G Model DRUPOL-1714; No of Pages International Journal of Drug Policy xxx (2016) xxx–xxx Contents lists available at ScienceDirect International Journal of Drug Policy journal homepage: www.elsevier.com/locate/drugpo Research paper Economic vulnerability of methadone maintenance patients: Implications for policies on co-payment services Bach Xuan Tran a,b,*, Huong Thu Thi Phan c, Long Hoang Nguyen a,d, Cuong Tat Nguyen e, Anh Tuan Le Nguyen e, Tuan Nhan Le f,g, Carl A Latkin b a Institute for Preventive Medicine and Public Health, Hanoi Medical University, Hanoi, Viet Nam Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, United States Authority of HIV/AIDS Control, Ministry of Health, Hanoi, Viet Nam d School of Medicine and Pharmacy, Vietnam National University, Hanoi, Viet Nam e Institute for Global Health Innovation, Duy Tan University, Da Nang, Viet Nam f Foreign Trade University, Hanoi, Viet Nam g Center for AIDS Control, Hanoi Department of Health, Hanoi, Viet Nam b c A R T I C L E I N F O A B S T R A C T Article history: Received 18 October 2015 Received in revised form 29 November 2015 Accepted 21 January 2016 Background: Co-payment for methadone maintenance treatment (MMT) services is a strategy to ensure the financial sustainability of the HIV/AIDS programs in Vietnam In this study, we examined health services utilization and expenditure among MMT patients, and further explored factors associated with catastrophic health expenditure among affected households Methods: A multi-site cross-sectional study was conducted among 1016 patients in two epicentres: Hanoi and Nam Dinh province in 2013 Results: Overall, 8.2% and 28.7% respondents used inpatient and outpatient health care services in the past 12 months apart from receiving MMT There were 12.8% respondents experiencing catastrophic health expenditure given MMT is provided free-of-charge, otherwise 63.5% patients would suffer from health care costs MMT integrated with general health or HIV services may encourage health care services utilization of patients Patients, who were single, lived in the rural, had inpatient care and reported problems in Mobility were more likely to experience catastrophic health expenditure than other patient groups Conclusions: The health care costs are still financially burden to many drug users and remained over the course of MMT that implies the necessity of continuous supports from the program Scaling-up and decentralizing integrated MMT clinics together with economic empowerments for treated drug users and their families should be prioritized in Vietnam ß 2016 Elsevier B.V All rights reserved Keywords: Methadone Integrative services Health services Costs Catastrophic Vietnam Introduction People who inject drugs (PWIDs) are a key population at increased risk of HIV are considered a major driver of the explosion of HIV in Asia countries (Sharma, Oppenheimer, Saidel, Loo, & Garg, 2009) Recent data estimates that 4.5 million of PWIDs out of 13 million drug users (DUs) live in this region The financial burden of drug addiction involves not only the costs for this risk behavior but also huge costs for health care services and loss of productivity (Nguyen, Tran, Tran, Le, & Tran, 2014; Tran et al., 2012a, 2013a; * Corresponding author at: Hanoi Medical University, Viet Nam Tel.: +84 982228662 E-mail addresses: bach@hmu.edu.vn, bach@jhu.edu (B.X Tran) Tran, 2013) Murphy and Scott defined the ‘‘economic vulnerability as the exposure of a household to exogenous shocks related to the wider global economic crisis and subsequent adoption of austerity policies and the potential for diminishing life satisfaction and quality of life’’ (Enda Murphy, 2014) In this study, we focused on the burden of health care costs that affected PWIDs and their households In developing countries such as Vietnam, drug users may spend much higher than average household monthly income, which results in economic burden on households (Tran et al., 2013a) In addition, it has also been a barrier for those who acquired HIV to access and use health care services (Tomori et al., 2014) Illicit drug use is known to reduce significantly adherence to and outcomes of antiretroviral treatment Therefore, opioid substitution treatment for DUs plays an indispensable