Mixed Multinomial Logit Model Research Articles

A Comprehensive Business Location Choice Model Leveraging Machine Learning in Systematic Choice Set

This study develops a comprehensive two-stage location choice framework for business establishments as part of a goods movements modeling. This study aims to formulate a systematic methodology for investigating the location choice of business establishments within Halifax Regional Municipality. This study presents a novel approach that leverages machine learning techniques to generate a systematic choice set, thereby improving the representation of realistic and reasonable location alternatives. Info Canada Business Establishments dataset 2022 is employed to achieve the aim of this study. Combining an unsupervised machine learning technique with the mixed multinomial logit model facilitates a data-driven approach to enhance the precision and robustness of business establishment location choice models. This approach possesses the potential to unveil latent patterns and heterogeneity among potential choice alternatives that may remain obscured when utilizing a conventional multinomial logit model. This thorough analysis offers robust insights into the factors influencing the location choice of business establishments. The findings obtained from this comprehensive study suggest that wholesalers prioritize proximity to highways and positions within business parks for their operations while avoiding higher population density and central business district proximity. Transportation businesses seek larger sites and locations near highways, favoring clustering with related transport companies and valuing accessibility and cost-effectiveness over proximity to business parks or rural settings. The findings of this study could provide valuable insights for commercial vehicle and goods movement modeling, business location strategies, and policymaking concerning sustainable urban development.

A Model-Free Approach for Solving Choice-Based Competitive Facility Location Problems Using Simulation and Submodularity

This paper considers facility location problems in which a firm entering a market seeks to open facilities on a subset of candidate locations so as to maximize its expected market share, assuming that customers choose the available alternative that maximizes a random utility function. We introduce a deterministic equivalent reformulation of this stochastic problem as a maximum covering location problem with an exponential number of demand points, each of which is covered by a different set of candidate locations. Estimating the prevalence of these preference profiles through simulation generalizes a sample average approximation method from the literature and results in a maximum covering location problem of manageable size. To solve it, we develop a partial Benders reformulation in which the contribution to the objective of the least influential preference profiles is aggregated and bounded by submodular cuts. This set of profiles is selected by a knee detection method that seeks to identify the best tradeoff between the fraction of the demand that is retained in the master problem and the size of the model. We develop a theoretical analysis of our approach and show that the solution quality it provides for the original stochastic problem, its computational performance, and the automatic profile-retention strategy it exploits are directly connected to the entropy of the preference profiles in the population. Computational experiments on existing and new benchmark sets indicate that our approach dominates the classical sample average approximation method on large instances of the competitive facility location problem, can outperform the best heuristic method from the literature under the multinomial logit model, and achieves state-of-the-art results under the mixed multinomial logit model. We characterize a broader class of problems, which includes assortment optimization, to which the solving methodology and the analyses developed in this paper can be extended. History: Accepted by Andrea Lodi, Area Editor for Design & Analysis of Algorithms—Discrete. Funding: This research was supported by Fonds de Recherche du Québec-Nature et Technologies and Institut de Valorisation des Données through scholarships to R. Legault. E. Frejinger was partially supported by the Canada Research Chair program [Grant 950-232244]. Supplemental Material: The online appendices are available at https://doi.org/10.1287/ijoc.2023.0280 .

Students’ commuting habits to the university: Transportation choices during the Covid-19 era

This paper seeks to analyse the drivers behind students’ commuting choice in the context of a medium-sized public university (University of Urbino Carlo Bo) in Italy. The study accounts for changes in commuting preferences and choices occurring during the Covid-19 pandemic. The results are based on a 2020 survey on students analysed by means of a mixed multinomial logit model and a latent class model. The University of Urbino is an interesting case study for several reasons. First, it has a higher number of enrolled students (approximately 15,000) than there are residents in the municipality (less than 15,000). Second, Urbino is located far from main roads and transport infrastructures. Third, there are commuting options to and from the city, meaning that local transport policies have a relevant impact on the entire territory. Personal characteristics, distance from home, and price of the transportation mode influence the choice of students. The estimates for the two post-Covid-19 scenarios showed no particular changes in students’ transport habits, except when the number of household members was taken into account. The study provides valuable insights into the attitudes towards change in transportation choices that have recently emerged among a specific student population after an extended lockdown, that is now faced with making decisions marked by evident uncertainty about the possible developments of the Covid-19 virus.

