Comparative Study On The Comparison Of Different Antiviral Therapy Schemes And Health Economics In Patients With Chronic Hepatitis B After The Implementation Of New Medical Insurance Policy In Beijing

Effect of the partial reimbursement implementation of antivirals for Hepatitis B virus in Beijing:comparative effectiveness analysis and cost-effectiveness analysis based on a retrospective cohort studyBackgroundHepatitis B virus (HBV) infection is a global public health problem, especially in China. Leading to a high rate of incidence and mortality from development of compensated cirrhosis (CC), decompensated cirrhosis (DCC) and hepatocellular carcinoma (HCC), chronic hepatitis B (CHB) infection produces a great economic burden to patients, families and society. Antiviral treatment is one of the effective ways to prevent the development of CC, DCC and HCC after HBV infection. But the essential prerequisite is the long-term adherence. One of the important factors related to adherence is the affordability of the patients for the medical care and antiviral drugs, which can be strongly influenced by the reimbursement scheme. In Beijing, all antiviral agents including Interferon and nucleos(t)ide analogues (NAs) must be paid out-of-pockets by the patients themselves before Julyl,2011. After then, all the antivirals for HBV have been listed in the National Reimbursement Catalogs of Drugs for Basic Medical Insurance. Patients could receive a 75%-85% reimbursement of the cost between a deductible of 1,800 yuan and a ceiling of 20,000 yuan. Whether the antiviral utilization, medication adherence and treatment outcome can be improved after the implementation of the new partial reimbursement policy have not been evaluated.AimsThis study aims to assess the economic burden of HBV related diseases and determine the impact of the partial reimbursement on antiviral utilization, medication adherence and treatment outcome for CHB patients in Beijing. The results would offer economic evidence for further improvement of the reimbursement policy.MethodsDemographic characteristics (including patients’ID, birth date, gender and insurance type), medication cost (including any expenditure from registration, lab tests, imaging tests, medications, hospital stays and hospital supplies), routine biochemical and virological detection results and antiviral prescription information (including drugs’name dose, usage and prescription date) of HBV related patients who were referred to Beijing You’an Hospital during Feb 14,2008 and Dec 31,2012 were collected from the hospital’s electronic database. The contents of the study were as follows:1 Cost and antiviral therapy analysis for HBV infection related patients in Beijing during 2008 and 2012:Multiple cross-sectional studies were conducted to assess the economic burden and antiviral treatment patterns of HBV related diseases, including CHB, CC, DCC and HCC from 2008 to 2012. The yearly tendency of the cost and antiviral treatment patterns through the study period were also explored.2 Impact of the new reimbursement policy on cost and antiviral therapy for CHB patients in Beijing:Multiple cross-sectional studies were also conducted among CHB outpatients and inpatients between February 14,2008 and December 31,2012. Propensity score (PS) matching was used to adjust baseline characteristics and disease severity for the patients in different years. Medical care costs and their composition, antivirals utilization and the proportion of different antiviral agents were compared among different years to explore the impact of the reimbursement.3 Impact of the new reimbursement policy on NAs treatment adherence for CHB patients in Beijing:Two retrospective cohorts of CHB outpatients referred to Beijing You’an Hospital were enrolled to explore the impact of the reimbursement on antiviral treatment adherence. Cohort 1 consisted of all CHB outpatients referred to Beijing You’an Hospital during the period of Jan 1,2010 to Dec 31,2010, followed up till Jun 30,2011. Cohort 2 composed of all CHB outpatients during July 1,2011 to June 30,2012, followed up until Dec 31,2012. PS matching was used to adjust the factors which might influence the antiviral therapy between the two cohorts, including patient’s age, sex, insurance type and disease severity. Antiviral utilization and medication adherence, including medication possession ratio (MPR) during the treatment period and persistence rate at different time point of treatment were compared among the patients with medical insurance (PMI) and those paid out-of-pocket (PPO) in the two cohorts. Multivariate logistic regression was used to adjust the baseline characteristic and disease severity to determine the impact of the reimbursement on antiviral utilization and adherence, respectively. Questionnaire survey was performed to confirm the validity of the electronic database.4 Comparative effectiveness analysis of ADV and ETV based on real-world data:the impact of the antiviral therapy adherence on the treatment outcome:Another retrospective cohort study was undertaken to determine the impact of antiviral medication adherence on treatment outcome. CHB NA-naive patients during Jan 1, 2009 and Dec 31,2012, who received adefovir (ADV), which was the most common used for CHB patients before the reimbursement, or entecavir (ETV), which was the most frequently used among PMI after the reimbursement, were enrolled. HBV DNA response, Hepatitis B e antigen (HBeAg) seroconversion and alanine aminotransferase (ALT) normalization for the specific antivirals among the patients with different medication adherence were calculated, respectively. COX regression was used to explore the impact of the