北台灣美沙冬治療的效益分析

全文

(1)National Taiwan Normal University Department of Health Promotion and Health Education Master Thesis. Ying-Chun, Chou. The Effectiveness of Methadone Maintenance Treatment in Northern Taiwan. Tony S.-H. Lee, Advisor. August 2012.

(2)

(3) 致. 謝. 沒想到終於到了完成碩士論文的這一天了，一路走來，其實經歷了許多挫折及困難，若是沒有老師、家人及朋友的包容及支持，其實不知道自己是否能在此時完成碩士論文。在此，我要感謝許多在撰寫論文期間幫助過我的人。首先，我必須非常誠摯感謝指導教授思賢老師不厭其煩的指導，謝謝老師給予我機會並帶領我認識成癮科學、藥癮治療，老師對研究的嚴謹態度讓我獲益匪淺，每週的論文討論時間也都使我獲益良多，還得感謝老師提供我接觸國外的學者、參與計畫及投稿至國外期刊的機會及經驗，豐富了我研究所的學習。再者，感謝口試委員李志恆老師、蔡偉德老師及施淑芳老師在論文寫作期間的指導，並於口試期間提供我許多精闢的意見，讓我知道自己在研究的不足之處，也了解未來自己需再努力的地方。此外，感謝導師景美老師在研究所期間聆聽我生活及課業上的困難並提供協助及建議；謝謝思賢老師辦公室的助理、學長姊及學弟妹們，不管是在學術的討論、計畫的事情或是論文的寫作，總是共同砥礪、共同克服困難，很開心與大家共事；也要謝謝研究所同學在課業及生活上的幫助及經驗分享。最後，一定要感謝我的父母支持我、資助我唸研究所，讓我在就讀研究所期間衣食無虞，不需擔心生活的花費來源，讓我可以專心唸書、寫論文，我愛你們！也要謝謝弟弟的祝福，期望弟弟退伍後能找到好工作。還要謝謝我的男朋友昭先，陪我走過風風雨雨，總能在我生氣難過時給予我支持及鼓勵；也要感謝昭先的父母關心我的學業，並且時常幫我加油打氣。因篇幅有限，無法在此一一列出。謝謝每一個在我就讀研究所期間，曾經幫助過我的人，在此獻上最誠摯的感謝及祝福。盈君. 謹誌. 民國 101 年 8 月. i.

(4) ABSTRACT THE EFFECTIVENESS OF METHADONE MAINTENANCE TREATMENT IN NORTHERN TAIWAN YING-CHUN, CHOU. Objective: This paper examined the costs and the effectiveness before the Methadone Maintenance Treatment (MMT) and after one-year follow-up of MMT from patients’ perspective in four hospitals in northern Taiwan.. Methods: The participants in this study were recruited from the population of outpatients with heroin addiction enrolled at methadone maintenance treatment program in four hospitals in northern Taiwan. We adopted costs and effectiveness from patients’ perspective. The WHOQOL-BREF and the employment status were administered at baseline and at one-year follow-up to calculate the incremental quality of life (QoL) gained and the wages differences. The averted incarceration costs were computed by the estimates from government reports or literatures and the criminal records from Ministry of Justice.. Results: The implementation of MMT program cost were 37,989, 38,992 and 40,195 NT dollars from 2009 to 2011; however, without the implementation of MMT program, the cost was 672,954. The measurements for benefit were quality of life, employment, and criminal activities. The cost per QoL (Quality ii.

(5) of Life) in physical health domain was the highest of four domains (24,214 NT dollars) and the cost per QoL in psychological health domain was the lowest (10,280 NT dollars). The unemployment rate decreased after one-year follow-up; however, the wages per hour slightly decreased. Compared with the year of 2004 to 2007, the intervention of MMT decreased criminal offenses and incarceration costs.. Conclusions: During one-year follow-up of MMT, this study has shown that methadone maintenance treatment is cost-effective in terms of quality of life improvement, increased employment and averted incarceration costs.. Keyword: cost, effectiveness, employment, methadone, quality of life. iii.

(6) CONTENTS 致謝. ⅰ. Abstract. ⅱ. Tables. 1. Chapter 1: Introduction. 3. Chapter 2: Literature Review. 6. 2.1 Consequences of Injection Drug Use. 6. 2.2 Methadone Maintenance Treatment. 7. 2.3 The Cost-Effectiveness of Methadone Maintenance Treatment. 9. Chapter 3: Method. 12. 3.1 Research Questions. 12. 3.2 Hypotheses. 12. 3.3 Research Design. 13. 3.4 Participants. 14. 3.5 Date Collection Instruments and Materials. 14. 3.6 Data Analysis. 18. 3.6.1 Cost analysis. 18 iv.

(7) 3.6.2 Effectiveness analysis. 34. 3.6.2.1 Quality of Life. 34. 3.6.2.2 Employment. 36. 3.6.2.3 Criminal Activity. 36. Chapter 4: Results. 39. Chapter 5: Discussions. 71. 5.1 Discussions. 71. 5.2 Limitations. 74. 5.2.1 Cost. 75. 5.2.2 Effectiveness. 77. Chapter 6: Conclusions. 79. Acknowledgements. 80. Reference. 81. v.

(8) Table Table 1. Summary of estimates and sources. 16. Table 2. The categories of costs. 18. Table 3. Medical Costs. 20. Table 4. Traffic Costs. 27. Table 5. Drug Costs. 30. Table 6. HIV medical treatment costs in Taipei city, New Taipei city, Keelung city and Ilan county 33. Table 7. Participant socio-demographic characteristics and drug use history at baseline 40 Table 8. The total costs of implementation and non-implementation of Methadone Maintenance Treatment. 45. Table 9. Quality of life in four domains. 46. Table 10. Cost per QoL in 4 domains. 49. Table 11. Working status prior to and after 12-month methadone maintenance treatment 51 Table 12. Participants opinions on the effects of employment on methadone maintenance treatment and variations in wages 53 Table 13. Mean wages per hour at baseline and after one-year follow-up 57. Table 14. Values to calculate criminal costs prior to the implementation of 1.

(9) methadone maintenance treatment. 59. Table 15. Penalties for criminal cases after implementation of methadone maintenance treatment 62 Table 16. The number of averted convicted offenses prior to methadone maintenance treatment. 64. Table 17. The averted incarceration costs prior to methadone maintenance treatment 66 Table 18. 2009 ~2011 Incarceration Costs per person per day in Taiwan. 2. 68.

(10) Chapter One Introduction Recent years there has seen increased attention given to drug use in the psychiatry literature. Several studies (C.-Y. Chen, Wu, Su, Chou, & Lin, 2010; Godfrey, Stewart, & Gossop, 2004) have noted that drug use not only damages health and life quality but also raises social cost and imposes burden on the whole society. It is estimated that about 230 million people, or 5 percent of the world’s adult population have used an illicit drug at least once in 2010 (United Nations Office on Drugs and Crime (UNODC), 2012). World Health Organization (W.H.O.) (2010) reported that at least 15.3 million people who have drug use disorders, and it is estimated that 13.5 million people take opioids, including 9.2 million who use heroin. From the above reports, heroin, cocaine and other drugs kill around 0.2 million people each year, shattering families and bringing misery to thousands of other people.. In addition, using drug by non-sterilized syringes is an important cause of the spread of HIV in the United States (Des Jarlais, 1991). Since 1998, UNAIDS (UNAIDS, 1998) has pointed out that of the 10 regions of the World, there were 6 regions (including North Africa & Middle East, East Asia & Pacific, Latin America, Eastern Europe & Central Asia, Western Europe, North America, and Australia & New Zealand) that have been noticed transmission through injecting drug use is one of the main modes of transmission for adults living with HIV/AIDS. In Taiwan, the incidence of HIV had increased steadily before 2004; however, the incidence of HIV in 3.

(11) 2005 rapidly increased twice as much as in 2004. In Taiwan, due to the HIV prevalence in injection drug users rose from 620 cases in 2004 to 2417 cases in 2005 and 1833 cases in 2006. In 2005, Taiwan Department of Health initiated “Harm Reduction and Methadone Maintenance Treatment Pilot Program.” By providing sterile syringes, HIV/AIDS education, and methadone maintenance treatment, from the point of view of injection drug users, they not only have the alternatives to face the problems, but also have the chance to avoid previous patterns of drug use, improve survival and pursue better lives. From the societal view, implementing the harm reduction programs can reduce criminal activity and reduce the burden on the public security.. However, while heroin users stay in the treatment, there is a great percentage that heroin users relapse after treatment. Because of the normal condition that heroin users going back and forth about whether to receive treatment or keep using heroin (DiClemente, 1999; O’Toole, Pollini, Ford, Bigelow, 2006), NIDA (2011) has suggested that drug addiction has come to be regarded as a chronic disease, thus how much money has to spend has been an important task for policy makers and health professions. WHO (2010) reviewed research articles and showed that for every dollar invested in drug treatment for drug abusers, 7 dollars are saved in health and social costs. Among all drug treatments, methadone maintenance treatment is a relatively effective for treating opioid dependence. Methadone is a prescribed drug substitute for opiate, and it is a rigorously well-tested medication that has 4.

