The Spectrum of Chronic Hepatitis C Virus Infection in the Virginia Correctional System: Development of a Strategy for the Evaluation and Treatment of Inmates with HCV

The American Journal of Gastroenterology
Volume 100 Issue 2 Page 313  - February 2005

Richard K. Sterling, M.D. 1 , Robert S. Brown, Jr., M.D. 1 , Charlotte M. Hofmann, R.N. 1 , Velimir A. Luketic, M.D. 1 , R. Todd Stravitz, M.D. 1 , Arun J. Sanyal, M.D. 1 , Melissa J. Contos, M.D. 1 , A. Scott Mills, M.D. 1 , Vernon Smith, M.D. 1 , and Mitchell L. Shiffman, M.D. 1
1Hepatology Section and Division of Pathology, Virginia Commonwealth University Health System, Richmond, Virginia; Center for Liver Disease and Transplantation, Columbia University College of Physicians and Surgeons, New York, New York; and the Virginia Department of Corrections, Richmond, Virginia

ABSTRACT
BACKGROUND AND OBJECTIVE:  Chronic hepatitis C virus (HCV) is common in the inmate population of the United States. Long-standing HCV can progress to cirrhosis, which can contribute to significant morbidity and mortality. However, those inmates with histologically mild disease are unlikely to develop liver-related morbidity or mortality during their period of incarceration. Our objective was to develop an economic strategy for evaluation and treatment of inmates with chronic HCV.

METHODS AND MEASURES:  A retrospective cohort analysis of 302 inmates within the Virginia Department of Corrections (VDOC) who underwent liver biopsy for chronic HCV at the Virginia Commonwealth University Health System between 1998 and 2002 was performed. The data from this analysis was to utilized to develop a cost model for treatment of chronic HCV in this population based upon biochemical or histologic criteria. We used the perspective of the VDOC using actual costs paid to providers, hospitals, and pharmacies. The primary endpoint was cost-effectiveness of HCV treatment.

RESULTS:  Eighty percent of inmates with chronic HCV were genotype 1, 49% had a normal value for serum ALT at the time of evaluation, 30% had no fibrosis, and 24% had bridging fibrosis or cirrhosis. The cost to evaluate and treat 100 consecutive inmates with peginterferon and ribavirin regardless of serum ALT and liver histology was calculated to be $1,775,900 or $35,500 per sustained virologic response (SVR). Although the cost declined by 50% if only those patients with an elevated serum ALT were treated, 45% of those inmates with varying degrees of fibrosis, and 21% with cirrhosis would not have received therapy utilizing this scenario. In contrast, the cost of performing liver biopsy and treating only those patients with any degree of fibrosis was $1,367,043; a savings of slightly more than $400,000 per 100 patients evaluated. The overall cost of treatment was most influenced by the price of peginterferon and ribavirin, which declined as the histologic criteria utilized for treatment increased.

CONCLUSIONS:  A strategy in which inmates with chronic HCV are evaluated and a decision regarding treatment is based upon either biochemical or histologic criteria, which appears to balance both the health-care rights of the inmate and the impact of treating this disease on the financial and other resources of the correctional system.

Clinical Characteristics of HCV in the Inmate Cohort
The study population consisted of 302 consecutive inmates who underwent liver biopsy between October 1998 when this program was initiated and July 2002. The mean age of the cohort was 41 yr, 91% were male, and 51% were Caucasian. Mean ALT was 94 IU/L and 49% had a normal value for ALT at the time of the liver biopsy. All patients were anti-HCV positive and all of those tested were positive for HCV RNA. Fourteen patients with a history of an elevated ALT and positive anti-HCV did not undergo HCV RNA determination. However, since the histology obtained from these patients was consistent with chronic HCV, these patients were included within the cohort. Genotype was available in 90% of subjects; 80% were genotype 1. The mean HAI score was 7.03; 30% had no fibrosis and 24% had advanced fibrosis, including bridging fibrosis or cirrhosis.

Patients with a normal ALT at the time of biopsy were slightly older (42 vs 39 yr; p< 0.001), had significantly lower scores for total HAI (6.2 vs 8.3; p< 0.001), inflammation (5.7 vs 6.7; p< 0.001), and fibrosis (0.9 vs 1.6; p< 0.001) compared to patients with an elevated serum ALT. However, the sensitivity, specificity, positive predictive value, and negative predictive value of a single normal ALT to identify patients without fibrosis on liver biopsy was poor; only 70%, 53%, 22%, and 90%, respectively.

Economic Analysis Decision Model
The cost to simply treat 100 representative patients without a biopsy was calculated to be $1,775,900. Given the genotype distribution of this population and assuming that a SVR to peginterferon and ribavirin would occur in 42% of patients with HCV genotype 1 and 82% with genotypes 2 or 3, the cost per sustained responder was calculated to be $35,517. If only those patients with an elevated serum ALT at the time of their evaluation (51% of the cohort) were treated, the total cost would be $905,709 and the cost per SVR would be unaffected.

