纵向研究中控制时依混杂的G方法

传统分析方法不能有效地控制纵向研究中的时依混杂以得到无偏因果效应估计值。本研究解释了纵向研究中正确控制时依混杂的必要性,概述了现有控制时依混杂的G方法——参数g-formula、逆概率加权和G估计,并通过比较它们的优缺点和适用情况,为研究者在纵向研究中估计因果效应提供参考。     

Long-term clinical outcome and effect of glycyrrhizin in 1093 chronic hepatitis C patients with non-response or relapse to interferon

Objective. Patients with chronic hepatitis C who do not respond to interferon can be treated with glycyrrhizin to reduce disease activity. The objective of this study was to evaluate the effect of glycyrrhizin on the incidence of hepatocellular carcinoma (HCC) during long-term follow-up after non-response to interferon. Material and methods. We analyzed individual patient data of all consecutive patients treated with interferon in 12 major Japanese hospitals between 1990 and 1995 who showed no sustained response. Results. The study comprised 1093 patients. During a mean follow-up of 6.1±1.8 years, 107 patients developed HCC. The Cox regression analysis with time-dependent variables showed that older age, male gender, higher alanine aminotransferase (ALAT) and higher fibrosis stage were significantly associated with a higher risk of developing HCC. Response to glycyrrhizin, defined as ALAT?p<0.01). G-estimation, used to correct for ALAT as the confounder, showed no significant benefit of glycyrrhizin in the overall study population. Conclusions. This study provides some evidence to show that interferon non-responder patients with chronic hepatitis C and fibrosis stage 3 or 4 may have a reduced incidence of HCC if glycyrrhizin therapy leads to normalization of ALAT levels.