A cross-sectional study of the clinical characteristics of hospitalized children with community-acquired pneumonia in eight eastern cities in China

BACKGROUND: Community-acquired pneumonia in children is common in China. To understand current clinical characteristics and practice, we conducted a cross-sectional study to analyze quality of care on childhood pneumonia in eight eastern cities in China. METHODS: Consecutive hospital records between January 1, 2010 and December 31, 2010 were collected from 13 traditional Chinese medicine (TCM) and western medicine (WM) hospitals in February, May, August, and November (25 cases per season, 100 cases over the year), respectively. A predesigned case report form was used to extract data from the hospital medical records. RESULTS: A total of 1298 cases were collected and analyzed. Symptoms and signs upon admission at TCM and WM hospitals were cough (99.3% vs. 98.6%), rales (84.8% vs. 75.0%), phlegm (83.3% vs. 49.1%), and fever (74.9% vs. 84.0%) in frequency. Patients admitted to WM hospitals had symptoms and signs for a longer period prior to admission than patients admitted to TCM hospitals. Testing to identify etiologic agents was performed in 1140 cases (88.4%). Intravenous antibiotics were administered in 99.3% (595/598) of cases in TCM hospitals and in 98.6% (699/700) of cases in WM hospitals. Besides, Chinese herbal extract injection was used more frequently in TCM hospitals (491 cases, 82.1%) than in WM hospitals (212 cases, 30.3%) (p < 0.01). At discharge, 818 cases (63.0%) were clinically cured, with a significant difference between the cure rates in TCM (87.6%) and WM hospitals (42.0%) (OR = 9.8, 95% confidence interval (CI): 7.3 ~ 12.9, p < 0.01). Pathogen and previous medical history were more likely associated with the disappearance of rales (OR = 7.2, 95% CI: 4.8 ~ 10.9). Adverse effects were not reported from the medical records. CONCLUSIONS: Intravenous use of antibiotics is highly prevalent in children with community-acquired pneumonia regardless of aetiology. There was difference between TCM and WM hospitals with regard to symptom profile and the use of antibiotics. Intravenous use of herbal injection was higher in TCM hospitals than in WM hospitals. Most of the cases were diagnosed based on clinical signs and symptoms without sufficient confirmation of aetiology. Audit of current practice is urgently needed to improve care.