Research from JAMA — Early vs Late Tracheotomy for Prevention of Pneumonia in Mechanically Ventilated Adult ICU Patients — A Randomized Controlled Trial

To compare percutaneous ethanol injection (PEI), the standard approach which has been used for many years to treat early non-surgical hepatocellular carcinoma (HCC) in cirrhotic patients, and radiofrequency ablation (RFA), which has become an interesting alternative. A randomized trial was carried out on 139 cirrhotic patients in Child-Pugh classes A/B with 1-3 nodes of HCC (diameter 15-30 mm), for a total of 177 lesions. Patients were randomized to receive RFA (n=70) or PEI (n=69). The primary end-point was complete response (CR) 1 year after the percutaneous ablation of all HCC nodes identified at baseline. Secondary end-points were: early (30-50 days) CR, complications, survival and costs. In an intention-to-treat analysis, 1-year CR was achieved in 46/70 (65.7%) and in 25/69 (36.2%) patients treated by RFA and PEI, respectively (p=0.0005). For lesions >20 mm in diameter, there was a larger CR rate in the RFA group (68.1% versus 26.3%). An early CR was obtained in 67/70 (95.7%) patients treated by RFA compared with 42/64 (65.6%) patients treated by PEI (p=0.0001). Complications occurred in 10 and 12 patients treated by RFA and PEI, respectively. The overall survival rate was not significantly different in the RFA versus PEI arm (adjusted hazard ratio=0.88, 95% CI: 0.50-1.53). There was an incremental health-care cost of 8286 euro for each additional patient successfully treated by RFA. The 1-year CR rate after percutaneous treatment of early HCC was significantly better with RFA than with PEI but did not provide a clear survival advantage in cirrhotic patients.

This article aims to describe the various approaches in multivariable modelling of healthcare costs data and to synthesize the respective criticisms as proposed in the literature. We present regression methods suitable for the analysis of healthcare costs and then apply them to an experimental setting in cardiovascular treatment (COSTAMI study) and an observational setting in diabetes hospital care. We show how methods can produce different results depending on the degree of matching between the underlying assumptions of each method and the specific characteristics of the healthcare problem. The matching of healthcare cost models to the analytic objectives and characteristics of the data available to a study requires caution. The study results and interpretation can be heavily dependent on the choice of model with a real risk of spurious results and conclusions.