Obesity is linked to disease severity in moderate to severe atopic dermatitis-Data from the prospective observational TREATgermany registry.

There are conflicting data on a potential association between obesity and atopic dermatitis (AD). The purpose of this study was to investigate the relationship between obesity and AD disease severity. Patients from the TREATgermany registry cohort were divided into three groups according to their body mass index (BMI). Due to low numbers, underweight patients (BMI <18.5 kg/m2) were excluded from the analysis. Physician- and patient-reported disease severity scores as well as additional phenotypic characteristics were evaluated for association with BMI. Generalized linear mixed models and multinomial logit models, respectively, were applied to investigate the association of BMI, age, sex and current systemic AD treatment with disease severity. This study encompassed 1416 patients, of which 234 (16.5%) were obese (BMI ≥30 kg/m2). Obesity was associated with lower educational background and smoking. Otherwise, obese and non-obese AD patients had similar baseline characteristics. Increased BMI was associated with higher oSCORAD (adjusted β: 1.24, 95% CI: 1.05-1.46, p = 0.013) and Patient-oriented eczema measure (POEM) (adjusted β: 1.09, 95% CI: 1.01-1.17, p = 0.038). However, the absolute difference in the overall oSCORAD was small between obese and non-obese AD patients (Δ oSCORAD = 2.5). Allergic comorbidity was comparable between all three groups, with the exception of asthma which was more pronounced in obese patients (p < 0.001). In this large and well-characterized AD patient cohort, obesity is significantly associated with physician- and patient-assessed measures of AD disease severity. However, the corresponding effect sizes were low and of questionable clinical relevance. The overall prevalence of obesity among the German AD patients was lower than in studies on other AD cohorts from different countries, which confirms previous research on the German population and suggests regional differences in the interdependence of AD and obesity prevalence.

Consumer Preferences for a Healthcare Appointment Reminder in Australia: A Discrete Choice Experiment.

It is essential to consider the evidence of consumer preferences and their specific needs when determining which strategies to use to improve patient attendance at scheduled healthcare appointments. This study aimed to identify key attributes and elicit healthcare consumer preferences for a healthcare appointment reminder system. A discrete choice experiment was conducted in a general Australian population sample. The respondents were asked to choose between three options: their preferred reminder (A or B) or a 'neither' option. Attributes were developed through a literature review and an expert panel discussion. Reminder options were defined by four attributes: modality, timing, content and interactivity. Multinomial logit and mixed multinomial logit models were estimated to approximate individual preferences for these attributes. A scenario analysis was performed to estimate the likelihood of choosing different reminder systems. Respondents (n = 361) indicated a significant preference for an appointment reminder to be delivered via a text message (β = 2.42, p < 0.001) less than 3 days before the appointment (β = 0.99, p < 0.001), with basic details including the appointment cost (β = 0.13, p < 0.10), and where there is the ability to cancel or modify the appointment (β = 1.36, p < 0.001). A scenario analysis showed that the likelihood of choosing an appointment reminder system with these characteristics would be 97%. Our findings provide evidence on how healthcare consumers trade-off between different characteristics of reminder systems, which may be valuable to inform current or future systems. Future studies may focus on exploring the effectiveness of using patient-preferred reminders alongside other mitigation strategies used by providers.

A Random Consideration Set Model for Demand Estimation, Assortment Optimization, and Pricing

Random Consideration Set Model We operationalize a microfounded consumer choice model—the random consideration set (RCS) choice model of Manzini and Mariotti [Manzini P, Mariotti M (2014) Stochastic choice and consideration sets. Econometrica 82(3):1153–1176]—that captures the limited attention of consumers, assuming that purchases are based on fixed preference orderings with consideration sets formed from independent attentions. We provide a condition for uniquely identifying model parameters and design an efficient algorithm for model parameters estimation. We offer a greedy-like algorithm for assortment optimization, adaptable for optimal assortment subject to cardinality constraint or discovering efficient sets. We extend the model to consider random product preferences, with a 1/2 performance-guaranteed approximation algorithm. Using data from a major U.S. airline, we find that the RCS model outperforms the mixed multinomial logit model in approximately half of the markets, particularly with smaller, less varied data sets.

