Diagnosis of COPD in hospitalised patients

Abstract

Objective: To examine the quality of COPD diagnosis in hospitalised patients.

Material and methods: Retrospective multicentre cross-sectional audit review of the clinical histories of patients discharged with a diagnosis of COPD. The diagnosis of COPD was considered correct (DxC) in cases where the combination of a bronchial obstruction (FEV1/FVC<70%) and smoking (>10 packets/year) could be documented. In the rest of the cases the diagnosis was considered deficient (DxD). A DxC in at least 60% of patients was required to be considered an acceptable quality health care diagnosis. Demographic data such as, smoking, spirometry, the specialist who discharged the patient (P: Pneumologist; MS: Medical Specialty; CS: Surgical Specialty), and health care level (hospital complexity; low (H1), intermediate (H2) and high (H3)).

Results: A total of 840 cases were analysed (718 males, 122 females); mean age (SD) 73 (10), from 10 hospitals (3 H1, 4 H2, 3 H3). A DxD was obtained in 597 (71.1%), due to either lack of spirometry (538, 64%) or smoking criteria (319, 38%), (P<0.001). Only two of the ten hospitals complied with the criteria of an acceptable quality health care diagnosis. Significant differences (P<0.0001) were seen on comparing DxC and DxD by health care level (DxC: 56.2% in H1, 29.9% in H2, 20.9% in H3), and by specialist (DxC: 47.6% en P, 24.6% in SP, 17.4% in MS). A multivariate analysis associated DxC with the male sex, H1 and pneumology reports.

Conclusions: 1. The quality health care for the diagnosis of COPD is deficient. 2. The lack of spirometry is the most common cause of DxD.