Corruption and informal sector households' participation in health insurance in Sierra Leone.

Lack of credibility and trust in fund managers has been highlighted as one of the key reasons why people do not join health insurance schemes in low- and middle-income countries, especially in Africa. This work investigates the impact of corruption on households' willingness to participate and pay for health insurance in Sierra Leone. A discrete choice experiment (DCE) method was used to elicit households' willingness to participate in a health insurance scheme with different attributes. The data were collected from 1458 representative households working in the informal sector of the Northern and Western regions. We explore the relationship between household characteristics and experienced (respectively, perceived) corruption with binary and ordered logit models. We use a Mixed Logit model to estimate the association between corruption and participation in a Health Insurance Scheme (HIS) and households' willingness to pay for a HIS. We find that corruption decreases participation in a public HIS and the willingness to pay for it. Our results highlight the perverse spillover effects of corruption. Not only does corruption hinder the effectiveness of healthcare systems and, thus, worsen health outcomes. It also undermines the willingness to pay for them, jeopardizing the sustainability of healthcare systems in the countries that need them most.

Unmet Parenthood Goals, Health-Related Quality of Life and Apparent Irrationality: Understanding the Value of Treatments for Infertility.

An increasing number of prospective parents are experiencing infertility along with associated negative impacts on mental health and life satisfaction that can extend across a network of individuals and family members. Assistive reproductive technologies (ART) can help prospective parents achieve their parenthood goals but, like any health technology, they must demonstrate acceptable 'value for money' to qualify for public funding. We argue that current approaches to understanding the value of ART, including quality-adjusted life-year (QALY) gains based on changes in health-related quality of life (HRQOL) and, more often, cost per live birth, are too narrow to capture the full impact of unmet parenthood goals and ART. We see a fundamental disconnect between measures of HRQOL and broader measures of wellbeing associated with met and unmet parenthood goals. We also suggest that simple concepts such as 'patient' and 'carer' are of limited applicability in the context of ART, where 'spillovers' extend across a wide network of individuals, and the person receiving treatment is often not the infertile individual. Consideration of individual and societal wellbeing beyond HRQOL is necessary to understand the full range of negative impacts associated with unmet parenthood goals and the corresponding positive impacts of successful ART. We suggest moving towards a wellbeing perspective on value to achieve a fuller understanding of value and promote cross-sector allocative efficiency.

Discontinuation of performance-based financing in primary health care: impact on family planning and maternal and child health.

Performance-based financing (PBF) is advocated as an effective means to improve the quality of care by changing healthcare providers' behavior. However, there is limited evidence on its effectiveness in low- and middle-income countries and on its implementation in primary care settings. Evidence on the effect of discontinuing PBF is even more limited than that of introducing PBF schemes. We estimate the effects of discontinuing PBF in Egypt on family planning, maternal health, and child health outcomes. We use a difference-in-differences (DiD) model with fixed effects, exploiting a unique dataset of six waves of spatially constructed facility-level health outcomes. We find that discontinuing performance-based incentives to providers had a negative effect on the knowledge of contraceptive methods, iron supplementation during pregnancy, the prevalence of childhood acute respiratory infection, and, more importantly, under-five child mortality, all of which were indirectly targeted by the PBF scheme. No significant effects are reported for directly targeted outcomes. Our findings suggest that PBF can induce permanent changes in providers' behavior, but this may come at the expense of non-contracted outcomes.

The economic burden of congenital Zika Syndrome in Brazil: an overview at 5 years and 10 years.