The cost of treating chronic HCV according to a liver biopsy-directed strategy where liver histologic severity is utilized to determine which patients in the cohort would receive treatment is illustrated in Fig. 1. Fifteen percent of patients had minimal liver disease histologically (defined as a Knodell score of less than 5 and no fibrosis). If these patients were not treated, the cost of performing liver biopsy on 100 inmates to identify and treat the remaining 85% would be $1,651,200 or $38,851 for each patient with a SVR. Thus, this biopsy-directed strategy would be associated with a cost savings of $124,700 for the 100 patients. The incremental cost associated with treating all patients would be $3,334 for each additional sustained responder, all of whom would have had histologically mild disease and unlikely to progress during their period of incarceration.

As the histologic cutoff for treatment is increased, the cost of treating HCV declined stepwise (Fig. 1). For example, if patients without any fibrosis on liver biopsy were not felt to require treatment, only 70% of the cohort, those with evidence of portal or more advanced fibrosis on biopsy, would receive therapy and the cost would decline to $1,367,043 per 100 patients or $4,088 per SVR. Further cost savings would occur if only patients with bridging fibrosis or cirrhosis are treated.

 

Sensitivity Analyses
One- and two-way sensitivity analyses demonstrated that the cost of the evaluation and treatment was highly dependent upon only two factors: the price of the medication and either the biochemical or histologic criteria utilized to select for treatment. The costs of medications were varied from $700 to $3,340 (Table 2). The cost savings of the liver biopsy-directed strategy declined with the cost of the medication and approached zero when the drug cost of peginterferon and ribavirin reached $700 per month (Fig. 2). Conversely, the cost savings associated with the liver biopsy strategy also declined as the histologic criteria for treatment was reduced. The cost was neutral when liver biopsy excluded only 8% of patients from treatment. Given the histologic spectrum of HCV in the inmate cohort, this would occur with a liver biopsy demonstrating a Knodell score of only 2 points and no fibrosis.

The cost of treating HCV was less responsive to the percentage of patients with HCV genotype 1, the percentage of African-Americans, and their response rate. This is because drug costs drive the per-patient costs in the overall model and the percentage of patients selected for therapy is determined by either the biochemical or histologic findings. In contrast, the percentage of African-Americans, their response, and the prevalence of patients with HCV genotype 1 drive the SVR rate; not the percentage of patients who receive antiviral therapy. Thus, if the proportion of African-Americans in a particular correctional facility varied from 15% to 80% or the percentage of patients with genotype 1 varied from 60% to 90%, little effect on the overall cost of treatment would be observed; cost savings ranged from only $1,150 to $1,300 per patient, very similar to the base case scenario. Similarly, if the overall rate of SVR within each genotype varied from a low of 30% (genotype 1 SVR of 20% and nongenotype 1 SVR of 60%) to a high of 75% (genotype 1 of 70% and non-1 of 90%), the cost savings changed only slightly from a low of $1,200 per patient to a high of $2,000 per patient. The incremental costs per sustained responder did increase from $1,693, for patients with a high response rate, to $5,145 in patients with a low response rate when compared to the no-biopsy strategy. Varying the SVR rate for African-Americans did not affect the overall cost of treatment but did lower the cost per SVR. Assuming that the rate for SVR in African-Americans was only 10%, the cost per sustained response was $5,145, which declined to $1,693 if the rate of SVR was the same as that observed in Caucasians. Discounting the second year had minimal impact on the overall costs as very few of the costs accrued once the patients had undergone liver biopsy and had completed therapy. Reanalysis of the data with a discount rate of 5% for the second year was performed and had no impact on the conclusions (data not shown).

AUTHOR DISCUSSION
Infection with HCV is the most common chronic disease affecting our nation's 1.8 million inmates (11-14). With a prevalence of 12-64%, anywhere from 200,000 to 1.2 million of the incarcerated persons in this country have chronic HCV infection (12, 13). Since HCV can lead to cirrhosis, liver cancer, and contribute to morbidity and mortality, it is important to evaluate inmates with chronic HCV infection and consider treatment when appropriate. Unfortunately, the correctional systems of this country have largely failed to address this issue. In a 1999 survey of state correctional facilities, only 10 of 37 (27%) reported that they tested inmates for HCV and only 4 (11%) had developed a plan for inmates who tested positive (3). Since then various advocacy groups, health-care agencies, and inmates themselves have lobbied state and federal legislators regarding the potential health consequences to inmates from HCV infection. In response, many correctional facilities have started to explore ways in which to evaluate and treat inmates with chronic HCV infection. A national meeting regarding this issue was recently sponsored by the Centers for Disease Control and Prevention and the National Institute of Health.