role in international HIV/AIDS prevention strategies http://dx.doi.org/10.1016/j.drugpo.2016.01.017 0955-3959/ß 2016 Elsevier B.V All rights reserved Please cite this article in press as: Tran, B X., et al Economic vulnerability of methadone maintenance patients: Implications for policies on co-payment services International Journal of Drug Policy (2016), http://dx.doi.org/10.1016/j.drugpo.2016.01.017 G Model DRUPOL-1714; No of Pages B.X Tran et al / International Journal of Drug Policy xxx (2016) xxx–xxx Currently, methadone maintenance treatment (MMT) has been used as an effective therapy for people dependent on opioids (Burns et al., 2015; Tran et al., 2012b, 2012c, 2012d; Zhou & Zhuang, 2014) In Vietnam, patients have been receiving MMT freeof-charge, however, since international fundings for HIV/AIDS is decreasing rapidly, resource mobilization using co-payment is considered (Tran, 2013; Tran et al., 2012c; Tran, Nguyen, Phan, Nguyen, & Latkin, 2015) Finding from prior reviews demonstrated that MMT minimized the demands of opioid use; crime activities, HIV-related risk behaviors and diseases, as well as promoted HIV/ AIDS care services access and improved quality of life (Fareed et al., 2011; Pilgrim, McDonough, & Drummer, 2013; Sun et al., 2015; Wang, Wouldes, & Russell, 2013; Weimer & Chou, 2014) Among drug users living with HIV, MMT also helped reduce healthcare services use and out-of-pocket (OOP) health expenditure (Tran & Nguyen, 2013) Thus, implementing and scaling-up MMT program has been considered a cost-effective intervention in both developed and developing countries (Hsiao et al., 2015; Keshtkaran, Mirahmadizadeh, Heidari, & Javanbakht, 2014; Roncero et al., 2015; Tran et al., 2012c, 2012d, 2012e; Tran, Nguyen, Phan, et al., 2015) The Vietnam Ministry of Health set a target for providing MMT services for 80,000 drug users by 2015 (Nguyen, Nguyen, Pham, Vu, & Mulvey, 2012) During the period of this study, some provinces, for instance, Hai Phong, there were few MMT sites applied the co-payment schemes that patients pay for averagely US$0.5 per day For other health care services, there was no subsidy and patients or their health insurers are supposed to pay their OOPs money for the services (Tran et al., 2013a) The public health system in Vietnam is organised into three levels: central, provincial and grassroots (Tran, Van Hoang, & Nguyen, 2013) Primary medical care services are provided throughout the country, however, high reliance on out-of-pocket financing for health exposes households to potential catastrophic expenditures and creates inequality in access to care (Do, Oh, & Lee, 2014) Current estimates showed OOP payments accounted for 30– 70% total health expenditure in various settings (Hoang Lan, Laohasiriwong, Stewart, Tung, & Coyte, 2013; Nguyen et al., 2012b; Nguyen, Ivers, Jan, & Pham, 2015; Pham, Kizuki, Takano, Seino, & Watanabe, 2013; Tran et al., 2013a) The percentage of public expenditure in health in the total health spending in Vietnam was estimated to be 40% (Mitra, Palmer, Mont, & Groce, 2015) Model to deliver MMT services varies across settings such as stand-alone or integrating with other health care services (Tran, Nguyen, Phan, et al., 2015) When freestanding model emphasizes the role of confidential services, combination programs facilitate the variety of health service utilized by drug users (Tran et al., 2012a, 2012e) The linkage of MMT service and general health care services were mentioned in the late 1980s, since previous reports suggested the high prevalence of co-morbidities, low frequency of health care use and daily clinic visit to uptake MMT among DUs (Weddington, 1990) By combining different components of health care service into a single site; or providing referral between them, these models give a chance to address the unmet needs of DU for medical services (Tran et al., 2012b; Tran, Nguyen, Phan, et al., 2015) The effectiveness of linked services have been well documented, including promote health care utilization, improve health outcome and treatment adherence (Tran & Nguyen, 2013; Tran et al., 2012b; Tran, Nguyen, Phan, et al., 2015) This model also reduces the health care cost of communities as well as the duplication of services and their administration