In economics, benefit-cost methods estimate the financial advantages of an intervention against its costs. These methods have already been extended to the economic evaluation of substance use disorder (SUD) treatment programs (1, 2). Drug Abuse Treatment Cost Analysis Program (DATCAP) and Client DATCAP are known tools for standardizing cost estimates of interventions from the perspective of a community and clients, respectively (3, 4). To evaluate the financial benefits of SUD treatment programs, standard frameworks have also been developed which consider the monetary benefits of treatment as prevented tangible costs of SUD following treatment interventions (5-8), such as saving health-related costs, alleviating the burden on the judicial system, and averting productivity reductions. Furthermore, reductions in clinical severity of SUD, as measured by the Addiction Severity Index (ASI), calculated in monetary values, have been applied to measure the benefits of treatment intervention (1, 2). French (9) classified the costs of addiction treatment into three groups: 1) cost of illness, emphasizing the costs of SUD as an illness such as reduced productivity of patients and the costs of their mortality and morbidity; 2) cost of averting behavior, focusing on costs of behaviors attempting to minimize the consequences of drug use, such as changing the place of residence or purchasing personal defense devices in high-prevalence areas; and 3) intangible costs of drug use, such as bystanders' pain and suffering, family disruption, and reduced public security and social welfare. These external costs can be measured by the utility valuation method. Evidently, the measurement of the benefits of SUD treatment must not be limited to avoided tangible costs but should be extended to intangible costs (10).
By definition, willingness to pay (WTP) is the maximum price that a person is willing to pay for an additional unit of applied product or service (11). Measurement of WTP, which was initially restricted to classical economic studies, has gradually expanded to public health (12) and drug use (13) policy studies. Nowadays, WTP is an appropriate method for measuring SUD treatment benefits (14-16). It measures the intangible outcomes of treatments such as improved social safety and individual well-being (12). Most WTP studies have focused on taxpayers’ characteristics (15, 16) or family viewpoints (17-19), whereas Cartwright (5) points out the need for considering treatment value from consumers' point of view. Moreover, while most WTP studies have focused on demographic factors, the association between addiction severity and WTP has rarely been measured (20, 21).
In 2017, the number of drug users in Iran was estimated at 2,800,000, among whom 1,300,000 were registered clients on methadone maintenance therapy (MMT) or in abstinence-based residential services (RFs) (22). Contrary to MMT as a self-referral and voluntary service, residential services in Iran are often sought under the force of the family or by court referral (23). The six-month relapse rate for RFs has been reported to be as high as 85% (24); for the MMT service, it is between 20 and 69%, with an average of 30% (25-28). In Iran, treatment costs in both programs are either paid by clients themselves or by their families with no public funding or support. In our 2017 study (29), we examined WTP from cost payers’ point of view and showed that attitudes towards different aspects of drug use and its treatment are important when deciding on paying for addiction treatment. In this research as the secondary analysis from the same study, we assessed the relationship of cost payers' income and SUD severity with their tendency to pay for treatment. Since economic studies on SUD treatment in Iran are limited (30, 31), we aimed to better understand the economics of drug use treatment in an Iranian sample by measuring WTP for addiction treatment.

3. Based on the contingency valuation (CV) method (15), we first presented the respondents with a variety of fixed payment ceilings (cost per day) for MMT clinics and RFs to select from. Then, we asked them whether the price of the service differed from the current one and what price would be the maximum they would pay above which they would give up their treatment. Next, the respondents were directly asked an open question regarding their maximum WTP to double-check their declared WTP. As noted before, in most cases, the actual person who paid for the treatment in RFs was a close family member, while it was the patients themselves in MMT clinics.

