Jui-Chen Yang - Academia.edu (original) (raw)

Papers by Jui-Chen Yang

Such deferrals are likely complicate matters later; indeed, neglecting the interests of consumers... more Such deferrals are likely complicate matters later; indeed, neglecting the interests of consumers has often derailed the reform process.

Infectious disease such as malaria, dengue, and diarrhea that are spread by vectors such as mosqu... more Infectious disease such as malaria, dengue, and diarrhea that are spread by vectors such as mosquitoes and flies are rife in the much of the developing world, potentially impacting more than two-third of the world's population. Environmental conditions that favor the transmission of such diseases are, in turn, affected by human activity such as deforestation, livestock rearing, irrigated farming, migration, road construction, dam-building, and water and sanitation infrastructure provision through their impact on the survival and abundance of disease vectors. Therefore, the prevention and control of these diseases relies on the interplay of public policies (e.g., reducing microbial pathogen load in public water supplies) and private choices (e.g., safe storage, treatment, and handling of drinking water and food inside the house). As described, externalities play important roles in disease dynamics, demonstrating the need for public policies such as subsidies/taxes, information and technical assistance to achieve socially desirable outcomes. This paper presents an application in public economics that tests analytical models of economic epidemiology. The first part of the paper reviews and synthesizes the early literature that has focused on the behavioral basis of disease control and prevention. The second part presents empirical applications where we use the analytical models to shape the econometric analyses, permitting perhaps the first tests of untested hypotheses about the effectiveness of disease control interventions and how they related to private disease control. We draw on several data sets, including a large multi-year, multi-scale on going study from rural India to build econometric models of disease outcomes and public interventions. We conclude with a discussion of policy implications and conceptual and empirical research extensions.

Clinical Journal of The American Society of Nephrology, Sep 1, 2022

Background and objectives Approximately 20% of deceased donor kidneys are discarded each year in ... more Background and objectives Approximately 20% of deceased donor kidneys are discarded each year in the United States. Some of these kidneys could benefit patients who are waitlisted. Understanding patient preferences regarding accepting marginal-quality kidneys could help more of the currently discarded kidneys be transplanted. Design, setting, participants, & measurements This study uses a discrete choice experiment that presents a deceased donor kidney to patients who are waiting for, or have received, a kidney transplant. The choices involve trade-offs between accepting a kidney today or a future kidney. The options were designed experimentally to quantify the relative importance of kidney quality (expected graft survival and level of kidney function) and waiting time. Choices were analyzed using a random-parameters logit model and latent-class analysis. Results In total, 605 participants completed the discrete choice experiment. Respondents made trade-offs between kidney quality and waiting time. The average respondent would accept a kidney today, with 6.5 years of expected graft survival (95% confidence interval, 5.9 to 7.0), to avoid waiting 2 additional years for a kidney, with 11 years of expected graft survival. Three patient-preference classes were identified. Class 1 was averse to additional waiting time, but still responsive to improvements in kidney quality. Class 2 was less willing to accept increases in waiting time for improvements in kidney quality. Class 3 was willing to accept increases in waiting time even for small improvements in kidney quality. Relative to class 1, respondents in class 3 were likely to be age ≤61 years and to be waitlisted before starting dialysis, and respondents in class 2 were more likely to be older, Black, not have a college degree, and have lower Karnofsky performance status. Conclusions Participants preferred accepting a lower-quality kidney in return for shorter waiting time, particularly those who were older and had lower functional status.

Obstetrical & Gynecological Survey, Dec 1, 2019

BACKGROUND: Randomized trials have reported conflicting findings on survival for advanced-stage o... more BACKGROUND: Randomized trials have reported conflicting findings on survival for advanced-stage ovarian cancer treated with primary debulking surgery (PDS) versus neoadjuvant chemotherapy with interval debulking; surgical complications and mortality are higher with PDS. We assessed women's preferences for tradeoffs related to this important clinical decision. METHODS: Ovarian cancer patients were recruited to complete a discrete-choice experiment (DCE) consisting of 8 choice tasks presenting experimentally designed treatment alternatives in terms of treatment order, extent of surgery including risk of ostomy, chance of death from surgical complications (1%-10%), readmission for surgical complications (5%-50%), progression-free survival (1-3 years), and overall survival (3-5 years). Randomparameters logit regression was applied to model participants' choices as a function of attribute levels. RESULTS: A total of 101 ovarian cancer survivors completed the DCE survey; of these participants, 30% were receiving chemotherapy at the time, and 33% had prior recurrence. Overall survival was of greatest importance to participants (36/100), followed by risk of readmission due to complications (23/100), progression-free survival (19/100), surgical mortality (16/100), extent of surgery (4/100), and order of surgery and chemotherapy (2/100). Overall, the participants would tolerate a 15-percentage point increase in risk of major complications (95% confidence interval [CI], 3%-29%) or a 4-percentage point increase in the risk of surgical mortality (95% CI, 2%-13%) in order to increase their expected overall survival from 3 to 3.5 years. CONCLUSIONS: Patients would accept a moderately higher risk of perioperative complications and surgical mortality in exchange for substantial gains in survival. These quantitative findings provide clinicians with a framework to discuss preferences with patients and to incorporate preferences into clinical trial design.