In the Commonwealth of Virginia, this problem was addressed through a series of meetings between representatives of the VDOC and our Hepatology group. The primary goal was to develop a plan that would respect both the inmate's right to health care and the financial and health-care resources available to our states' correctional facilities. Two ways of identifying inmates with chronic HCV were discussed; a policy of mandatory screening for all inmates currently incarcerated and for new inmates as they entered the correctional facility; and a selective process whereby any inmate would be screened for HCV upon their request. The more complicated issue was to develop and implement a plan for evaluation and treatment of those inmates who tested positive for HCV infection. To accomplish this we considered the natural history of chronic HCV, the risk of the inmate to develop morbidity or mortality from chronic HCV during their period of incarceration, and the length of time it required for the inmate to be screened, evaluated, and undergo treatment for HCV. We concluded that a rational plan that met our stated goals, could only be developed after the demographic, virologic, and histologic spectrum of chronic HCV in the inmate population was understood.

The natural history of chronic HCV has been elucidated through a variety of studies over the past decade (15-21). It is now well accepted that about 20-30% of patients with chronic HCV infection will develop cirrhosis over 20-30 yr following acute infection. However, it has also been recognized that a similar percentage of patients with chronic HCV infection will have histologically mild disease, progress very slowly or not at all, and therefore not develop cirrhosis or any other morbidity from HCV during their lifetime (30). Liver biopsy has been utilized to identify those patients with mild disease and at low risk for developing cirrhosis (31). Many of these patients were asymptomatic and it has been argued that such patients derive no significant long-term health benefit from treatment (30).

Before a rationale strategy could be implemented, there was a need to know the spectrum of disease in our population (26). The majority of patients were male and a high proportion (49%) had a normal ALT at the time of biopsy. Importantly, 15% had "histologically mild disease" which we defined as an HAI <5 and the absence of fibrosis and 30% had no evidence of fibrosis. Either of these groups would be at extremely low risk to develop progressive liver disease during their period of incarceration, which in Virginia and many other states average just 5 yr. Conversely, 24% had advanced fibrosis (either bridging fibrosis or cirrhosis). Given the sensitivity of 70% and specificity of 53%, our data show the limitation of serum ALT at the time of biopsy to predict histology in this population.

Data from the HCV inmate cohort was then utilized to develop economic models to estimate the cost of evaluating and treating HCV in this population. Compared to a strategy in which all patients with chronic HCV were simply treated without histologic evaluation, the biopsy-directed strategy would save the VDOC $124,700 per 100 HCV positive patients if those with mild disease (an HAI score of <5 and any fibrosis) were not treated. The cost savings would increase to $408,857 per 100 HCV positive patients if only those patients with fibrosis on liver biopsy were treated. Since all patients not offered therapy with this strategy have mild liver disease and/or no fibrosis, these cost savings are unlikely to either jeopardize the overall health of the inmate while incarcerated or affect their ability to respond to treatment if this required at some point in the future. Defining a histologic cutoff of portal fibrosis or even greater degrees of liver injury would limit treatment to an even more select group of inmates and may appear to be even more cost advantageous to the correctional system. However, such a strategy is probably not justified as this may adversely affect the inmate and correctional system by increasing the risk of morbidity and mortality from advanced liver disease. Therefore, it is believed that a strategy in which all patients with chronic HCV undergo liver biopsy and only those with a histologically significant liver disease undergo therapy strikes an equitable balance between the health-care rights of the inmate and the impact that this disease will have on the financial and health-care resources of our nation's prison systems. Assuming the histologic spectrum of chronic HCV is similar throughout the U.S. correctional system, adopting this strategy would reduce the cost of treating the estimated 500,000 inmates with HCV by $620 million.

The cost savings of the biopsy and selective treatment strategy are closely related to the cost of medication and the histologic criteria utilized for treatment. If the percentage of patients with mild disease declined to less than 8% of the total HCV population, or the cost of peginterferon and ribavirin were reduced to only $700 per month, a strategy favoring the treatment of all patients regardless of histologic severity would be cost neutral and liver biopsy would not be necessary. Although previous studies have concluded that the use of liver biopsy may not be cost-effective in the management of patients with chronic HCV (32-34), such studies contained many biased assumptions. For example, the definition of mild hepatitis that did not warrant therapy was so minor in these previous studies that less than 10% of HCV patients failed to meet the treatment criteria. In addition, these studies grossly overestimated the risk of developing cirrhosis in patients with mild HCV and assumed that these patients would never be reconsidered for therapy at any point in the future. Finally, these studies calculated costs based upon utilizing standard interferon with or without ribavirin administered for 48 wk to all patients (32, 35). This treatment is considerably less costly than peginterferon and ribavirin is today and does not take into account the current practice of discontinuing therapy in nonresponders after 24 wk of treatment. We therefore believe that our cost analysis is more appropriate for the inmate population being discussed today. Furthermore, because recent evidence suggests that patients with normal ALT respond similarly to those with elevated ALT (36, 37), the presence or absence of an elevated ALT should not affect the analysis. Because the model is highly sensitive to the costs of medications, lower drug costs to the VDOC would likely make treatment more available to a broader range of inmates.