(12) been used successfully to treat opiate dependence in the United States for more than 40 years. In Taiwan, methadone maintenance treatment was initiated in 2005 and has already been provided to opioid dependence in more than 100 hospitals in 2012.. Although the costs or effectiveness of methadone maintenance treatment has been widely investigated; however in Taiwan, whether methadone maintenance treatment is a cost-effective treatment or should have been supported by the government is still unknown and under-inverstigated.. The purpose of this study is to calculate the costs and to determine the benefits or the effectiveness of implementation and non-implementation of methadone maintenance treatment in northern Taiwan. The economic costs and benefits or effectiveness in this analysis were determined from a patient perspective. This study also attempts to contribute to the averted costs and cost-effective ratio by conducting an economic evaluation and to better inform whether to continue the methadone maintenance treatment in the future.. 5.

(13) Chapter Two Literature Review 2.1 Consequences of Injection Drug Use. Injection drug use results in physical, mental and social problems, or even leads to death.. First, individuals who use drugs often have one or more accompanying medical issues, including physical and mental disorders (NIDA, 2010). Although different drugs have different physical effects, drug use causes damage to the brain and nervous system (NIDA, 2010). Drug use not only contributes to damaging users’ brain, but also infection with HIV (Human immunodeficiency virus) and HCV (Hepatitis C virus) which especially has a significant impact on the public health. Des Jarlais (1991) found that injection drug use is an important cause of the spread of HIV in the United States. Once injection drug users are infected with HIV or HCV, the society has to afford subsequent medical care costs. In addition, NIH (2010) showed that drug use and mental disorders often co-exist. In some cases, mental diseases may precede addiction; in other cases, drug use may trigger mental disorders.. Drug use not only harms individual health but severely affects quality of life. Drug use are associated with multiple negative consequences in a broad range of areas of functioning including physical or psychological problems, personal safety, social relations, roles and obligations, and work 6.

(14) (APA, 1994).. Drug use also imposes a huge burden on social security and national productivity and takes a heavy toll on different societies (C. Y. Chen & Lin, 2009). Drug use adds significant costs to communities, including violent and property crimes, prison expenses, court and criminal costs, emergency room visits, child abuse and neglect, lost child support, foster care and welfare costs, reduced productivity, unemployment, and victimization (NIDA, 2010). Due to the acquisition of the high price of drug, drug users might commit crimes or use violence under the influence of intoxicating substances, rather than earn wages by seeking employment (Allen, 2005; Basu, Paltiel, & Pollack, 2008). In the United States, drug use costs the economy over $600 billion dollars annually in increased health care costs, crime, and lost productivity (NIDA, 2011). The cost to society of drug use in the year 2002 was $181 billion among it and $107 billion is associated with drug-related crime (NIDA, 2006).. 2.2 Methadone Maintenance Treatment. There are several effective medications for the treatment of opioid use, such as methadone, buprenorphine, diacetylmorphine, and naltrexone. Over the last few decades, methadone has been shown an advantage over the other medications since it is popular, cheaper and has strong research evidence (Dobbs, 1971; Hubbard et al., 1989). Because methadone is approved for use 7.

(15) in heroin users in Taiwan and it is cheaper than other medications that patients and government could both afford, we examine if the methadone maintenance treatment is cost-effective in this study. Methadone is a prescribed drug substitute for opiate, and it is a rigorously well-tested medication that has been used successfully to treat opiate dependence in the United States for more than 40 years (NIDA, 2010). Despite the various forms of methadone, including diskettes, tablets, oral solution, liquid concentrate and powder, most countries utilize oral solution because of its safety and simplicity (NIDA, 2010). Besides, the initial dosing of methadone is guided by the severity of the patient’s opiate dependence, drug use history, the results of urine test and medical examination (NSW Health Department, 1999). Patients who participate in the methadone maintenance treatment program have to take methadone once a day because the elimination half-life of methadone averages 24 to 36 hours (CE Inturrisi, 1973; Kreek, 1973). The goal of methadone maintenance treatment is to reduce and even eliminate opiate use among users by stabilizing them on methadone for as long as is necessary to help them avoid returning to previous patterns of drug use.. In Taiwan, Department of Health have proposed “Harm Reduction and Methadone Maintenance Treatment Program” in 2005. Until 2010, there are 100 hospitals in Taiwan has provided methadone maintenance treatment services (CDC Taiwan, 2011). Not every drug users have chance to participate methadone maintenance treatment program. During incarceration, offenders usually went through enforced drug abstinence while receiving relatively 8.

(16) limited medical intervention and treatment.. 2.3 The Cost-Effectiveness and Cost-Benefits of Methadone Maintenance Treatment. According to Folland, Goodman, & Stano (2007), Cost-benefit analysis (CBA) provides a good method for evaluating public projects and all new federal regulations must undergo CBA since 1981. (Folland, Goodman, & Stano, 2007) There has been a dramatic proliferation of research concerned the cost-effectiveness of methadone maintenance treatment (Barnett, 1999; Masson et al., 2004; Zaric GS, 2000).. However, in studying the effectiveness of mortality and quality of life, several studies have suggested the effectiveness of methadone maintenance treatment may provide a modest survival advantage compared with detoxification. The cost per life-year gained is $16,976 (Masson, et al., 2004). Heroin users with access to methadone maintenance treatment has an incremental cost-effectiveness ratio of $5915 per life-year gained (Barnett, 1999). Cost per quality-adjusted life-month for physical domain was 2227.55 EUR; for psychological domain was 1879.50 EUR; for social domain was 5467.64 EUR; and for environmental domain was 4626.47 EUR. Costs per QALM and QALY for total quality of life in the maintenance program were 2864 EUR and 34368 EUR (Vanagas, Padaiga, & Bagdonas, 2010). In all, limited literature has been published on the cost-effectiveness approach in 9.

(17) mortality and quality of life.. There is only one study used Monte Carlo simulation model to predict wages after treatment. The mean earnings per month per individual significantly rises from $371 before treatment to $1005 after treatment. While considerable attention has been paid in the past to research issues related to employment after participating methadone maintenance treatment, a literature on issues of enhancement of wages and value of life has emerged only very slowly and in a more scattered way.. Several studies (French, Dunlap, Zarkin, McGeary, & McLellan, 1997; Godfrey, et al., 2004; Gossop, 2001; Healey, Knapp, Marsden, Gossop, & Stewart, 2003) have noted that crime reduction is one of the economic benefits of methadone maintenance treatment. Flynn et al (2003) showed that 394 subjects completing up to 1 year of treatment had benefits due to reduced crime rates of US $8797 (benefit-cost ratio of 3.06:1). Godfrey et al (2004) also found that at 2 years following intake to treatment, the study intervention generated savings in crime costs of US$ 3,037,996 or a benefit ratio of 15:1. For every US$1.7 spent on opiate dependence treatment, US$5 was saved due to lower crime rates (Gossop, 2001). Although substantial studies have been performed on the cost-benefits of methadone maintenance treatment, there has been little research conducted in Taiwan and there have been few attempts to estimate how much money heroin users have to pay to improve their quality of life and value of life. Taiwan government has been spending a lot of 10.

(18) money – 300,008,000 NT dollars in 2008, on the drug use treatment (CDC Taiwan, 2008). With the growing pressures for cost-containment, distributing limited resources should refer to the cost-effectiveness and the support for continuing the implementation of the policy from citizens.. 11.

(19) Chapter Three Method 3.1 Research Questions 1. Does methadone maintenance treatment improve quality of life in heroin users after methadone maintenance treatment? How much improvement do heroin users get from methadone maintenance treatment? 2. Does methadone maintenance treatment increase the wages of employment in heroin users after methadone maintenance treatment? How much benefits in wages of employment do heroin users get from methadone maintenance treatment? 3. Does methadone maintenance treatment reduce criminal activity after methadone maintenance treatment? How much benefits of criminal activity does the society get from methadone maintenance treatment? 3.2 Hypotheses 1. Heroin users will improve quality of life after methadone maintenance treatment. 2. The unemployment will decrease and the wages of employment will increase after methadone maintenance treatment.. 3. Criminal activity will reduce after methadone maintenance treatment 12.

