Background Idiopathic Pulmonary Fibrosis (IPF) is a fatal respiratory illness with limited treatment options.
We previously estimated the UK prevalence of IPF using The Health Improvement Network (THIN) data. We have now further investigated this using General Practise Research Database (GPRD) data.
Aims Calculate the prevalence and incidence of IPF diagnoses in the GPRD in 2010.
Calculate the prevalence of IPF diagnoses in the GPRD in 2007, to compare to the THIN 2007 findings.
Build a matrix of epidemiological data, to better target future public health resources.
Methods Descriptive cross-sectional study, using UK GPRD data to calculate the prevalence and incidence of IPF diagnoses in 2010.
Patients with a Read/OXMIS code corresponding to IPF diagnosis were identified; H563.00 Idiopathic Fibrosing Alveolitis (IFA), H563z00 Idiopathic Fibrosing Alveolitis NOS (IFA NOS), H563.12 Cryptogenic Fibrosing Alveolitis (CFA), H563100 Diffuse Pulmonary Fibrosis (DPF). To investigate coding variability, IPF diagnoses were classified as broad (all codes), narrow (IFA/IFA NOS) and IFA+CFA
Patients with a first record of IPF during 2010 were classed as incident. Patients with a first record of IPF prior to or during 2010 were classed as prevalent.
Results In 2010 in the GPRD, IPF was most commonly diagnosed in males (56.5%) and ≥65 year olds (80.8%). Most IPF was coded as DPF (prevalence: 39.4 [37.6–41.3] per 100,000 persons; incidence: 9.6 [8.8–10.6] per 100,000 person-years).
For the broad definition, IPF prevalence was 50.7 (48.6–52.8) per 100,000 persons, with an incidence of 11.0 (10.1–12.0) per 100,000 person-years. For the narrow definition, IPF prevalence was 10.1 (9.2–11.1) per 100,000 persons, with an incidence of 1.5 (1.1–1.9) per 100,000 person-years. For the IFA+CFA definition, IPF prevalence was 13.3 (12.3–14.4) per 100,000 persons, with an incidence of 1.6 (1.3–2.0) per 100,000 person-years.
In 2007, compared to THIN, the GPRD IPF point prevalence was lower in both males and females, for all diagnosis definitions (Figure 1).
Conclusions IPF rates differ amongst UK healthcare databases; this may be due to GP coding variability, which is an inherent limitation of database studies. A prospective IPF registry is essential to characterise this orphan disease, to better target future public health resources.