BACKGROUND: The aim of this paper is to estimate the economic burden of children with congenital Zika Syndrome (CZS) in Brazil over 5-10 years. METHODS: We conducted a modelling study based on data collected in a case-control study in Brazil, including children with CZS (cases) and typically developing children (controls), born in 2015 and 2016. In total, 484 participants were recruited in two sites, Recife and Rio de Janeiro. Social and economic information was collected in a survey from the carers of cases and controls, and detailed healthcare utilisation was recorded for each child in the Rio de Janeiro cohort prospectively in a database. We used this information to estimate the cost per child with severe, moderate and no CZS and incremental cost per child with severe and moderate versus no CZS from a disaggregated societal perspective. These estimates were incorporated into an economic burden model to estimate the incremental burden of the CZS epidemic in Brazil over 5 years and 10 years. FINDINGS: The societal cost per child with severe CZS was US$50 523 to 10 years of age (born in 2015 and 2016), substantially higher than the costs for moderate CZS (US$29 283) and without CZS (US$12 331). The incremental economic burden of severe versus no CZS in Brazil over 10 years was US$69.4 million from the household and US$129.0 million from the government perspective. For moderate CZS, these figures amounted to US$204.1 million and US$86.6 million. Over 10 years, 97% of the total societal economic cost of severe CZS is borne by the government, but only 46% for moderate CZS. INTERPRETATION: The economic burden of CZS is high at the household, provider and government levels. The compensatory government payments helped to alleviate some of the additional costs incurred by families with a child qualifying for the disability benefits, and could be scaled to include the children with moderate CZS.

A pilot economic evaluation of a feasibility trial for SUpporting wellbeing through PEeR-Befriending (SUPERB) for post-stroke aphasia.

OBJECTIVES: To explore the feasibility of a full economic evaluation of usual care plus peer-befriending versus usual care control, and potential cost-effectiveness of peer-befriending for people with aphasia. To report initial costs, ease of instruments' completion and overall data completeness. DESIGN: Pilot economic evaluation within a feasibility randomised controlled trial. SETTING: Community, England. PARTICIPANTS: People with post-stroke aphasia and low levels of psychological distress. INTERVENTION: All participants received usual care; intervention participants received six peer-befriending visits between randomisation and four months. MAIN MEASURES: Costs were collected on the stroke-adapted Client Service Receipt Inventory (CSRI) for health, social care and personal out-of-pocket expenditure arising from care for participants and carers at 4- and 10-months post-randomisation. Health gains and costs were reported using the General Health Questionnaire-12 and the EQ-5D-5L. Mean (CI) differences for costs and health gains were reported and uncertainty represented using non-parametric bootstrapping and cost-effectiveness acceptability curves. RESULTS: 56 participants were randomised. Mean age was 70.1 (SD 13.4). Most (n = 37, 66%) had mild and many (n = 14; 25%) severe aphasia. There was ≥94% completion of CSRI questions. Peer-befriending was higher in intervention arm (p < 0.01) but there were no significant differences in total costs between trial arms. Peer-befriending visits costed on average £57.24 (including training and supervision costs). The probability of peer-befriending being cost-effective ranged 39% to 66%. CONCLUSIONS: Economic data can be collected from participants with post-stroke aphasia, indicating a full economic evaluation within a definitive trial is feasible. A larger study is needed to demonstrate further cost-effectiveness of peer-befriending.

Adjustment with aphasia after stroke: study protocol for a pilot feasibility randomised controlled trial for SUpporting wellbeing through PEeR Befriending (SUPERB).

BACKGROUND: Despite the high prevalence of mood problems after stroke, evidence on effective interventions particularly for those with aphasia is limited. There is a pressing need to systematically evaluate interventions aiming to improve wellbeing for people with stroke and aphasia. This study aims to evaluate the feasibility of a peer-befriending intervention. METHODS/DESIGN: SUPERB is a single blind, parallel group feasibility trial of peer befriending for people with aphasia post-stroke and low levels of psychological distress. The trial includes a nested qualitative study and pilot economic evaluation and it compares usual care (n = 30) with usual care + peer befriending (n = 30). Feasibility outcomes include proportion screened who meet criteria, proportion who consent, rate of consent, number of missing/incomplete data on outcome measures, attrition rate at follow-up, potential value of conducting main trial using value of information analysis (economic evaluation), description of usual care, and treatment fidelity of peer befriending. Assessments and outcome measures (mood, wellbeing, communication, and social participation) for participants and significant others will be administered at baseline, with outcome measures re-administered at 4 and 10 months post-randomisation. Peer befrienders will complete outcome measures before training and after they have completed two cycles of befriending. The qualitative study will use semi-structured interviews of purposively sampled participants (n = 20) and significant others (n = 10) from both arms of the trial, and all peer befrienders to explore the acceptability of procedures and experiences of care. The pilot economic evaluation will utilise the European Quality of life measure (EQ-5D-5 L) and a stroke-adapted version of the Client Service Receipt Inventory (CSRI). DISCUSSION: This study will provide information on feasibility outcomes and an initial indication of whether peer befriending is a suitable intervention to explore further in a definitive phase III randomised controlled trial. TRIAL REGISTRATION: ClinicalTrials.gov identifier NCT02947776, registered 28th October 2016.