(20) 3.3 Research Design. This study used a 12-month prospective design conducted in four hospitals in northern Taiwan between the years of 2008 to 2009 and each participant was interviewed semi-annually for 3 times. Due to the diversity of time point before versus after the methadone maintenance treatment, the study employed an implementation and non-implementation design of methadone maintenance treatment to eliminate difference. In addition, there is no other widely-acceptable treatment which participants are as many as methadone treatment in Taiwan, thus we do not compare the effectiveness of different treatments;. we. compare. the. effectiveness. of. implementation. and. non-implementation instead.. This study applied the economic evaluation procedures from French’s study (French, Salome, Sindelar, & McLellan, 2002). First, a patient-specific cost analysis was performed using the unit cost of medication for each of four hospitals. Second, we propose the following two different methods to calculate economic benefit and effectiveness. On one hand, the research design for implementation applied a 12-month longitudinal survey design in this study to collect self-administered questionnaires from those who have participated methadone maintenance treatment at baseline and 12 months. On the other hand, the research design for non-implementation used a data collection approach to estimate the figures if those participants had not attended methadone maintenance treatment. Third, the cost and benefit 13.

(21) figures were combined to derive cost-benefit or cost-effectiveness ratio.. 3.4 Participants. The participants in this study were selected from the population of outpatients with heroin addiction enrolled at methadone maintenance treatment program in four hospitals in northern Taiwan, including Taipei City, Taipei County, Keelung City and Ilan County. The primary criteria for selecting participants were that they were heroin dependent users and signed an informed consent. The methadone maintenance treatment participants have to go to the nearest hospital to take methadone once a day and they were required to do urine toxicology screening at baseline and 12 months. The study was reviewed and approved by the institutional review board (IRB) of the Taipei Medical University (approval number: P960205) and Taipei City Hospital (approval number: TCHIRB-970404-E).. 3.5 Date Collection Instruments and Materials. This study applied WHOQOL-BREF questionnaire which is a standardized measure designed to predict 4 domains of quality of life, including physical, psychological, social and environmental domain. The employment status and wages were collected from self-reported data at baseline and one-year follow-up. Criminal records were provided by the Ministry of Justice judiciary records. The data of non-methadone maintenance 14.

(22) treatment were collected from the government reports or literatures.. Table 1 summarizes each estimates and sources of costs and effectiveness. We adopted the estimates from the government reports which were more reliable, such as Department of Health, Ministry of Justice and Ministry of Economic Affairs. Few estimates that could not be found in government reports were strictly chosen from the research literatures. We will give detailed illustration of how each estimate performed in the calculation in 3.6 Data Analysis.. 15.

(23) Table 1. Summary of estimates and sources Esimates First-visit diagnostic fees Registration fees Prescription fees Examination fees Supportive psychotherapy Family therapy, group therapy or other therapies Healthcare services fees Case management services fees. Source 行政院衛生署疾病管制局 99-101 年度愛滋防治替代治療補助計畫行政院衛生署疾病管制局 99-101 年度愛滋防治替代治療補助計畫行政院衛生署疾病管制局 99-101 行政院衛生署疾病管制局 99-101 行政院衛生署疾病管制局 99-101 行政院衛生署疾病管制局 99-101 行政院衛生署疾病管制局 99-101 行政院衛生署疾病管制局 99-101. 年度愛滋防治替代治療補助計畫年度愛滋防治替代治療補助計畫年度愛滋防治替代治療補助計畫年度愛滋防治替代治療補助計畫年度愛滋防治替代治療補助計畫年度愛滋防治替代治療補助計畫. Average scooter fuel consumption (km / per liter). 行政院衛生署疾病管制局 99-101 年度愛滋防治替代治療補助計畫經濟部能源局車輛油耗指南(2008~2010). Unleaded gasoline prices (#92 and #95) in Taipei. 經濟部能源局油價資訊管理與分析系統 (http://www.moeaboe.gov.tw/oil102/). Medicine service fees. Metropolitan area (NT dollar/ per liter) Drug Costs. 法務部(1995) 毒品犯罪型態及相關問題之研究 p.120~121. Total HIV infection medical expenses in (including. 行政院衛生署全民健康保險醫療統計：門、住診合計就醫總醫療費用統計 ---- 按戶籍縣市別分. inpatient and outpatient). http://www.doh.gov.tw/CHT2006/DM/DM2_2.aspx?now_fod_list_no=10905&class_ no=440&level_no=3 Total number of patients. 行政院衛生署全民健康保險醫療統計：門、住診合計患者人數統計 ---- 按戶籍縣市別分 16.

(24) http://www.doh.gov.tw/CHT2006/DM/DM2_2.aspx?now_fod_list_no=10905&class_ no=440&level_no=3 Ratio of infected HIV by injecting drugs. 疾病管制局(2011) 73-100 年底 HIV 感染人數危險因子統計. Probability heroin users commit a crime. http://www2.cdc.gov.tw/ct.asp?xItem=37500&ctNode=2695&mp=220 李思賢,吳憲璋,黃昭正,王志傑,石倩瑜 (2010)毒品罪再犯率與保護因子研究. The total number of offenses solved/cracked The total number of offenses known over all the. 行政院主計處(2009) 98 年中華民國統計年鑑行政院主計處(2009) 98 年中華民國統計年鑑. mentioned crimes considered The percentage of arrested individuals who were. 法務部(2007) 96 犯罪狀況及分析. convicted The percentage of convicted individuals who were. 行政院主計處(2009) 98 年中華民國統計年鑑. incarcerated. Average costs per day of incarceration. 刑法 320, 325 & 335 條及毒品危害防制條例第 10 條余萬能胡文琳李志恆(1998) 台灣區藥物濫用社會成本推估初報. Total Legal budgets of Correctional Institutions. 98 年至 100 年法務部預算及決算書(矯正機關). Total Accommodated persons in Correctional. http://www.moj.gov.tw/ct.asp?xItem=102271&/ct.asp?xItem=102271&ctNode=2809 5&mp=001 98 年至 100 年法務部預算及決算書矯正機關總收容人數. Institutions. http://www.moj.gov.tw/site/moj/public/MMO/moj/stat/new/index3.pdf. Average days of incarceration per heroin user. 17.

(25) 3.6 Data Analysis. 3.6.1 Cost analysis. We adopted the perspective of patients. The patient cost data was estimated by conducting a survey of unit cost of medication (which included initial patient assessment and initial medical services, individual, couple, family or group counseling, methadone dosing, urine collection). We also collected information of travel distance, average number of daily working hours, and the income of the patient at baseline and 12 months. All costs were measured in NT dollars.. Table 2. The categories of costs Direct Costs. Indirect Costs. Methadone implementation . Medical Costs  First-visit intake diagnostic fees   . . Registration fees Prescription fees Examination fees (Urine test, HIV test, etc.)  Counseling fees  Case management fees  Methadone medication fees Traffic Costs. Non-methadone implementation . . Drug/heroin Costs. 18. HIV (Human immunodeficiency virus) medical treatment/care costs.

(26) Methadone implementation. Direct Costs. 1. Medical Costs. Medical costs were collected primarily from the regulation for governing payments on prevention, testing, and treatment of HIV which was formulated and announced by the Department of Health in 2011. Medical costs consist of: first-visit diagnostic fees, registration fees, prescription fees, healthcare services fees (including. health. counseling),examination. education, fees. case. management. and. (including. amphetamine. tests,. morphine tests, HIV (Human immunodeficiency virus) tests, GOT (glutamate oxaloacetate transaminase) tests, GPT (Glutamic Pyruvic Transaminase) tests, r-GT (r-Glutamyl Transpeptidase) tests, HBsAg (Hepatitis B Surface Antigen) tests, anti-HBs (Hepatitis B surface antibody) tests, anti-HCV (Hepatitis C antibody) tests, RPR/VDRL (rapid plasma regain/ venereal disease research laboratory) tests, Chest X-ray tests and EKG (electrocardiogram)), counseling and psychotherapy fees, and medication services fees.. 19.