cost (Kresina, Bruce, Lubran, & Clark, 2008), therefore, it may improve the efficiency of the service delivery system However, the performance of different integrative models in diverse settings as well as patients’ responses has not been examined In Vietnam, there has been evidence that services quality and socio-cultural and economic factors, rather than geographical barriers, may affect the use of health services in both general and HIV-related populations (Duong, Binns, & Lee, 2004; Nguyen et al., 2015b, 2015c; Tran, Ohinmaa, Nguyen, Nguyen, & Nguyen, 2011; Tran, Van Hoang, et al., 2013) Thus, comparing across models for MMT delivery can provide insights for improving the efficiency of the MMT program HIV epidemic in Vietnam is recognized in a concentrated stage, which is primarily driven by unsafe sex with commercial sex workers and illicit drug injection (Tran et al., 2012a; Tran, Nguyen, & Pham, 2014) It is estimated that about 180,000 people are using illicit drug in the country by 2012, of those, 20–50% were contracted HIV/AIDS (National Institute of Hygiene and Epidemiology, 2011) The MMT program has been prioritized in the National HIV/AIDS Strategy and rapidly scaled up nationwide at a daily cost of US$1 per patient (Tran, 2013; Tran et al., 2012e) Recent data reported an approximate of 15,500 DUs enrolling for treatment, with 26.8% were PWIDs (Tran, 2013) Vietnam Ministry of Health targeted covering 80,000 DUs on the program in 2015 The impact of MMT on health services utilization and OOP payments in Vietnam are mentioned in the previous investigation analysis, however, the sample included only those living with HIV/AIDS (Tran & Nguyen, 2013) To date, none of the literature analysed health services utilization and expenditure of drug users over the course of MMT or examined the role of different service delivery models on these outcomes of interest This study assessed health service use and OOP health spending of MMT patients in MMT clinics with and without other general health or HIV/AIDS services; and further explored factors associated with catastrophic health expenditure among this patient group Materials and methods Study settings and sampling A multi-site cross-sectional study was conducted in two Vietnamese epicentres: Hanoi and Nam Dinh from January to August 2013 (Tran, Nguyen, Phan, et al., 2015) The selection of provinces were purposive, in consultation with the Vietnam Authority of HIV/AIDS Control, that included a setting with new MMT sites (Nam Dinh) and a setting with other sites since the first national pilot (Ha Noi) (Tran et al., 2012a, 2012e) These areas were amongst those with the greatest HIV epidemic in northern Vietnam Five selected clinics were classified to two delivery models comprised: stand-alone (Provincial AIDS Centre – Nam Dinh) and integrated into general healthcare facilities (Xuan Truong district health center in Nam Dinh provinces; Tu Liem and Long Bien district health centers and Ha Dong regional polyclinic in Hanoi) At these clinics, patients have been receiving MMT free-ofcharge The selection was based on following criteria: (1) These clinic had been providing MMT services; (2) including provincial-, regional- and district-level clinics and (3) having sufficient patients for the study In this sample we also considered the involvement of rural and urban sites which located in the rural district (Xuan Truong) or urban cities (Table 1) The inclusion criteria for participants were: (1) respondents were 18 years or older; (2) enrolling MMT programs or having requested to participate the program; (3) agreeing to sign in written informed consents and (4) having capacity to answer a 30–45 interview All patients met criteria and went to selected clinics during study period were invited to participate in the study A convenient sample of 1016 participants was recruited in the study Measures and instruments Data was collected by master students and medical doctors with extensive experience with IDU and MMT Patients were Please cite this article in press as: Tran, B X., et al Economic vulnerability of methadone maintenance patients: Implications for policies on co-payment services International Journal of Drug Policy (2016), http://dx.doi.org/10.1016/j.drugpo.2016.01.017 G