Data Collection
The questionnaires were completed and graded by a trained interviewer during a face-to-face interview.
Statistical Analysis
The two samples of MMT clinics and RFs could not be analyzed together because services in the two facilities were different in nature, i.e., outpatient pharmacological intervention vs. residential non-pharmacological service, respectively. Furthermore, in most cases, the actual person who paid for the treatment in RFs was a family member, while in outpatient clinics, it was the patients themselves. Therefore, considering different services and different supply and demand markets in the two samples, we recruited two separate regression models. For the non-parametric data, we also used Kruskal-Wallis and Mann-Whitney U statistical tests to examine differences between the opposing groups. To analyze the predictors of WTP, we used cost payers’ income and the addiction severity scales as predictor variables. After checking the absence of collinearity in the variables, by using a backward regression equation, we calculated the best model specification.

Discussion

We designed a study based on the contingency valuation approach to apply WTP for addiction treatment as a practical method in the evaluation of intangible costs of drug use and addiction treatment monetary benefits (18). In a previous study (29), we showed that cost payers' attitudes towards different aspects of drug use and its treatment play an important role in WTP for addiction treatment. In this research, as a secondary analysis from the same study, we considered patients' addiction severity and cost payers' economic status as other deciding factors.
This study showed that compared to the MMT clinics, WTP was higher in residential programs. Compared to MMT clients, those admitted to RFs were significantly younger, more educated, and had a higher RUI. In terms of financial status, RF clients were often dependent on their families who paid for the treatment. However, at MMT clinics, while the average client had an employment-based income, the household family income was lower. Therefore, the significantly higher WTP for drug use treatment in RFs, compared to MMT clinics, might simply be regarded as a function of income. In other words, families with higher incomes prefer to send their addicted members to RFs to ease their own minds, even for a short time (23).
Given the nature of single payment upon admission to RFs, compared to the monthly payment in MMT clinics, it appears that families preferred the more expensive single-shot strategy of detoxification as a magic solution to drug use despite the higher relapse rate in this model. This preference for short-term detoxification treatment is similar to previous findings of a recent Iranian study on RFs (23).
Despite the reasonable expectation that LTUI should follow the same pattern as RUI, the lack of a significant difference in LTUI between the two groups may be attributed to the lower age of clients in RFs. The higher RUI in RFs clients may reflect poly-drug use with more high-risk use in RFs patients that results in more serious harm in a shorter period. However, as this harmful pattern lasted shorter in generally younger RFs clients, when compared to long-term traditional use of opioids with a less risky pattern in MMT clients, the harm-related difference became insignificant (represented in the non-significant difference of LTUI) between the two groups. Based on the average age of the MMT clients which was higher than that of the other group, one could add the possibility that individuals with a high-risk pattern of drug use have failed to survive to more advanced ages and, therefore, MMT clients should naturally have a low-risk pattern of drug use and lower LTUI.
As revealed by our regression model for MMT clinics, average monthly income had a significant association with WTP, followed by LTUI. Furthermore, a negative significant association between clients’ LTUI and their houses' surface area was observed in MMT clients. In other words, the longer the clients’ history of drug use, the worse their economic status. Evidently, in this study, we used the average surface area of the house as a representative of economic indexes; however, the use of overcrowding—the ratio of persons to floor space in square feet (35)—could also have been applied as a better representative. Our data showed that in MMT clinics, when the clients are paying for the treatment themselves and not through a third party, WTP is much more dependent on economic status. This conclusion is in line with former studies (36, 37) which found affordability to be a key determinant of retention in MMT. As shown in Table 3, the average monthly income of the MMT clients was $312.50—almost equal to the official monthly minimum wage for the same year ($282.50). Besides, according to the Statistical Center of Iran, the average nominal cost of Iranian households in 2017 was $686.50 compared to the $764.47 nominal income (38). One can, therefore, conclude that since there is almost no margin left for treatment costs ($32 per month or 14% of the minimum wage) in the case of MMT patients, this could be a reasonable explanation for the cardinal role of economic status in the WTP of this group of patients.
Regarding items measured in ASI, drug use status, legal status, and medical status sub-scales have been shown to be key elements for calculating the financial benefits of drug use treatment (2). We found that intangible costs such as deteriorated legal and medical status were associated with a lower WTP for treatment. Our findings are consistent with those of previous studies in the United States (20) and Vietnam (39) on the association between clients’ health conditions and WTP for methadone treatment.
Our findings revealed that in RFs, WTP is mainly associated with the cost payers' average income, which is consistent with a previous argument (16), justifying the role of income with more maneuverability in paying for drug use treatment costs. In a previous study (40) in Norway, after excluding two groups of respondents (those not believing in the efficacy of treatment and protest zeroes who believed the government is responsible for paying for addiction treatment), the income elasticity of WTP was calculated as 0.75; this means that for every percent of increase in income, the WTP for the addiction treatment was raised by 0.75 %. Moreover, it has been shown (23) that desperate families who pay for treatment might use RFs as a solution to improve their own mental health. In fact, if they can afford the costs, they use RFs as a means of secluding the drug user from the community and the family. Therefore, WTP in RFs appears to be a factor of short-term drug use severity with no attention to the effectiveness of the program. Although there was no correlation between drug use severity and WTP in our sample from MMT clinics, both RUI and LTUI were correlated with WTP in RFs. We believe that the homogeneity of the drug use index in MMT clinics obscured such a correlation, while a wide variety of drug types and poly-drug use in RFs was a factor that led to such a correlation. Furthermore, the absence of such an association in our regression model might be due to the effect of other factors not measured in our study. Since we reported the first estimation of WTP for SUD in Iran, we believe our findings should be re-examined by future studies.
Assuming that only 15% of individuals with SUD undergo treatment each year, it is of policy-making importance to measure WTP for the priority assessment of subsidization or insurance coverage of different treatment programs. Using the price elasticity of demand for MMT and WTP for the service, Bishai (20) provided a model for allocating optimal subsidization of MMT and suggested the necessity of higher subsidy allocation for clients with lower WTP. According to our findings where WTP was lower in MMT settings, MMT should be prioritized over detoxification treatments whenever an incentive policy for addiction treatment is considered (such as subsidization or insurance premium). Other factors such as patients' drug use status, legal status, and health status, as well as cost payers' economic status, should also be taken into account. The other point to consider is that according to this study, WTP was higher in residential programs, yet residential treatment is typically shorter. It would be advised to look not at WTP/day at a specific time point, but at WTP over a longer period or a lifetime WTP.