(27) Table 3. Medical Costs Unit Cost (NT dollar). First-visit diagnostic fees. 12-Month Costs (NT dollar). 2011. 2010. 2009. 2011. 2010. 2009. 2000. 1500. 4600. 2000. 1500. 4600. First-visit diagnostic fees in 2009 includes supportive psychotherapy and all required examination fees.. Registration fees. 150. 150. 150. 2250. 2250. 2250. The registration fees were calculated once a week for the first month and once a month for the successive 11 months.. Prescription fees. 300. 300. 300. 4500. 4500. 4200. In 2009, the prescription fees were calculated once a week from the second week to the fourth week and once a month for the successive 11 months. In 2010 and 2011, the prescription fees were calculated once a week for the first month and once a month for the successive 11 months.. Examination fees HIV test. 400. 400. 400. 1200. 1200. 800. The HIV testing fees were calculated at baseline, 6 month and 12 month, except the HIV testing fees were included in the first-visit diagnostic fees at baseline in 2009.. 20.

(28) Urine drug test. 300. 300. 300. 900. 900. 600. (Morphine). The morphine testing fees were calculated at baseline, 6 month and 12 month, except the morphine testing fees were included in the first-visit diagnostic fees at baseline in 2009.. Urine drug test. 300. 300. 300. 900. 900. 600. (Amphetamine). The Amphetamine testing fees were calculated at baseline, 6 month and 12 month, except the Amphetamine testing fees were included in the first-visit diagnostic fees at baseline in 2009.. HBsAg test. 250. 250. 250. 750. 750. 500. The HBsAg testing fees were calculated at baseline, 6 month and 12 month, except the HBsAg testing fees were included in the first-visit diagnostic fees at baseline in 2009.. Anti-HBs test. 300. 300. 300. 900. 900. 600. The Anti-HBs testing fees were calculated at baseline, 6 month and 12 month, except the Anti-HBs testing fees were included in the first-visit diagnostic fees at baseline in 2009.. Anti-HCV test. 350. 350. 350. 1050. 1050. 700. The Anti-HCV testing fees were calculated at baseline, 6 month and 12 month, except the. 21.

(29) Anti-HCV testing fees were included in the first-visit diagnostic fees at baseline in 2009.. TPHA test. 400. 400. 400. 1200. 1200. 800. The TPHA testing fees were calculated at. (Treponema pallidum. baseline, 6 month and 12 month, except the. hemagglutination). TPHA testing fees were included in the first-visit diagnostic fees at baseline in 2009.. SGOT blood test. 50. 50. 50. 150. 150. 100. The SGOT blood testing fees were calculated at baseline, 6 month and 12 month, except the SGOT blood testing fees were included in the first-visit diagnostic fees at baseline in 2009.. SGPT blood test. 50. 50. 50. 150. 150. 100. The SGPT blood testing fees were calculated at baseline, 6 month and 12 month, except the SGPT blood testing fees were included in the first-visit diagnostic fees at baseline in 2009.. r-GT test. 70. 70. 70. 210. 210. 140. The r-GT testing fees were calculated at baseline, 6 month and 12 month, except the r-GT testing fees were included in the first-visit diagnostic fees at baseline in 2009.. 22.

(30) Chest X-ray test. 300. 300. 300. 900. 900. 600. The Chest X-ray testing fees were calculated at baseline, 6 month and 12 month, except the Chest X-ray testing fees were included in the first-visit diagnostic fees at baseline in 2009.. EKG. 150. 150. 0. 450. 450. 0. (Electrocardiography). Supportive psychotherapy. The EKG fees were calculated at baseline, 6 month and 12 month, except in 2009.. 300. 300. 300. 3600. 3600. 3300. The supportive psychotherapy fees were calculated once a month, except that the supportive psychotherapy fees in the first month were included in the first-diagnostic fees in 2009. Family therapy, group. 3200. 3200. 3200. 3200. 3200. 3200. therapy or other therapies Healthcare services fees. Family therapy, group therapy or other therapies fees were calculated once every year.. 500. 0. 0. 5500. 0. 0. (including health education,. The healthcare services fees were calculated once a month except the first month in 2011.. case management and counseling) Case management services. 0. 5000. 5000. 0. fees. 5000. 5000. The case management services fees were calculated once a year except the first month in 2009 and 2010.. 23.

(31) Medicine service fees. 20. 20. 20. 7300. 7300. 7300. The medicine service fees were calculated once a day.. Total. 37110. 24. 36110. 35390.

(32) From a patient’s perspective, the total required costs in Table 3 that a patient entered into the methadone maintenance treatment and visited the hospital for taking their medication every day had to pay NTD 35390, NTD 36110 and NTD 37110 annually from 2009 to 2011, respectively, which increased slightly by NTD 720 and NTD 1720 in 2010 and 2011, compared with the costs in 2009. The unit costs of all items in these three years were largely identical but with minor differences, such as first-visit diagnostic fees, EKG fees, healthcare services fees and case management services fees. In 2009, the first-visit diagnostic fees already included supportive counseling and psychotherapy fees and all required examination fees (therefore supportive psychotherapy fees and all the examination fees were charged only for the rest of 11 months due to double counting except that the EKG fees were not listed in 2009), yet the first-visit diagnostic fees in 2010 and 2011were substantially lower. The registration fees were charged at each visit and the fees were regulated by legislation that the maximum fees of each visit could not exceed NTD150; therefore, we adopted NTD 150 so that the maximum likelihood estimate could be derived. In 2009 and 2010, the case management fees were NTD 5000, but there were no case management fees in 2011. Instead, the healthcare services fees in 2011 included not only the case management but also health education and counseling. Therefore, the healthcare services fees in 2011 were NT5500, slightly higher than the case management fees in 2009 and 2010. Also, we assumed that patients went to the hospital 25.

(33) for taking methadone regularly, so the maximum medicine service fees would be NT7300.. 2. Traffic Costs. Participants in the methadone maintenance treatment ride scooters to the regular nearby hospitals and took methadone once everyday. The reasons why we included the traffic costs can be explained as follows: the most likely explanation is that there were two hospitals in counties which could not be easily arrived through public transportation; for another, some heroin users had jobs to do before or after taking methadone. Therefore, travelling distance through scooters from each heroin users’ houses to the hospitals were collected and also average gasoline consumed per scooter per day and the gasoline prices (including unleaded gasoline #92 and unleaded gasoline #95) were obtained from Bureau of Energy, Ministry of Economic Affairs in order to calculate traffic costs..  Travelling distance (km, single trip) ÷ Average scooter fuel consumption (km / per liter) × Average gasoline prices in Taipei Metropolitan area (NT dollar/ per liter) × 2 (Round-trip). 26.

(34) Table 4. Traffic Costs Year. Cost per person per day (NT dollar) Unleaded Gasoline #92. Unleaded Gasoline #95. Unleaded Gasoline #92. Unleaded Gasoline #95. 7.03 7.80 8.36. 7.21 7.99 8.55. 2564.35 2847.54 3050.66. 2632.71 2915.90 3119.02. 2009 2010 2011 . Cost per person per year (NT dollar). Travelling distance (km, single trip) ÷ Average scooter fuel consumption (km / per liter) × Average gasoline prices in Taipei Metropolitan area (NT dollar/ per liter) × 2 (Round-trip). 27.

(35) In Table 4, the daily routine visits to hospitals for taking methadone medication could produce traffic costs from home to hospital or from workplace to hospital. The traffic costs of a patient who had to go to hospitals for taking methadone everyday for a year were NT 2564.35, NT 2847.54 and NT 3050.66 counted on a basis of unleaded gasoline #92 annual price and NT 2632.71, NT 2915.90 and NT 3119.02 counted on a basis of unleaded gasoline #95 annual price from 2009 to 2011 since the gasoline prices fluctuated all the time. The traffic cost per person per year increased by almost NT500 dollars from 2009 to 2011; in other words, the traffic cost increased by nearly 20% within 2 years (18.95% and 18.47% respectively of unleaded gasoline price).. Non-methadone implementation. Direct Costs. 1. Drug Costs. This study assumed that, prior to the implementation of the methadone maintenance treatment, the 599 participants had not received the methadone; that is, they had habits of using heroin. According to the study from Ministry of Health conducted in 1995, we assumed that the 599 participants’ monthly heroin expenses probability distribution were same as the probability distribution in 28.

(36) the study. Then, we calculated each range (probability) of the number of people on heroin use which multiplied by the mean number of expenses. Also, we summed up the monthly expenses on heroin use in each range and came out the total monthly expenses on heroin use among all 599 participants.. 29.