Economic evaluation of the use of non-vitamin K oral anticoagulants in patients with atrial fibrillation on antiplatelet therapy: a modelling analysis using the healthcare system in the Netherlands.

AIMS: Non-vitamin K oral anticoagulants (NOACs) have consistently demonstrated superior efficacy in terms of stroke prevention and safety in terms of bleeding over vitamin K antagonist (VKA) in patients with non-valvular atrial fibrillation (AF). The potential use of NOACs in AF patients requiring antiplatelet therapy (APT) has only been assessed in small meta-analyses reporting consistent benefits of NOACs over VKAs. However, the prescription costs of NOACs are higher than those of VKAs. The aim of his study was to estimate the cost-effectiveness (CE) of NOACs compared to VKAs in patients with non-valvular AF also requiring APT with the Dutch healthcare system used as a surrogate of many European healthcare systems. METHODS AND RESULTS: A decision tree was constructed to analyse the CE of NOACs compared to VKAs in patients with non-valvular AF with an indication for APT over a horizon of 1 year. Beside the base-case analysis, univariate probabilistic sensitivity and two sensitivity analyses were performed: first, we assessed the impact of VKA home monitoring; second, we varied the NOACs price assuming patent expiration. Use of NOACs instead of VKA is associated with a health gain of 0.0171 quality-adjusted life years (QALYs) and with an incremental cost of €357, resulting in an incremental cost-effectiveness ratio of €20 919, which is almost equal to the generally accepted CE threshold of €20 000 used in the Netherlands. The probability that NOACs are cost-effective at a conservative willingness-to-pay threshold of €20 000 per QALY was 50%. Introducing home monitoring increased VKAs costs so much that NOACs became the dominant option (less costly and more effective). Price drops associated to patent expiration of NOACs increased its CE. CONCLUSION: This analysis suggests that the use of NOACs is a cost-effective alternative of VKAs in patients with AF needing APT. Our findings in the Netherlands healthcare system are probably consistent with other European populations.

The suitability of a DRG casemix system in the Maltese hospital setting.

The healthcare system in Malta is financed through global budgets and healthcare is provided free at the point of use. This paper is a first attempt to examine the feasibility of introducing a Diagnosis Related Groups casemix system for Malta, not necessarily for payment and funding purposes, but as a tool to help describe, manage and measure resource use. This is particularly challenging in view of the constraints and characteristics of a small state country. The study evaluates the applicability of the MS-DRG (Version 27.0) Grouper to describe acute hospital activity on the island. The classification of 151,615 admissions between 2009-2011 resulted in 636 DRG categories. Around half of these DRGs accounted for 99% of the total activity at the hospital, while 296 DRG categories had fewer than 15 cases over the period. Patient length of stay is used to explain resource use and the Coefficient of Multiple Determination obtained was of 0.19 (improving to 0.25 when a number of trimming algorithms were applied). A good proportion of the resulting DRGs had a Coefficient of Variation, which indicates a low degree of variability within the obtained DRG groups. This presents good evidence to support the introduction of a DRG system in Malta particularly in view of the recent drive towards more public-private partnerships and legislation on cross-border patient treatment.

Consequences of Biomarker Analysis on the Cost-Effectiveness of Cetuximab in Combination with FOLFIRI as a First-Line Treatment of Metastatic Colorectal Cancer: Personalised Medicine at Work.

