Search

The obesity epidemic is a significant public policy issue facing the international community, resulting in substantial costs to individuals and society. Various policies have been suggested to reduce and prevent obesity, including those informed by standard economics (a key feature of which is the assumption that individuals are rational) and behavioral economics (which identifies and harness deviations from rationality). It is not known which policy interventions taxpayers find acceptable and would prefer to fund via taxation. We provide evidence from a discrete choice experiment on an Australian sample of 996 individuals to investigate social acceptability of eight policies: mass media campaign; traffic light nutritional labeling; taxing sugar sweetened beverages; prepaid cards to purchase healthy food; financial incentives to exercise; improved built environment for physical activity; bans on advertising unhealthy food and drink to children; and improved nutritional quality of food sold in public institutions. Latent class analysis revealed three classes differing in preferences and key respondent characteristics including capacity to benefit. Social acceptability of the eight policies at realistic levels of tax increases was explored using post-estimation analysis. Overall, 78% of the sample were predicted to choose a new policy, varying from 99% in those most likely to benefit from obesity interventions to 19% of those least likely to benefit. A policy informed by standard economics, traffic light labeling was the most popular policy, followed by policies involving regulation: bans on junk food advertising to children and improvement of food quality in public institutions. The least popular policies were behaviorally informed: prepaid cards for the purchase of only healthy foods, and financial incentives to exercise.

Background Most Australian Aboriginal children are on track with their development, however, the prevalence of children at risk of or with a developmental or behavioural problem is higher than in other children. Aboriginal child development data mostly comes from remote communities, whereas most Aboriginal children live in urban settings. We quantified the proportion of participating children at moderate and high developmental risk as identified by caregivers’ concerns, and determined the factors associated with developmental risk among urban Aboriginal communities. Methods Study methods were co-designed and implemented with four participating urban Aboriginal Community Controlled Health Services in New South Wales, Australia, between 2008 and 2012. Caregiver-reported data on children Results Of 725 children in SEARCH with PEDS data (69% of eligible), 405 (56%) were male, and 336 (46%) were aged between 4.5 and 8 years. Using PEDS, 32% were at high, 28% moderate, and 40% low/no developmental risk. Compared with low/no risk, factors associated with high developmental risk in a mutually-adjusted model, with additional adjustment for study site, were male sex (OR 2.42, 95% confidence intervals 1.62–3.61), being older (4.5 to Conclusion In SEARCH, 40% of urban Aboriginal children younger than 8 years were at no or low developmental risk. Several factors associated with higher developmental risk were modifiable. Aboriginal community-driven programs to improve detection of developmental problems and facilitate early intervention are needed.

Background/aimsTo validate a preference-based Diabetic Retinopathy Utility Index (DRU-I) using discrete choice experiment (DCE) methods and assess disutilities associated with vision-threatening DR (VTDR: severe non-proliferative DR, proliferative DR and clinically significant macular oedema) and associated vision impairment.MethodsThe DRU-I comprises five quality-of-life dimensions, including Visual symptoms, Activity limitation/mobility, Lighting and glare, Socio-emotional well-being and Inconvenience, each rated as no, some, or a lot of difficulty. The DRU-I was developed using a DCE comprising six blocks of nine choice sets which, alongside the EuroQoL-5D (EQ-5D-3L) and Vision and Quality of Life (VisQoL) utility instruments, were interviewer-administered to participants. To ensure the DRU-I was sensitive to severe disease, we oversampled patients with VTDR. Data were analysed using conditional logit regression.ResultsOf the 220 participants (mean±SD age 60.1±11.3 years; 70.9% men), 57 (29.1%) and 139 (70.9%) had non-VTDR and VTDR, respectively, while 157 (71.4%), 20 (9.4%) and 37 (17.3%) had no, mild or moderate/severe vision impairment, respectively. Regression coefficients for all dimensions were ordered as expected, with worsening levels in each dimension being less preferred (theoretical validity). DRU-I utilities decreased as DR severity (non-VTDR=0.87; VTDR=0.80; p=0.021) and better eye vision impairment (none=0.84; mild=0.78; moderate/severe=0.72; p=0.012) increased. DRU-I utilities had low (r=0.39) and moderate (r=0.58) correlation with EQ-5D and VisQoL utilities, respectively (convergent validity).DiscussionThe DRU-I can estimate utilities associated with vision-threatening DR and associated vision impairment. It has the potential to assess the cost-effectiveness of DR interventions from a patient perspective and inform policies on resource allocation relating to DR.