Blockchain-enabled electric vehicle orderly charging optimization

The increasing prevalence of electric vehicles (EVs) urges charging station operators (CSOs) to expand their networks. However, currently CSOs operate independent EV charging service platforms separately, which poses inconvenience for EV users and hinders the efficient utilization of charging infrastructure. This paper introduces a consortium blockchain-enabled EV charging service platform (CB-EVCSP), which leverages novel orderly charging guidance models (OCGMs) encoded as smart contracts to integrate and coordinate the efforts of CSOs. The CB-EVCSP not only mitigates trust and security concerns but also enhances the utilization of charging facilities. Both multinomial logit (MNL) and mixed MNL models are developed to characterize the interaction between charging station recommendations and EV users’ choice behaviors, where a coefficient of variation based fairness criterion is used to ensure distribution fairness. We provide equivalent reformulations and design efficient constraint generation algorithms. Numerical experiments are conducted to validate the effectiveness of the proposed models and algorithms.

Studying shared bike route choice behavior using a bike-sharing system in Germany

The objective of this research is to explore the determinants of shared bike route choice in both urban and rural settings in Germany. To this end, a stated preference (SP) survey on route choice was conducted across several German cities. Participants were drawn from the service area of a regional bike-sharing system (BSS) in Germany. This area included five major cities and several mid-sized and smaller municipalities. The survey includes responses from both users and non-users of the BSS. Mixed multinomial logit models were used in analysis of the SP data. The study also calculated the willingness to pay (WTP) values for selected attributes. Significant influences on route choice included access and egress time, ride time, and ride cost. Significant random heterogeneity was found, especially for ride cost. Further heterogeneity was reported for interactions with sociodemographic attributes. Street type, surface, and bike infrastructure also had significant effects on route choice. Linkages between infrastructural preferences and respondents’ ages were evident. Although WTP values were strongly influenced by random heterogeneity in ride cost perception, most WTP values were comparable to the values obtained in other studies.

Prognostic Information on Progression to Dementia: Quantification of the Impact on Quality of Life.

The increasing interest in early identification of people at risk of developing dementia, has led to the development of numerous models aimed at estimating the likelihood of progression from mild cognitive impairment (MCI) to dementia. It is important to study both the need for and possible outcomes related with such prediction models, including the impact of risk predictions on perceived quality of life (QoL). This study aimed to quantify the impact that receiving a risk prediction on progression from MCI to dementia has on QoL. A Discrete Choice Experiment (DCE) and Time Trade Off (TTO) study were performed. Participants completed choice tasks related to dementia prognosis while imagining having MCI. We collected DCE data by an online survey, and TTO data via videoconferencing interviews. DCE data were analyzed using a mixed multinomial logit model and were anchored to a health state utility scale using mean observed TTO valuations. 296 people participated in the DCE and 42 in the TTO. Moderate and high predicted dementia risks were associated with decrements in utility (-0.05 and -0.18 respectively), compared to no prognostic information. Low predicted risk was associated with an increase in utility (0.06), as well as the availability of medication or lifestyle interventions (0.05 and 0.13 respectively). This study shows a significant impact of dementia risk predictions on QoL and highlights the importance of caution when sharing information about expected MCI disease courses.

Preferences and willingness to pay for personalized nutrition interventions: Discrete choice experiments in Europe and the United States