medication adherence on HBV DNA response and the difference of treatment outcome between ADV and ETV after adjusting baseline characteristics and disease severity, respectively.5 Cost-effectiveness of ADV and ETV therapy with different medication adherence for CHB patients in Beijing:A 6-health-state 1-year-cycle Markov model based on a cohort of 100035-year CHB patients was built to compare the cost and effectiveness of the life-time therapy with different medication adherence of ADV or ETV. Medical costs of CHB patients were calculated based on the same population from the section 4. Medical costs for the other health states, including CC, DCC and HCC, were calculated based on the patients during the same period from You’an hospital. The quality adjusted life year (QALY) of CHB, CC, DCC and HCC patients were derived from our previous study and the literature review from the studies about QALY in Chinese or other Asian populations. The annual transition probabilities between different disease status depended on the HBV DNA response of patients with different medication adherence in the "Part 4" study and other similar related researches. We also carried out both the one-way sensitive analysis and probabilistic sensitive analysis due to the uncertainty of the parameters in the model.Results1 Cost and antiviral therapy analysis for HBV infection related patients in Beijing during 2008 and 20121.1 Medical costs for HBV related patients were significant burdens and increased greatly with disease progression. Annual medical costs for CHB, CC, DCC and HCC outpatients were 3788.3 yuan,4599.7 yuan,3830.8 yuan and 11913.9 yuan, respectively, these accounted for 10.4%,12.6%,10.5% and 32.7% of the annual per capita GDP in 2012 in Beijing, and that of inpatients were 14527.4 yuan (39.8%), 40775.7 yuan (111.8%),32330.1 yuan (88.7%) and 65682.5 yuan (180.1%), respectively. Besides, a yearly increase of total costs were observed for both outpatients and inpatients from 2008 to 2012, especially for that in 2012;1.2 Cost composition showed that antiviral medication costs accounted for the 60% of the total cost for CHB outpatients, and that were 30%-40% for CC and DCC patients, less than 20% for HCC patients; among inpatients, that of CHB patients were 14.1% while no more than 5% for CC, DCC and HCC patients;2 Impact of the new reimbursement policy on cost and antiviral therapy for CHB patients in Beijing2.1 After adjusting patients’characteristics by PS matching among different years for CHB patients who visited Beijing You’an hospital during 2008 and 2012, the annual total costs for PMI after the half year of the implementation of the reimbursement was 1200 yuan more than that in the half year before the implementation, and that of 2012 had a 1300 yuan increase than the previous year; for PMI inpatients, that in the second half year of 2011 were 50% higher than the first half of 2011; in PPO, however, no significantly increase was observed during the 5 years;2.2 Antiviral utilization analysis showed that after the implementation of the reimbursement, the antiviral utilization had rapidly increased form 40% to 50%; especially for PMI, which had increased from 49% to 65.9%, while that of PPO was still around 40%;2.3 Utilization of specific antiviral agents showed that before the implementation of the reimbursement, ADV was most commonly used either for PMI or PPO, the utilization was 51.9% and 53.3%, respectively; after that, the utilization of the first-line agent, ETV, had increased to 50.7% by the year 2012 among PMI and ETV became the most frequently used for them; for PPO, however, the utilization of ADV was still higher than 50% and was still the most preferred agent.3 Impact of the new reimbursement policy on NAs treatment adherence for CHB patients in Beijing3.1 After PS matching for the patients’characteristics between the two cohorts before and after the reimbursement implementation, antiviral utilization analysis showed that the reimbursement could significantly increase the utilization for PMI, which increased from 57.4% to 75.9%, while for PPO, only a 2% increase was observed;3.2 Further analysis showed that the utilization of ADV were 56.8% and 60.3% for PMI and PPO before the reimbursement implementation, respectively, more than the other agents; after then, the utilization of ETV of PMI had increased from 30.4% to 51.6% and ETV became the most commonly used for them, while that of PPO was only 41.8%, lower than ADV(56.6%), and the most frequently used was still ADV for them;3.3 Multivariate logistic regression was further used to determine the impact of the reimbursement on antiviral utilization by adjusting patients’characteristics. Our results showed that even adjusting the covariates, an increased utilization could still observed after the implementation of the reimbursement with a odds ratio (OR) of 1.075 (95% CI:1.008-1.146); meanwhile, the reimbursement had a larger impact for PMI than PPO with the interaction OR of 2.194 (95% CI:1.979-2.432);3.4 Medication adherence analysis had found a 5% increase in MPR after the reimbursement among PMI and an increase of under 2% among PPO. Further analysis of MPR among different agents showed that the MPR of ETV had significantly increased from 79.7% to 89.1% in PMI, much higher than that of PPO, which increased from 81.1% to 85.4%; for the other agents, however, no significantly change were observed after the reimbursement. About 71% of the patients had more than 80% MPR in each cohort before the reimbursement. This increased to 79.2% (P<0.0001) and 73.1%(P=0.0228) for PMI and PPO, respectively;3.5 We also used multivariate logistic regression to explore the impact of the reimbursement