Aim To describe socio-demographic patterns of asthma prevalence in urban Aboriginal children and quantify associations between asthma prevalence and pre-natal maternal and current carer smoking. Methods Analyses used carer-reported survey data for 1290 urban Aboriginal children aged 2-17 years from the Study of Environment on Aboriginal Resilience and Child Health. Multilevel log-binomial regression was used to estimate asthma prevalence ratios (PRs) for child- and family-level socio-demographic factors, pre-natal maternal smoking and current carer smoking. Smoking-related PRs were compared with general-population estimates derived from meta-analyses of published cross-sectional data. Results Overall, 33.9% of children had ever had asthma, and 12.9% had received treatment for asthma in the past month. Prevalence estimates declined with increasing household income and increasing household size (posterior probabilities of decreasing trend >0.98), while children exposed to pre-natal maternal smoking had a higher risk of asthma ever than unexposed children (PR 1.18 (95% credible interval 1.00-1.40)). Recently treated asthma prevalence was not significantly associated with pre-natal maternal (0.98 (0.71-1.41)) or current carer smoking (0.97 (0.68-1.37)); however, there was substantial uncertainty in our PR estimates, and 95% credible intervals contained general-population estimates derived from the meta-analyses (1.37 (1.17-1.65) for pre-natal smoking, 1.28 (1.15-1.44) for current parental or household smoking). Conclusion Among urban Aboriginal children in the Study of Environment on Aboriginal Resilience and Child Health cohort, asthma prevalence declines as household income and household size increase, while children exposed to pre-natal maternal smoking are at increased risk of ever having asthma. Our results emphasise the importance of reducing smoking in Aboriginal communities, particularly among pregnant women.

To determine the degree of agreement of diagnoses by audiologists and otolaryngologists of otitis media (OM) in Aboriginal children.Cross-sectional study of agreement between diagnoses.Study of Environment on Aboriginal Resilience and Child Health (SEARCH), a prospective cohort study of Aboriginal children attending four Aboriginal Community Controlled Health Services in New South Wales (three metropolitan, one regional) during 2008-2012.1310 of 1669 SEARCH participants (78.5%; mean age, 7.0 years; SD, 4.4 years) were assessed and received a diagnosis from one of five experienced audiologists. Test results (but not case histories) were forwarded to one of three otolaryngologists for blinded independent assessment.Agreement of OM diagnoses by audiologists and otolaryngologists at ear and child levels; correctness of audiologist diagnoses (otolaryngologist diagnosis as reference).Paired diagnoses by audiologists and otolaryngologists were available for 863 children at the child level and 1775 ears (989 children) at the ear level. Otolaryngologists diagnosed OM in 251 children (29.1%), including 11 (1.3%) with tympanic membrane perforation, and in 396 ears (22.3%), including 12 (0.7%) with perforation. Agreement between audiologists and otolaryngologists for OM at the ear level was 92.2% (κ = 0.78; 95% CI, 0.74-0.82), and at the child level 91.7% (κ = 0.81; 95% CI, 0.77-0.85). No otolaryngologist-diagnosed perforation was missed by audiologists. Among 1000 children triaged by an audiologist, there would be 45 false positives and 30 false negatives when compared with assessments by an otolaryngologist, with no missed perforations.There was substantial agreement between audiologists' and otolaryngologists' diagnoses of OM in a high prevalence population of Aboriginal children. In settings with limited access to otolaryngologists, audiologists may appropriately triage children and select those requiring specialist review.