BACKGROUND: Therapies may be more efficacious when targeting a patient subpopulation with specific attributes, thereby enhancing the cost-effectiveness of treatment. In the CRYSTAL study, patients with metastatic colorectal cancer (mCRC) were treated with cetuximab plus FOLFIRI or FOLFIRI alone until disease progression, unacceptable toxic effects or withdrawal of consent. OBJECTIVE: To determine if stratified use of cetuximab based on genetic biomarker detection improves cost-effectiveness. METHODS: We used individual patient data from CRYSTAL to compare the cost-effectiveness, cost per life-year (LY) and cost per quality-adjusted LY (QALY) gained of cetuximab plus FOLFIRI versus FOLFIRI alone in three cohorts of patients with mCRC: all randomised patients (intent-to-treat; ITT), tumours with no detectable mutations in codons 12 and 13 of exon 2 of the KRAS protein ('KRAS wt') and no detectable mutations in exons 2, 3 and 4 of KRAS and exons 2, 3 and 4 of NRAS ('RAS wt'). Survival analysis was conducted using RStudio, and a cost-utility model was modified to allow comparison of the three cohorts. RESULTS: The deterministic base-case ICER (cost per QALY gained) was £130,929 in the ITT, £72,053 in the KRAS wt and £44,185 in the RAS wt cohorts for cetuximab plus FOLFIRI compared with FOLFIRI alone. At a £50,000 willingness-to-pay threshold, cetuximab plus FOLFIRI has a 2.8, 20 and 63% probability of being cost-effective for the ITT, KRAS wt and RAS wt cohorts, respectively, versus FOLFIRI alone. CONCLUSION: Screening for mutations in both KRAS and NRAS may provide the most cost-effective approach to patient selection.

Willingness to pay for health insurance in the informal sector of Sierra Leone.

PURPOSE: The objective of this project is to study the willingness to pay (WTP) for health insurance (HI) of individuals working in the informal sector in Sierra Leone, using a purposely-designed survey of a representative sample of this sector. METHODS: We elicit the WTP using the Double-Bounded Dichotomous Choice with Follow Up method. We also examine the factors that are positively and negatively associated with the likelihood of the respondents to answer affirmatively to joining a HI scheme and to paying three different possible premiums, to join the HI scheme. We additionally analyze the individual and household characteristics associated with the maximum amount the household is willing to pay to join the HI scheme. RESULTS: The results indicate that the average WTP for the HI is 20,237.16 SLL (3.6 USD) per adult but it ranges from about 14,000 SLL (2.5 USD) to about 35,000 SLL (6.2 USD) depending on region, occupation, household and respondent characteristics. The analysis of the maximum WTP indicates that living outside the Western region and working in farming instead of petty trade are associated with a decrease in the maximum premium respondents are WTP for the HI scheme. Instead, the maximum WTP is positively associated to being a driver or a biker; having secondary or tertiary education (as opposed to not having any); the number of pregnant women in the household; having a TV; and, having paid for the last medical requirement. CONCLUSIONS: In summary, the various analyses show that a premium for the HI package could be set at approximately 20,000 SLL (3.54 USD) but also that establishing a single premium for all individuals in the informal sector could be risky. The efficient functioning of a HI scheme relies on covering as much of the population as possible, in order to spread risks and make the scheme viable. The impact of the various population characteristics raises the issue of how to rate premiums. In other words, setting a premium that may be too high for a big proportion of the population could mean losing many potential enrollees and might have viability consequences for the operation of the scheme.

Social and economic impacts of congenital Zika syndrome in Brazil: Study protocol and rationale for a mixed-methods study.