There is both a cost and risk to society when potentially dangerous inmates are transported from a correctional facility to a hospital for evaluation and treatment. Furthermore, it may simply not be feasible to implement a biopsy-directed strategy in all correctional settings. Because of the wide fluctuation in costs to transport inmates to our facility among the correctional facilities, we did not include cost of transportation into our analysis. Moreover, since transportation costs would be applied to each visit, the higher costs of treating all inmates would make the biopsy-directed strategy more cost-effective. An alternative approach could be to utilize serum ALT as a marker of more severe liver injury and simply treat all inmates with HCV as long as they have an elevated ALT. Serum ALT was not used as a criteria for biopsy. In our analysis, 49% of inmates in this cohort had a normal serum ALT at the time of biopsy. Previous studies have demonstrated that up to 25-33% of patients with chronic HCV may have a persistently normal serum ALT and that liver histology in such patients is in general milder than that observed in patients with elevated ALT (36). However, despite having a persistently normal ALT, fibrosis was present in about 33% of these patients upon histologic analysis. In the present cohort, a strategy of treating only those inmates with an elevated ALT would have provided therapy to 23% of patients with no fibrosis who would otherwise not receive treatment through the biopsy-directed approach, and not offer therapy to 64% of patients with normal ALT who would have been found to have varying degrees of fibrosis and even cirrhosis upon histologic analysis (Table 4). Furthermore, although treating only those inmates with an elevated ALT might be the least costly scenario for the DOC, it would not improve the cost per sustained response. Inmates with normal ALT were not followed over time. Therefore, it is not known how many developed an elevated ALT in follow-up and did not factor into the analysis. The use of noninvasive tests to predict histologic severity has gained recent attention (38, 39). However, the clinical utility of these tests in the correctional setting, given the high proportion with normal ALT, needs further testing. Thus, it is felt that a biopsy-directed strategy, if feasible, is preferable to serum ALT in deciding which patients with chronic HCV require treatment.

Large clinical trials have demonstrated that about 33-45% of patients achieve a SVR following treatment with interferon and ribavirin (40, 41). This has increased to 54-56% with peginterferon and ribavirin (27-29). However, SVR is significantly lower in African-Americans when compared to that observed in all other races (22-24). As a result, the expected SVR in the inmate population, where roughly half the patients with HCV are African-American, would be expected to be significantly lower than reported in large clinical trials where less than 10% of the study population were African-Americans (27-29, 40, 42). While this would increase the cost of treatment per sustained responder, this would not increase the overall cost of treatment as this is driven by medication costs and the percentage of patients selected for treatment. Finally, we did not factor compliance with therapy into our analysis. Although compliance with therapy will certainly impact on response, a recent report of HCV treatment in the correctional setting found excellent tolerability of interferon and ribavirin without the need for dose reduction or discontinuation (43).

There are several limitations to this analysis. Due to its retrospective design, it may not represent the true spectrum of liver diseases in the correctional system. Because not all state facilities refer to our center, not all HCV-infected patients identified by correctional centers are referred for evaluation, and patients otherwise appropriate may have refused evaluation and treatment, selection bias may have affected our results. Because patients with obvious liver decompensation were not included, the true spectrum of HCV-related disease severity may be underestimated. However, as these individuals would not be considered for HCV therapy, it should not affect this analysis. Nevertheless, these are unavoidable and true of other studies. Furthermore, this analysis does not take into account the long-term benefit of HCV therapy, namely reduced morbidity and mortality resulting in increased quality adjusted life years (QALY). However, because data on the impact of HCV therapy on QALY in the general population is limited at best, and in the incarcerated population is unavailable, we chose to focus solely on the economic impact to the correctional system. Similarly, we did not take into account long-term outcome and follow-up after therapy or the long-term benefits to society related to reduced morbidity, and mortality from chronic HCV if inmates were HCV RNA negative at the time of release. We also did not include the costs of complications of liver biopsy. Given the rarity of significant events (such as bleeding), these costs would not have a significant impact on the model.

In the state of Virginia, the decision to treat is ultimately left to the DOC and the individual facilities. Consequently, comorbid conditions and psychiatric contraindications for therapy need to be included. If these steps are taken, treatment can be successful in the correctional setting (43, 44). In order to allow inmates to complete therapy, a period of 2 yr of remaining incarceration was chosen. In support of this, in our initial experience, 14% of those responding to therapy were paroled and did not return to clinic for follow-up and therefore their benefit of therapy is unknown (43).

In summary, it is proposed that treatment of HCV in the inmate population be directed by a biopsy strategy where the histologic cutoff for therapy is no fibrosis. Such a strategy appropriately balances the health-care rights of the inmate and the financial and health-care limitations of the correctional system. In the absence of liver biopsy, treating those inmates with elevated ALT is an acceptable, but a less-effective alternative. The high percentage of African-Americans in this country's penal system will theoretically reduce the expected SVR, and therefore increase the cost of successful therapy. However, this would not be expected to increase the overall cost of treating HCV in this setting. Recently, a treatment program has been implemented utilizing these guidelines for those HCV patients incarcerated within the Commonwealth of Virginia. The results of this treatment program are being carefully monitored and will be available in the near future.