Objective: Despite being disproportionately affected by injury, little is known about factors associated with injury in Aboriginal children. We investigated factors associated with injury among urban Aboriginal children attending four Aboriginal Community Controlled Health Services in New South Wales, Australia. Methods: We examined characteristics of caregiver‐reported child injury, and calculated prevalence ratios of ‘ever‐injury’ by child, family, and environmental factors. Results: Among children in the cohort, 29% (n=373/1,303) had ever broken a bone, been knocked out, required stitches or been hospitalised for a burn or poisoning; 40–78% of first injuries occurred at home and 60–91% were treated in hospital. Reported ever‐injury was significantly lower (prevalence ratio ≤0.80) among children who were female, younger, whose caregiver had low psychological distress and had not been imprisoned, whose family experienced few major life events, and who hadn't experienced alcohol misuse in the household or theft in the community, compared to other cohort members. Conclusions: In this urban Aboriginal child cohort, injury was common and associated with measures of family and community vulnerability. Implications for public health: Prevention efforts targeting upstream injury determinants and Aboriginal children living in vulnerable families may reduce child injury. Existing broad‐based intervention programs for vulnerable families may present opportunities to deliver targeted injury prevention.

This article describes the second stage of an empirical comparison of the performance of designs for a discrete choice experiment. Six designs were chosen to represent the range of construction techniques that are currently popular for choice experiments, with some of the designs incorporating into the design generation process prior knowledge of the parameters gained from the previous stage of this experiment. Each design had 320 respondents, each of whom completed 16 choice sets. The results indicate that efficient designs constructed using several different strategies all identify various types of heterogeneity with similar levels of precision. Specifying the right model to best describe the underlying preferences of respondents in each sample may then become the limiting factor in the estimation of more complex generalized multinomial models, rather than the design per se.

A key objective of discrete choice experiments is to obtain sufficient quantity of high quality choice data to estimate choice models to be used to explore various policy/clinically relevant issues. This paper focuses on a relatively new form of choice experiment, ‘Best Worst Discrete Choice Experiments’ (BWDCEs) and their relevance to health research as a new way to meet such an objective. We explain what BWDCEs are, how and when to apply them and we present several analytical approaches to model the resulting data. We demonstrate this preference elicitation approach in an empirical application exploring preferences of 898 members of the general public in Edmonton and Calgary, Canada for treatment of cardiac arrest occurring in a public place and show the gains achieved compared to traditional analysis of first best data. We suggest that BWDCEs are a valuable way to investigate preferences in the health sector and discuss implications for task design, analysis and areas for future research. Refereed/Peer-reviewed

Davidson and Beaver (1977) extended the Bradley-Terry model to incorporate the possible effect of position within a choice set on the choices made in paired comparisons experiments. We further extend the Davidson-Beaver result to choice sets of any size and show, under a mild restriction, that designs optimal for the multinomial logit model are still optimal. Designs balanced for carry-over effects of all orders can be used to construct designs with a diagonal information matrix for attribute effects. The theoretical results are derived assuming equal merits and we discuss the possible consequences of assuming unequal merits in an example. Copyright © Taylor & Francis Group, LLC.

QALYs are increasingly being utilized as a health outcome measure to calculate the benefits of new treatments and interventions within cost-utility analyses for economic evaluation. Cost-utility analyses of adolescent-specific treatment programmes are scant in comparison with those reported upon for adults and tend to incorporate the views of clinicians or adults as the main source of preferences. However, it is not clear that the views of adults are in accordance with those of adolescents on this issue. Hence, the treatments and interventions most highly valued by adults may not correspond with those most highly valued by adolescents. Ordinal methods for health state valuation may be more easily understood and interpreted by young adolescent samples than conventional approaches. The availability of young adolescent-specific health state values for the estimation of QALYs will provide new insights into the types of treatment programmes and health services that are most highly valued by young adolescents.The first objective of this study was to assess the feasibility of applying best-worst scaling (BWS) discrete-choice experiment (DCE) methods in a young adolescent sample to value health states defined by the Child Health Utility 9D (CHU9D) instrument, a new generic preference-based measure of health-related quality of life developed specifically for application in young people. The second objective was to compare BWS DCE questions (where respondents are asked to indicate the best and worst attribute for each of a number of health states, presented one at a time) with conventional time trade-off (TTO) and standard gamble (SG) questions in terms of ease of understanding and completeness.A feasibility study sample of consenting young adolescent school children (n16) aged 1113 years participated in a face-to-face interview in which they were asked to indicate the best and worst attribute levels from a series of health states defined by the CHU9D, presented one at a time. Participants were also randomly allocated to receive additional conventional TTO or SG questions and prompted to indicate how difficult they found them to complete.The results indicate that participants were able to readily choose 'best' and 'worst' dimension levels in each of the CHU9D health states presented to them and provide justification for their choices. Furthermore, when presented with TTO or SG questions and prompted to make comparisons, participants found the BWS DCE task easier to understand and complete.The results of this feasibility study suggest that BWS DCE methods are potentially more readily understood and interpretable by vulnerable populations (e.g. young adolescents). These findings lend support to the potential application of BWS DCE methods to undertake large-scale health state valuation studies directly with young adolescent population samples. © 2011 Adis Data Information BV. All rights reserved.

