The high comorbidity burden of the hepatitis C virus infected population in the United States

Hepatitis C virus (HCV) is the etiology of the most common chronic bloodborne infection in the United States (US), affecting at least 4.1 million people, of whom 3.2 million are chronically infected [1]. Chronic infection can lead to progressive development of liver fibrosis to cirrhosis, end-stage liver disease, and hepatocellular carcinoma (HCC) in 5-20% of patients within the first two decades of initial infection [2]. Complications of chronic HCV are projected to increase dramatically over the next few decades [3], placing an ever greater burden on health systems of a disease where the annual health care costs are already estimated at over one billion dollars [4]. Known co-factors for the progression of liver disease include duration of infection, male gender, high-levels of alcohol consumption, hepatitis B virus (HBV) infection, and human immunodeficiency virus (HIV) infection. Other data show that metabolic factors such as steatosis, being overweight or obese may also contribute to progression [5]. Several of these factors are comorbid conditions that may also impact HCV treatment eligibility and outcomes. Other conditions that are known contraindications for HCV treatment are severe depressive disorders, hyperthyroidism, hypertension, heart failure, coronary heart disease, diabetes, and pulmonary disease [2].

The current standard-of-care for chronic HCV treatment is combination pegylated-interferon and ribavirin with the addition of boceprevir or teleprevir for genotype 1 patients only [6]. Among patients on pegylated-interferon and ribavirin only, treatment is only efficacious in approximately 55% of treatment-naïve patients [7]. About 75% of treated patients experience one or more side effects requiring discontinuation or dose modification of treatment [8]. Specifically, antiviral treatment may exacerbate certain conditions such as those with anaemia, thrombocytopenia or leukopenia, which can further complicate treatment outcomes by increasing the number and severity of side-effects [9]. Consequently, many patients are excluded from treatment, or have treatment deferred or compromised because of their comorbid conditions.

To maximize the benefit of current and future HCV therapeutic regimens, a greater understanding of the most frequent comorbidities associated with HCV is needed. Successfully managing these comorbidities may lead to increased treatment eligibility and improved outcomes. No studies, to our knowledge have systematically profiled comorbidities among a general HCV population. The objectives of this study were: (1) to systematically estimate the prevalence of the top 25 comorbidities in an HCV population; (2) to compare whether these most common HCV comorbidities differ between receiving antiviral treatment during the study period; and (3) to compare whether these most common HCV comorbidities differ between an HCV infected and uninfected control population.

To our knowledge, this is the first study to systematically profile non-selected comorbidities in an HCV population. Nearly every HCV patient in our study had at least one comorbidity, and in general, these comorbidities were more common in the overall HCV infected case patients compared to the uninfected controls. In addition, comorbidities covered a wide range of diseases and symptoms affecting a number of body systems. Our findings substantiate that HCV is a systemic disease with many extra-hepatic features which highlight the urgency to treat these extra-hepatic conditions regardless of the liver disease itself. Individualized anti-viral treatment decisions for HCV infection should weigh the potential risks and benefits of both hepatic and extra-hepatic manifestations to increase the likelihood of treatment success.

Our insured study population reported a 47% current treatment rate, which is comparable to treatment rates (10-51%) in other US medical claims database studies [20‐22]. There are multiple reasons why HCV patients may not be treated which include disease stage, patient choice and lack of access to liver specialists [23]. Non-liver specialists may decide that the patient is ineligible for treatment based on a list of contraindications, whereas liver specialists may otherwise safely treat the patient with relative contraindications (i.e. patients with cirrhosis and moderate thrombocytopenia) since most treatment regimens are individualized to achieve maximal sustained virological response. Treatment rates appear to be declining in the US, and it is estimated that from 2002-2030, only 14.5% of liver-related deaths caused by HCV will be prevented [24]; therefore, it is urgent to understand what barriers may exist that are preventing treatment uptake as the burden of HCV morbidity and mortality increases.

Up to one-third of our HCV population had a comorbidity that could potentially make them ineligible for antiviral HCV treatment. This indicates that a high proportion of the HCV population could potentially receive effective treatment if their comorbid conditions could be managed. A recent review has reported that the benefits of treatment can potentially reach a wider HCV population by including targeted groups such as active drug users and persons those with psychiatric illnesses [25]. These patients were able to complete antiviral therapy, achieve a sustained virological response, and did not develop adverse events that would require discontinuation of treatment. However, it was recommended that individualized management of these patients could be integrated within addiction and treatment services [26]. These studies suggest that some comorbidities can be effectively managed, increasing access to HCV treatment with successful outcomes.