This study gives insight into what intervention-related factors are crucial for using personalized nutrition (PN) interventions, as well as what the general population is willing to pay for PN. This was done by focusing on two different types of PN (i.e., PN advice and personalized meals) in two discrete choice experiments (DCEs). The DCEs were conducted in four European countries and the United States, including at least 500 respondents per country aged 18–65 years. Panel mixed multinomial logit models were used to evaluate the preferences. Results show that for both types of PN in all countries, the total expenditure on nutrition was the most crucial factor when choosing a PN intervention. The participation rate for specific hypothetical scenario’s varied but was considered high overall (maximum 81 % for ‘PN advice’ and 87 % for ‘personalized meals’ in Spain). Moreover, highest willingness to pay estimates were found for six kilograms of weight loss. For example, Polish respondents were willing to spend an extra 25.78 euros per week for ‘personalized meals’ for a 4-month period to lose six kilograms. Our models showed preference heterogeneity between, but also within, the different countries. In conclusion, this study showed that people seem willing to pay for and participate in PN interventions. Since PN interventions may improve health outcomes, policymakers should consider subsidizing some of the costs, financially incentivizing PN interventions or introducing commitment lotteries to encourage uptake. More research is needed to study heterogeneity in preferences.

Environmental impact of business-to-consumer e-commerce: Does it matter to consumers?

The rapid development of business-to-consumer e-commerce has sparked interest in the environmental impact of the last-mile logistics of goods purchased online. The distribution stage, with orders from many different manufacturers in far-flung locations and the need for the fastest possible parcel delivery, has become complicated, expensive and environmentally damaging. The aim of this paper is to investigate consumers' willingness to pay for reducing or offsetting the emissions generated by their online purchases and the acceptability of alternative collection methods in order to reduce the environmental impact caused. We surveyed 1204 Italians using a discrete choice experiment and analysed the data using multinomial logit (MNL) and mixed multinomial logit (MMNL) models. The results show that providing information on the amount of pollution reduction/offset and the type of project implemented significantly increased consumers' willingness to pay to reduce the environmental impact of last-mile logistics. Our sample was willing to pay up to €0.88 to fully offset the environmental impact caused by the delivery received and an additional amount of €0.17 if the funds raised would be used for reforestation projects. We also found that consumers' socio-demographic characteristics and online shopping habits influenced both the willingness to pay and the acceptability of choosing more sustainable logistic settings regarding delivery time and location. Eco-conscious customers were more likely to accept longer delivery times and non-home delivery points. The minimum discount required to accept these delivery options ranged from €0.20 to €0.80. Our findings are new to the literature and are relevant not only to businesses seeking to integrate sustainability and efficiency into delivery services but also to policy makers seeking to mitigate the environmental impacts of e-commerce and freight.

Patient preferences for heart valve disease intervention.

This study aimed to determine how patients trade-off the benefits and risks of two different types of procedures used to treat heart valve disease (HVD). It also aimed to determine patients'preferences for HVD treatments (predicted uptake)and the relative importance of each treatment attribute. A discrete choice experiment (DCE) was conducted in Australia and Japan with patients who required a heart valve procedure. Patients were stratified into three categories: no prior procedure experience, minimally invasive procedure experienceand invasive procedure experience. DCE attributes included risk of mortality; risk of stroke; needing dialysis; needing a new pacemaker; valve durability; independence 1 month after surgery; and out-of-pocket expenses. Participants chose between two hypothetical labelled approaches to therapy ('invasive procedure' and 'minimally invasive procedure'), with a separate opt-out included. A mixed multinomial logitmodel was used to analyse preferences. The DCE was completed by 143 Australian and 206 Japanese patients. Both populations demonstrated an overall preference for the minimally invasive procedure over the invasive procedure. All attributes tested significantly predicted choice and were important to patient decision-making. However, patients'choices were most influenced by the durability of the valve and the likelihood of independence postprocedure, irrespective of their prior procedure experience. Differences in preference were observed between Australian and Japanese patients; valve durability was the most important attribute among Australian patients, while Japanese patients emphasised regaining independence postsurgery. Risk of mortality was less important relative to other key attributes in Japan; however, it remained significant to the model. HVD patients prefer a minimally invasive procedure over an invasive procedure, irrespective of prior treatment experience. Key attributes contributing to treatment preferences are valve durability and faster recovery. These results can be used to help inform healthcare decision-makers about what features of heart valve procedures patients value most. People with lived experience of HVD were included in multiple stages of the design phase of this research. First, patients and doctors were consulted by taking part in qualitative interviews. The qualitative interviews helped inform which treatment attributes to include in the DCEbased on what was important to those with lived experience and those who help make treatment decisions on behalf of patients. Qualitative interview participants also assisted with the framing of questions in the online survey to ensure the terminology was patient-friendly and relevant to those with lived experience. Following qualitative interviews, the DCE attribute list was agreed on in expert consultation with a steering committee, which included patient representatives and treating physicians (interventional cardiologists, cardiothoracic surgeons). The survey was also pilot tested with a small sample of patients and minor adjustments were made to the wording to ensure it was appropriate and meaningful to those with lived experience of HVD.