 

Limitation

Our sample was recruited from the Tehran metropolitan area; therefore, the results should be generalized with caution. Because of the payment system for drug use treatment in Iran in which patients and their families pay for treatment, we regarded WTP by people in treatment and third-party payers to be of the same value, an approach that needs further research.

 

Conclusion

WTP is a practical tool for evaluating SUD treatment programs. As long as addiction treatment programs follow conventional market rules where payment is out-of-pocket, cost payers' economic status plays a key role in preparedness for purchasing treatment services. Patients' severity of SUD could be another key factor determining the WTP for treatment. To understand the nature of the illegal drug market and its supply and demand for addiction treatment programs, the WTP for treatment could be applied in drug policy-making (20, 41) as a recommended research direction.

 

Acknowledgements

TThis study was a secondary analysis of a pervious study (29) as part of the “Addiction Studies” PhD dissertation titled “Modeling Economics of Drug Use in Iran” at the School of Advanced Technologies, Tehran University of Medical Sciences. The authors would like to thank all the participants for their cooperation.

 

Authors' contributions

Samaneh Ahmadian_Moghadam: Writing - Original Draft. Formal analysis, Investigation, Resources, Data Curation, Project administration, Software, Validation.
Emran M Razaghi: Writing - Review & Editing, Visualization, Supervision, Project administration, Conceptualization, Methodology, Funding acquisition.
Ali Mazyaki: Conceptualization, Methodology, Software.

 

Ethics approval and consent to participate

The researchers pledged the confidentiality of information. The respondents filled out and signed a written consent form. The ethics standards of the study were approved by the Tehran University of Medical Sciences IRB (Code # 9121457002). The authors declare that they have no conflicting interests regarding this study.

 

Availability of data and materials

The datasets used and/or analyzed in the current study are available from the corresponding author on reasonable request

 

Funding

This project was funded by the Tehran University of Medical Sciences.

 

Conflicts of Interest

The authors declare that they have no competing interests.