Our findings confirm previous studies as we consistently observed the known comorbidities associated with HCV antiviral treatment and disease among our top ranked list, highlighting that our systematic evaluation supports how common these conditions are among the HCV infected population. However, three conditions, benign neoplasms, genitourinary symptoms, and viral infections, were identified in our ranked list and have not been previously reported to be associated with treatment or disease progression. The prevalence of benign neoplasms were mainly of the large bowel and skin (trunk and unspecified sites), which could possibly be explained by metastasis to these sites among HCV patients with HCC but its pathophysiology is unclear and warrants further study [27, 15]. In our study, the prevalence of HCC was low, but we cannot exclude the possibility that this may be an underestimation if patients were diagnosed prior to the study period. HCV patients with genitourinary symptoms were mainly of hematuria and proteinuria [28] which could be clinical manifestations of glomerular kidney disease that is known to be associated with cryoglobulinemia and membranoproliferative glomerulonephritis [29]. For the unexpected finding of viral infections, the significance is difficult to interpret given the lack of specificity as almost half the cases were coded as non-specific viral infections.

Our HCV infected population was more likely to report comorbidities than the uninfected controls. This may be explained in part by the higher-risk or unhealthy behavior (e.g., alcohol and drug users) of those infected with HCV, making them more vulnerable to acquire new morbidities than the uninfected population. Indeed, prevalence of alcohol use and drug-use were much higher among the HCV infected compared to the uninfected population (alcohol; 7% vs. < 1%; drug use 15% vs. 3%, respectively). In addition, HCV-infected patients with health insurance may have closer follow-up care and higher rates of screening for other diseases than the uninfected, hence increasing the reporting of conditions. HCV-infected patients may also have more comorbidities because of the illness itself, which may place them at higher risk for other conditions. However, we identified disorders of lipid metabolism to be significantly lower among the HCV population compared to the uninfected population. This finding is not clearly understood since HCV infection is associated with enhanced lipogenesis that may lead to the pathological development of steatosis and metabolic syndromes such as insulin resistance, obesity, and HCC [18].

Our study has several limitations. Firstly, as we used a medical claims database, we did not have liver histology data to confirm stage of disease and thus cannot exclude the possibility of case ascertainment bias as the selection of cases required at least two clinical ICD-9 HCV diagnosis codes for (e.g. patients with more advanced liver disease or on anti-viral treatment would have more clinical visits). Secondly, since data were limited on HCV genotypes (< 1%) and unavailable for racial groups, we were unable to match on these factors in our analyses although HCV disease/treatment efficacy is known to vary by genotype and race [30]. Thirdly, we relied on clinical coding for disease stage so the potential for misclassification of HCV diagnosis and liver disease stage cannot be ruled out. In addition, we cannot exclude the possibility of inconsistent coding or misclassification of comorbidities in our study; for example, COPD, which did not rank in our top 25 list, could have been misclassified under lower respiratory disease which ranked 5. Similarly, cutaneous conditions such as lichen planus and porphyria cutanea tarda which are commonly seen extrahepatic manifestations did not rank in the top 25, however, it is possible that these dermatologic conditions could be misclassified under allergic reactions [rank 18], which was mainly composed of unspecified dermatitis (65.9% of HCV patients with allergic reactions). Furthermore, given the cross-sectional design of this study, we were unable to establish the temporal sequence of events, i.e. whether a comorbidity, such as anemia deficiency, allergic reactions or eye disorders occurred prior to treatment, during treatment, or were treatment side effects. Lastly, patients captured in a medical claims database may be a biased selection of the HCV population as it does not reflect the population who do not have access to healthcare, so estimates of these comorbidities should be extrapolated with caution to the general HCV population. However, our results could be generalized to the privately insured population. According to population-based data from the National Health and Nutrition Examination Survey (NHANES), 61% of those chronically infected with HCV had health insurance and among those insured, half of them had private health insurance [31]. Despite these limitations, our study includes a much broader HCV population than previous studies, which were limited to select populations of US veterans [14, 32, 33], dialysis patients [34], psychiatric patients [35] and drug users [36]; and represents an insured population that would have access to treatment if comorbidities were able to be managed.

In summary, this study identified a wide range of highly prevalent comorbidities, which may have implications for HCV prognosis, management and treatment success. With over 4 million Americans currently infected with HCV, it is one of the most pressing public health challenges. Effective strategies to manage and control comorbidities among chronic HCV patients may result in increased access to treatment, thereby reducing the future burden of HCV disease and its manifestations.