(37) Table 5. Drug Costs. 0~4999 5000~9999 10000~19999 20000~29999 30000~49999 50000~99999 100000 Total. Ratio of Monthly expenses on heroin use 7.4% 8.2% 9% 11.9% 16% 22.9% 24.6%. Number of people Monthly expenses on heroin use 44.326 49.118 53.91 71.281 95.84 137.171 147.354 599 IDUs. 30. Monthly expenses on heroin use 110,815 368,385 808,650 1,782,025 3,833,600 10,287,825 14,735,400 31,926,700 NT. Average monthly expenses on heroin use per person. 53,300 NT/IDU.

(38) According to a study from Ministry of Justice (1995), we assumed that the 599 participants’ monthly heroin expenses probability distribution were same as the probability distribution in this study. In Table 5, the number of people of monthly expenses on heroin use was 599 multiplying the ratio in each range of monthly expenses. For example, 599×7.4%＝44.326. Then, the number of people on heroin use in each range multiplied by the mean number of expenses (44.326×NT2,500＝NT110,815). Also, we summed up the monthly expenses on heroin use in each range and came out the total monthly expenses on heroin use among all 599 participants was NT31,926,700. The average monthly expenses on heroin use was calculated as NT31,926,700÷599＝NT53,300 per person.. Indirect Costs. 1. HIV (Human immunodeficiency virus) medical treatment costs. Heroin users could be infected with HIV (Human immunodeficiency virus) through non-sterile needles and syringes. Because prior to the implementation of methadone maintenance treatment, heroin users could go to doctors for HIV treatment and were subsidized the payments from the National Health Insurance, Department of Health. The total HIV infection medical expenses (including inpatient and outpatient) and total number of patients in Taipei city, New Taipei 31.

(39) city, Keelung city and Ilan county were collected from the National Health Insurance database to calculate average HIV medical expenses per patient. However, not all the HIV infection patients were infected by injecting drugs, the average HIV medical expenses per patient should multiply the ratio of total HIV infection injected drug cases to total HIV infection cases in order to calculate average HIV infection medical expenses by injecting drugs per patient..  Average HIV infection medical expenses by injecting drugs per patient = Average HIV infection medical expenses per patient × Ratio of infected HIV injecting drugs. 32.

(40) Table 6. HIV medical treatment costs in Taipei city, New Taipei city, Keelung city and Ilan county. Year 2007. 2006. 2005. 2004. 2003. 2002. 2001. 2000. 1999. 122,111.49. 129,572.81. 135,280.54. 146,845.61. 140,595.58. 76,016.24. 23,735.02. 18,763.45. 29,590.36. 38.22%. 62.82%. 71.51%. 40.76%. 9.30%. 2.35%. 1.07%. 1.52%. 0.42%. 46,671.01. 81,397.64. 96,739.11. 59,854.27. 13,075.39. 1,786.38. 253.96. 285.20. 124.28. Average HIV infection medical expenses per patient Ratio of infected HIV by injecting drugs Average HIV infection medical expenses by injecting patient. drugs. per. The values of average HIV infection medical expenses per patient were retrieved from the National Health Insurance, Department of Health. Average HIV infection medical expenses per patient: Total HIV infection medical expenses (including inpatient and outpatient) ÷total number of patients Ratio of infected HIV by injecting drugs: the ratio of total HIV infection injected drug cases to total HIV infection cases. Average HIV infection medical expenses by injecting drugs per patient = Average HIV infection medical expenses per patient × Ratio of infected HIV injecting drugs. 33.

(41) The reason why the methadone maintenance treatment was taken place in Taiwan was because the number of HIV cases of injecting heroin had risen abruptly during the years 2003-2005. Before the implementation of methadone maintenance treatment, the participants who had been infected with HIV might have to pay the medical expenses for HIV treatment. Therefore, in Table 6 we calculated that prior to the methadone maintenance treatment, the average HIV infection medical expenses by injecting drugs per year per patient was NT33,354.14 during 1999-2007, which the maximum was NT96,739.11 in 2005. Before 2002, the average HIV infection medical expenses by injecting drugs per year per patient was lower than NT1800; however, its expenses had been soaring above the level of NT46,000, by more than 25 folds during 2004 to 2007.. 3.6.2 Effectiveness analysis. 3.6.2.1 Quality of life.. Because the patients had taken methadone for a period of time prior to entering this research, we applied a test to examine if quality of life had been improved from the time of enrollment to the baseline interview. The results show that there is no significant relationship between the length of taking methadones and quality of life. Therefore, we adopted the estimate of quality of life at the baseline as an estimate of non-implementation.. 34.

(42) Because the WHOQOL-100 is too lengthy, we applied the WHOQOL-BREF Taiwan version for practical use which also included two questions in domain 3 (Do you think you are respected?) and domain 4 (Can you eat food when you want to?) in order to measure quality of life. Then, we used SPSS for checking, cleaning and computing domain scores which steps are as follows (WHO, 1996):. 1. Recode all 28 items from a range of 1-5 scale in which 1 means poorly and 5 means extremely. 2. Reverse 3 negatively phrased items (Q3, Q4 and Q26) 3. Compute each domain scores Domain 1: Q3, Q4, Q10, Q15, Q16, Q17, Q18 Domain 2 :Q5, Q6, Q7, Q11, Q19, Q26 Domain 3 : Q20, Q21, Q22 Domain 3 (Taiwan) : Q20, Q21, Q22, Q27 Domain 4 : Q8, Q9, Q12, Q13, Q14, Q23, Q24, Q25 Domain 4 (Taiwan) : Q8, Q9, Q12, Q13, Q14, Q23, Q24, Q25, Q28 4. Delete cases with > 20% missing data. 5. Where an item is missing, the mean of other items in the domain is substituted. Where more than two items are missing from the domain, the domain score should not be calculated (with the exception of domain 3, where the domain should only be calculated if ≦1 item is missing) 6. Convert raw scores to transformed scores both to a 4-20 scale and a 0-100 scale. 35.

(43) The four domain scores represent an individual’s perception of quality of life in each domain. Higher scores represent higher quality of life.  Effectiveness = Incremental quality of life gained ＝ methadone implementation QOL scores － non-methadone implementation QOL scores.  Cost-per-QOL= 3.6.2.2. Employment. Benefits＝. 3.6.2.3. Criminal activity.. Benefits＝ Averted incarceration costs ＝. Probability accused heroin user is convicted of crime × Probability 36.

(44) convicted heroin user is incarcerated for crime × days of incarceration per heroin user × costs per day of incarceration. Non-methadone implementation. Individuals might commit crimes in order to struggle for heroin, the common crimes heroin users commit comprise theft, abrupt taking, robbery, piracy criminal conversion, fraudulent, and breach of trust, taking, and usury, which were categorized as “property crimes.” In addition, the other usual crime is possessing and using heroin, which were classified as “drug-related crimes.” Therefore, we assumed that if the 599 participants have not attended methadone maintenance treatment prior to the implementation of methadone maintenance treatment, some heroin users might commit crimes, assigning the rates of recidivism 12.5%, 33.24%, 46.02% and 65.7% within 6 months, 12 months, 24 months and 32 months respectively to the probability of committing crimes. Those who have committed crimes might be arrested by the police, which the probability of being arrested for commit property crimes and drug-related crimes were calculated respectively as follows: the total number of offenses solved/cracked divided by the total number of offenses known over all the mentioned crimes considered. And for those who were arrested, they might be accused for committing crimes; then for those who were accused, they might be convicted of offenses afterward and incarcerated in prisons, 37.

(45) the probability that the heroin users have being sentenced for property crimes and drug-related crimes were calculated as follows: the probability arrested heroin user is accused for committing a crime times the probability accused heroin user is convicted of crime times the percentage of convicted individuals who were incarcerated.. Methadone implementation. The criminal records were acquired by submitting petition to the Ministry of Justice for asking to provide heroin users’ serving prison sentence periods during the implementation of methadone maintenance treatment. Only criminal cases investigated completely from August, 2008 to May, 2010 were included. Then, we summed up the number of criminal cases, terms of sentence and fines which also categorized by the varieties of criminals of cases. Last, we multiplied the average terms of sentence (days) by the average incarceration budget per prisoner per day during 2009 to 2011.. 38.

(46) Chapter Four Results This study aimed at calculating the costs and effectiveness from government reports, literatures and self-reported data from heroin users. This chapter starts with a summary of the characteristics of the participants. The discussion is followed by a detailed presentation of results relating to each cost and effectiveness. The first section illustrates the prices in each category of costs and adds all of them. The second section displays the self-reported scores of quality of life and cost per QoL in each domain. The third section demonstrates the variation of employment wages between baseline interview and one-year follow-up. The final section shows the avoided incarceration costs before and after the methadone maintenance treatment.. 39.