To investigate the relationship between asthma control and psychosocial outcomes in pregnant women with asthma.Secondary analysis (N = 221) of a randomized controlled trial of treatment adjustments, based on fractional exhaled nitric oxide versus clinical guideline-based algorithms. Psychosocial variables included generic and asthma-specific quality of life (SF12, AQLQ-M), illness perceptions (BIPQ), perceived control (PCAQ), perceived risk of side effects (PRSE) and anxiety (STAI-6). Asthma control was defined as controlled (Asthma Control Questionnaire (ACQ7) ≤1.5 at randomization and end of study), improved (ACQ7 1.5 at randomization and ≤1.5 at end of study) and unimproved (ACQ7 1.5 at end of study). Regression models were fitted for each psychosocial measure at the end of the study, with adjustment for baseline values and smoking status, with predictor variable asthma control.Women with unimproved asthma had poorer physical (SF12, p = 0.012) and asthma-specific quality of life across all domains (AQLQ-M, p ≤ 0.012) compared to women with controlled asthma. They believed that they had less control over their asthma (PCAQ total p = 0.014), had more symptoms and that their illness had a greater effect on their emotions and their lives in general (BIPQ identity, consequences, concern, emotional response p ≤ 0.015). Women with improved asthma control had significantly lower AQLQ-M breathlessness (p = 0.048) and lower total scores (p = 0.04) than women with controlled asthma.Pregnant women who are not able to get control of their asthma symptoms may experience worse quality of life and are likely to have more negative perceptions about their condition.

In this article we establish the choice sets in the D-optimal design for a choice experiment for testing main effects and for testing main effects and two-factor interactions, when there are k attributes, each with two levels, for choice set size m. We also give a method to construct optimal and near-optimal designs with small numbers of choice sets.

In this paper we give a construction for four factor orthogonal main effect plans (OMEPs) and an interchange algorithm to give four factor OMEPs with various different numbers of repeated runs.

In this paper we study orthogonal main effect plans with four factors, A table of such designs, where each factor has at most 10 levels, and there are at most 40 runs, is generated. We determine the spectrum of the degrees of freedom of pure error for these designs.

How many choice sets respondents will answer is a critical issue in how much data a researcher has for analysis. We used 66 separate surveys that ask respondents, from an opt-in web panel, sequences of preference questions about consumer products to study design factors that influence the rate of completing the entire sequence of questions comprising a discrete choice experiment. We do this by systematically varying the number of choice sets, the number of alternatives respondents were asked to consider, the nature of the list of attributes of each alternative and the type of statistical design. Completion rates systematically varied with the factors explored, but perhaps the key finding is that completion rates are reasonably high in all cases. We found that completion rates are relatively unaffected by asking more questions (choice sets), but they decline as one includes more alternatives. Expected time to complete a survey often plays a key role in the cost of web-based panels, so we also look at how the preceding factors impact completion times. Practical implications for applied research using opt-in web panels are discussed. Refereed/Peer-reviewed