The Patient: Patient-Centered Outcomes Research, Aug 11, 2023

Background Treatment options for patients with metastatic hormone-sensitive prostate cancer (mHSP... more Background Treatment options for patients with metastatic hormone-sensitive prostate cancer (mHSPC) have broadened, and treatment decisions can have a long-lasting impact on patients' quality of life. Data on patient preferences can improve therapeutic decision-making by helping physicians suggest treatments that align with patients' values and needs. Objective This study aims to quantify patient preferences for attributes of chemohormonal therapies among patients with mHSPC in the USA, Canada, and the UK. Methods A discrete-choice experiment survey instrument was developed and administered to patients with high-and veryhigh-risk localized prostate cancer and mHSPC. Patients chose between baseline androgen-deprivation therapy (ADT) alone and experimentally designed, hypothetical treatment alternatives representing chemohormonal therapies. Choices were analyzed using logit models to derive the relative importance of attributes for each country and to evaluate differences and similarities among patients across countries. Results A total of 550 respondents completed the survey (USA, 200; Canada, 200; UK, 150); the mean age of respondents was 64.3 years. Treatment choices revealed that patients were most concerned with treatment efficacy. However, treatmentrelated convenience factors, such as route of drug administration and frequency of monitoring visits, were as important as some treatment-related side effects, such as skin rash, nausea, and fatigue. Patient preferences across countries were similar, although patients in Canada appeared to be more affected by concomitant steroid use. Conclusion Patients with mHSPC believe the use of ADT alone is insufficient when more effective treatments are available. Efficacy is the most significant driver of patient choices. Treatment-related convenience factors can be as important as safety concerns for patients. Key Points for Decision Makers Preferences of patients with prostate cancer were derived from choices between experimentally designed treatment alternatives. Patients preferred treatment options with the highest efficacy and ones without risks of serious infections. Patients valued treatment convenience as much as treatment-related side effects.

Value in Health, Jun 1, 2023

Value in Health, Jun 1, 2023

medRxiv (Cold Spring Harbor Laboratory), May 29, 2023

doi: medRxiv preprint NOTE: This preprint reports new research that has not been certified by pee... more doi: medRxiv preprint NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.

The Millennium Development Goals reflect the world's collective hope and resolve to reverse a par... more The Millennium Development Goals reflect the world's collective hope and resolve to reverse a particularly pernicious, pervasive, and persistent set of problems in much of the world: high rates of diarrhea (the number one killer of small children), insufficient water and sanitation, and seemingly unsafe and myopic behaviors. Environmental health policies related to water and sanitation (W&S) must address the usual efficiency criteria (e.g., externalities), but also significant equity concerns. Health, time, and energy costs fall disproportionately on the poor, women and children. While there is an extensive literature on the appropriate theoretical model for evaluating welfare impacts, these have been rarely used in estimating gains from environmental health interventions both because adequate data for such evaluations are scarce and because few practitioners have explored how the textbook elegance of these models correspond to field realities. Thus, our paper uses a large-scale community-demand-driven (CDD) water supply project to (a) describe the challenges of welfare estimation using revealed preference data on multiple inputs and outputs, and (b) showcase a unique combination of propensity-score 'pre-matching' and rich panel data for estimating welfare impacts of a multi-dimensional environmental health project. Three years after project initiation, we found that the CDD project had had a significant impact on reported use of taps (13% increase) and toilets (7% increase). Diarrhea incidence fell significantly during the evaluation period in both project and control villages, suggesting weak health impacts of the project. In terms of economic welfare, we derive several empirical regularities related to illness and coping costs. Overall, our estimates of benefits indicate savings in coping costs equivalent to 5% of monthly expenditures, suggesting potentially significant gains from rural water and sanitation policies.

The views expressed in this paper are those of the author(s) and do not necessarily reflect the v... more The views expressed in this paper are those of the author(s) and do not necessarily reflect the views or policies of the Asian Development Bank.

Value in Health, Oct 1, 2021

BACKGROUND 'Hope' is a construct in patient-centered value frameworks, but few studies ha... more BACKGROUND 'Hope' is a construct in patient-centered value frameworks, but few studies have attempted to measure the value of hope separately from treatment-related gains in quality of life and survival to support its application in economic evaluation. OBJECTIVE To generate quantitative information on the "value of hope". METHODS We designed a discrete-choice experiment in which treatment alternatives varied the probability of achieving 10-year survival, expected survival as the weighted sum of short-term and long-term survival, health status, and out-of-pocket cost. Two-hundred patients with cancer or history of cancer recruited by Cancer Support Community each completed 10 choice questions. We used mixed-logit and latent-class models to analyze the choice data. RESULTS Relative to fixed survival periods of two, three or five years with 0% chance of 10-year survival, participants positively valued treatments with 5% and 10% chances of 10-year survival. However, participants negatively valued a 20% chance of 10-year survival that required an offsetting 80% chance of shorter survival. This finding was particularly strong when expected survival was two years. Compared to a 0% chance, dollar-equivalent values of 5% and 10% chances of long-term survival were 5,975and5,975 and 5,975and12,421, respectively, independent of health status or expected survival. The corresponding value for 20% versus 0% chance of long-term survival was negative. Latent-class analysis revealed 4 groups with distinct preference patterns. CONCLUSIONS Our findings affirm positive value for hope independent of expected survival and health status. However, this finding does not universally hold in all situations nor across all groups.

