scholarly journals Patients’ Preferences for Artificial Intelligence Applications Versus Clinicians in Disease Diagnosis During the SARS-CoV-2 Pandemic in China: Discrete Choice Experiment

10.2196/22841 ◽  
2021 ◽  
Vol 23 (2) ◽  
pp. e22841
Author(s):  
Taoran Liu ◽  
Winghei Tsang ◽  
Fengqiu Huang ◽  
Oi Ying Lau ◽  
Yanhui Chen ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Discrete Choice Experiment ◽  
Discrete Choice ◽  
Choice Experiment ◽  
Latent Class ◽  
Latent Class Model ◽  
Disease Diagnosis ◽  
Class Model ◽  
Class 1 ◽  
Time Accuracy

Background Misdiagnosis, arbitrary charges, annoying queues, and clinic waiting times among others are long-standing phenomena in the medical industry across the world. These factors can contribute to patient anxiety about misdiagnosis by clinicians. However, with the increasing growth in use of big data in biomedical and health care communities, the performance of artificial intelligence (Al) techniques of diagnosis is improving and can help avoid medical practice errors, including under the current circumstance of COVID-19. Objective This study aims to visualize and measure patients’ heterogeneous preferences from various angles of AI diagnosis versus clinicians in the context of the COVID-19 epidemic in China. We also aim to illustrate the different decision-making factors of the latent class of a discrete choice experiment (DCE) and prospects for the application of AI techniques in judgment and management during the pandemic of SARS-CoV-2 and in the future. Methods A DCE approach was the main analysis method applied in this paper. Attributes from different dimensions were hypothesized: diagnostic method, outpatient waiting time, diagnosis time, accuracy, follow-up after diagnosis, and diagnostic expense. After that, a questionnaire is formed. With collected data from the DCE questionnaire, we apply Sawtooth software to construct a generalized multinomial logit (GMNL) model, mixed logit model, and latent class model with the data sets. Moreover, we calculate the variables’ coefficients, standard error, P value, and odds ratio (OR) and form a utility report to present the importance and weighted percentage of attributes. Results A total of 55.8% of the respondents (428 out of 767) opted for AI diagnosis regardless of the description of the clinicians. In the GMNL model, we found that people prefer the 100% accuracy level the most (OR 4.548, 95% CI 4.048-5.110, P<.001). For the latent class model, the most acceptable model consists of 3 latent classes of respondents. The attributes with the most substantial effects and highest percentage weights are the accuracy (39.29% in general) and expense of diagnosis (21.69% in general), especially the preferences for the diagnosis “accuracy” attribute, which is constant across classes. For class 1 and class 3, people prefer the AI + clinicians method (class 1: OR 1.247, 95% CI 1.036-1.463, P<.001; class 3: OR 1.958, 95% CI 1.769-2.167, P<.001). For class 2, people prefer the AI method (OR 1.546, 95% CI 0.883-2.707, P=.37). The OR of levels of attributes increases with the increase of accuracy across all classes. Conclusions Latent class analysis was prominent and useful in quantifying preferences for attributes of diagnosis choice. People’s preferences for the “accuracy” and “diagnostic expenses” attributes are palpable. AI will have a potential market. However, accuracy and diagnosis expenses need to be taken into consideration.

scholarly journals Patients’ Preferences for Artificial Intelligence Applications Versus Clinicians in Disease Diagnosis During the SARS-CoV-2 Pandemic in China: Discrete Choice Experiment (Preprint)

2020 ◽  
Author(s):  
Taoran Liu ◽  
Winghei Tsang ◽  
Fengqiu Huang ◽  
Oi Ying Lau ◽  
Yanhui Chen ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Discrete Choice Experiment ◽  
Discrete Choice ◽  
Choice Experiment ◽  
Latent Class ◽  
Latent Class Model ◽  
Disease Diagnosis ◽  
Class Model ◽  
Class 1 ◽  
Time Accuracy

