Incidence and Prevalence of Non-Idiopathic Pulmonary Fibrosis (IPF) Progressive Fibrosing Interstitial Lung Disease (PF-ILD): A Systematic Literature Review and Meta-Analysis

OBJECTIVES: To synthesize the published evidence on the incidence and prevalence of non-IPF PF-ILD within adults and assess regional variations.

METHODS: MEDLINE®, Embase, and the Cochrane Database of Systematic Reviews were searched from 2000-01-01 to 2023-11-07 for all English-language studies reporting incidence or prevalence of non-IPF PF-ILD. A DerSimonian and Laird random-effects model was used to calculate pooled weighted-mean incidence and prevalence estimates from studies reporting sufficient data.

RESULTS: Of 3,823 abstracts, five studies were included in the meta-analysis (4 reported prevalence and incidence, 1 reported prevalence only). All studies were retrospective and used administrative databases. Study periods ranged from 2010–2019. Pooled global incidence across four studies was 10.9 per 100,000 persons (95% confidence interval [95%CI]=1.3–20.5; median=6.8). In Europe, pooled incidence across two studies was 6.7 per 100,000 (95%CI=2.2–11.3; median=6.8). Reported incidence rates were 2.4 and 27.6 per 100,000 in two studies from South Korea and the United States (US), respectively. Pooled global prevalence across five studies was 37.0 per 100,000 (95%CI=23.6–50.5; median=46.4). Within Europe, pooled prevalence was 30.2 per 100,000 from two studies (95%CI=0.0–61.9; median=30.2); reported prevalence estimates were 6.4 in a South Korean study, and 57.8 and 60.7 in two US studies.

CONCLUSIONS: This small evidence base showed higher incidence and prevalence in the US versus other countries at the population level. Non-IPF PF-ILD is associated with various risk factors and complex development; pooled global estimates suggest that it is a rare disorder. Wide variability in study estimates of non-IPF PF-ILD from the US versus other regions is likely explained by: (i) non-validated algorithms to identify PF-ILD from claims databases, (ii) various types of denominators, and (iii) possible inclusion of few IPF cases in one study. Further standardization in the definition of non-IPF PF-ILD is required for more accurate incidence and prevalence estimates.