Journal of Bone and Joint Surgery, American Volume, Oct 8, 2020

Background: Total knee arthroplasty (TKA) is a common treatment for end-stage knee osteoarthritis... more Background: Total knee arthroplasty (TKA) is a common treatment for end-stage knee osteoarthritis but is associated with increased complication rates compared with unicompartmental knee arthroplasty (UKA). UKA offers better functional outcomes but is associated with a higher risk of revision. The purpose of this study was to apply good-practice, statedpreference methods to quantify patient preferences for benefit-risk tradeoffs associated with arthroplasty treatments for end-stage knee osteoarthritis. Methods: A discrete-choice experiment was developed with the following attributes: chance of complications, functional ability, awareness of the knee implant, and chance of needing another operation within 10 years. Patients included those aged 40 to 80 years with knee osteoarthritis. A pivot design filtered respondents into 1 of 2 surveys on the basis of selfreported functional ability (good compared with fair or poor) as measured by the Oxford Knee Score. Treatment-preference data were collected, and relative attribute-importance weights were estimated. Results: Two hundred and fifty-eight completed survey instruments from 92 males and 164 females were analyzed, with 72 respondents in the good-function cohort and 186 in the fair/poor-function cohort. Patients placed the greatest value or relative importance on serious complications and rates of revision in both cohorts. Preference weights did not vary between cohorts for any attribute. In the good-function cohort, 42% of respondents chose TKA and 58% chose UKA. In the fair/poor-function cohort, 54% chose TKA and 46% chose UKA. Conclusions: Patient preferences for various treatment attributes varied among patients in a knee osteoarthritis population. Complication and revision rates were the most important factors to patients, suggesting that physicians should focus on these areas when discussing treatments. The proportion of patients who chose UKA suggests that the current trend of increased UKA utilization is aligned with patient preferences. Clinical Relevance: Systematic elicitation of patient preferences for knee arthroplasty procedures, which lays out evidence-based risks and benefits of different treatments, indicates a larger subset of the knee osteoarthritis population may prefer UKA than would be suggested by the current rates of utilization of the procedure. Arthroplasty treatment should align with patient preferences and eligibility criteria to better deliver patient-centered care. Disclosure: This study was funded by a grant from Zimmer Biomet. Zimmer Biomet played no role in the study. On the Disclosure of Potential Conflicts of Interest forms, which are provided with the online version of the article, one or more of the authors checked "yes" to indicate that the author had a relevant financial relationship in the biomedical arena outside the submitted work (including relationships with Zimmer Biomet) (http://links.lww.com/JBJS/G141).

Journal of Clinical Oncology, May 20, 2019

5558 Background: Maintenance therapy with PARP inhibitors has become prevalent in treating ovaria... more 5558 Background: Maintenance therapy with PARP inhibitors has become prevalent in treating ovarian cancer. However, the preferences of women with ovarian cancer regarding the risks, side effects and benefits afforded by maintenance therapies are largely unstudied. Methods: A discrete choice experiment was designed to elicit the preferences of women with ovarian cancer regarding 6 attributes (levels presented in parentheses) relevant to the decision for maintenance PARP inhibitor therapy versus surveillance: (1) overall survival (36, 38, 42 months); (2) progression-free survival (15, 17, 21 months); (3) nausea (none, mild, moderate); (4) fatigue (none, mild, moderate); (5) probability of death from myelodysplastic syndrome/acute myelogenous leukemia (MDS/AML) (0%, 1%, 5%, 10%); and (6) monthly out-of-pocket cost ($0, 50,50, 50,500, $1,000). Educational material was provided, with embedded questions to test respondents’ understanding. Participants chose between 2 variable clinical scenarios and a constant scenario representing a treatment break, with multiple iterations. Random-parameters logit regression was applied to model participants’ choices as a function of attribute levels. Results: Of 150 women with ovarian cancer recruited, 95 were eligible and completed the survey. The mean age was 62, 48% had recurrent ovarian cancer, and 13% were currently taking a PARP. Participants always significantly preferred better clinical outcomes to worse (except between 0% and 1% risk of MDS/AML), preferred low out-of-pocket cost, and disliked the idea of a treatment break. Participants valued overall survival most (average importance weight 25 out of 100 total), followed by monthly out-of-pocket cost (24), risk of death from MDS/AML (18), nausea (15), PFS (10) and fatigue (8). On average, participants would tolerate a 2% additional risk of MDS/AML in exchange for 2 additional months of PFS and 7% additional risk of MDS/AML in exchange for 6 additional months of PFS. Conclusions: Women with ovarian cancer are willing to accept the side effects of PARP maintenance therapy and a higher than clinically observed risk of MDS/AML in exchange for clinically observed levels of improvement in PFS.

