estimates may be confounded by underlying autoimmune diseases, which should be considered when designing observational studies using real-world data. Consistent with prior literature, accurately estimating the true prevalence of DI-ILD remains challenging due to multiple drug exposures, overlapping clinical and radiological features with other ILD subtypes, and the absence of standardized diagnostic criteria³.

Exposure to drugs associated with DI-ILD

The notable difference between the number of patients formally diagnosed with DI-ILD and those exposed to associated drugs may reflect both the complexity of identifying drug-induced cases in clinical practice, though further research is needed to confirm causality and clarify diagnostic patterns. Among patients diagnosed with ILD, a significant proportion (42.9%) had been exposed to at least one drug known to be associated with DI-ILD. Furthermore, 12.9% of these patients were exposed to three or more associated drugs, while 26.9% had exposure to two drugs, raising the risk of pulmonary toxicity. This finding is consistent with international studies, which estimate the DI-ILD prevalence as a rare condition. However, prevalence estimates may vary depending on healthcare system awareness and diagnostic practices³. For instance, studies from Japan and Europe have identified a higher frequency of DI-ILD, possibly due to more systematic drug surveillance or broader access to diagnostic modalities^3,11.

The significant associated drug exposure percentage combined without diagnostic confirmation underscores the diagnostic complexity of DI-ILD⁹. Clinically, DI-ILD remains a diagnosis of exclusion, as it lacks distinct clinical, radiological, or histopathological features that can clearly differentiate it from other causes of ILD^17,18. A thorough assessment begins with a detailed medical history, documenting drug use including prescription and over-the-counter drugs as well as occupational or environmental exposures⁸. Consequently, true DI-ILD cases may be overlooked or misattributed, particularly in settings lacking comprehensive diagnostic protocols or multidisciplinary healthcare teams^17,18. This highlights a critical issue in pharmacoepidemiological research “misclassification bias”; where non–drug-related ILD cases may be mistakenly attributed to drug exposure or vice versa. Such misclassification can significantly affect study outcomes, leading to underestimation or overestimation of drug risk^17,18. Therefore, careful attention to case definitions, exposure windows, and diagnostic verification is essential when designing pharmacoepidemiological studies to ensure valid assessment of drug safety signals and to minimize the risk of erroneous associations.

Demographics and risk profile

The demographic profile of patients, predominantly older adults with a slight female predominance and a moderate comorbidity burden, aligns with known risk factors for DI-ILD^17–19. Notable comorbidities among the DI-ILD group included hypertension, diabetes mellitus, congestive heart failure, and connective tissue disease—conditions that often require long-term pharmacologic treatment, thereby increasing cumulative drug exposure and risk for DI-ILD⁹. Mortality was marginally higher in the DI-ILD subgroup, suggesting a possible association with worse outcomes, although causality cannot be definitively established due to the observational nature of the study¹⁰.

Common associated drugs

Many ILD patients had recent exposure to drugs linked with DI-ILD, particularly immunosuppressants reflecting both the underlying disease burden and potential for drug-induced toxicity^20–23. Antiarrhythmics like hydralazine (12.2%) and amiodarone (6.5%), along with antineoplastic agents such as methotrexate (4.3%) and cyclophosphamide (1.5%), were also among the more commonly prescribed agents with known pulmonary toxicity^24–26. Although biologic DMARDs like rituximab (3.2%) and etanercept (1.8%) were less frequently observed, their presence highlights the ongoing diagnostic challenge of distinguishing drug-induced from disease-related ILD, particularly in patients with rheumatologic conditions^9,27–29. Traditional chemotherapeutics such as bleomycin, busulfan, and carmustine were rarely reported (< 0.5%), possibly reflecting modern shifts away from these high-toxicity regimens in oncology^30–32.

A substantial proportion of immunosuppressive drug exposure occurred in patients with underlying autoimmune diseases, which are themselves well-established causes of ILD, thereby limiting potential causal attribution between drug exposure and lung injury in this subgroup³³. Subgroup analyses excluding patients with autoimmune disease showed that several commonly implicated immunosuppressants remained frequent exposures, although their relative contribution was attenuated, highlighting the complexity of separating disease-related and drug-related factors (Supplementary Table 3).

While agents like amiodarone, nitrofurantoin, and bleomycin have well-established pulmonary toxicity, some immunosuppressants carry dual roles used to treat ILD but also implicated in rare cases of drug-induced lung injury. This duality complicates causality assessment in DI-ILD and highlights the importance of carefully evaluating the timing and context of drug exposure^10,24. Although causal relationships cannot be established from prescription data alone, these findings underscore the need for systematic temporal drug history review in ILD patients^9,17.

