Health care utilization and costs associated with direct-acting antivirals for patients with substance use disorders and chronic hepatitis C

BACKGROUND: Patients with substance use disorders (SUD) and chronic hepatitis C virus infection (HCV) have limited access to direct-acting antivirals (DAAs) due to multilevel issues related to providers (eg, concern about reinfection); patients (eg, refusal); payers (eg, prior authorization); and health system structure, although clinical guidelines recommend timely DAA treatment for patients with SUD and HCV. Effects of DAAs on real-world health care utilization and costs among these patients is unknown. OBJECTIVE: To compare changes in medical service utilization and costs related to liver, SUD, and all-cause morbidity in patients with SUD and HCV treated with DAAs (DAA group) vs not treated with DAAs (non-DAA group). METHODS: We conducted a retrospective cohort study using MarketScan Commercial and Medicare Supplemental Claims databases (2012-2018) for newly diagnosed HCV treatment-naive adults with SUD. We used difference-in-differences analyses, stratified by cirrhosis status, to determine the adjusted ratio of rate ratio (RoRR) to assess the difference in the relative changes from the pre- to posttreatment periods between the 2 groups. RESULTS: 6,266 patients with SUD and HCV were identified. Of these patients who also had cirrhosis (n = 607), 49% (n = 298) initiated DAA therapy for HCV, whereas of those without cirrhosis (n = 5,659), 22% (n = 1,219) initiated DAAs. For patients with cirrhosis (n = 607), the liver-related costs decreased by $6,213 (95% CI = −$8,571, −$3,856) for the DAA group and $1,585 (95% CI = −$4,659, $1,490) for the non-DAA group. The relative decreases in the rate of liver-related costs were larger for the DAA group than for the non-DAA group, and the relative changes between groups were significantly different (RoRR = 0.37, 95% CI = 0.19-0.73). There was no difference in the relative changes after DAAs in the rate of SUD-related visits/costs or all-cause costs between the 2 groups. For patients without cirrhosis (n = 5,659), a similar association was observed. Besides, the relative decreases in the rate of SUD-related emergency department (ED) visits (RoRR = 0.54, 95% CI = 0.38-0.77); SUD-related long-term care visits (RoRR = 0.30, 95% CI = 0.13-0.73); all-cause ED visits (RoRR = 0.75, 95% CI = 0.64-0.88); and all-cause long term-care visits (RoRR = 0.36, 95% CI = 0.18-0.72) were larger in the DAA group than in the non-DAA group. CONCLUSIONS: DAAs are associated with a significant decrease in the rate of SUD-related ED visits and liver-related costs without increasing the rate of all-cause costs among patients with SUD and HCV, suggesting that the benefits of DAAs extended beyond liver-related outcomes, especially in this disadvantaged population.

**: We defined patients as having a clinical condition if they had one inpatient claim or two outpatient claims on separate dates within one year before the index date, using ICD-9-CM and ICD-10-CM ^: Pretreatment follow-up time was estimated from the date when patients were either first diagnosed with SUD or HCV, whichever came later, until the index date. If the first SUD diagnosis was after the index date, pretreatment follow-up time was estimated from the first HCV diagnosis until the index date.

Supplementary
^^: Posttreatment follow-up was estimated from the index date to the end of enrollment, or December 31, 2018. ‡ CVD includes cerebrovascular disease, coronary artery disease, and peripheral vascular disease. -515 (-1573, 542) HCV = hepatitis C virus; SUD = substance use disorders; DAA = all-oral direct-acting antivirals; ED = emergency department; SNF = skilled nursing facility; NA = not available; RoRR = ratio of rate ratio. *Results were derived from the stabilized inverse probability of treatment weights weighted difference-in-differences approach after controlling for age (18-35, 36-50, 51-64 and >64 years), sex at the index date; clinical conditions including the type of SUDs, diabetes, hypertension, dyslipidemia, cardiovascular disease, chronic kidney disease, chronic obstructive pulmonary disease, human immunodeficiency virus infection, schizophrenia/bipolar disorder, depression, pregnancy, and other liver diseases (i.e., alcoholic liver disease, nonalcoholic fatty liver disease, hepatitis B infection, hepatitis A infection) during the year before the index date; as well as Current SUD status, medication for opioid use disorder or alcohol use, and pretreatment follow uptime.

Supplementary
&: Due to the small sample size for the outcome, convergence was not achieved. *Results were derived from the stabilized inverse probability of treatment weights weighted difference-in-differences approach after controlling for age (  The excluded total pretreatment and posttreatment all-cause medical costs outliers were: costs>= 681842 in patients with SUD and chronic HCV without cirrhosis group (n =108); The maximum cost is $15376600 for one patient.

Mean Adjusted
Patients with newly diagnosed HCV from January 1, 2013 to December 31, 2018 (n = 211536) Excluded patients who are younger than 18 years old (n = 692) Excluded patients who were not continuously enrolled 12 months before and 6 months after the first HCV diagnosis (n = 169651) Excluded patients who did not have pharmacy benefit identified (n = 6030) Excluded patients who had decompensated cirrhosis (n = 801), hepatocellular carcinoma (n = 252) or liver transplant (n = 263) during baseline* Patients with HCV aged ≥18 years and continuously enrolled 12 months before and 6 months after the HCV diagnosis (n = 33978) Excluded patients who received HCV medication 12 months before the first HCV diagnosis (n = 480) Excluded patients had SUD 180 days after the first HCV diagnosis (n = 2176) Exclude patients who used interferon-based therapy after the date of newly diagnosis of HCV (n = 984) Excluded patients who had decompensated cirrhosis (n = 812), hepatocellular carcinoma (n=348) or liver transplant (n = 87) after the HCV diagnosis and before the DAA initiation* Excluded patients without a SUD (n = 22843) Treatment-naïve patients with SUD and HCV (n = 6385) Final cohort (n = 6266) & Excluded more than or equal to 99% percentile of the total pretreatment and posttreatment all-cause medical costs outliers (n = 119) ^ Excluded patients with stabilized IPTW more than (or equal to) 10 (n = 0) &: only 2 pretreatment total all-cause medical services costs were negative; 1 posttreatment total all-cause medical services cost was negative. We reset that patient's cost to zero. A.

Supplementary Figure 3. Adjusted Medical Service Costs Among the Stabilized IPTW Weighted Patients With SUD and Chronic HCV and Cirrhosis: (A) SUD-related Costs. (B) All-cause Costs.
Abbreviations: HCV = hepatitis C virus; SUD = substance use disorders; DAA = all-oral direct acting antivirals; RoRR = ratio of rate ratio Results were derived from the stabilized inverse probability of treatment weights weighted difference-in-differences approach after controlling for age (18-35, 36-50, 51-64 and >64 years), sex at the index date; clinical conditions including the type of SUD, diabetes, hypertension, dyslipidemia, cardiovascular disease, chronic kidney disease, chronic obstructive pulmonary disease, human immunodeficiency virus infection, schizophrenia/bipolar disorder, depression, pregnancy, and other liver diseases (i.e., alcoholic liver disease, nonalcoholic fatty liver disease, hepatitis B infection, hepatitis A infection) during the year before the index date; as well as current SUD status, medication for opioid use disorder or alcohol use, and pretreatment follow uptime. Costs were adjusted to 2018 U.S. dollars using an annual 3% inflation rate.