Social Science & Medicine, Aug 1, 2012

Over 5 billion people worldwide are exposed to unsafe water. Given the obstacles to ensuring sust... more Over 5 billion people worldwide are exposed to unsafe water. Given the obstacles to ensuring sustainable improvements in water supply infrastructure and the unhygienic handling of water after collection, household water treatment and storage (HWTS) products have been viewed as important mechanisms for increasing access to safe water. Although studies have shown that HWTS technologies can reduce the likelihood of diarrheal illness by about 30%, levels of adoption and continued use remain low. An understanding of household preferences for HWTS products can be used to create demand through effective product positioning and social marketing, and ultimately improve and ensure commercial sustainability and scalability of these products. However, there has been little systematic research on consumer preferences for HWTS products. This paper reports the results of the first state-of-the-art conjoint analysis study of HWTS products. In 2008, we conducted a conjoint analysis survey of a representative sample of households in Andhra Pradesh (AP), India to elicit and quantify household preferences for commercial HWTS products. Controlling for attribute non-attendance in an error components mixed logit model, the study results indicate that the most important features to respondents, in terms of the effect on utility, were the type of product, followed by the extent to which the product removes pathogens, the retail outlet and, the time required to treat 10 L. Holding all other product attributes constant, filters were preferred to combination products and chemical additives. Department stores and weekly markets were the most favorable sales outlets, followed by mobile salespeople. In general, households do not prefer to purchase HWTS products at local shops. Our results can inform the types of products and sales outlets that are likely to be successful in commercial HWTS markets in AP, as well as the influence of different pricing and financing strategies on product demand and uptake.

Journal of choice modelling, Sep 1, 2015

•A meta-simulation strategy analyzes what affects utility-difference precision.•Meta-data are obt... more •A meta-simulation strategy analyzes what affects utility-difference precision.•Meta-data are obtained from 34 individual DCE patient-preference data sets.•Utility-difference precision is regressed on sample size and study characteristics.•Study-design features affect precision more than experimental-design efficiency.•An empirical sample-size power equation is derived for DCE health studies.

Value in Health, May 1, 2018

The Journal of Allergy and Clinical Immunology: In Practice, May 1, 2021

BACKGROUND Recently developed peanut desensitization treatment reduces the incidence of allergic ... more BACKGROUND Recently developed peanut desensitization treatment reduces the incidence of allergic reactions, the anxiety associated with the risk of accidental exposure, and the burden of precautionary behavior. Eliciting parent preferences for tradeoffs involving treatment effectiveness, tolerability, costs, and convenience quantifies the burden of juvenile peanut allergy and the perceived value of peanut desensitization therapies. OBJECTIVE This study aimed to understand heterogeneity in parents' treatment preferences and the role of personal characteristics in explaining differences. METHODS An Internet-based, discrete-choice experiment survey was administered to a national sample of 500 parents of children 4 to 17 years of age with peanut allergy to quantify parents' preferences for peanut desensitization therapies for their children. Latent-class, mixed-logit analysis estimated relative-importance coefficients for groups of participants with distinctly different preferences. RESULTS Parents' choice patterns fell into one of four preference subgroups: 1) Cost-sensitive, 2) Pro-treatment (but Side Effect-Averse), 3) Trader, and 4) Inconsistent. Mode of administration had little relative importance across all subgroups. Characteristics associated with belonging to a given preference subgroup included parent age, child age, income, parent perception of child risk and ability to manage allergic reactions, past allergic reactions, and changes in precautionary behaviors post-treatment. CONCLUSION We found distinct differences in parent preferences for tradeoffs involving effectiveness, tolerability, and costs of peanut desensitization treatments. Parents' treatment preferences help quantify the burden of juvenile peanut allergy and the perceived value of new therapies. Such information can inform patient-centric clinical and regulatory decision-making.

Journal of Developmental and Behavioral Pediatrics, 2016

Objective: Fragile X syndrome (FXS) is the most common inherited form of intellectual disability.... more Objective: Fragile X syndrome (FXS) is the most common inherited form of intellectual disability. The objective of this study was to determine the relative importance that caregivers place on improving different phenotypic traits observed in males with FXS to better understand the greatest medical needs for developing and evaluating FXS treatments. Method: Fragile X syndrome caregivers (n 5 614) compared hypothetical treatments in a discrete-choice experiment. The treatments varied in their effects on 6 outcomes associated with FXS: learning and applying new skills, explaining needs, controlling behavior, taking part in new social activities, caring for oneself, and paying attention. The relative importance was calculated for improving severe or moderate levels of disability and transformed to a 10-point scale. Relative importance was also quantified by patient age group (child, adolescent, and adult). Results: Most important to caregivers were controlling behavior (10.0) and caring for oneself (9.9). Least important was taking part in new social activities (4.2). A partial improvement in controlling behavior or self-care was more important than full resolution of the least important disabilities. This was consistent across age groups. Improvements from severe to moderate disability were more important than from moderate to no disability. Conclusion: Caregivers expressed strong preferences for improvement in self-care and behavioral control, independent of the age of the individual with FXS. These data may be helpful when designing studies to test the efficacy of FXS treatments because small treatment effects on very important outcomes may be valued more than large treatment effects on less valued outcomes.

Value in Health, May 1, 2018

S11 strong progressive design of Pharmaceutical benefits, Pharmaceutical spending over 0-11 years... more S11 strong progressive design of Pharmaceutical benefits, Pharmaceutical spending over 0-11 years of age favored the poor only insignificantly (concentration index-0.067, 95% CI-0.155, 0.023). The Medical spending on the other hand favored the rich slightly (concentration index 0.043, 95% CI 0.028, 0.058), despite its design to be predominantly income-neutral. The total Medical and Pharmaceutical government spending-sum of the two schemes-was slightly pro-rich (concentration index 0.035, 95% CI 0.018, 0.052). ConCluSionS: The income-related distributions of the publicly funded, universal, Medical and Pharmaceutical benefits for children in Australia slightly deviated from their designs regarding income, which may reflect differences in parents' healthcare-seeking preferences and access or barriers to health. Results will inform the design and reform of universal health insurance schemes especially for young children.