Clinical and regulatory implications

These findings underscore the need for clinicians to remain vigilant for DI-ILD, investigating any new respiratory symptoms after starting therapy and to adopt frequent assessment (e.g., checklists for high-risk drugs, scheduled pulmonary monitoring, and multidisciplinary case review) to improve detection and management⁹. Regulators, notably the Saudi Food and Drug Authority (SFDA), can support these efforts by requiring clear DI-ILD warnings on high-risk drug labels, mandating periodic safety updates, integrating pharmacovigilance alerts into electronic health records, and establishing a national DI-ILD registry for real-time surveillance. Educational campaigns and guideline development led by the SFDA could further enhance awareness among prescribers, ensuring early recognition and intervention to reduce preventable lung injury.

Strengths and limitations

This study’s major strength lies in its use of a large, real-world, national dataset that captures prescribing patterns and diagnostic trends across multiple healthcare settings in Saudi Arabia. It is one of the first studies to characterize the epidemiology of drug-induced interstitial lung disease (DI-ILD) in this region, utilizing data from the Saudi Real-world Evidence Network (SRWEN) encompassing over 2.8 million individuals.

However, these findings should be interpreted in light of several limitations. Reliance on ICD-10 diagnostic codes may result in misclassification or underestimation of DI-ILD cases, particularly where clinical documentation is incomplete or ambiguous. In addition, the cross-sectional design and lack of detailed clinical data (including imaging findings, symptom onset, and physician adjudication) prevent establishment of a clear temporal relationship between drug exposure and ILD diagnosis, limiting causal inference. Although drug exposure is a necessary precondition for DI-ILD, not all exposed individuals will develop the condition; therefore, exposure-based prevalence estimates should be interpreted as indicators of potential risk rather than confirmed diagnoses.

Importantly, a substantial proportion of drug exposure, particularly immunosuppressive therapy, occurred in patients with underlying autoimmune diseases, which are themselves can cause of ILD. This introduces confounding by indication and further complicates attribution of lung injury to medication exposure alone, even when subgroup analyses excluding autoimmune disease were performed. Moreover, several drugs identified in this study have dual roles, being used therapeutically in ILD while also rarely implicated in drug-induced lung injury, necessitating cautious interpretation of exposure-based associations.

Finally, the use of routinely collected health records may introduce documentation errors, incomplete capture of medication exposure, and other biases inherent to secondary data analysis. Collectively, these factors may lead to both under- and over-estimation of DI-ILD prevalence and drug associations. In addition, because this study reflects Saudi Arabia’s healthcare infrastructure, prescribing practices, and coding systems, the generalizability of these findings to other settings may be limited.

Future directions

Future research should focus on prospective cohort studies with longitudinal follow-up to better establish causality between specific drug exposures and the development of DI-ILD. Integration of clinical, radiographic, and biomarker data may help distinguish DI-ILD from other ILD subtypes. Development and validation of clinical scoring tools or diagnostic algorithms may aid clinicians in identifying suspected DI-ILD cases more reliably. Leveraging electronic health records and pharmacovigilance databases could also facilitate the creation of predictive models to stratify patients by DI-ILD risk based on drug combinations, dosing patterns, and comorbid profiles. Moreover, studies exploring immune pathways, genetic predispositions, and sex-related differences may uncover biologic mechanisms that explain differential susceptibility among patients.

Conclusions

This study provides a real-world assessment of drug-induced interstitial lung disease in Saudi Arabia, demonstrating that although clinically coded DI-ILD represents a small proportion of diagnosed ILD cases, exposure to drugs associated with pulmonary toxicity is common. These findings highlight the need for enhanced diagnostic awareness, systematic documentation of medication histories, and careful clinical evaluation of drug exposure when assessing patients with ILD. Given the diagnostic challenges and potential for misclassification, exposure-based estimates should be interpreted cautiously and viewed as indicators of potential risk rather than confirmed causality. Strengthening pharmacovigilance efforts, multidisciplinary collaboration, and regionally tailored clinical and regulatory guidance may support earlier recognition and more informed management of suspected DI-ILD in routine practice.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

AMA: conception of the work, methods, analysis and interpretation of data, and manuscript writing. AAA: conception of the work, methods, and interpretation of data. MAA, NSA and ASA: contributed to study conception, data acquisition. All authors have made substantial contributions to editing and revising the manuscript.

Data availability

The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

The study was approved by Saudi Food and Drug Authority (SFDA) Institutional Review Board (Ref. No 2025_011). Very minimal risk as no intervention is involved in the study design. Subjects’ privacy and confidentiality were assured, no identifiers were collected.