Background. SF-6D utility weights are conventionally produced using a standard gamble (SG). SG-derived weights consistently demonstrate a floor effect not observed with other elicitation techniques. Recent advances in discrete choice methods have allowed estimation of utility weights. The objective was to produce Australian utility weights for the SF-6D and to explore the application of discrete choice experiment (DCE) methods in this context. We hypothesized that weights derived using this method would reflect the largely monotonic construction of the SF-6D. Methods. We designed an online DCE and administered it to an Australia-representative online panel (n = 1017). A range of specifications investigating nonlinear preferences with respect to additional life expectancy were estimated using a random-effects probit model. The preferred model was then used to estimate a preference index such that full health and death were valued at 1 and 0, respectively, to provide an algorithm for Australian cost-utility analyses. Results. Physical functioning, pain, mental health, and vitality were the largest drivers of utility weights. Combining levels to remove illogical orderings did not lead to a poorer model fit. Relative to international SG-derived weights, the range of utility weights was larger with 5% of health states valued below zero. Conclusions. DCEs can be used to investigate preferences for health profiles and to estimate utility weights for multi-attribute utility instruments. Australian cost-utility analyses can now use domestic SF-6D weights. The comparability of DCE results to those using other elicitation methods for estimating utility weights for quality-adjusted life-year calculations should be further investigated. Refereed/Peer-reviewed

This paper describes an empirical comparison of the performance of four designs for a discrete choice experiment. These designs were chosen to represent the range of construction techniques that are currently popular for choice experiments when no prior knowledge of the parameters is available. Each design had 320 respondents who each completed 16 choice sets. The results suggest that for the multinomial logit model (MNL) the design that is used at this stage is fairly unimportant. As the sample size gets smaller, however, differences between the designs become apparent. We also analysed the results using four different models which accommodate preference heterogeneity. We find that any of these models are able to predict choices more accurately for born in-sample and out-of-sample than the MNL model for the designs used here, and that the differences across designs arc larger for models with more parameters, although preliminary results suggest the gain appears to depend on the underlying preference structure. © Grace Scientific Publishing, LLC.

Background: Stroke rehabilitation is moving towards more intense therapy models that incorporate technologies such as robotics and computer games. It is unclear how acceptable these changes will be to stroke survivors, as little is known about which aspects of rehabilitation programmes are currently valued. Discrete choice experiments are a potential approach to assessing patient preferences, as they reveal the characteristics of programmes that are most important to consumers. Methods: A discrete choice experiment was presented as a face-to-face interview to assess the priorities and preferences of stroke survivors (n = 50, mean age 72 years) for alternative rehabilitation service configurations. The discrete choice experiment was presented to the participants while they were on the stroke rehabilitation ward (approximately 3–4 weeks following stroke). Results: Participants were highly focused on recovery and expressed strong preferences for therapy delivered one-to-one, but they did not favour very high intensity programmes (6 hours per day). While the attitudinal statements indicated high levels of agreement for programmes to incorporate the latest technology, the results from the discrete choice experiment indicated that participants were averse to computer-delivered therapy. Conclusion: Whilst rehabilitation therapy is highly valued, stroke survivors exhibited stronger preferences for low-intensity programmes and rest periods. High-intensity therapy protocols or approaches dependent on new technologies will require careful introduction to achieve uptake and acceptability. Refereed/Peer-reviewed

Background Interactive video games such as the Nintendo Wii Fit are increasingly used as a therapeutic tool in health and aged care settings however, their acceptability to older people is unclear. The aim of this study was to determine the acceptability of the Nintendo Wii Fit as a therapy tool for hospitalised older people using a discrete choice experiment (DCE) before and after exposure to the intervention. Methods A DCE was administered to 21 participants in an interview style format prior to, and following several sessions of using the Wii Fit in physiotherapy. The physiotherapist prescribed the Wii Fit activities, supervised and supported the patient during the therapy sessions. Attributes included in the DCE were: mode of therapy (traditional or using the Wii Fit), amount of therapy, cost of therapy program and percentage of recovery made. Data was analysed using conditional (fixed-effects) logistic regression. Results Prior to commencing the therapy program participants were most concerned about therapy time (avoiding programs that were too intensive), and the amount of recovery they would make. Following the therapy program, participants were more concerned with the mode of therapy and preferred traditional therapy programs over programs using the Wii Fit. Conclusions The usefulness of the Wii Fit as a therapy tool with hospitalised older people is limited not only by the small proportion of older people who are able to use it, but by older people's preferences for traditional approaches to therapy. Mainstream media portrayals of the popularity of the Wii Fit with older people may not reflect the true acceptability in the older hospitalised population.