Willingness to pay for changes in travel time and work time: A stated choice experiment of US commuters

Studying the tradeoffs between commuting, working, and money—including the willingness to pay (WTP) for changes to travel time and work time—is of interest in both travel behavior research and transportation practice. Our study's objectives were to: (1) quantify WTP for travel time and work time changes; (2) measure preference heterogeneity in these time values; and (3) examine sensitivities to changes in income and travel cost. First, we conducted a stated choice experiment that added work time and income to traditional travel time and cost attributes, collecting data from 675 US adult commuters in late 2020. Second, we estimated pseudo panel mixed multinomial logit models with random and systematic preference heterogeneity. Average WTP was larger for travel time changes than for work time changes, although values varied across personal, household, and commute/work characteristics. Overall and on the margin, US commuters appear to dislike commuting more than they dislike working, a finding which differs from previous studies in Chile and Austria. Although future work could improve the realism of the choice experiment, our research offers a simple stated preference method to help work towards advancing an understanding of the values and tradeoffs involving both working and commuting.

Effects of proactive and reactive health control measures on public transport preferences of passengers – A stated preference study during the COVID-19 pandemic

The outbreak of COVID-19 has caused unprecedented disruptions to public transportation services, passengers' mode choice preferences, and their behavioral pattern. Recent studies have updated the systemic shift from shared travel to private mode due to the perceived higher health risk of shared mobility. However, findings from these studies may not be applicable to public transport-oriented cities with very low rate of private car ownership. In the long term when we arrive at a stabilized ‘New Normal’, the need to promote public transport is likely to be re-emphasized with high priority. This study presents insights into the factors affecting public transport (PT) preferences and potential modal shifts within various modes of PT when proactive and reactive health control measures are implemented. An online stated preference survey was conducted in Hong Kong, where public buses, public light buses, MTR, and taxis are the popular modes of public transport. The regret-based panel mixed multinomial logit model was applied in this study, aiming to capture individual’s regret aversion psychology in the context of the COVID-19 pandemic. The results of our random regret model suggest that travelers generally have lower regrets for the chosen mode when having passenger temperature screening, COVID tracing app, and more frequent disinfection, while seem to be unaware of the in-vehicle air quality. Interaction effects between health control measures and personal characteristics (e.g., age and risk attitudes) were also considered. Trip characteristics, including in-vehicle crowding level, fares, in-vehicle travel time, waiting time at a station, walking time, and trip purpose, were found to influence passengers' choices of travel modes. Policy simulations of possible changes in public transport modal preferences in the presence of various health control measures have also been undertaken. The findings suggest that overall health control measures are appreciated by passengers and do align with promotion of a return to public transport use. However, their effects are quite moderate and should be selectively used for different transport modes. Increasing service frequency can be a promising solution, which reduces crowding and waiting time of passengers using public transport modes, and reduces virus transmission risk. This study contributes to a better understanding of public acceptance and preference toward health control measures in public transport, and call for in-depth cost-benefit analysis in the related fields so that better response can be made in possible future public health crisis.

An Exact Method for (Constrained) Assortment Optimization Problems with Product Costs

We study the problem of optimizing assortment decisions in the presence of product-specific costs when customers choose according to a multinomial logit model. This problem is NP-hard, and approximate solutions methods have been proposed in the literature to obtain both lower and upper bounds in a tractable manner. We propose the first exact solution method for this problem and show that provably optimal assortments of instances with up to 1,000 products can be found, on average, in about 2/10 of a second. In particular, we propose a bounding procedure to enhance an approximation method originally proposed by Feldman and Topaloglu and provide tight lower and upper bounds at a fraction of a second. We show how these bounds can be used to effectively identify an optimal assortment. We also describe how to adapt our approach to handle cardinality or space/resource capacity constraints on the assortment as well as assortment optimization under a mixed-multinomial logit model. In both cases, our solution method provides significant computational boosts compared with exact methods from the literature.

