Scribd Logo
Upload

Welcome to Scribd!

  • 2002 Aggarwal NMJI 15 320 Interferon Costs

    Uploaded by

    Marcela Murafa
    12 views8 pages
    d

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document
    d

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    Download as pdf
    12 views8 pages

    2002 Aggarwal NMJI 15 320 Interferon Costs

    Uploaded by

    Marcela Murafa
    d

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    Download as pdf
    of 8
    320 ‘THE NATIONAL. MEDICAL JOURNAL OF INDIA Original Articles VOL. 15, NO. 6, 2002 Treatment of chronic hepatitis B with interferon-alpha: Cost-effectiveness in developing countries RAKESH AGGARWAL, U.C. GHOSHAL, S. R. NAIK ABSTRACT ‘Background. Treatment with interferon-alpha (IFN) bas ‘been shown to be cost-effective in developed countries. How: ‘ever, cast-effectiveness in developing countries such as india has ‘ot been studed, Methods. Using the Markov transitional probability model, ‘We studied two cohorts of young patients (30 years of age) with chronic hepatitis B, one untreated and the other treated with ‘interferon (IFN), 5 million units daity for 16 weeks, wich ‘evidence of viral replication and chronic hepatitis, but not irthosis, and were followed up over a 30-year period. Rates of disease progresion, efficacy of IFN and quality of life associated With various disease states were estimated from the available literature. Direct costs were estimated using Indian prices of IFN and frm the usual costs of medical treatment in india based on ‘expert opinion, Unrelated mortality rates were modelled on age- ‘specific death rates of the general population. The efficacy of JFN- as judged in terms of extra fe-years and quafty-achsted fife “years (QALY) gained, and marginal cost-effectiveness and cost- “utility. Several sensitivity analyses, both undiscounted and with dlscounted analyses, were done. Results, At the end of the 30-year period, fewer patients in ‘the FN-treated group developed cirrhosts or decompensated -cirthosis, or were dead. The average life span of the treated cohort was 25.14 years, a gain of 0.6 years over the untreated cohort (24,54 years}, The QALY lived by the ovo cohorts were 23,69 and 22.75 years, respectively, representing a gain of 0.94 years for the IFN-treated group. The cost incurred by the ‘IFN-reated group was Rs 300 000, and that for the untreated cohort was Rs 40700, a substantial difference. Using the baseline estimates, undiscounted costs per year of life gained and per QALY gained were Rs 432 500 and Rs 276 900, respec- ‘dvely; these estimates are 20.5 and 13.1 thmes the per capita sss national income of the Indian population. Sensitivity analyses showed that changes in various parameters led to only minor changes in these estimates. Use of discounting led to an increase in marginal cost per life-year or QALY gained. Conctasion. In developing countries with a low per capita lacome, IFN therapy for chronic hepatitis B may not be cost- Sanjay Gand Posigrndunts nna of Mado Seems, Trucknow 226014, Utes Pradesh India [RAKESH AGGARWAL. .€. GHOSHAL, S.R.NAIK Degurtnetol Gastconelogy CCorespondenceto RAKESH AGGARWAL: rakesh@sgpglacin (© The Datonal Mesa! Jour of bn 2002, effective. A careful consideration of costeffectiveness fs there- fore essential before instituting IFN therapy In patients with, chronic hepatitis B In such populations. ‘Natl Med J India 2002;15:320-7 INTRODUCTION Chronic infection with hepatitis B virus (HBV) leads to chronic hepatitis, cithosis of the liver and hepatocetllar carcinoma (HCO), and is wesponsible for major morbidity and mortality. Interferon-apha CIPN}, which has immunomodulatory and enti Viral properties, has been shown to be effective in patients with chronic hepatitis B, IFN treatment leads toa significant increase in conversion of serum hepatitis Be antigen (HBeAg), « marker of viral replication, from positive to negative, clearance of HBV DNA from the scrum and normalization of serum aminotrans- erase levels >in a large meta-analysis of 16 published studies, mong adolt patients with chronic hepatitis B who were positive for HBoAg, 33% of those treated with IFN became HBeAg negative as compared 19 12% of placebo-treated controls“ For- ther, in that study, 7.8% of patients treated with IFN became Depatits B surface antigen (HBsAg) negative compared to 1.8% of controls, Successful IFN treatment has also been shown to be associated with improvement in liver histology.’ IFN therapy has been shown to be cost-effective for populations tiving in devel- ‘ped courie, with a roargnalcost-effectiveness of USS 12.000 per year of life gained in one study. and of US$ 2142-17 128in another study.? This treatment has thus become the standard in those countries." ‘A large proportion of the global burden of HBY infection is located in developing counties, For instance, in India, 3%~5% of the population is chronically infected with HY, with approxi mately 30-50 million chronie ceriers, accounting for nearly 125% ‘of the total global HBV borden." In recent years, IFN has been Widely used inthe treatment of paticts with chronic hepatitis B in Tndia*"* Some professional organizations and consensus _aroupshave recommended the use of IEN therepy forthe treatment ‘ofchronic hepatitis Bin Inia. These recommendations take into onsideraion the literature on costetfectiveness, albeit from Geveloped countries. Since cost estimates and income levels vary indifferent coun- tres, a costeffectiveness analysis for one population may not necessaily be applicable so other populations, espectally those with vecy different gross national products (GNP) and per capita incomes. Besides, possible variations inthe cost of and response 10 FN therapy in different racial groups andthe lack of a medical insurance system leading to “out-of-pocket” payment by the AGGARWAL eral. ? TREATMENT OF CHROKIC HEPATITIS 8 WITH INTERFERON-ALPHA 321 patients and their employers for medical costs, may also make extrapolation of cost-offectiveness data from developed countries to the developing worid inappropriate. Since no costelficacy or cost tity analysis of EN therapy has been dene in India or other developing countries till date, we decided to undertake such 2n analysis. Though our analysis is primarily directed at tbe Indian ‘population, we believe tha its results have a wider application in several developing countries, which have similar medieal costs and per eapite GNP. METHODS Modelling technique We used x decision analysis model to predict and compare the likely clinical and economic outcames in two hypothetical cohorts ‘of patients with chronic hepatitis B. One-cohort received IEN in a dose ofS eillon units (MU) daily for 16 weeks, whereas the other id not the two cohorts were similar inal other respects, Using 4 decision analysis software program (DATA for Healthcare release 3.5.5; TreeAge Software Inc, Williamstown, MA, USA), \weconstructed a Markov modeltoestimate the outcome and costs inthe wwe hypothetical cohorts, ‘The Markov transitional probability simulation model is a techaigue in which & hypothetical cohort i followed over time as ltssubjects move from one state of health tosnother.* We used the following states of health: chronic hepatitis B, compensated ‘cirthosis, decompensated eitrhosis, HCC and death (Fig. 1). OF the patients in x particular state of health, during each simulated ‘unit time peviod (say a yeas), specified proportions move to other A. Untreated cohort a= states, whereas the remaining continue in their original state of health. Death is un absorbing stale and persons entering this sate ‘eauiot move to another state, The likelibood of a patient moving. trom one state of health to enoiher wus specified as a set of probabitities (Fable D, In our model, the simulation continued for 30 years ‘Aveilable datain the published medical literature were used for ‘he natura history and progression of HBV infection, probability ‘of success with IFN therapy and quality of life associated with Lifferentelinica states. Due to thelack of such data from India, we used published dara from other populations for several of these parameters, For cost analysis, we used estimates based on Indian prices. ‘Tame 1. Probability rates of disease progression used for analysis Pastor Bice AKeratve sume _ einai Tow Hi Anaad rae of progression fom ‘Mepaitistocirmosis 02 nar 068 Cintosisiodecompensuedcinbosis 005 003.10 (Cinhosishepatocehuarcarcinoma oo oot oa Annual rate of decth among pions wits Desompessstelcibesis 020 019 040 Hepatoelslarearsnon ago 020 060 Ralcofremissimwitvnteriewontreamentas 0.30 0.10 Od eongaredtplaeho testes! Peccaraseontecalcals ying potency 100 Hepatoceliiar Chronic hepatitis 8 [compensated ecompensatea oe cats 8. interferon-treated cohort | Hepatocaliar Deccinpensated hepatitis 8 Chrono Compencate ¥ ‘bros ———__+ Fc 1. Sttes of bealth used in the decision model for chronic hepatitis B. The upper panel (pane) A) represents the untreated cohort and the Tower panel (pane! B) presents a cohort whick bus been teated with interferon, Each box represents a state of health, The solid arrows represent potemial change from one health state to another that may oceur due to hepatitis B vis jection during one year. Each change is associated with a paniculer specified probability (see Table I for probability estimites); the remaining patients continue ft the orginal state of heals axrows not shown), The dotied arrows represent deaths due to nrelted causes that may be ‘expected to occur their frequency is modellod on age-specitle morality rates of the Indiaa population, The dashed arrow represents a leansition toa “successfully treated” state in reypoase to interferon treatment. 