Incentive programs aimed at encouraging private landowners to set aside areas of forest for their conservation value have existed in Australia for more than two decades. Many programs restrict the use of the land by legal agreements or other means and some programs offer the landowner financial compensation. Most programs are based on voluntary entry by landowners. Programs available in Tasmania have added significantly to the total forest area conserved on private land. Nevertheless, in some regions more than 80 percent of land with conservation value remains unprotected and programs routinely fail to meet enrolment targets. This has resulted in considerable debate about the design of programs and has resulted in an increase in the amount of compensation offered and the introduction of more flexible conservation management options. In a limited number of situations, the option of forcing landowner entry into conservation incentive programs has also been considered. The objective of this study is to provide information for policy makers that hasn't existed before and that can be used in designing conservation programs. An improved understanding of landowners' decision framework, their motivation and the strength of their behavioral response will facilitate better forecasting of landowner participation decisions which may lead to an increase in landowner enrolment in programs. In this dissertation a conceptual model of landholders' participation choice is developed that combines a traditional utility maximisation framework with information about landowner attitudes. An empirical model of landowners' conservation incentive program choice is then developed. The model is estimated using stated preference data from a Best-Worst and a Choice survey. The responses to the Best-Worst survey, which was carried out first, were used to determine the choice set for the subsequent Choice survey. The Best-Worst survey was also used to explore differences between the perceived importance of program attributes by program designers and administrators, and landowners.

In 1970 Davidson generalised the Bradley-Terry model to allow respondents to say that the two options presented in a choice task were equally attractive. In this paper we extend this idea to the MNL model with m options in each choice set and we show that the optimal designs for the MNL model are also optimal in this setting. © 2009 Elsevier B.V. All rights reserved.

We show how to combine statistically efficient ways to design discrete choice experiments based on random utility theory with new ways of collecting additional information that can be used to expand the amount of available choice information for modeling the choices of individual decision makers. Here we limit ourselves to problems involving generic choice options and linear and additive indirect utility functions, but the approach potentially can be extended to include choice problems with non-additive utility functions and non-generic/labeled options/attributes. The paper provides several simulated examples, a small empirical example to demonstrate proof of concept, and a larger empirical example based on many experimental conditions and large samples that demonstrates that the individual models capture virtually all the variance in aggregate first choices traditionally modeled in discrete choice experiments.

This chapter walks the reader through the stages of a discrete choice experiment (DCE) application drawing on concepts introduced in Chapters 1 and 2. A case study eliciting women’s preferences for prenatal screening is used to illustrate the points (Ryan et al., 2005). It should be noted that this data was collected several years ago, and therefore adopted old methods of experimental design. The sample size is also small.We have chosen it because it demonstrates nicely the many potential uses of a DCE, it addressed a policy-relevant question at the time, and it represents one of the few studies in health economics where the scientists (geneticists) worked with the evaluators (economists) and implementers (obstetricians) to look at development, evaluation and implementation of prenatal screening programmes.

In discrete choice experiments, design decisions are crucial for determining data quality and costs. While high statistical efficiency designs are desirable, they may come at a price If they increase the cognitive burden for respondents. We address this problem by designing 44 experiments that systematically vary numbers of attributes and attribute level differences. Our results for two product categories suggest that respondents systematically are less consistent In answering choice questions as statistical efficiency increases. This relationship holds regardless of the number of attributes and is statistically significant even if one accommodates preference heterogeneity. Implications for practice and future research are discussed. © 2008 by JOURNAL OF CONSUMER RESEARCH, Inc.