An exploration of preferences for soil health practices in potato production

Given the renewed interest in promoting healthy soils in agricultural production via practices such as cover crops, decreased fumigations, or biological fertilizers, this article explores U.S. potato farmers' preferences for soil health-related production practices. We conduct a discrete choice experiment (DCE) under which hypothetical practices varied based on their effect on pests, fertilizer requirements, irrigation needs, additional soil testing/education, and expected net returns. Surveys were deployed to potato growers using convenience sampling online—through practitioner journals—and in person—at industry meetings. Standard descriptive statistics of the sample are presented and the DCE experiment was analyzed using a mixed multinomial logit model. Results suggest that when considering a new practice, farmers are highly sensitive to changes in net returns (p ​< ​0.001), and somewhat responsive to changes in fertilizer requirements, irrigation, and pests (p ​< ​0.01). However, potato growers appear indifferent to changes in workload, education, or soil testing requirements. While potato growers’ responses to the survey suggest they are highly concerned with soil health, conventional practices (e.g., 2–3 years rotations, standard tillage, fumigation) are still common. Additionally, we find that complexity in implementation, uncertainty in economic benefits, and capital constraints are the most common barriers to adopting novel soil health practices.

Preferences in the Design and Delivery of Neurodevelopmental Follow-Up Care for Children: A Systematic Review of Discrete Choice Experiments.

Neurodevelopmental disorders are a significant cause of morbidity. Early detection of neurodevelopmental delay is essential for timely diagnosis and intervention, and it is therefore important to understand the preferences of parents and clinicians for engaging with neurodevelopmental surveillance and follow-up care. Discrete choice experiment (DCE) may be an appropriate method for quantifying these preferences. This review systematically examined how DCEs have been designed and delivered in studies examining neurodevelopmental care of children and identified the preferred attributes that have been reported. PubMed, Embase, CINAHL, and Scopus databases were systematically searched. Studies were included if they used DCE to elicit preferences for a neurodevelopmental follow-up program for children. Two independent reviewers conducted the title and abstract and full-text screening. Risk of bias was assessed using a DCE-specific checklist. Findings were presented using a narrative synthesis. A total of 6618 records were identified and 16 papers were included. Orthogonal (n=5) and efficient (n=5) experimental designs were common. There was inconsistent reporting of design-related features. Analysis was primarily completed using mixed logit (n=6) or multinomial logit (n=3) models. Several key attributes for neurodevelopmental follow-up care were identified including social, behavioral and emotional support, therapy, waiting time, and out-of-pocket costs. DCE has been successfully used as a preference elicitation method for neurodevelopmental-related care. There is scope for improvement in the design and analysis of DCE in this field. Nonetheless, attributes identified in these studies are likely to be important considerations in the design and implementation of programs for neurodevelopmental care.

Empowering revealed preference survey with a supplementary stated preference survey: demonstration of willingness-to-pay estimation within a mode choice case

Mode choice models play a pivotal role in transport demand modelling and help transport planners, engineers and researchers with policy and infrastructure investment evaluation. Recent mode choice studies primarily use revealed preference (RP) data to reflect individuals’ true behaviour. However, this may not be the best practice, given the lack of information in RP data. This study uses a nonlinear utility specification for a multinomial logit mode choice model development using high-quality travel data collected by a GPS-based smartphone application complemented by stated preference (SP) data. The model results highlight the impact of sociodemographic variables on mode choice behaviour and individuals’ willingness-to-pay (WTP) when the model is jointly developed compared to stand-alone SP and RP models. The main message of this study is that in addition to collecting RP, which is a reliable and unbiased source of data, collecting complementary SP data is beneficial as it provides information that is not otherwise available in RP data. This may include a proper variation in the public transport cost variable as demonstrated in this study. Moreover, to better understand the travellers' behaviour regarding the trade-off between time and cost a mixed multinomial logit (MMNL) model in the willingness to pay space is developed on the SP data. capturing the unobserved heterogeneity within the estimated WTPs, the MMNL model outputs reveal a higher variation in WTP of car in-vehicle travel time compared to bus in-vehicle travel time.