on medication adherence by adjusting patients’characteristics and disease severity. It was observed that the medication adherence was still significantly higher after the implementation of the reimbursement even after adjusting the covariates with the OR of 1.162 (95% CI:1.047-1.290). And the reimbursement can significantly improve adherence for PMI than PPO with the interaction OR of 1.422 (1.220-1.657). Besides, the adherence of PMI was lower than existing patients or PPO, the interaction OR was 0.820 (95% CI:0.703-0.957);4 Comparative effectiveness analysis of ADV and ETV based on real-world data: the impact of the antiviral therapy adherence on the treatment outcome4.1 Treatment outcome analysis of ADV and ETV naive patients from 2009 to 2012 showed that ETV demonstrated higher HBV DNA response rate than ADV, regardless of patients’HBeAg status. The HBV DNA response rate for ETV and ADV treated HBeAg positive patients at the 6,12 and 24 month were 47.0% vs. 14.6%,63.7% vs.25.2% and 71.1% vs.34.3%(P<0.0001), respectively, and that of HBeAg negative patients were 76.6% vs.48.5%,85.9% vs.62.8% and 93.1% vs. 71.4%, respectively (P<0.0001). Even adjusting for covariates, patients with ETV would still achieve higher HBV DNA response with the hazard ratio (HR) of 3.266 (2.507-4.256) and 2.147 (1.801-2.560) for HBeAg positive and HBeAg negative patients, respectively;4.2 Compared to ADV, HBeAg positive patients with ETV therapy could also achieve higher HBeAg seroconversion rate, the rates at the 6,12 and 24 month of treatment were 4.1% vs.0.6%,12.0% vs.3.4% and 29.0% vs.18.1%(P=0.003), respectively;4.3 Further analysis of the impact of medication adherence on HBV DNA response showed that patients with better medication adherence could achieve higher HBV DNA response. Among HBeAg positive patients, the HBV DNA response rates at 6, 12,18 and 24 month of ETV treated patients with MPR≥0.8 were 51.8%,74.0%, 79.6% and 80.8%, respectively, significantly higher than those with poorer adherence (P<0.0001); for those with ADV, however, no significantly difference was observed among patients with different adherence. In HBeAg negative patients, significantly higher response rate were also observed among patients with better adherence, both for those with ETV and ADV therapy;4.4 After adjusting for patients’age, gender, insurance type and disease severity, higher HBV DNA response rate was still observed among ETV treated patients with MPR≥0.8, the HR were 2.152 (1.433-3.233) and 1.48 (1.063-2.06) for HBeAg positive and HBeAg negative patients, respectively;5 Cost-effectiveness of ADV and ETV therapy with different medication adherence for CHB patients in Beijing5.1 A cost-effectiveness analysis was performed by a 1-year-cycle Markov model in a 35-year cohort with a discount rate of 3%. After 40 cycles, the cost-effectiveness ratio of ADV treated patients with MPR<0.5、0.5-0.8 and≥0.8 were 5298.7 yuan/ QALY,6592.9 yuan/QALY and 8745.6 yuan/QALY, and were 6112.6 yuan/QALY, 9192.3 yuan/QALY and 11699.7 yuan/QALY for ETV therapy, respectively. Patients in the highest ETV adherence group gained the largest QALY Compared with ADV or lower ETV adherence strategy, the highest ETV adherence strategy had an incremental cost-effectiveness ratio of 84381.5 yuan/QALY, which was less than the common reference thresholds based on per capita GDP (87,091 yuan). ETV therapy with MPR≥0.8 was considered the most cost-effectiveness strategy at an acceptable cost according to the current economic conditions in Beijing;5.2 One way sensitive analysis showed that the discounting rate had the greatest impact on the model. The cost and QALY decreased with the increase of the discounting rate. ETV therapy with MPR≥0.8 strategy was still the most cost-effectiveness option with the range of the discounting rate of 0% to 5%;5.3 In the probabilistic sensitive analysis, after 10000 times of Monte Carlo simulation, the probability of being the dominant strategy for ETV therapy with MPR≥0.8 increased with the raising the willingness to pay (WTP). As the WHO suggested, we set the WTP threshold at three times of the per capita GDP and the probability of being the dominant strategy for ETV therapy with MPR≥0.8 was much higher than the other strategies. The same conclusion we can make that ETV therapy with MPR≥0.8 was the most cost-effectiveness strategy.Conclusions and suggestions1 Medical costs increase with disease progression for CHB patients and may lead to a high economic burden to the individuals, families and the whole society. Early antiviral therapy should be carried out for CHB patients to prevent or delay the hepatic HBV progression and to offer significant savings;2 The reimbursement policy has an optimistic impact on antiviral medication utilization and adherence, especially for the first-line agent of ETV;3 Compared to ADV, which was the most commonly used antiviral agent among CHB patients before the reimbursement implementation, the first-line antiviral agent of ETV, which was most preferred after the implementation among PMI, could achieve higher treatment outcome. Also, patients with good medication adherence could reach a higher HBV DNA response at an early time. Thus, increasing the reimburse proportion for the antiviral agents cost to increase the patient’s affordability for the antiviral therapy, will contribute to keeping well medication adherence during therapy therefore optimistic outcome would be achieved in the long run;4 The partial reimbursement might effectively improve antiviral treatment outcomes by increasing antiviral utilization, especially the usage of the first line agent, as well as adherence for CHB patients at an acceptable cost.