(47) Table 7. Participant socio-demographic characteristics and drug use history at baseline (n=599) n(%). Mean(SD). Gender Male Female Age Under 30 31-40 41-50. 521(87) 78(13) 42.01(9.17) 48(8.0) 248(41.4) 178(29.7). 51-60 More than 60 Hospital Location Taipei City Keelung City New Taipei City. 105(17.5) 20(3.3). Ilan County Marriage Married Unmarried Divorced Other. 89(14.9). Missing Value Education Less than or equal to 6 years. 245(40.9) 153(25.5) 112(18.7). 147(24.5) 290(48.4) 104(17.4) 6(1.0) 52(8.7) 17.45(90.16) 40. 79(13.2).

(48) 7-9 years More than 9 years Missing Value Age of First-time Heroin Intake ≦18. 298(49.7) 217(36.3) 5(0.8) 27.27(7.36) 42(7.0) 332(55.4) 98(16.4) 28(4.7) 58(9.7). 18-30 31-40 ＞40 Missing Value HIV Positive Negative Missing Value HCV Positive Negative. 71(11.9) 503(84.0) 25(4.2) 513(85.6) 39(6.5). Missing Value Morphine Positive Negative Missing Value First-time Visit Dosage (mg/day) ≦20. 47(7.8) 382(63.8) 185(30.9) 32(5.3) 38.88(20.34) 86(14.4) 41.

(49) 21-40 41-60 61-80 ≧81. 380(63.4) 61(10.2) 45(7.5) 24(4.0). Missing Value Baseline methadone dosage (mg/day) ≦20. 3(0.5) 49.55(31.56) 102(17.0) 159(26.5) 145(24.2) 92(15.4). 21-40 41-60 61-80 ≧81. 69(11.5) 32(5.3). Missing Value. 42.

(50) Table 7 summarizes the participant characteristics at the baseline. In total, 599 participants were interviewed, 40.9% of the participants were from the hospital in Taipei city, 25.5% from Keelung city, 18.7% from New Taipei city and 14.9% from the Ilan county. Of these 599 participants, 87% of the participants were males (n=521), and the 13% were females (n=78). The average age of the participants was 42 years old, approximately 50% of the participants were under 40 years old (n=296) and 29.7% were at the ages of 41-50 (n=178). Almost half of the participants (49.7%) received 7-9 years of education (n=298), and 36.3% received more than 9 years (n=217). Approximately 50% of the participants were unmarried (n=290), and other marital status such as married and divorced accounted for 24.5% and 17.4% respectively. In terms of patterns of using heroin, the average age at first-time heroin use was 27.27 years old and more than half of the participants (55.4%) started to use heroin at the age of 18-30.. 84% (n=503) of the participants were tested HIV-negative at baseline; however, 85.6% (n=513) of them were tested HCV-positive at baseline. The average methadone dose at intake was 38.88 mg per day, with 63.4% (n=380) of the participants were prescribed 21-40 mg and 21.7% (n=130) of the participants were prescribed more than 40 mg at the starting point of methadone maintenance treatment. Since the interviews were conducted after the participants entering the methadone maintenance treatment, the baseline dosage was recorded as soon as the interviews were finished. The average doses of methadone increased to 49.55 mg, with more than half of 43.

(51) the participants (51.1%) were prescribed more than 40 mg.. 44.

(52) Table 8. The total costs of implementation and non-implementation of Methadone Maintenance Treatment. Methadone maintenance treatment implementation 2011. 2010. 2009. Direct Costs. 37,110. 36,110. 35390. Traffic Costs. 3,084.84. 2881.72. 2598.53. Total Costs. 40,194.84. 38,991.72. 37,988.53. Non-methadone maintenance treatment implementation Drug Costs. 639,600. HIV medical treatment costs Total Costs. 33,354.14 (96,739.11, 124.28) 672,954.14 (736,339.11, 639,724.28). 45.

(53) Table 9. Quality of life in four domains Domain 1. Domain 2. Domain 3. Domain 3 Taiwan. Domain 4. Domain 4 Taiwan. (Physical Health). (Psychological). (Social relationship). (Social relationship). (Environment). (Environment).  4-20 scale. 13.04(2.43). 12.04(2.61). 12.49(3.08). 12.22(3.00). 12.35(2.79). 12.60(2.73).  0-100 scale. 56.52(15.22). 50.25(16.29). 53.04(19.26). 51.39(18.78). 52.18(17.42). 53.72(17.06).  4-20 scale. 13.31(2.37). 12.67(2.82). 13.06(2.96). 12.94(2.89). 12.80(2.73). 13.00(2.66).  0-100 scale. 58.18(14.83). 54.16(17.63). 56.61(18.50). 55.89(18.09). 55.00(17.08). 56.30(16.65).  4-20 scale. 0.27. 0.63. 0.57. 0.72. 0.45. 0.4.  0-100 scale. 1.66. 3.91. 3.57. 4.5. 2.82. 2.58. Baseline (n=189). 12-month (n=189). Incremental quality of life gained. 46.

(54) In Table 9, The four domain scores represent an individual’s perception of quality of life in each domain. Higher scores represent higher quality of life. At baseline of the interview, the highest score is 56.52 (13.04) in physical health domain and the least score is 50.25 (12.04) in psychological domain, which were the same orders after the 12-month follow-up. After the 12-month follow-up, the highest score is 58.18 (13.31) also in physical health domain and the least score is 54.16 (12.67) in psychological domain as well. The results suggested that heroin users perceived their quality of life in physical health was better than other domain of health. However, the results showed that heroin users perceived their psychological health was not better than any other domain of health.. The social relationship domain scored 53.04 at baseline and 56.61 at the 12-month follow-up, respectively, which is greater than domain 3 (Taiwan version) score (51.39 at baseline and 55.89 at the 12-month follow-up) either at baseline or after the 12-month follow-up; however, environment domain score (52.18 at baseline and 55.00 at the 12-month follow-up) is smaller than domain 4 (Taiwan version) score (53.72 at baseline and 56.30 at the 12-month follow-up) even at baseline or after the 12-month follow-up.. In terms of incremental quality of life gained, we subtracted the quality of life scores at baseline from the scores after the 12-month follow-up. Domain 3 (Taiwan version) scores increased the most, from 51.39 to 55.89, which increased by 4.5. However, domain 1 scores 47.

(55) increased the least, from 56.52 to 58.18, which increased by 1.66. The results showed that the heroin users perceived their quality of life score in social relationship (Taiwan version) domain had had lower than other domain scores at baseline (with the exception of domain 2) which denoted that the quality of life in social relationship had more growing spaces to facilitate progress. Nevertheless, the results showed that the heroin users perceived their quality of life score in physical health domain had had highest score than other domain scores at baseline which denoted that the quality of life in physical health had more difficulties to achieve further improvement.. 48.

(56) Table 10. Cost per QoL in 4 domains Domain 1. Domain 2. Domain 3. Domain 3. Domain 4. Taiwan (Physical Health). Incremental QoL gained Cost per QoL. (Psychological). (Social. (Social. relationship). Relationship). Domain 4 Taiwan. (Environment). (Environment). 1.66. 3.91. 3.57. 4.5. 2.82. 2.58. 24,213.76. 10,280.01. 11,259.06. 8,932.19. 14,253.49. 15,579.40. Cost per QoL ＝ Total Methadone Maintenance Treatment Costs ÷ Incremental QoL gained We applied total methadone maintenance treatment costs 40194.84 NT dollars in 2011 so as to simplify the calculations as shown above.. 49.

(57) As shown in the first row of Table 10, we derived incremental quality of life gained from the Table 9. Then we applied total methadone maintenance treatment costs in 2011 which was 40194.84 NT dollars in order to simplify the calculations as shown above. This may suggest that investing on physical health domain needs more money (less cost-effective); however, investing the social relationships domain needs less money.. 50.

(58) Table 11. Working status prior to and after 12-month methadone maintenance treatment (n=196) Prior to Methadone Maintenance Treatment After 12-month Methadone Maintenance Treatment Yes No Total. Yes. No. 76(38.8) 31(15.8) 108(54.6). 38(19.4) 51(26.0) 89(45.4). 51. Total. 115(58.2) 82(41.8) 196.