Informing a target product profile for rapid tests to identify HBV-infected pregnant women with high viral loads: a discrete choice experiment with African healthcare workers

BackgroundElimination of mother-to-child transmission of hepatitis B virus (HBV) requires infant immunoprophylaxis and antiviral prophylaxis for pregnant women with high viral loads. Since real-time polymerase chain reaction (RT-PCR), a gold standard for assessing antiviral eligibility, is neither accessible nor affordable for women living in low-income and middle-income countries (LMICs), rapid diagnostic tests (RDTs) detecting alternative HBV markers may be needed. To inform future development of the target product profile (TPP) for RDTs to identify highly viremic women, we used a discrete choice experiment (DCE) and elicited preference and trade-off of healthcare workers (HCW) in Africa between the following four attributes of fictional RDTs: price, time-to-result, diagnostic sensitivity, and specificity.MethodsThrough an online questionnaire survey, we asked participants to indicate their preferred test from a set of two RDTs in seven choice tasks with varying levels of the four attributes. We used mixed multinomial logit models to quantify the utility gain or loss generated by each attribute. We attempted to define minimal and optimal criteria for test attributes that can satisfy ≥ 70% and ≥ 90% of HCWs, respectively, as an alternative to RT-PCR.ResultsA total of 555 HCWs from 41 African countries participated. Increases in sensitivity and specificity generated significant utility and increases in cost and time-to-result generated significant disutility. The size of the coefficients for the highest attribute levels relative to the reference levels were in the following order: sensitivity (β = 3.749), cost (β = -2.550), specificity (β = 1.134), and time-to-result (β = -0.284). Doctors cared most about test sensitivity, while public health practitioners cared about cost and midwives about time-to-result. For an RDT with 95% specificity, costing 1 US$, and yielding results in 20 min, the minimally acceptable test sensitivity would be 82.5% and the optimally acceptable sensitivity would be 87.5%.ConclusionsAfrican HCWs would prefer an RDT with the following order of priority: higher sensitivity, lower cost, higher specificity, and shorter time-to-result. The development and optimization of RDTs that can meet the criteria are urgently needed to scale up the prevention of HBV mother-to-child transmission in LMICs.

Preferences for Adjuvant Immunotherapy in Adults with Resected Stage III Melanoma-A Discrete Choice Experiment.

This study aimed to quantify adult preferences for adjuvant immunotherapy for resected melanoma and the influence of varying levels of key attributes and baseline characteristics. A D-efficient design generated 12 choice tasks for two alternative treatments, adjuvant immunotherapy or no adjuvant immunotherapy. Recruitment to the online discrete choice experiment (DCE) occurred via survey dissemination by eight Australian melanoma consumer and professional groups, targeting adults with resected stage III melanoma, considering or having received adjuvant immunotherapy. The DCE included six attributes with two to three levels each, including 3-year risk of recurrence, mild, permanent and fatal adverse events (AEs), drug regimen and annual out-of-pocket costs. A mixed multinomial logit model was used to estimate preferences and calculate marginal rates of substitution and marginal willingness to pay (mWTP). The DCE was completed by 116 respondents, who chose adjuvant immunotherapy over no adjuvant immunotherapy in 70% of choice tasks. Respondents preferred adjuvant immunotherapy when associated with reduced: probabilities of recurrence, permanent and fatal AEs, and out-of-pocket costs. mWTP for an absolute reduction of 1% in 3-year risk of recurrence was less for respondents with lower rather than higher incomes, AU$794 (US$527) and AU$2190 (US$1454) per year. Respondents accepted an additional 4% chance of a permanent AE to reduce their absolute risk of 3-year recurrence by 1%. Respondents were willing to accept an extra 2% chance of 3-year recurrence to lower their chance of a fatal AE by 1%. Almost three-quarters of respondents chose adjuvant immunotherapy over no adjuvant immunotherapy, preferring treatment that improved efficacy and safety. Findings may inform decisions about access to adjuvant immunotherapy following surgery for melanoma.