Global concern broke out in late 2015 as thousands of children in Brazil were born with microcephaly, which was quickly linked to congenital infection with Zika virus (ZIKV). ZIKV is now known to cause a wider spectrum of severe adverse outcomes-congenital Zika syndrome (CZS)-and also milder impairments. This study aimed to explore the social and economic impacts of CZS in Brazil. Data was collected through mixed methods across two settings: Recife City and Jaboatão dos Guararapes in Pernambuco State (the epicentre of the epidemic), and the city of Rio de Janeiro (where reports of ZIKV infection and CZS were less frequent). Data was collected May 2017-January 2018. Ethical standards were adhered to throughout the research. In-depth qualitative interviews were conducted with: mothers and other carers of children with CZS (approximately 30 per setting), pregnant women (10-12 per setting), men and women of child-bearing age (16-20 per setting), and health professionals (10-12 per setting). Thematic analysis was undertaken independently by researchers from at least two research settings, and these were shared for feedback. A case-control study was undertaken to quantitatively explore social and economic differences between caregivers of a child with CZS (cases) and caregivers with an unaffected child (controls). We aimed to recruit 100 cases and 100 controls per setting, from existing studies. The primary caregiver, usually the mother, was interviewed using a structured questionnaire to collect information on: depression, anxiety, stress, social support, family quality of life, health care and social service use, and costs incurred by families. Multivariable logistic regression analyses were used to compare outcomes for cases and controls. Costs incurred as a result of CZS were estimated from the perspective of the health system, families and society. Modelling was undertaken to estimate the total economic burden of CZS from those three perspectives.

The impact of the Great Recession on health-related risk factors, behaviour and outcomes in England.

This paper examines the impact that the Great Recession had on individuals' health behaviours and risk factors such as diet choices, smoking, alcohol consumption, and Body Mass Index, as well as on intermediate health outcomes in England. We exploit data on about 9000 households from the Health Survey for England for the period 2001-2013 and capture the change in macroeconomic conditions using regional unemployment rates and an indicator variable for the onset of the recession. Our findings indicate that the recession is associated with a decrease in the number of cigarettes smoked - which translated into a moderation in smoking intensity - and a reduction in alcohol intake. The recession indicator itself is associated with a decrease in fruit intake, a shift of the BMI distribution towards obesity, an increase in medicines consumption, and the likelihood of suffering from diabetes and mental health problems. These associations are often stronger for the less educated and for women. When they exist, the associations with the unemployment rate (UR) are nevertheless similar before and after 2008. Our results suggest that some of the health risks and intermediate health outcomes changes may be due to mechanisms not captured by worsened URs. We hypothesize that the uncertainty and the negative expectations generated by the recession may have influenced individual health outcomes and behaviours beyond the adjustments induced by the worsened macroeconomic conditions. The net effect translated into the erosion of the propensity to undertake several health risky behaviours but an exacerbation of some morbidity indicators. Overall, we find that the recession led to a moderation in risky behaviours but also to worsening of some risk factors and health outcomes.

Adaptation to health states: Sick yet better off?

Healthcare funding decisions in the UK rely on health state valuations of the general public. However, it has been shown that there is disparity between the valuation of the impact of hypothetical conditions on health and the reported health by those experiencing them. Patients' adaptation to health states is among the most common explanations for this discrepancy. Being diagnosed with a disease appears to affect individual perception of health over time so that better subjective health may be reported over a disease trajectory. This paper examines adaptation to health states using a longitudinal dataset. We use four waves of the British Cohort Study (BCS70), which tracks a sample of British individuals since birth in 1970 and contains information on self-assessed health (SAH), morbidity, and socioeconomic characteristics. We implement a dynamic ordered probit model controlling for health state dependence. Results are supportive of the existence of adaptation: Time since diagnosis has a positive impact on SAH. Moreover, adaptation happens over relatively long durations. We do not find significant results proving different adaptation paths for patients reporting prior better SAH. The analysis by specific conditions generally supports the existence of adaptation, but results are statistically significant only for a subset of conditions.

Costs of testing for ocular Chlamydia trachomatis infection compared to mass drug administration for trachoma in the Gambia: application of results from the PRET study.