 

INTRODUCTION
Hepatitis C virus (HCV) infection is one of the most common causes of chronic liver disease in the United States. The most recent National Health and Nutritional Educational Survey (NHANES III), estimated that nearly 1.8% of the general population has been exposed to this virus and that 2.7 million persons are chronically infected (1). However, this survey did not include the estimated 1.8 million inmates who reside within this nation's correctional institutions. One of the most common risk factors for HCV is illicit drug use; approximately 83% of this country's estimated 2 million intravenous drug users will become incarcerated at some point during their lifetime (2). It is therefore not surprising that 12-64% of the inmate population test positive for HCV (3-13). As a result, HCV is now the most common health problem affecting the inmate population. This far exceeds the prevalence of infection with human immunodeficiency virus (HIV), diabetes mellitus, asthma, and hypertension, which affects only 1.5-15% of inmates respectively (11-14).

Chronic HCV can progress to cirrhosis (15-18), which may lead to significant morbidity and mortality (19-21). However, inmates with histologically mild disease are unlikely to develop liver-related morbidity during their period of incarceration, which in Virginia, like many other states, averages only 5 yr (2). When compared to the general population, a disproportionately higher percentage of inmates are African-American, and recent studies have suggested that progression of chronic HCV to cirrhosis may be slower in these persons (22). Several studies have also demonstrated that the response to interferon therapy is significantly reduced in African-Americans (23-24). Given the cost and morbidity associated with therapy, understanding the spectrum of chronic HCV in the inmate population is an important first step toward designing a treatment strategy for these patients in a setting with limited financial and social support.

In 1998, representatives form the Virginia Department of Corrections (VDOC) and the Hepatology Section at the Virginia Commonwealth University Health System (VCUHS) initiated a program designed to determine the spectrum of chronic HCV in the inmate population of Virginia and to estimate the resources that would be required by the VDOC to care for these patients and develop criteria for treatment of HCV. This manuscript describes the demographic, biochemical, virologic, and histologic spectrum of the disease, and utilizes this data to develop an economic model to predict the cost of several possible treatment strategies for inmates with chronic HCV.

Identification and Evaluation of Inmates with HCV
A voluntary screening program to identify communicable viral diseases (HCV, hepatitis B virus (HBV), and HIV) was initiated by the VDOC. Those inmates who tested positive for anti-HCV were next offered a complete evaluation by the medical staff at each correctional facility. This included a medical history, physical examination, and blood tests for the following: alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin (TB), albumin, blood urea nitrogen (BUN), serum creatinine (Cr), white blood count (WBC), hemoglobin, platelet count, and prothrombin time (PT). HCV RNA testing was performed by a commercially available polymerase chain reaction (PCR) assay. Those inmates who tested positive for HCV RNA, had a platelet count of greater than 70,000/ml, a PT within the limits of normal (INR of <1.4), a serum Cr <2.0 mg/dl, and no current or previous evidence of hepatic decompensation (variceal bleeding, ascites, or hepatic encephalopathy) were then offered the opportunity to undergo liver biopsy. Inmates who had less than 2 yr of incarceration remaining within their sentence, or were likely to achieve parole within this time frame, were not offered further evaluation or treatment. The rationale for this policy was to insure that the evaluation, treatment, and follow-up necessary to achieve a sustained virologic response (SVR) in the majority of patients, which in most cases requires 18-24 months, could be achieved. This was explained to those inmates in this situation.

Those inmates who wished to undergo further evaluation for chronic HCV and be considered for treatment were transported to a secure inmate unit at the Medical College of Virginia Hospitals (MCVH) to undergo consultation. This included additional testing to exclude coexistent liver disease such as autoimmune hepatitis, hemochromatosis, alpha-1-antitrypsin deficiency, or Wilson's disease; anti-HIV and HBV surface antigen were also obtained if not previously performed by the VDOC. Quantitative HCV RNA testing was performed by a PCR assay (Amplicor, Roche Molecular Systems, Branchberg, NJ) and HCV genotype by Inno-Lipa (InnoGenetics, Tarrytown, NY). However, since the values used to report HCV RNA levels changed since the inception of this program, a qualitative rather than quantitative value is reported. Percutaneous liver biopsy was then performed on all patients felt to be potential candidates for interferon and ribavirin therapy. The biopsy was not performed if the patient had a history of decompensated cirrhosis (variceal hemorrhage, ascites, or hepatic encephalopathy); a Child-Pugh score of greater than 6; renal insufficiency (serum creatinine > 2 mg/L), a platelet count of <70,000/cc or an INR >1.4. Patients were returned to their respective correctional facility after a period of observation. All liver biopsy specimens were assessed and scored according to the histologic activity index (HAI) of Knodell et al. (25). The results of this consultation and testing were communicated to the Medical Director of the VDOC (VS) and the staff physician at the respective correctional facility.

