Several databases of epidemiologic studies in patients with idiopathic pulmonary fibrosis (IPF) have been analyzed in the Western community. However, few studies have been reported in Asia. The objective of this study was to investigate the epidemiology of IPF in Taiwan. We collected and analyzed patients with IPF from the Taiwan National Health Insurance Research Database from 2001 to 2011. We estimated the annual incidence and cumulative prevalence of IPF and mean survival time of patients and determined the causes of death. The annual incidence rates of IPF remained stable after 2005, ranging from 0.7 to 1.3 cases per 100,000 people per year, whereas the cumulative prevalence rates increased steadily from 3.1 to 6.4 cases per 100,000 people per year during 2006–2011 based on a narrow case definition. Men older than 75 years had higher incidence compared with other age groups. The mean survival after diagnosis was 6.9 years. Old age, male sex, and respiratory hospitalization were associated with shorter survival time after diagnosis. Both the incidence and prevalence rates of IPF were lower in Taiwanese patients than Western ones. Moreover, the survival time was higher in the Asian population compared with the Western population. These results may suggest the heterogeneity of the IPF definition in different study populations and geographic locations.

Background: The prognostic factors for hospital mortality in patients with idiopathic pulmonary fibrosis (IPF) have not been characterised well. We aimed to determine the impact of respiratory hospitalisation on patient outcomes and the prognostic predictors for hospital mortality.Methods: Data from patients with IPF was collected and analysed from two cohorts, TaiwanNational Health Insurance Research Database (NHIRD) between 2001 and 2011 and a national medical centre (NMC)between 2001and 2015. Results: 122 patients in NHIRD cohort were recruited. The median survival was lower in patients with respiratory hospitalisation (25/122, 29.8%) compared with non-respiratory hospitalisation(59/122, 48.4%) in NHIRD cohort (2.3 years vs non-respiratory 9 years, p=0.007). In NMC cohort (N=47), patients with acute respiratory deterioration (ARD) hospitalisation (30/47,63.8%) compared with non-ARD hospitalisation (17/47, 36.2%) had poor survival (0.9 monthsvs 29.6 months,p<0.001). The differences in survival between two cohorts were associated with age, causes of hospitalisation, mechanical ventilation and comorbidities. The patients hospitalised for ARD had statistically higher GAP score, lower oxygenation, higher C-reactive protein levels, higher neutrophil counts and lower serum albumin level compared to those non-ARD hospitalisation. For both in-hospital and 1 year mortality, multivariate analysis adjusting for patient characteristics confirmed the independent predictive factors were steroid pulse therapy, acute exacerbation and mechanical ventilation. Conclusions: The respiratory hospitalisation negatively impacted patient outcomes, and the independent prognostic factors may be utilised to refine the management in patients with IPF.