Such deferrals are likely complicate matters later; indeed, neglecting the interests of consumers... more Such deferrals are likely complicate matters later; indeed, neglecting the interests of consumers has often derailed the reform process.

Infectious disease such as malaria, dengue, and diarrhea that are spread by vectors such as mosqu... more Infectious disease such as malaria, dengue, and diarrhea that are spread by vectors such as mosquitoes and flies are rife in the much of the developing world, potentially impacting more than two-third of the world's population. Environmental conditions that favor the transmission of such diseases are, in turn, affected by human activity such as deforestation, livestock rearing, irrigated farming, migration, road construction, dam-building, and water and sanitation infrastructure provision through their impact on the survival and abundance of disease vectors. Therefore, the prevention and control of these diseases relies on the interplay of public policies (e.g., reducing microbial pathogen load in public water supplies) and private choices (e.g., safe storage, treatment, and handling of drinking water and food inside the house). As described, externalities play important roles in disease dynamics, demonstrating the need for public policies such as subsidies/taxes, information and technical assistance to achieve socially desirable outcomes. This paper presents an application in public economics that tests analytical models of economic epidemiology. The first part of the paper reviews and synthesizes the early literature that has focused on the behavioral basis of disease control and prevention. The second part presents empirical applications where we use the analytical models to shape the econometric analyses, permitting perhaps the first tests of untested hypotheses about the effectiveness of disease control interventions and how they related to private disease control. We draw on several data sets, including a large multi-year, multi-scale on going study from rural India to build econometric models of disease outcomes and public interventions. We conclude with a discussion of policy implications and conceptual and empirical research extensions.

Clinical Journal of The American Society of Nephrology, Sep 1, 2022

Background and objectives Approximately 20% of deceased donor kidneys are discarded each year in ... more Background and objectives Approximately 20% of deceased donor kidneys are discarded each year in the United States. Some of these kidneys could benefit patients who are waitlisted. Understanding patient preferences regarding accepting marginal-quality kidneys could help more of the currently discarded kidneys be transplanted. Design, setting, participants, & measurements This study uses a discrete choice experiment that presents a deceased donor kidney to patients who are waiting for, or have received, a kidney transplant. The choices involve trade-offs between accepting a kidney today or a future kidney. The options were designed experimentally to quantify the relative importance of kidney quality (expected graft survival and level of kidney function) and waiting time. Choices were analyzed using a random-parameters logit model and latent-class analysis. Results In total, 605 participants completed the discrete choice experiment. Respondents made trade-offs between kidney quality and waiting time. The average respondent would accept a kidney today, with 6.5 years of expected graft survival (95% confidence interval, 5.9 to 7.0), to avoid waiting 2 additional years for a kidney, with 11 years of expected graft survival. Three patient-preference classes were identified. Class 1 was averse to additional waiting time, but still responsive to improvements in kidney quality. Class 2 was less willing to accept increases in waiting time for improvements in kidney quality. Class 3 was willing to accept increases in waiting time even for small improvements in kidney quality. Relative to class 1, respondents in class 3 were likely to be age ≤61 years and to be waitlisted before starting dialysis, and respondents in class 2 were more likely to be older, Black, not have a college degree, and have lower Karnofsky performance status. Conclusions Participants preferred accepting a lower-quality kidney in return for shorter waiting time, particularly those who were older and had lower functional status.

Obstetrical & Gynecological Survey, Dec 1, 2019

BACKGROUND: Randomized trials have reported conflicting findings on survival for advanced-stage o... more BACKGROUND: Randomized trials have reported conflicting findings on survival for advanced-stage ovarian cancer treated with primary debulking surgery (PDS) versus neoadjuvant chemotherapy with interval debulking; surgical complications and mortality are higher with PDS. We assessed women's preferences for tradeoffs related to this important clinical decision. METHODS: Ovarian cancer patients were recruited to complete a discrete-choice experiment (DCE) consisting of 8 choice tasks presenting experimentally designed treatment alternatives in terms of treatment order, extent of surgery including risk of ostomy, chance of death from surgical complications (1%-10%), readmission for surgical complications (5%-50%), progression-free survival (1-3 years), and overall survival (3-5 years). Randomparameters logit regression was applied to model participants' choices as a function of attribute levels. RESULTS: A total of 101 ovarian cancer survivors completed the DCE survey; of these participants, 30% were receiving chemotherapy at the time, and 33% had prior recurrence. Overall survival was of greatest importance to participants (36/100), followed by risk of readmission due to complications (23/100), progression-free survival (19/100), surgical mortality (16/100), extent of surgery (4/100), and order of surgery and chemotherapy (2/100). Overall, the participants would tolerate a 15-percentage point increase in risk of major complications (95% confidence interval [CI], 3%-29%) or a 4-percentage point increase in the risk of surgical mortality (95% CI, 2%-13%) in order to increase their expected overall survival from 3 to 3.5 years. CONCLUSIONS: Patients would accept a moderately higher risk of perioperative complications and surgical mortality in exchange for substantial gains in survival. These quantitative findings provide clinicians with a framework to discuss preferences with patients and to incorporate preferences into clinical trial design.