The population for this study included 302 consecutive inmates who underwent liver biopsy for evaluation of chronic HCV and met eligibility criteria was recently described (26). Patients with evidence of steatohepatitis on liver biopsy were not excluded. All inmates referred for biopsy were included in the cohort analysis irrespective of time of incarceration remaining. This experience was reviewed and approval to analyze the data obtained from these inmates was granted by the Office of Research Subjects Protection at the VCUHS.

Economic Analysis Utilized to Evaluate Treatment Strategies
A decision analysis model was developed to compare the cost of evaluating and treating 100 representative inmates with peginterferon and ribavirin according to several different strategies. This model was constructed utilizing the demographic, biochemical, virologic, and histologic data from 302 inmates who underwent liver biopsy; and the virologic response to treatment at weeks 24, 48, and 72 reported for patients with various HCV genotypes from large multicenter controlled trials (27-29). The various input variables utilized to construct the model are listed in Table 2. It was assumed that patients with HCV genotypes 2 and 3 and those HCV genotype 1 patients who failed to respond to therapy (nonresponders) would only receive 24 wk of treatment. Patients with HCV genotype 1 who responded to treatment and became HCV RNA undetectable by week 24 would be treated for 48 wk. For purposes of analysis, the primary end point was SVR, defined as an undetectable HCV RNA 24 months following completion of therapy.

The cost of evaluating and treating patients are listed in Table 3. The actual payments made by the insurance carrier for the VDOC for hospital care required to perform the liver biopsy, laboratory testing required to monitor therapy, physician costs, and the actual cost of peginterferon and ribavirin that would be paid to the pharmacy by the VDOC were obtained. Because of the wide fluctuations, the cost to transport inmates from various correctional facilities around the state to MCVH were not included in the analysis. The model was constructed to assume the perspective of the payer (the VDOC) and thus evaluated direct costs only. Since all patients were inmates, loss of work or work productivity was not assessed. No discounting of costs was performed since the evaluation and treatment would occur within just 1-2 yr. The cost of medication was varied during the sensitivity analysis.

A decision tree was constructed using Excel (Microsoft Corporation, Redmond, WA), from data in the literature to estimate the probability of virologic response at various decision time points (weeks 24, 48, and 72). The cost of evaluating and treating HCV was then calculated according to several different approaches as follows: treating all patients without a liver biopsy, treating only those patients with an elevated serum ALT without performing a liver biopsy, and finally according to a strategy in which liver biopsy and examination of liver histology was utilized to define which patients had sufficient liver injury from chronic HCV to warrant treatment. For each treatment strategy, the probability of virologic response at each decision time point was determined and the total cost of each treatment strategy calculated. The incremental cost of each treatment strategy was calculated by comparing the cost to treat 100 representative patients without a liver biopsy to treating only selected patients according to either the biochemical or histologic criteria. The incremental effectiveness of each treatment strategy was calculated by comparing the estimated number of sustained virologic responders in each group. The cost-effectiveness of a particular treatment strategy was calculated by dividing the incremental cost by the incremental effectiveness.

