**3. Results**

#### *3.1. Patient Characteristics at Baseline*

We enrolled 56 patients who met the inclusion criteria: this included 29 subjects in the in-hospital group and 27 subjects in the out-of-hospital group (as shown in Figure 1). The baseline characteristics are shown in Table 1. The median age was 44.5 years and 92.9% of patients were male. All the subjects in the in-hospital group actively used drugs at enrollment, while only 44.4% of those in the out-of-hospital were PWID (*p* < 0.001). Eleven out of 27 patients referring to out-of-hospital service were experiencing homelessness, whereas only one patient (a 51 years-old female) within the in-hospital setting was homeless, at the time of study participation. All of the patients included in this study had a positive history of intravenous drug use (current or previus).

**Table 1.** Baseline patients' characteristics, represented for total population and sorted by in-hospital and out-of-hospital setting where chronic hepatitis C was managed.


1 assessed by transient elastography (FibroScan®), 2 variable described only for those patients with documented diagnosis of liver cirrhosis (*n* = 8). Abbreviations: BMI, body mass index; PWID, people who inject drugs; OST, opioid substitute therapy; IFN, interferon; RBV, ribavirin.

> An overall of 71.4% of individuals (40/56) used OST, with a lower percentage in the out-of-hospital setting rather than the comparison setting (*p* = 0.003). Psychiatric comorbidity was found in 26.8% (15/56) of patients; 58.8% (30/56) of subjects were infected with HCV genotype 1. Five out of fifty-six patients (8.9%) had F3 fibrosis according to Metavir score, while 15.7% (8/56) had documented liver cirrhosis: two out of these eight subjects with an advanced liver disease had decompensated cirrhosis (B8 Child-Pugh class). There was a statistically significant difference in creatinine values between the two groups, with higher levels among those who were treated in the standard in-hospital setting (*p* = 0.003). Thirteen patients (24,5%) were HCV-HIV coinfected: characteristics of this particular subset of patients are shown in Table 2.

**Figure 1.** Flow chart of study enrollment.



Abbreviation: ART, antiretroviral therapy; NRTI, nucleoside reverse transcriptase inhibitors; NNRTI, non- nucleoside reverse transcriptase inhibitors; INSTI, integrase strand transfer inhibitors; PI, protease inhibitors.

#### *3.2. Primary and Secondary Outcomes*

In our study population, 33 out of 56 patients started therapy with DAAs. The most used HCV regimen was Glecaprevir/Pibrentasvir (73% treated for 8 weeks, 9% for 12 weeks). The remaining patients received therapy with Sofosbuvir/Velpatasvir. All of the 33 patients who started DAAs (corresponding to 60% of the study population) completed treatment with DAAs, with no difference between groups. However, when we analyzed the rate of retention in care (defined as DAAs treatment start and completion, as described in Section 2) among the total study population (56 patients), we observed a higher rate of retention in care in the out-of-hospital group than in standard in-hospital setting (*p* = 0.001), Figure 2A. The expected waiting time was significantly longer in subjects referring to standard in-hospital services (*p* < 0.001), in comparison with the other group (Figure 2B). Among the 33 patients who were treated for Hepatitis C, 93.9% achieved SVR 12 (31/33), with similar SVR12 rates among the two groups (Table 3). The two patients (one in each of the two groups) did not achieve sustained virological response: one experienced a relapse after four weeks from the end of treatment (in-hospital group) and one was diagnosed with HCV reinfection over the follow-up (out-of-hospital group). At the univariate analysis, retention in care was associated only with the out-of-hospital managemen<sup>t</sup> (*p* = 0.002) and with a shorter expected waiting time (*p* = 0.003), as shown in Table 4. At the multivariate analysis, when we included the covariate "expected waiting time" in the model with

"out-of-hospital management" as an exposure variable, the out-of-hospital managemen<sup>t</sup> did not remain statistically significant as a predictor of retention in care (O.R. 099, *p* = 0.69), while the "expected waiting time" showed a definite trend for association with retention in care, although not still significant (O.R. 0.65, *p* = 0.08). This could potentially sugges<sup>t</sup> that our primary outcome (i.e., retention in care) might be driven by a shorter expected waiting time rather than the setting where patients were managed. When we analyzed the association of parameters with retention in care considering only the 41 patients who were actively using intravenous drugs at time of enrollment, we found that a greater retention in care rate was achieved among those treated out of the hospital (58%) than in the hospital (38%), although not statistically significant (*p* = 0.31). At the univariate analysis, we did not observe any variable associated with our primary outcome, although a shorter waiting time seemed to sugges<sup>t</sup> a higher chance to complete DAAs therapy (Exp (B) 0.995, CI 95% 0.99;1, *p* = 0.055).

**Figure 2.** Patients treated with DAAs in our population. Retention in care rates among patients treated for HCV in hospital and out of hospital Panel (**A**); expected days of waiting before DAAs treatment start in the standard in-hospital setting group and in the out-of-hospital setting group panel (**B**).



1 completion of DAAs treatment; 2 time between the first medical visit and the scheduled DAAs treatment initiation. Abbreviation: SVR12, sustained virological response 12 weeks after end of treatment.

> Overall, 37 patients accessed the established out-of-hospital service from January through June 2019 and were all screened for the study. All of them were past or current intravenous drug users or homeless persons. For the comparison group, we considered all the intravenous drug users with detectable HCV RNA who accessed traditional inhospital service for a visit from May 2017 through August 2018, and we screened a total of 38 patients.



1 time between the first medical visit and the scheduled DAAs treatment initiation. Abbreviations: BMI, body mass index; OST, opioid substitute therapy; IFN, interferon; RBV, ribavirin.