[...] [T]he application of discrete choice experiments (DCEs) in health economics has seen an increase over the last few years. While the number of studies using DCEs is growing, there has been relatively limited consideration of experimental design theory and methods. Details of the development of the designed experiment are rarely discussed. Many studies have used small fractional factorial designs (FFDs), generated with commercial design software packages, e.g. orthogonal main effects plans (OMEPs), sometimes manipulated in ad hoc ways (e.g. randomly pairing up scenarios or taking one scenario from the design and combining it with every other scenario). Such approaches can result in designs with unknown statistical design properties, in particular with unknown correlations between parameter estimates.

© 2007 by John Wiley & Sons, Inc. All rights reserved. The most comprehensive and applied discussion of stated choice experiment constructions available. The Construction of Optimal Stated Choice Experiments provides an accessible introduction to the construction methods needed to create the best possible designs for use in modeling decision-making. Many aspects of the design of a generic stated choice experiment are independent of its area of application, and until now there has been no single book describing these constructions. This book begins with a brief description of the various areas where stated choice experiments are applicable, including marketing and health economics, transportation, environmental resource economics, and public welfare analysis. The authors focus on recent research results on the construction of optimal and near-optimal choice experiments and conclude with guidelines and insight on how to properly implement these results. Features of the book include: Construction of generic stated choice experiments for the estimation of main effects only, as well as experiments for the estimation of main effects plus two-factor interactions. Constructions for choice sets of any size and for attributes with any number of levels. A discussion of designs that contain a none option or a common base option. Practical techniques for the implementation of the constructions. Class-tested material that presents theoretical discussion of optimal design. Complete and extensive references to the mathematical and statistical literature for the constructions. Exercise sets in most chapters, which reinforce the understanding of the presented material. The Construction of Optimal Stated Choice Experiments serves as an invaluable reference guide for applied statisticians and practitioners in the areas of marketing, health economics, transport, and environmental evaluation. It is also ideal as a supplemental text for courses in the design of experiments, decision support systems, and choice models. A companion web site is available for readers to access web-based software that can be used to implement the constructions described in the book.

In this paper, we establish the optimal size of the choice sets in generic choice experiments for asymmetric attributes when estimating main effects only. We give an upper bound for the determinant of the information matrix when estimating main effects and all two-factor interactions for binary attributes. We also derive the information matrix for a choice experiment in which the choice sets are of different sizes and use this to determine the optimal sizes for the choice sets.

In this paper we establish the form of the optimal design for choice experiments in which attributes need not have the same number of levels for testing main effects only, when there are k attributes, and all choice sets are of size m. We give a construction for optimal and near-optimal designs with small numbers of choice sets. We derive the general form of the determinant of the information matrix for estimating main effects and two-factor interactions and derive the optimal designs for this situation in some special cases. © 2004 Elsevier B.V. All rights reserved.

Stated preference choice experiments are routinely used in many areas from marketing to medicine. While results on the optimal choice sets to present for the forced choice setting have been determined in a variety of situations, no results have appeared to date on the optimal choice sets to use when either all choice sets are to contain a common base alternative or when all choice sets contain a "none of these" option. These problems are considered in this paper. © 2004 Elsevier B.V. All rights reserved.

Paul Green inspired many of us who work in conjoint analysis and related areas, such as stated preference discrete choice experiments, and all of us who undertake research in understanding and modeling preferences have benefited from his work. Knowing Paul Green, his impact and contributions will continue during his “retirement.” Our thanks to him for so many of the advances that we now often take for granted, but without which we would not be where we are today.

This paper gives constructions for optimal and near-optimal sets of pairs for the estimation of main effects, and for the estimation of main effects and two factor interactions, in forced choice experiments in which all attributes have two levels. The number of pairs in the sets is much smaller than that in previously constructed optimal 2-level choice experiments. © 2002 Elsevier B.V. All rights reserved.

This comprehensive collection brings together more than 50 contributions from some of the most influential researchers in health economics. It authoritatively covers theoretical and empirical issues in health economics, with a balanced range of material on equity and efficiency in health care systems, health technology assessment and issues of concern for developing countries. This thoroughly revised second edition is expanded to include four new chapters, while all existing chapters have been extensively updated.