BACKGROUND: Mass drug administration (MDA) treatment of active trachoma with antibiotic is recommended to be initiated in any district where the prevalence of trachoma inflammation, follicular (TF) is ≥ 10% in children aged 1-9 years, and then to continue for at least three annual rounds before resurvey. In The Gambia the PRET study found that discontinuing MDA based on testing a sample of children for ocular Chlamydia trachomatis(Ct) infection after one MDA round had similar effects to continuing MDA for three rounds. Moreover, one round of MDA reduced disease below the 5% TF threshold. We compared the costs of examining a sample of children for TF, and of testing them for Ct, with those of MDA rounds. METHODS: The implementation unit in PRET The Gambia was a census enumeration area (EA) of 600-800 people. Personnel, fuel, equipment, consumables, data entry and supervision costs were collected for census and treatment of a sample of EAs and for the examination, sampling and testing for Ct infection of 100 individuals within them. Programme costs and resource savings from testing and treatment strategies were inferred for the 102 EAs in the study area, and compared. RESULTS: Census costs were $103.24 per EA plus initial costs of $108.79. MDA with donated azithromycin cost $227.23 per EA. The mean cost of examining and testing 100 children was $796.90 per EA, with Ct testing kits costing $4.80 per result. A strategy of testing each EA for infection is more expensive than two annual rounds of MDA unless the kit cost is less than $1.38 per result. However stopping or deciding not to initiate treatment in the study area based on testing a sample of EAs for Ct infection (or examining children in a sample of EAs) creates savings relative to further unnecessary treatments. CONCLUSION: Resources may be saved by using tests for chlamydial infection or clinical examination to determine that initial or subsequent rounds of MDA for trachoma are unnecessary.

Economic evaluation for ophthalmologists.

PURPOSE: This review will describe the basic concepts of economic evaluation, using examples from the ophthalmic literature. The aim is to provide readers with knowledge about the fundamentals of economic evaluation to enable them to read papers critically, make healthcare funding and planning decisions, and understand the economic evaluation of interventions. REVIEW: Ophthalmic services are often constrained by a lack of funding, and this is true in both high- and low-income countries. Ophthalmology also competes with other healthcare specialities for funding. Economic evaluation is used to identify the most efficient way of allocating and planning the use of these scarce resources among alternative activities. An economic evaluation is typically conducted by comparing two or more interventions in terms of their effectiveness and their cost. Cost is the value of all resources used in the intervention. Effects, or consequences of the intervention, can be measured in monetary terms, through disease-specific outcomes or using health-related quality of life measures. The four most commonly used types of economic evaluations are cost-minimization analysis, cost-benefit analysis, cost-effectiveness analysis and cost-utility analysis. There are a variety of intended and unintended consequences of a health intervention, and so the consequences of the intervention may be positive or negative. Economic models, such as decision trees and Markov models, calculate effectiveness and costs, taking into account all the consequences of the intervention, including complications. Uncertainties in the parameters of the models can be expressed through sensitivity analyses and confidence intervals. CONCLUSIONS: Economic evaluation may be used to identify the most efficient way of allocating scarce resources among alternative activities. Its use, if standardized in all areas, can improve the quality of care while enhancing efficiency and thereby enabling more programs to be funded.

Value to smokers of improved cessation products: evidence from a willingness-to-pay survey.

The present study demonstrated the use of willingness to pay to value hypothetical new smoking cessation products. Data came from a baseline survey of participants in a clinical trial of medications for smoking cessation (N=356) conducted in New Haven, Connecticut. We analyzed individual willingness to pay for a hypothetical tobacco cessation treatment that is (a) more effective than those currently available and then (b) more effective and attenuates the weight gain often associated with smoking cessation. A majority of the respondents (n=280; 84%) were willing to pay for the more effective treatment, and, of those, 175 (63%) were willing to pay more if the increased effectiveness was accompanied by attenuation of the weight gain associated with smoking cessation. The present study suggests the validity of using willingness-to-pay surveys in assessing the value of new smoking cessation products and products with multifaceted improvements. From these data, we calculated estimates of the value of a quit. For the population studied, this survey suggests a substantial market for more effective smoking cessation treatments.

Creating an aggregate outcome index: cost-effectiveness analysis of substance abuse treatment.