REFERENCES
1.          Alter MJ, Kruszon-Moran D, Nainan OV, et al. The prevalence of hepatitis C virus infection in the United States, 1988 through 1994. N Engl J Med 1999;341: 556-62.
CrossRef Abstract MEDLINE Abstract ISI Abstract CSA Abstract
2.          Brown BS, Beschner GM. Handbook on risk of AIDS: Injection drug users and sexual partners. Wesport (CT): Greenwood Press, 1993: 338.
3.          Spaulding A, Greene C, Davidson K, et al. Hepatitis C in state correctional facilities. Prev Med 1999;28: 92-100.
CrossRef Abstract MEDLINE Abstract ISI Abstract
4.          Holsen DS, Harthug S, Myrmel H. Prevalence of antibodies to hepatitis C virus and association with intravenous drug use and tattooing in a national prison in Norway. Eur J Clin Microbiol Infect Dis 1993;12: 673-6.
MEDLINE Abstract ISI Abstract CSA Abstract
5.          Schwarz TF, Dobler G, Gilch S, et al. Hepatitis C and arboviral antibodies in the island populations of Mauritius and Rodrigues. J Med Virol 1984;44: 379-83.
MEDLINE Abstract CSA Abstract
6.          Butler TG, Dolan KA, Ferson MJ, et al. Hepatitis B and C in New South Wales prisons: Prevalence and risk factors. Med J Aust 1997;166: 127-30.
MEDLINE Abstract ISI Abstract CSA Abstract
7.          Malliori M, Sypsa V, Psichogiou M, et al. A survey of bloodborne viruses and associated risk behaviors in Greek prisons. Addiction 1998;93: 243-51.
Synergy Abstract MEDLINE Abstract ISI Abstract CSA Abstract
8.          Anastassopoulou CG, Paraskevis D, Sypsa V, et al. Prevalence patterns and genotypes of GB virus C/hepatitis G virus among imprisoned intravenous drug users. J Med Virol 1998;56: 246-52.
CrossRef Abstract MEDLINE Abstract ISI Abstract CSA Abstract
9.          Vlahov D, Nelson KE, Quinn TC, et al. Prevalence and incidence of hepatitis C virus infection among male prison inmates in Maryland. Eur J Epidemiol 1993;9: 566-9.
CrossRef Abstract MEDLINE Abstract ISI Abstract CSA Abstract
10.          Ruiz JD, Molitor F, Sun RK, et al. Prevalence and correlates of hepatitis C virus infection among inmates entering the California Correctional System. West J Med 1999;170: 156-60.
MEDLINE Abstract ISI Abstract CSA Abstract
11.          Kendig NE. Hepatitis C management in correctional facilities. Prev Med Manag Care 2000;1: S33-41.
12.          Hammett TM, Harmon P, Rhodes W. The burden of infectious disease among inmates and releases from correctional facilities. Prepared for The National Commission on Correctional Health Care-National Institute of Justice. A Health status of soon-to-be-released inmates@ project. Presented at National Conference on Correctional Health; 1999; Ft. Lauderdale, FL, as a report in: Spotlight: Revealing the need for change: An interview with Robert Greifinger. HIV Education Prison Project News 1999;2: 4-5.
13.          Weinbaum C, Lyerla R, Margolis HS. Prevention and control of infections with hepatitis viruses in correctional settings. Centers for disease control and prevention. MMWR Recomm Rep 2003;52(RR-1):1-36.
MEDLINE Abstract
14.          Reindollar RW. Hepatis C and the correctional population. Am J Med 1999;107: 100S-103S.
CrossRef Abstract MEDLINE Abstract CSA Abstract
15.          Seeff LB, Hollinger B, Alter HJ, et al. Long-term mortality and morbidity of transfusion-associated non-A, non-B, and type C hepatitis: A National Heart, Lung, and Blood Institute collaborative study. Hepatology 2001;33: 455-63.
CrossRef Abstract MEDLINE Abstract ISI Abstract
16.          Tong MJ, El-Farra NS, Reikes AR, et al. Clinical outcomes after transfusion-associated hepatitis C. N Engl J Med 1995;332: 1463-6.
CrossRef Abstract MEDLINE Abstract ISI Abstract CSA Abstract
17.          Strader DB, Seeff LB. The natural history of chronic C hepatitis infection. Eur J Gastroenterol Hepatol 1996;8: 324-8.
MEDLINE Abstract ISI Abstract
18.          Poynard T, Bedossa P, Opolon P, et al. Natural history of liver fibrosis progression with chronic hepatitis C. Lancet 1997;349: 825-32.
CrossRef Abstract MEDLINE Abstract ISI Abstract CSA Abstract
19.          Fattovich G, Giustina G, Degos F, et al. Morbidity and mortality in compensated cirrhosis type C: A retrospective follow-up study of 384 patients. Gastroenterology 1997;112: 463-72.
MEDLINE Abstract ISI Abstract
20.          Thomas DL, Astemborski J, Rai RM, et al. The natural history of hepatitis C virus infection. Host, viral, and environmental factors. J Am Med Assoc 2000;284: 450-6.
CrossRef Abstract MEDLINE Abstract ISI Abstract CSA Abstract
21.          Khan MH, Farrell GC, Byth K, et al. Which patients with hepatitis C develop liver complications. Hepatology 2000;31: 513-20.
CrossRef Abstract MEDLINE Abstract ISI Abstract
22.          Howell C, Jeffers L, Hoofnagle JH. Hepatitis C in African Americans: Summary of a workshop. Gastroenterology 2000;119: 1385-96.
MEDLINE Abstract ISI Abstract
23.          Reddy KR, Hoffnagle JH, Tong MJ, et al. Racial differences in response to therapy with interferon in chronic hepatitis C. Hepatology 1999;30: 787-93.
CrossRef Abstract MEDLINE Abstract ISI Abstract