322 “THE NATIONAL MEDICAL JOURNAL OF INDIA ‘Using hese estimates andthe Markov model, weestimated the average mime of yeas lived, the average musaber of quality- adjusted ie-years (QALY; amieasure which integrates quality of Jie (QOL fer each year lived and provides a composite estimae oflife span ascorrected for morbiity du to disease] ved, andthe average medical costs incutred by patents in the two cohorts From these data, we calculated the marginal east-ffectiveness an marginal ostuility of IPN therapy hese values indicate the ‘et amountof money spent to gain each ext yearof hie and cach extra QALY, respectively, with lower values indicating greater beneficofieatment. These velues were compared with annval per capita GNP, the amount that a person may beexpected 0 con ‘ote to the evonomy and eam during an average year Patient poputation We considered each cohost as consisting of young adult patients (BO years of age), with chronic HBV infection (HBsAg posiivetor atleast 6 months), Pationts were assumed to have elevated serum aminotransferase activity (more than 1. times the upper limit of npnmal) fort least 6 months, evidence of active viral replication (positive HBV DNA in the serum) and a histological diagnosis of chronic hepatitis but no evidence of ciethosis. Pationts with ‘corxistent hepatitis C virus, hepatitis D virus or human inmuno- deficiency virus (ATV) infection were excluded, This clinica ad laboratory profile represents the sitwation for which there is ‘curreniy « consensus on treatment with IFN.” Disease progression rates ‘Our mode! includes five stages of progression (Table I: From chronic hepatitis B to compensated citshosis, From compensated ctzhosis to decompensated cisthosiss From compensated eitshosis © HCC: From decompensated cirhosis to death; and From HCC to death, For each of these, we based our estimates on previously published reports” ‘The annual incidence of cirrhosis among patients with chronic hepatitis Bas been estimated tobe between 1.085% and 2.24%per year” and 3% per year.” We assumed 2 baseline rate of develop- iment of compenssied citthosis smong patients with chronic hepatitis B of 2% per year with a range of 1%-39% “The probability of patients with compensated cirshosis die to [HRV infection developing decompensation ha been estimate as 2.34-10% per year.'**! Most prognostic models of HBY infee- tion haveused aneaal figures midway imthisrange,j.e.5%,"5.6%° and $420" We used a baseline annual rate of development of decompensated cithosis among. patients with cihosts of $%, with a range af 310%, ‘Te probability of patients with compensated cirrhosis devel- ‘oping HCC has been estimate as | 5%-2,8% per yeat."=" Previ- ‘ous prognostic models of HBV infection have used rates of 2% 2.466: some analyses have in fact ignored the occurrence of HCC.? We used an annual rate of 2% forthe development of HCC Inpasients wih eishosis, with arunge of £%—3%%; developmentof HCC among patients with chronic hepatitis B but no cirebosis as ignored. The mortality tale among patients with HCC was as- sumed to be 40% per year, with a range of 20%-60%, ‘Thetisk of death among patients with decompensated cierhusis has been estimated to vary from 13% to 23% per year."*" How- ‘ever, in one analysis, an annual rale of 39% was used. We assuined arate of 20% per year with a rmge of 10%%40%, VoL. 15, NO. 6, 2002 Affect of EN treatment In two previous studies, patients similar to those included in ont ccobost, with IF «rcatment, had an incremental oss of HBeAg of 21% (IEN-tteated patients 32%, placebo controls 12%)" and 36.5% CEN-treated 45.6%, placobw controis 9.1%). Weassumed 12 30% incremental efficacy for IFN treatment (Table 1; we also assumed that no patient jn the untreated eohort would lose HBeAg, Since long term date on patients treated with FN are limited, we assumed that loss of HBeAg would eliminate the isk of cizthosis, decompensated circhosis and death, Duration of follow up “Thetso cohorts weze followed up for 30-year i.e. til the cohort members reached 60 years of age or died of deompensated

    You might also like