The Patient: Patient-Centered Outcomes Research, Aug 11, 2023

Background Treatment options for patients with metastatic hormone-sensitive prostate cancer (mHSP... more Background Treatment options for patients with metastatic hormone-sensitive prostate cancer (mHSPC) have broadened, and treatment decisions can have a long-lasting impact on patients' quality of life. Data on patient preferences can improve therapeutic decision-making by helping physicians suggest treatments that align with patients' values and needs. Objective This study aims to quantify patient preferences for attributes of chemohormonal therapies among patients with mHSPC in the USA, Canada, and the UK. Methods A discrete-choice experiment survey instrument was developed and administered to patients with high-and veryhigh-risk localized prostate cancer and mHSPC. Patients chose between baseline androgen-deprivation therapy (ADT) alone and experimentally designed, hypothetical treatment alternatives representing chemohormonal therapies. Choices were analyzed using logit models to derive the relative importance of attributes for each country and to evaluate differences and similarities among patients across countries. Results A total of 550 respondents completed the survey (USA, 200; Canada, 200; UK, 150); the mean age of respondents was 64.3 years. Treatment choices revealed that patients were most concerned with treatment efficacy. However, treatmentrelated convenience factors, such as route of drug administration and frequency of monitoring visits, were as important as some treatment-related side effects, such as skin rash, nausea, and fatigue. Patient preferences across countries were similar, although patients in Canada appeared to be more affected by concomitant steroid use. Conclusion Patients with mHSPC believe the use of ADT alone is insufficient when more effective treatments are available. Efficacy is the most significant driver of patient choices. Treatment-related convenience factors can be as important as safety concerns for patients. Key Points for Decision Makers Preferences of patients with prostate cancer were derived from choices between experimentally designed treatment alternatives. Patients preferred treatment options with the highest efficacy and ones without risks of serious infections. Patients valued treatment convenience as much as treatment-related side effects.

Value in Health, Jun 1, 2023

Value in Health, Jun 1, 2023

medRxiv (Cold Spring Harbor Laboratory), May 29, 2023

doi: medRxiv preprint NOTE: This preprint reports new research that has not been certified by pee... more doi: medRxiv preprint NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.

The Millennium Development Goals reflect the world's collective hope and resolve to reverse a par... more The Millennium Development Goals reflect the world's collective hope and resolve to reverse a particularly pernicious, pervasive, and persistent set of problems in much of the world: high rates of diarrhea (the number one killer of small children), insufficient water and sanitation, and seemingly unsafe and myopic behaviors. Environmental health policies related to water and sanitation (W&S) must address the usual efficiency criteria (e.g., externalities), but also significant equity concerns. Health, time, and energy costs fall disproportionately on the poor, women and children. While there is an extensive literature on the appropriate theoretical model for evaluating welfare impacts, these have been rarely used in estimating gains from environmental health interventions both because adequate data for such evaluations are scarce and because few practitioners have explored how the textbook elegance of these models correspond to field realities. Thus, our paper uses a large-scale community-demand-driven (CDD) water supply project to (a) describe the challenges of welfare estimation using revealed preference data on multiple inputs and outputs, and (b) showcase a unique combination of propensity-score 'pre-matching' and rich panel data for estimating welfare impacts of a multi-dimensional environmental health project. Three years after project initiation, we found that the CDD project had had a significant impact on reported use of taps (13% increase) and toilets (7% increase). Diarrhea incidence fell significantly during the evaluation period in both project and control villages, suggesting weak health impacts of the project. In terms of economic welfare, we derive several empirical regularities related to illness and coping costs. Overall, our estimates of benefits indicate savings in coping costs equivalent to 5% of monthly expenditures, suggesting potentially significant gains from rural water and sanitation policies.

The views expressed in this paper are those of the author(s) and do not necessarily reflect the v... more The views expressed in this paper are those of the author(s) and do not necessarily reflect the views or policies of the Asian Development Bank.

Value in Health, Oct 1, 2021

BACKGROUND 'Hope' is a construct in patient-centered value frameworks, but few studies ha... more BACKGROUND 'Hope' is a construct in patient-centered value frameworks, but few studies have attempted to measure the value of hope separately from treatment-related gains in quality of life and survival to support its application in economic evaluation. OBJECTIVE To generate quantitative information on the "value of hope". METHODS We designed a discrete-choice experiment in which treatment alternatives varied the probability of achieving 10-year survival, expected survival as the weighted sum of short-term and long-term survival, health status, and out-of-pocket cost. Two-hundred patients with cancer or history of cancer recruited by Cancer Support Community each completed 10 choice questions. We used mixed-logit and latent-class models to analyze the choice data. RESULTS Relative to fixed survival periods of two, three or five years with 0% chance of 10-year survival, participants positively valued treatments with 5% and 10% chances of 10-year survival. However, participants negatively valued a 20% chance of 10-year survival that required an offsetting 80% chance of shorter survival. This finding was particularly strong when expected survival was two years. Compared to a 0% chance, dollar-equivalent values of 5% and 10% chances of long-term survival were 5,975and5,975 and 5,975and12,421, respectively, independent of health status or expected survival. The corresponding value for 20% versus 0% chance of long-term survival was negative. Latent-class analysis revealed 4 groups with distinct preference patterns. CONCLUSIONS Our findings affirm positive value for hope independent of expected survival and health status. However, this finding does not universally hold in all situations nor across all groups.