This article proposes a method of calculating a practical index of improvement for conducting a cost-effectiveness analysis of substance abuse treatment. While the methodology of cost-effectiveness analysis necessitates the use of a single outcome measure, substance abuse treatment can produce multiple, important outcomes that must be considered in policy decisions about funding and alternative treatments. Thus, there is a need to aggregate the variety of outcomes from substance abuse treatment into a single index. The proposed index is a preference-weighted sum of outcomes using data from the Addiction Severity Index. The methodology and use of the index for conducting a cost-effectiveness analysis are illustrated using Addiction Severity Index data from a controlled quasi-experimental field study of case management.

Cost effectiveness of disulfiram: treating cocaine use in methadone-maintained patients.

Converging evidence suggests that disulfiram is a promising treatment for cocaine dependence. We study the cost-effectiveness of providing disulfiram to methadone-maintained opioid addicts in a randomized clinical trial setting. Our economic evaluation is based on a double blind clinical trial in which 67 cocaine-dependent methadone-maintained opioid-dependent subjects were randomized to get the additional treatment of disulfiram or placebo in a 12-week trial. Outcome measures used are the number of days of cocaine use and grams of cocaine per week. Cost measures used are the cost of providing standard methadone treatment and the incremental cost of adding disulfiram to the standard treatment. Cost measures of standard and disulfiram-enhanced treatment were collected retrospectively from the provider. Results from this cost-effectiveness analysis imply that, even though disulfiram increases slightly the cost of methadone treatment, its increase in effectiveness may be important enough to warrant its addition for treating cocaine dependence in methadone-maintained opiate addicts.

Cost-effectiveness analysis of addiction treatment: paradoxes of multiple outcomes.

This paper identifies and illustrates the challenges of conducting cost-effectiveness analysis (CEA) of addiction treatments given the multiple important outcomes of substance abuse treatment (SAT). Potential problems arise because CEA is intended primarily for single outcome programs, yet addiction treatment results in a variety of outcomes such as reduced drug use and crime and increased employment. Methodological principles, empirical examples, and practical advice are offered on how to conduct an economic evaluation given multiple outcomes. An empirical example is provided to illustrate some of the conflicts in cost-effectiveness (CE) ratios that may arise across the range of outcomes. The data are from the Philadelphia Target Cities quasi-experimental field study of standard versus "enhanced" (e.g. case management and added social services) drug treatment. Outcomes are derived from of the Addiction Severity Index (ASI), while cost data were collected and analyzed using the Drug Abuse Treatment Cost Analysis Program (DATCAP). While the results are illustrative only, they indicate that cost-effectiveness ratios for each of several different outcomes can produce conflicting implications. These findings suggest that multiple outcomes should be considered in any economic analysis of addiction treatments because focusing on a single outcome may lead to inadequate and possibly incorrect policy inferences. However, incorporating multiple outcomes into a CEA of addiction treatment is difficult. Cost-benefit analysis (CBA) may be a preferable and more appropriate approach in some cases.

Burning a hole in the budget: tobacco spending and its crowd-out of other goods.

Smoking is an expensive habit. Smoking households spend, on average, more than $US1000 annually on cigarettes. When a family member quits, in addition to the former smoker's improved long-term health, families benefit because savings from reduced cigarette expenditures can be allocated to other goods. For households in which some members continue to smoke, smoking expenditures crowd-out other purchases, which may affect other household members, as well as the smoker. We empirically analyse how expenditures on tobacco crowd-out consumption of other goods, estimating the patterns of substitution and complementarity between tobacco products and other categories of household expenditure. We use the Consumer Expenditure Survey data for the years 1995-2001, which we complement with regional price data and state cigarette prices. We estimate a consumer demand system that includes several main expenditure categories (cigarettes, food, alcohol, housing, apparel, transportation, medical care) and controls for socioeconomic variables and other sources of observable heterogeneity. Descriptive data indicate that, comparing smokers to nonsmokers, smokers spend less on housing. Results from the demand system indicate that as the price of cigarettes rises, households increase the quantity of food purchased, and, in some samples, reduce the quantity of apparel and housing purchased.