Health Related Quality of Life in Patients with Chronic Diseases

Abstract

Background and Aim: Assessment of health-related quality of life (HRQL) is important for patients with chronic liver diseases (CLD). Our aim was to explore the impact and predictors of HRQL in CLD. Methods: Two prospectively collected HRQL databases with Short-Form 36 (SF-36) data (The US Cohort and the Italian Cohort) were used for this study. Scores for each of SF-36 scales (PF - physical functioning, RP - role functioning, BP - bodily pain, GH - general health, VT - vitality, SF - social functioning, RE - role emotional and MH -mental health, MCS- mental component score, PCS- physical component score) were compared between different types of CLD and other clinical variables. SF-36 scales were compared using Wilcoxon tests and Spearman's rank coefficient was used as a measure of age-HRQL correlations. For all tests and correlation estimations, p-values not exceeding the 0.05 threshold of significance after Benjamini-Hochberg correction were considered to be statistically significant. Regression models for SF-36 scales and summary scores were generated by stepwise (bi-directional) selection. The median difference between actual and predicted SF-36 scale scores was measured. Optimistic estimation, using the same cohort was used for training and testing (assuming a representative sample), followed by ten-fold cross-validation (C-V). All analyses were performed using S-Plus software. Results: Data was available for 1103 patients with chronic liver diseases. Clinical data included: age 54.2±12.0 years, 40% female, 69% with cirrhosis. Analysis revealed that age correlated (p<0.05) with worsening HRQL on every scale of the SF-36. Female CLD patients had more HRQL impairments in PF, RP, BP, GH, VT and MH scales of SF-36 (Δ scale score: 6.6-10.7, p<0.05). Furthermore, cirrhotic patients had more impairment of HRQL in every scale of SF-36 (Δ scale score: 6.6-43.0, p<0.05). In terms of diagnostic groups, non-alcoholic fatty liver disease (NAFLD) patients showed consistently more HRQL impairment than other liver diseases. In developing models predicting different HRQL scores, PCS scores was predicted by female gender, increasing age, diagnosis of alcoholic liver disease or NAFLD, presence of cirrhosis and being of Italian cohort (p-value = 1.9E-45). Model predicting MCS scores included, female gender, carrying the diagnosis of autoimmune hepatitis, cholestatic liver disease or NAFLD as well as presence of cirrhosis and being of Italian cohort (p-value