Journal of Bone and Joint Surgery, American Volume, Oct 8, 2020

Background: Total knee arthroplasty (TKA) is a common treatment for end-stage knee osteoarthritis... more Background: Total knee arthroplasty (TKA) is a common treatment for end-stage knee osteoarthritis but is associated with increased complication rates compared with unicompartmental knee arthroplasty (UKA). UKA offers better functional outcomes but is associated with a higher risk of revision. The purpose of this study was to apply good-practice, statedpreference methods to quantify patient preferences for benefit-risk tradeoffs associated with arthroplasty treatments for end-stage knee osteoarthritis. Methods: A discrete-choice experiment was developed with the following attributes: chance of complications, functional ability, awareness of the knee implant, and chance of needing another operation within 10 years. Patients included those aged 40 to 80 years with knee osteoarthritis. A pivot design filtered respondents into 1 of 2 surveys on the basis of selfreported functional ability (good compared with fair or poor) as measured by the Oxford Knee Score. Treatment-preference data were collected, and relative attribute-importance weights were estimated. Results: Two hundred and fifty-eight completed survey instruments from 92 males and 164 females were analyzed, with 72 respondents in the good-function cohort and 186 in the fair/poor-function cohort. Patients placed the greatest value or relative importance on serious complications and rates of revision in both cohorts. Preference weights did not vary between cohorts for any attribute. In the good-function cohort, 42% of respondents chose TKA and 58% chose UKA. In the fair/poor-function cohort, 54% chose TKA and 46% chose UKA. Conclusions: Patient preferences for various treatment attributes varied among patients in a knee osteoarthritis population. Complication and revision rates were the most important factors to patients, suggesting that physicians should focus on these areas when discussing treatments. The proportion of patients who chose UKA suggests that the current trend of increased UKA utilization is aligned with patient preferences. Clinical Relevance: Systematic elicitation of patient preferences for knee arthroplasty procedures, which lays out evidence-based risks and benefits of different treatments, indicates a larger subset of the knee osteoarthritis population may prefer UKA than would be suggested by the current rates of utilization of the procedure. Arthroplasty treatment should align with patient preferences and eligibility criteria to better deliver patient-centered care. Disclosure: This study was funded by a grant from Zimmer Biomet. Zimmer Biomet played no role in the study. On the Disclosure of Potential Conflicts of Interest forms, which are provided with the online version of the article, one or more of the authors checked "yes" to indicate that the author had a relevant financial relationship in the biomedical arena outside the submitted work (including relationships with Zimmer Biomet) (http://links.lww.com/JBJS/G141).

Journal of Clinical Oncology, May 20, 2019

5558 Background: Maintenance therapy with PARP inhibitors has become prevalent in treating ovaria... more 5558 Background: Maintenance therapy with PARP inhibitors has become prevalent in treating ovarian cancer. However, the preferences of women with ovarian cancer regarding the risks, side effects and benefits afforded by maintenance therapies are largely unstudied. Methods: A discrete choice experiment was designed to elicit the preferences of women with ovarian cancer regarding 6 attributes (levels presented in parentheses) relevant to the decision for maintenance PARP inhibitor therapy versus surveillance: (1) overall survival (36, 38, 42 months); (2) progression-free survival (15, 17, 21 months); (3) nausea (none, mild, moderate); (4) fatigue (none, mild, moderate); (5) probability of death from myelodysplastic syndrome/acute myelogenous leukemia (MDS/AML) (0%, 1%, 5%, 10%); and (6) monthly out-of-pocket cost ($0, 50,50, 50,500, $1,000). Educational material was provided, with embedded questions to test respondents’ understanding. Participants chose between 2 variable clinical scenarios and a constant scenario representing a treatment break, with multiple iterations. Random-parameters logit regression was applied to model participants’ choices as a function of attribute levels. Results: Of 150 women with ovarian cancer recruited, 95 were eligible and completed the survey. The mean age was 62, 48% had recurrent ovarian cancer, and 13% were currently taking a PARP. Participants always significantly preferred better clinical outcomes to worse (except between 0% and 1% risk of MDS/AML), preferred low out-of-pocket cost, and disliked the idea of a treatment break. Participants valued overall survival most (average importance weight 25 out of 100 total), followed by monthly out-of-pocket cost (24), risk of death from MDS/AML (18), nausea (15), PFS (10) and fatigue (8). On average, participants would tolerate a 2% additional risk of MDS/AML in exchange for 2 additional months of PFS and 7% additional risk of MDS/AML in exchange for 6 additional months of PFS. Conclusions: Women with ovarian cancer are willing to accept the side effects of PARP maintenance therapy and a higher than clinically observed risk of MDS/AML in exchange for clinically observed levels of improvement in PFS.