Model DRUPOL-1714; No of Pages B.X Tran et al / International Journal of Drug Policy xxx (2016) xxx–xxx Table Study settings and sample size Level Settings Site Name Type of services Sample size Province District (rural) District (urban) District (urban) District (urban) Nam Dinh City Xuan Truong District Tu Liem District Long Bien District Ha Dong District Provincial AIDS Centre (PAC) District Health Centre (DHC) District Health Centre (DHC) District Health Centre (DHC) Regional Polyclinic (RPC) MMT + VCT MMT + VCT + ART + GH MMT + VCT + ART + GH MMT + VCT + ART + GH MMT + GH 270 151 201 184 210 VCT: voluntary HIV counseling and testing ART: antiretroviral treatment; GH: General healthcare (apart from ART, VCT, MMT) invited to a confidential room, which was designated for face-toface interview Their self-reported information was collected using a structured questionnaires comprising: socioeconomic status, health status, health-related quality of life household monthly expenditure, health service utilization and OOP of healthcare expenditure (Tran & Nguyen, 2013; Tran et al., 2011; Tran, Ohinmaa, & Nguyen, 2012) Health related quality of life was measured using the EuroQol-five dimensions-five level (EQ-5D-5L) which has been widely used in HIV studies in Vietnam (Tran et al., 2011, 2012f; Tran, Nguyen, Ohinmaa, Maher, Nong, & Latkin, 2015) The descriptive system includes five domains: Mobility, Self-care, Usual activities, Pain/Discomfort and Anxiety/Depression with five levels of response: no problems, slight problems, moderate problems, severe problems, and extreme problems, giving 3125 health states with respective single indexes Household monthly expenditure was computed including two components: recurring expenses in the last 30 days (e.g food, utility, rent, education and others) and non-recurring expenses in the last 12 months (e.g construction, health care, furniture, travels, community events and others) To calculate OOP health expenditure per capita, the sum of healthcare expenses was divided to the total number of family members (Tran et al., 2013a; Nguyen et al., 2014; Tran, Nong, Maher, Nguyen, & Luu, 2014) Catastrophic health expenditure was defined as the proportion of health care payments over 10% of the total households’ expenditure (Lu, Chin, Li, & Murray, 2009; Saito et al., 2014; Wagstaff & van Doorslaer, 2003) Health service utilization of respondents included inpatient and outpatient care in the last 12 months (regular outpatient clinic visits for ARV and MMT medications were excluded) This measure was frequently applied in prior national surveys (Tran, 2012; Tran, 2013; Tran & Nguyen, 2012; Tran et al., 2011, 2013a, 2013c; Tran, Nguyen, Nguyen, Hoang, & Hwang, 2013; Tran, Nguyen, Do, Nguyen, & Maher, 2014) Cost per inpatient and outpatient visit were estimated by recalling the expenditure of patients in the last use of health services Interviewers guided patients to list all services and procedures of the last health care, and associated cost items Patients then estimated the costs by activity while the interviewers triangulated them with the total costs Unit costs comprised two elements: (1) medical expenditure (non-ARV and MMT drugs, lab tests, hospital fees and other) and (2) non-medical spending (transportations, accommodation and special meals if any) Total OOP payments for healthcare services of MMT patients were estimated by multiplying the frequency of healthcare service utilization by the average costs per visit Statistical analysis Data was analysed using STATA version 12.0 To examine the difference of those characteristics between delivery models, Student t-test and Chi-squared test were used Household monthly income, expenditure, the rate of healthcare services utilization and costs per inpatient and outpatient visits were presented in both means (95% CI) Multiple logistic regression was used to determine correlates of experiencing catastrophic OOP health expenditure and having any inpatient or outpatient health service use in the past year Multiple linear regression was used to determine correlates of OOP health expenditure (unit: 1000 Vietnam dong) We applied a stepwise forward model building strategy that selected variables based on the log-likelihood ratio test at a p-value