(59) In Table 11, at baseline of the interviews, there were 599 participants completed the questionnaires. After 12 months, there were 196 participants left (32.72%). Working status among 64.8% of participants remained the same prior to and after 12-month methadone maintenance treatment, 38.8% of the participants stayed in employment and 26% of the participants remained unemployed. About 55% of the participants (54.6%) reported that prior to the implementation of methadone maintenance treatment they had jobs, and after 12 month there was a higher percentage (58.2%) of participants working in workplace.. 52.

(60) Table 12. Participants opinions on the effects of employment on methadone maintenance treatment and variations in wages (n=196) Wages per hour n(%) Participants who had jobs prior to and after 12-month methadone maintenance treatment (n=76) Positive effect, no negative effect Negative effect, no positive effect No positive or negative effects Both positive and negative effects. Sum(Mean, SD) -566.03(-7.45, 104.36). 22(28.9) 8(10.5) 30(39.5) 16(21.1). Worked less than 40 hours a week Participants who did not have jobs prior to methadone maintenance treatment but had jobs after 12-month methadone maintenance treatment (n=38) Positive effect, no negative effect Negative effect, no positive effect. 15(19.74%) 11562.04(304.26, 890.34). 20(52.6) 3(7.9). No positive or negative effects Both positive and negative effects. 12(31.6) 3(7.9). Worked less than 40 hours a week Participants who had jobs prior to methadone maintenance treatment but being unemployed after 12-month methadone maintenance treatment (n=31). 8(29.03%) -14086.2(-454.39,639.98). Positive effect, no negative effect. 8(25.8) 53.

(61) Negative effect, no positive effect No positive or negative effects Both positive and negative effects Worked less than 40 hours a week. 2(6.5) 20(64.5) 1(3.2) 9(21.05%). Participants who had jobs prior to and after 12-month methadone maintenance treatment (n=51) Positive effect, no negative effect Negative effect, no positive effect No positive or negative effects Both positive and negative effects. 11(21.6) 5(9.8) 29(56.9) 6(11.8). 54.

(62) In Table 12, among 38 participants who did not have jobs prior to methadone maintenance treatment but had jobs after 12-month methadone maintenance treatment, there were more than half of them (52.6%) felt that methadone maintenance treatment brought about positive effects on their employment, with no negative effects. However, among those 31 participants who had jobs prior to methadone maintenance treatment but being unemployed after 12-month methadone maintenance treatment, a quarter of them (25%) thought methadone maintenance treatment brought about positive effects with no negative effects on their employment, and nearly 65% of them had no opinions on it. Those 127 participants who remained the same working status including being employed and being unemployed prior to and after 12-month methadone maintenance treatment, more than 20% of them (28.9% and 21.6%) felt that methadone maintenance treatment had positive effects without negative effects on their employment.. Among 38 participants who did not have jobs prior to methadone maintenance treatment but. had. jobs after 12-month. methadone. maintenance treatment, there were almost 30% participants (n=8) who worked at part-time jobs, which carried less than 40 hours a week. The percentage of working at part-time jobs which participants did not have jobs prior to methadone maintenance treatment but had jobs after one-year methadone maintenance treatment is higher than the percentage of working at part-time jobs in the other two scenarios (19.74% and 21.05%, respectively). 55.

(63) The wages of participants working prior to and after 12-month methadone maintenance treatment varied continuously with crucial events and period of time. Among 76 participants who had jobs prior to and after 12-month methadone maintenance treatment, the average wage loss was NT7.45 per participant per hour. Also, among those 31 participants who had jobs prior to methadone maintenance treatment but being unemployed after 12-month methadone maintenance treatment, the average wage loss was NT454.39 per participant per hour. Although some participants suffered the wage loss in these two scenarios, there were 38 participants who were unemployed prior to the methadone maintenance treatment but had jobs after 12-month methadone maintenance treatment producing average wage gains NT304.26 per participant per hour.. 56.

(64) Table 13. Mean wages per hour at baseline and after one-year follow-up. Participants who had jobs prior to. Participants who did not have. Participants who had jobs prior to. and after 12-month methadone. jobs prior to methadone. methadone maintenance. maintenance treatment (n=76). maintenance treatment but had. treatment but being unemployed. jobs after 12-month methadone. after 12-month methadone. maintenance treatment (n=38). maintenance treatment (n=31). Wages per hour at baseline 216.90 (127.17). 0 (0)＊. 278.52 (203.94)＊＊＊. 209.45 (159.87). 304.26 (890.34)＊. 0 (0)＊＊＊. Wages per hour after one-year follow-up ＊. ＊＊. p < .05，. ＊＊＊. p < .01，. p < .001. 57.

(65) In Table 13, the mean wages per hour of the baseline and after one-year follow-up in the scenario of having jobs prior to and after 12-month methadone maintenance treatment were NT 216.90 and NT209.45, which showed no difference. In contrast, there was a statistically significant difference in the other scenario of not having jobs prior to methadone maintenance treatment but having jobs after 12-month methadone maintenance treatment (p < .05) and of having jobs prior to methadone maintenance treatment but being unemployed after 12-month methadone maintenance treatment (p < .001).. 58.

(66) Table 14. Values to calculate criminal costs prior to the implementation of methadone maintenance treatment Year 2007. 2006. Probability heroin users commit a crime  Within 0.5 year  Within 1 year  Within 2 years  Within 2.5 years. Probability heroin user is arrested for committing a crime  Property crimes  Drug-related crimes Probability arrested heroin user is accused for committing a crime  Property crimes  Drug-related crimes. 2005. 2004. 2003. 12.5% 33.24% 46.02% 65.7%. 69.93%. 58.30%. 52.27%. 47.43%. 45.43%. 82.45%. 63.90%. 64.59%. 73.22%. 70.67%. 51.35%. 44.05%. 40.23%. 38.38%. 35.45%. 46.5%. 37.2%. 33.4%. 33.6%. 26.2%. Probability accused heroin user is convicted of. 59.

(67) crime . Property crimes  Drug-related crimes Probability convicted heroin user is incarcerated for crime  Property crimes  Drug-related crimes. 74.96% 66.18%. 71.83% 80.12%. 73.19% 73.64%. 76.49% 58.69%. 85.61% 77.35%. 35.04%. 40.97%. 41.78%. 45.05%. 42.93%. 37.11%. 50.60%. 48.75%. 74.77%. 47.24%. Average days of incarceration per heroin user . Property crimes  Drug-related crimes. Minimum. 6 months. Maximum. 5 years. Average costs per day of incarceration. 248 (231.1, 761.2). Property crimes include theft, abrupt taking, robbery, piracy criminal conversion, fraudulent, and breach of trust, taking, and usury. Drug-related crimes include possessing and using heroin Probability heroin user is arrested for committing a crime: the total number of offenses solved/cracked divided by the total number of offenses known over all the mentioned crimes considered Probability arrested heroin user is accused for committing a crime: the percentage of the total number of accused individuals divided by the total number individuals who were completely investigated Probability accused heroin user is convicted of crime: the percentage of the total number of convicted cases divided by the total number of accused individuals Probability convicted heroin user is incarcerated for crime: the percentage of the total number of newly incarcerated inmates divided the total number of convicted 60.

(68) cases Probability arrested heroin user is incarcerated for crime: the percentage of arrested individuals who were convicted times the percentage of convicted individuals who were incarcerated (the percentage of convicted individuals who were incarcerated= the total number of after-inmates divided by the total number of convicted cases). 61.

(69) Table 15. Penalties for criminal cases after implementation of methadone maintenance treatment Theft. Offenses of Criminal Conversion. Number of criminal cases investigated completely Total terms of sentence (Days) Total fines (NT dollars) Average terms of sentence per case (Days) Average fines per case (NT dollars). 65. Extortion. Abrupt Taking. 3. 3. 2. 78. 840. 1,440. 75. 10,120. 5. Harm. 40,000. 50,000. 119.46 153.85. Total criminal offenses. Offenses of. 7,765 10,000. Offenses of Causing Bodily. Offenses of. 280. 480. 8000. 37.5. 129.74 641.03. Only criminal cases investigated completely from August, 2008 to May, 2010 were included. The average criminal offenses were committed per year is 42.62 (78 offenses÷1.83 years). There were overall 51 people committed 78 criminal cases. In other words, average 1.53 criminal cases were investigated per person. The maximum term of sentence of a criminal case is 18 months, and the minimum term of sentence of a criminal case is 20 days.. 62.

(70) In Table 15, there were overall 51 individuals in this study committed 78 criminal offenses, which overall 42.62 criminal offenses were committed per year. To put it in another way, there were average 1.53 criminal cases were investigated for each individual. We summed up and averaged the total terms and fines of sentence of 78 criminal offenses. The total terms of sentence were 10, 120 days, which were 27.72 years, and the total fines were 50,000 NT dollars. The average terms of each criminal offense case were 129.74 days and the average fines per case were 641.03 NT dollars.. 63.