Social Science & Medicine, Aug 1, 2012

Over 5 billion people worldwide are exposed to unsafe water. Given the obstacles to ensuring sust... more Over 5 billion people worldwide are exposed to unsafe water. Given the obstacles to ensuring sustainable improvements in water supply infrastructure and the unhygienic handling of water after collection, household water treatment and storage (HWTS) products have been viewed as important mechanisms for increasing access to safe water. Although studies have shown that HWTS technologies can reduce the likelihood of diarrheal illness by about 30%, levels of adoption and continued use remain low. An understanding of household preferences for HWTS products can be used to create demand through effective product positioning and social marketing, and ultimately improve and ensure commercial sustainability and scalability of these products. However, there has been little systematic research on consumer preferences for HWTS products. This paper reports the results of the first state-of-the-art conjoint analysis study of HWTS products. In 2008, we conducted a conjoint analysis survey of a representative sample of households in Andhra Pradesh (AP), India to elicit and quantify household preferences for commercial HWTS products. Controlling for attribute non-attendance in an error components mixed logit model, the study results indicate that the most important features to respondents, in terms of the effect on utility, were the type of product, followed by the extent to which the product removes pathogens, the retail outlet and, the time required to treat 10 L. Holding all other product attributes constant, filters were preferred to combination products and chemical additives. Department stores and weekly markets were the most favorable sales outlets, followed by mobile salespeople. In general, households do not prefer to purchase HWTS products at local shops. Our results can inform the types of products and sales outlets that are likely to be successful in commercial HWTS markets in AP, as well as the influence of different pricing and financing strategies on product demand and uptake.

Journal of choice modelling, Sep 1, 2015

•A meta-simulation strategy analyzes what affects utility-difference precision.•Meta-data are obt... more •A meta-simulation strategy analyzes what affects utility-difference precision.•Meta-data are obtained from 34 individual DCE patient-preference data sets.•Utility-difference precision is regressed on sample size and study characteristics.•Study-design features affect precision more than experimental-design efficiency.•An empirical sample-size power equation is derived for DCE health studies.

Value in Health, May 1, 2018

The Journal of Allergy and Clinical Immunology: In Practice, May 1, 2021

BACKGROUND Recently developed peanut desensitization treatment reduces the incidence of allergic ... more BACKGROUND Recently developed peanut desensitization treatment reduces the incidence of allergic reactions, the anxiety associated with the risk of accidental exposure, and the burden of precautionary behavior. Eliciting parent preferences for tradeoffs involving treatment effectiveness, tolerability, costs, and convenience quantifies the burden of juvenile peanut allergy and the perceived value of peanut desensitization therapies. OBJECTIVE This study aimed to understand heterogeneity in parents' treatment preferences and the role of personal characteristics in explaining differences. METHODS An Internet-based, discrete-choice experiment survey was administered to a national sample of 500 parents of children 4 to 17 years of age with peanut allergy to quantify parents' preferences for peanut desensitization therapies for their children. Latent-class, mixed-logit analysis estimated relative-importance coefficients for groups of participants with distinctly different preferences. RESULTS Parents' choice patterns fell into one of four preference subgroups: 1) Cost-sensitive, 2) Pro-treatment (but Side Effect-Averse), 3) Trader, and 4) Inconsistent. Mode of administration had little relative importance across all subgroups. Characteristics associated with belonging to a given preference subgroup included parent age, child age, income, parent perception of child risk and ability to manage allergic reactions, past allergic reactions, and changes in precautionary behaviors post-treatment. CONCLUSION We found distinct differences in parent preferences for tradeoffs involving effectiveness, tolerability, and costs of peanut desensitization treatments. Parents' treatment preferences help quantify the burden of juvenile peanut allergy and the perceived value of new therapies. Such information can inform patient-centric clinical and regulatory decision-making.

Journal of Developmental and Behavioral Pediatrics, 2016

Objective: Fragile X syndrome (FXS) is the most common inherited form of intellectual disability.... more Objective: Fragile X syndrome (FXS) is the most common inherited form of intellectual disability. The objective of this study was to determine the relative importance that caregivers place on improving different phenotypic traits observed in males with FXS to better understand the greatest medical needs for developing and evaluating FXS treatments. Method: Fragile X syndrome caregivers (n 5 614) compared hypothetical treatments in a discrete-choice experiment. The treatments varied in their effects on 6 outcomes associated with FXS: learning and applying new skills, explaining needs, controlling behavior, taking part in new social activities, caring for oneself, and paying attention. The relative importance was calculated for improving severe or moderate levels of disability and transformed to a 10-point scale. Relative importance was also quantified by patient age group (child, adolescent, and adult). Results: Most important to caregivers were controlling behavior (10.0) and caring for oneself (9.9). Least important was taking part in new social activities (4.2). A partial improvement in controlling behavior or self-care was more important than full resolution of the least important disabilities. This was consistent across age groups. Improvements from severe to moderate disability were more important than from moderate to no disability. Conclusion: Caregivers expressed strong preferences for improvement in self-care and behavioral control, independent of the age of the individual with FXS. These data may be helpful when designing studies to test the efficacy of FXS treatments because small treatment effects on very important outcomes may be valued more than large treatment effects on less valued outcomes.

Value in Health, May 1, 2018

S11 strong progressive design of Pharmaceutical benefits, Pharmaceutical spending over 0-11 years... more S11 strong progressive design of Pharmaceutical benefits, Pharmaceutical spending over 0-11 years of age favored the poor only insignificantly (concentration index-0.067, 95% CI-0.155, 0.023). The Medical spending on the other hand favored the rich slightly (concentration index 0.043, 95% CI 0.028, 0.058), despite its design to be predominantly income-neutral. The total Medical and Pharmaceutical government spending-sum of the two schemes-was slightly pro-rich (concentration index 0.035, 95% CI 0.018, 0.052). ConCluSionS: The income-related distributions of the publicly funded, universal, Medical and Pharmaceutical benefits for children in Australia slightly deviated from their designs regarding income, which may reflect differences in parents' healthcare-seeking preferences and access or barriers to health. Results will inform the design and reform of universal health insurance schemes especially for young children.