One hundred sixty-two patients volunteered to participate in the study and completed at least 1 psychiatric assessment during PEG IFN/ribavirin treatment. Demographic and clinical data for the study sample are presented in . Men comprised 58% of the sample, and 88% of patients were non-Hispanic white. Mean age was 45.0 years (SD = 6.7; range, 18–70 years). Consistent with patterns of HCV infection in the United States,3
77% of patients were infected with genotype 1 viral strain. Seventy-six patients (47%) were randomly assigned to receive standard ribavirin dosing (800 mg/day), and 86 patients (53%) received weight-based dosing (800–1400 mg/day). Thirty-one patients (19%) endorsed a history of at least 1 prior episode of major depressive disorder, and 90 (56%) were positive for a history of substance abuse (substance abuse disorder in remission). At baseline, 2 patients met criteria for current major depressive disorder, and no patients met criteria for current drug or alcohol abuse. Due to the lack of patients with active depression or substance abuse, neither condition was included as a predictive factor in the statistical analyses.
Characteristics of 162 Study Participants
The mean SDS index score for the study population at baseline was 41.9 (SD = 9.9). This score is similar to the mean score reported for the U.S. population as a whole34
and is consistent with some,18,27,35,36
but not all,26,37,38
previous reports of depressive symptoms in patients with HCV not receiving IFN-alpha/ribavirin therapy. Of note, patients with a history of depression exhibited significantly higher mean baseline SDS index scores compared with patients without a depression history (48.7 [SD = 10.9] vs. 40.3 [SD = 9.0], t = 4.5, p < .0001). Finally, 47 patients (29%) were receiving an antidepressant at baseline, and an additional 36 patients (22%) started taking an antidepressant during PEG IFN/ribavirin treatment.
Twenty-two patients discontinued the study prior to week 24. Eleven patients withdrew consent or were lost to follow-up. Six patients discontinued due to IFN-alpha–related behavioral symptoms (e.g., depression/anxiety/fatigue), and 5 patients discontinued due to medical complications. There were no statistically significant differences between patients who dropped out and study completers in terms of age, gender, baseline SDS score, history of major depressive disorder, or history of substance abuse (data not shown),
As shown in , there was a significant main effect of time on mean SDS index scores (F = 40.29, df = 4,161; p < .001). Multiple pairwise comparisons of specific time points during the study revealed that depression scores were significantly elevated over baseline by week 4 (week 4 vs. baseline, t = 9.46, df = 161, p < .001) and remained increased through the 24th week of PEG IFN/ribavirin treatment, at which time mean depression scores were the highest (week 24 vs. week 4, t = 3.33, df = 161, p < .01; week 24 vs. week 8, t = 2.26, df = 161, p < .05; week 24 vs. week 12, t = 3.54, df= 161, p< .001). Of note, the point prevalence of moderate to severe depression (SDS index score ≥ 60) also significantly increased over the course of PEG IFN/ribavirin therapy (Cochran-Armitage Test, z = 3.8, p < .001) (). Whereas II (6.8%) of 162 patients exhibited symptoms consistent with moderate to severe depression at baseline, 37 (26.4%) of the 140 patients remaining at week 24 had an SDS index score ≥ 60. Over the entire 24-week treatment period, 38.9% of patients exhibited moderate to severe depressive symptoms during at least 1 assessment. Prevalence rates for mild depressive symptoms, defined as an SDS index score ≥ 50, also increased significantly during treatment, from 24.7% at baseline to a maximum of 55.7% at 24 weeks (Cochran-Armitage Test, z = 4.2, p < .001). Seventy-two percent of patients had an SDS index score ≥ 50 during at least 1 assessment over the study period.
Depression Scores (A) and Percentage of Patients Experiencing Moderate to Severe Depressive Symptoms (B) Among Patients With HCV Receiving PEG IFN/Ribavirin Therapya
Based on previous results from our group and others, we examined the relationship between baseline depressive symptoms and the maximal depression score during PEG IFN plus ribavirin treatment. As shown in , baseline depression scores were highly predictive of the maximal depression score during PEG IFN plus ribavirin treatment (simple regression analysis; Y = 0.55X + 32.7, p < .0001 for slope; R = 0.54, p < .0001).
Relationship Between Baseline and Maximum Depression Scores During PEG IFN/Ribavirin Treatment in 162 Patients With HCVa
To evaluate the contribution of other relevant factors to the development of moderate to severe depressive symptoms (SDS index score ≥ 60), logistic regression analysis was performed in patients who completed all 24 weeks of the study (N = 140) (see Method). Given the impact of baseline depression scores on the development of depression during PEG IFN/ribavirin treatment, logistic regression was initially conducted controlling for baseline SDS index score. Backward stepwise elimination was used to remove nonsignificant factors from the model. Only ribavirin dose assignment significantly predicted the development of moderate to severe depressive symptoms when baseline SDS index score was controlled for Patients who received weight-based dosing of ribavirin (800–1400 mg/day) were significantly more likely to develop moderate to severe depressive symptoms compared with patients who received standard dosing (800 mg/day) (OR = 2.4, 95% CI = 1.1 to 5.4, p < .05). Consistent with this finding, the mean SDS index score across the 24-week treatment period was significantly higher in the group randomly assigned to weight-based ribavirin versus the group receiving standard dose therapy (51.5 [SD = 9.3] vs. 46.8 [SD = 8.2], respectively, t = 3.1, p < .01). Patients randomly assigned to weight-based ribavirin began the study taking more ribavirin (mean starting dose = 13.5 mg/kg) than patients randomly assigned to standard dose therapy (mean starting dose = 9.8 mg/kg). Of note, somatic symptoms on the SDS scale (including sleep difficulties, altered appetite, weight loss, constipation, palpitations, and fatigue) were not significantly increased in patients receiving weight-based versus standard dose ribavirin (data not shown).
Factors that did not predict the development of depression during PEG IFN plus ribavirin treatment when controlling for baseline SDS index score included gender, age, viral genotype, history of substance abuse, history of depression, antidepressant use, or the need for dose reductions of either PEG IFN or ribavirin. Of the 140 patients considered in the logistic regression analysis, 19.3% (N = 27) required reduction in the dosage of PEG IFN, and 21.4% (N = 30) required reduction in ribavirin for a period of at least 1 week. Although history of major depressive disorder was eliminated from the regression model described above, patients with a depression history exhibited a significantly higher rate of moderate to severe depressive symptoms (SDS index score ≥ 60) during PEG IFN/ribavirin treatment than patients with no history of depression (64.5% vs. 32.8%, respectively; χ2 = 10.6, df = 1, p < .005). As previously mentioned, baseline depression scores were also higher in patients with a depression history, indicating that the increased rate of moderate to severe depressive symptoms in patients with a history of depression occurred through an association with elevated baseline depression status. Indeed, when logistic regression analysis was repeated without controlling for baseline SDS index score (using backward stepwise elimination of nonsignificant factors as noted above), a history of major depressive disorder emerged as a significant predictor of moderate to severe depression during PEG IFN/ribavirin therapy (OR = 3.3, 95% CI = 1.3 to 8.1, p < .01). As in the previous regression analysis, assignment to weight-based ribavirin remained a significant predictive factor in the model (OR = 2.7, 95% CI = 1.3 to 5.6, p < .01).