(71) Table 16. The number of averted convicted offenses prior to methadone maintenance treatment 2007. 2006. 2005. 2004. 2003. 71.5. 51.13. 41.87. 36.24. 32.07. 1.19. 1.20. 1.19. 1.21. 1.21. 85.09. 61.36. 49.83. 43.85. 38.80. 42.47. 18.74. 7.21. 1.23. - 3.82. 76.34. 47.33. 42.95. 48.98. 36.87. 1.02. 1. 1.02. 1.02. 1.03. Property crimes Estimated completely investigated individuals Number of offenses committed per person Estimated number of offenses committed Number of averted investigated offenses Drug-related crimes Estimated completely investigated individuals Number of offenses committed per person. 64.

(72) Estimated number of offenses committed Number of averted investigated offenses. 77.87. 47.33. 43.81. 49.96. 37.98. 35.25. 4.71. 1.19. 7.34. - 4.64. Estimated completely investigated individuals = 599 × Probability heroin users commit a crime × Probability heroin user is arrested for committing a crime × Probability arrested heroin user is accused for committing a crime Number of offenses committed per person (Property crimes): Number of individuals of investigated completely / Number of criminal cases investigated completely (http://www.moj.gov.tw/site/moj/public/MMO/moj/stat/%20monthly/m4.pdf 地方法院檢察署偵字案件終結情形) Number of offenses committed per person (Drug-related crimes): Number of individuals accused of drug-related crimes / Number of cases accused of drug-related crimes Estimated number of offenses committed = Estimated completely investigated individuals × Estimated completely investigated individuals Number of averted investigated offenses = Estimated number of offenses committed prior to methadone maintenance treatment－average criminal offenses were committed per year during methadone maintenance treatment. 65.

(73) Table 17. The averted incarceration costs prior to methadone maintenance treatment 2007. 2006. 2005. 2004. 2003. 42.47. 18.74. 7.21. 1.23. - 3.82. (464,032.49,. (229,410.80,. (91,712.37,. (17,631.00,. (-58,401.12,. 15,496,641.45). 7,661,310.53). 3,062,789.12). 588,797.85). -1,950,340.31). 35.25. 4.71. 1.19. 7.34. - 4.64. (360,121.70,. (79,430.05,. (17,770.83,. (133,986.18,. (-70,527.85,. 12,026,478.71). 2,652,613.73). 593,467.59). 4,474,548.28). -2,355,319.54). Property crimes Number of averted investigated offenses Averted incarceration costs Drug-related crimes Number of averted investigated offenses Averted incarceration costs. Averted incarceration costs = Number of averted investigated offenses × Probability accused heroin user is convicted of crime × Probability convicted heroin user is incarcerated for crime × Average costs per day of incarceration × Average days of incarceration per heroin user. 66.

(74) In Table 17, the averted investigated property offenses gradually increased from 2003 to 2007. However, the averted investigated drug-related offenses increased in 2004 but decreased in 2005 and 2006, abruptly increased in 2007.. In the part of property crimes, the results show a clear relationship that as the year goes by the averted minimum and maximum incarceration costs. increase. However,. the. averted. minimum and. maximum. incarceration costs in the part of drug-related crimes appeared not to be the same, both the minimum and maximum criminal costs abruptly increased in 2004 then decreased in 2005, and soon markedly increased in the next two years. Also, the averted incarceration costs in property crimes section were much greater than the averted incarceration costs in drug-related crimes.. 67.

(75) Table 18. 2009 ~2011 Incarceration Costs per person per day in Taiwan. Total Legal budgets of Correctional Institutions Total Accommodated persons in Correctional Institutions Incarceration Costs per person per day. 2011. 2010. 2009. 6,708,331,000. 6,809,714,000. 6,816,383,000. 64,864. 65,311. 63,875. 283.35. 285.66. 292.37. Incarceration Costs per person per day＝（Total Legal budgets of Correctional Institutions ÷ Total Accommodated persons in Correctional Institutions）÷ 365. 68.

(76) In Table 18, the total legal budgets of correctional institutions of Ministry of Justice decreased gently from 6.8 billion in 2009 to 6.7 billion in 2011. Since the total accommodated persons in correctional institutions increased in 2010 but decreased in 2011, the average incarceration cost per person per day decreased by less than 10 NT dollars.. 69.

(77) 70.

(78) Chapter Five Discussions 5.1 Discussions. This study employed an implementation and non-implementation design of methadone maintenance treatment to estimate the cost and effectiveness of methadone maintenance treatment from patient’s perspective. We found out that methadone maintenance treatment is a cost-effective treatment in terms of the improvement of quality of life, the decrease of unemployment, the decline of criminal activities and incarceration costs. We will begin our discussions by reviewing the results and then turn to the interpretations and suggestions.. The finding from the study shows that methadone maintenance treatment contributed to improvement in four domains of quality of life (social relationship domain the most and physical domain the least) (Barnett, 1999; Ha, 2010; Masson, et al., 2004; Vanagas, et al., 2010) The heroin users had rather lower scores in psychological health domain at baseline of methadone maintenance treatment which means they might have negative feelings or lower self-esteem; however, after the methadone maintenance treatment, the psychological domain scores increased. They probably felt more confident and had positive feelings. In contrast, the heroin users had rather higher scores in 71.

(79) physical health domain at baseline of methadone maintenance treatment which means they might have energy and might not have pain. Because they had good enough physical health, after one-year methadone maintenance treatment, the physical domain scores only increased slightly. In terms of cost per QoL, heroin users have to pay the most money on physical domain in order to achieve one QoL gained and pay the least money on psychological domain in order to achieve one QoL gained (Vanagas, et al., 2010). It is recommended that if the government plans to pump money into addiction field, it should invest money in the psychological issues first such as providing counseling during the treatment.. The improvement of quality of life in this study implied that methadone maintenance treatment has positive effects on heroin users’ quality of life. Although we used a validated and generic QoL measurement instrument which is suitable for addiction field, there are still other outcomes that could not be precisely measured in quantitative analysis.. In terms of the employment, the percentage of unemployment decreased after one-year follow-up, from 45.4% to 41.8%, which was same as the variation of unemployment levels (Godfrey, et al., 2004). However, the percentage of unemployment in this study was substantially lower than the percentage of unemployment in Godfrey’s study (from 88% at baseline to 83% after one year), which our study 72.

(80) might overestimate because the participants who finished both baseline and one-year follow-up interviews (n=196) had higher compliance of taking methadone and higher probability of going off to work.. The wages per hour among participants who had jobs prior to and after one-year methadone maintenance treatment and also the sum of wages variation decreased. The more likely explanation is the 2008 financial crisis which resulted in the collapse of large financial institutions, national government and stock markets around the world and this directly affected the job markets in Taiwan, especially most heroin users are blue-collar workers. As soon as the recession occurred, the Blue Collars had more chances to run into the employment problems. In addition, we found out from the analysis that there were heroin users worked at part-time jobs tended to lose their jobs during recession. Therefore, providing methadone for heroin users helps them keep away from drugs. More important, providing behavioral skills training for seeking jobs would be beneficial for heroin users.. In the part of property crimes, the results show a clear relationship that as the year goes by the averted minimum and maximum incarceration costs increase. However, the averted minimum and maximum incarceration costs in the part of drug-related crimes appeared not to be the same, both the minimum and maximum criminal costs abruptly increased in 2004 then decreased in 2005, and soon markedly increased in the next two years. A partial explanation 73.

(81) for this may lie in the fact that the Taiwan government initiated the methadone maintenance treatment program in 2005, resulting in the alternative substitutes of using heroin.. Finally, we found that methadone patients had to pay at least 35,390 NT dollars annually in order to receive methadone treatment. It is an economic burden for heroin users to pay considerable amount of money on the medical treatment, let alone that they are a subculture or minority group in society. Therefore, in terms of the sustainability and financial accessibility of the treatment, policy makers and health professions should consider the treatment costs from patients’ perspectives, lower the threshold and provide subsidies tailored to the methadone maintenance treatment.. 5.2 Limitations. This study has a number of limitations. The first limitation concerns the research design that we chose which we compared the costs and effectiveness prior to and posterior to the methadone maintenance treatment. Although the costs and the effectiveness could be compared based on theories, there were still a variety of uncertainties which could not be avoided from a practical point of view.. Second, the study was limited to four hospitals in northern 74.