24.          Wiley TE, Mika BP, McCarthy ME, et al. Pre-treatment differences between HCV-infected African Americans and non-African American patients (abstract). Hepatology 1999;30: 417.
25.          Knodell RG, Ishak KG, Black WC, et al. Formulation and application of a numerical scoring system for assessing histological activity in asymptomatic chronic active hepatitis. Hepatology 1981;1: 431-5.
MEDLINE Abstract ISI Abstract
26.          Sterling RK, Stravitz RT, Luketic VA, et al. A comparison of the spectrum of chronic hepatitis C virus between Caucasians and African Americans. Clin Gastroenterol Hepatol 2004;2: 469-73.
MEDLINE Abstract
27.          Manns MP, McHutchinson JG, Gordon SC, et al. Peginterferon-alfa-2b plus ribavirin compared with interferon alfa-2b plus ribavirin for initial treatment of chronic hepatitis C: A randomized trial. Lancet 2001;358: 958-65.
CrossRef Abstract MEDLINE Abstract ISI Abstract CSA Abstract
28.          Fried MW, Shiffman ML, Reddy KR, et al. Combination of peginterferon alfa-2a plus ribavirin in patients with chronic hepatitis C virus infection. N Engl J Med 2002;347: 975-82.
CrossRef Abstract MEDLINE Abstract ISI Abstract CSA Abstract
29.          Hadziyannis SJ, Sette H, Morgan T, et al. Peginterferon alfa-2a and ribavirin combination therapy in chronic hepatitis C. A randomized study of treatment duration and ribavirin dose. Ann Intern Med 2004;140: 346-55.
MEDLINE Abstract ISI Abstract
30.          Levine RA. Treating histologically mild chronic hepatitis C: Monotherapy, combination therapy, or tincture of time. Ann Intern Med 1998;129: 323-6.
MEDLINE Abstract ISI Abstract CSA Abstract
31.          Saadeh S, Cammell G, Carey WD, et al. The role of liver biopsy in chronic hepatitis C. Hepatology 2001;33: 196-200.
CrossRef Abstract MEDLINE Abstract ISI Abstract
32.          Wong JB, Poynard T, Ling MH, et al. Cost effectiveness of 24 or 48 weeks of interferon alfa-2b alone or in combination with ribavirin as initial treatment of chronic hepatitis C. Am J Gastroenterol 2000;95: 1524-30.
Synergy Abstract MEDLINE Abstract ISI Abstract
33.          Wong JB, Koff RS. Watchful waiting with periodic liver biopsy versus immediate empirical therapy for histologically mild chronic hepatitis C. A cost-effective analysis. Ann Intern Med 2000;133: 665-75.
MEDLINE Abstract ISI Abstract CSA Abstract
34.          Leigh JP, Bowles CL, Leistikow BN, et al. Costs of hepatitis C. Arch Intern Med 2001;161: 2231-7.
CrossRef Abstract MEDLINE Abstract ISI Abstract CSA Abstract
35.          Wong JB, Bennett WG, Koff RS, et al. Pretreatment evaluation of chronic hepatitis C. Risks, benefits, and costs. JAMA 1998;280: 2088-93.
CrossRef Abstract MEDLINE Abstract ISI Abstract CSA Abstract
36.          Bacon BR. Treatment of patients with hepatitis C and normal serum aminotransferase levels. Hepatology 2002;36: S179-84.
CrossRef Abstract MEDLINE Abstract ISI Abstract
37.          Zeuzum S, Diago M, Gane E, et al. International multicenter, randomized, controlled study for the treatment of patients with chronic hepatitis C and persistently normal ALT levels with peginterferon alfa-2a (40KD) and ribavirin. Hepatology 2003;38: 208A (abstract 106).
CrossRef Abstract
38.          Poynard T, McHutchinson J, Manns M, et al. Biochemical surrogate markers of liver fibrosis and activity in a randomized trial of peginterferon alfa-2b and ribavirin. Hepatology 2003;38: 481-92.
CrossRef Abstract MEDLINE Abstract ISI Abstract
39.          Wai CT, Greenson JK, Fontana RJ, et al. A simple noninvasive index can predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C. Hepatology 2003;38: 518-26.
CrossRef Abstract MEDLINE Abstract ISI Abstract
40.          McHutchinson JG, Gordon SC, Schiff ER, et al. Interferon alfa-2b alone or in combination with ribavirin as initial treatment for chronic hepatitis C. Hepatitis Interventional Therapy Group. N Engl J Med 1998;339: 1485-92.
CrossRef Abstract MEDLINE Abstract ISI Abstract CSA Abstract
41.          Poynard T, Marcellin P, Lee SS, et al. Randomized trial of interferon alfa 2b plus ribavirin for 48 weeks or for 24 weeks versus interferon alfa 2b plus placebo for 48 weeks for the treatment of chronic hepatitis C. International Hepatitis Interventional Therapy Group. Lancet 1998;352: 1426-32.
CrossRef Abstract MEDLINE Abstract ISI Abstract CSA Abstract
42.          McHutchinson JG, Poynard T, Pianko S, et al. The impact of interferon plus ribavirin on response to therapy in black patients with chronic hepatitis C. Gastroenterology 2000;119: 1317-23.
MEDLINE Abstract
43.          Sterling RK, Hofmann CM, Luketic VA, et al. Treatment of chronic hepatitis C virus in the Virginia Department of Corrections: Can compliance overcome racial differences to response. Am J Gastroenterol 2004;99: 866-72.
Synergy Abstract MEDLINE Abstract ISI Abstract
44.          Allen SA, Spaulding AC, Osei AM, et al. Treatment of chronic hepatitis C in a state correctional facility. Ann Intern Med 2003;138: 187-90.
MEDLINE Abstract