BACKGROUND Misdiagnosis, arbitrary charges, annoying queues, and clinic waiting times among others are long-standing phenomena in the medical industry across the world. These factors can contribute to patient anxiety about misdiagnosis by clinicians. However, with the increasing growth in use of big data in biomedical and health care communities, the performance of artificial intelligence (Al) techniques of diagnosis is improving and can help avoid medical practice errors, including under the current circumstance of COVID-19. OBJECTIVE This study aims to visualize and measure patients’ heterogeneous preferences from various angles of AI diagnosis versus clinicians in the context of the COVID-19 epidemic in China. We also aim to illustrate the different decision-making factors of the latent class of a discrete choice experiment (DCE) and prospects for the application of AI techniques in judgment and management during the pandemic of SARS-CoV-2 and in the future. METHODS A DCE approach was the main analysis method applied in this paper. Attributes from different dimensions were hypothesized: diagnostic method, outpatient waiting time, diagnosis time, accuracy, follow-up after diagnosis, and diagnostic expense. After that, a questionnaire is formed. With collected data from the DCE questionnaire, we apply Sawtooth software to construct a generalized multinomial logit (GMNL) model, mixed logit model, and latent class model with the data sets. Moreover, we calculate the variables’ coefficients, standard error, <i>P</i> value, and odds ratio (OR) and form a utility report to present the importance and weighted percentage of attributes. RESULTS A total of 55.8% of the respondents (428 out of 767) opted for AI diagnosis regardless of the description of the clinicians. In the GMNL model, we found that people prefer the 100% accuracy level the most (OR 4.548, 95% CI 4.048-5.110, <i>P</i>&lt;.001). For the latent class model, the most acceptable model consists of 3 latent classes of respondents. The attributes with the most substantial effects and highest percentage weights are the accuracy (39.29% in general) and expense of diagnosis (21.69% in general), especially the preferences for the diagnosis “accuracy” attribute, which is constant across classes. For class 1 and class 3, people prefer the AI + clinicians method (class 1: OR 1.247, 95% CI 1.036-1.463, <i>P</i>&lt;.001; class 3: OR 1.958, 95% CI 1.769-2.167, <i>P</i>&lt;.001). For class 2, people prefer the AI method (OR 1.546, 95% CI 0.883-2.707, <i>P</i>=.37). The OR of levels of attributes increases with the increase of accuracy across all classes. CONCLUSIONS Latent class analysis was prominent and useful in quantifying preferences for attributes of diagnosis choice. People’s preferences for the “accuracy” and “diagnostic expenses” attributes are palpable. AI will have a potential market. However, accuracy and diagnosis expenses need to be taken into consideration.

Tourists’ Willingness to Pay for holiday trip characteristics: A Discrete Choice Experiment

2020 ◽  
pp. 135481662095990
Author(s):  
David Boto-García ◽  
Petr Mariel ◽  
José Baños Pino ◽  
Antonio Alvarez
Keyword(s):  
Willingness To Pay ◽  
Discrete Choice Experiment ◽  
Discrete Choice ◽  
Choice Experiment ◽  
Latent Class ◽  
Latent Class Model ◽  
Welfare Loss ◽  
Northern Spain ◽  
Marginal Rates Of Substitution ◽  
Mode Of Transport

This article studies the marginal rates of substitution and Willingness to Pay for holiday trip characteristics. Using a Discrete Choice Experiment, we examine how much individuals from four cities in Northern Spain are willing to pay for accommodation, mode of transport, travel time and length of stay. We estimate a Latent Class Model that accounts for taste heterogeneity based on sociodemographic characteristics. The welfare loss due to a tourism daily tax is also examined. Our results show that respondents place positive utility to travelling by plane, high-quality accommodation and longer stays. Specifically, they are willing to pay €170 more for plane travelling with respect to the use of car, €120 for staying at a four-star hotel relative to an apartment and €760 for a 10-day trip relative to a 3-day one. A daily tax of €1 per person would produce a larger welfare loss in coastal destinations.

scholarly journals Who Gets the Last Bed? A Discrete-Choice Experiment Examining General Population Preferences for Intensive Care Bed Prioritization in a Pandemic

2021 ◽  
pp. 0272989X2199661
Author(s):  
Amelia E. Street ◽  
Deborah J. Street ◽  
Gordon M. Flynn
Keyword(s):  
Decision Making ◽  
Intensive Care ◽  
General Population ◽  
Discrete Choice Experiment ◽  
Discrete Choice ◽  
Choice Experiment ◽  
Latent Class ◽  
Latent Class Model ◽  
Latent Class Models ◽  
Bed Allocation

Objective To explore the key patient attributes important to members of the Australian general population when prioritizing patients for the final intensive care unit (ICU) bed in a pandemic over-capacity scenario. Methods A discrete-choice experiment administered online asked respondents ( N = 306) to imagine the COVID-19 caseload had surged and that they were lay members of a panel tasked to allocate the final ICU bed. They had to decide which patient was more deserving for each of 14 patient pairs. Patients were characterized by 5 attributes: age, occupation, caregiver status, health prior to being infected, and prognosis. Respondents were randomly allocated to one of 7 sets of 14 pairs. Multinomial, mixed logit, and latent class models were used to model the observed choice behavior. Results A latent class model with 3 classes was found to be the most informative. Two classes valued active decision making and were slightly more likely to choose patients with caregiving responsibilities over those without. One of these classes valued prognosis most strongly, with a decreasing probability of bed allocation for those 65 y and older. The other valued both prognosis and age highly, with decreasing probability of bed allocation for those 45 y and older and a slight preference in favor of frontline health care workers. The third class preferred more random decision-making strategies. Conclusions For two-thirds of those sampled, prognosis, age, and caregiving responsibilities were the important features when making allocation decisions, although the emphasis varies. The remainder appeared to choose randomly.

scholarly journals Non–Small Cell Lung Cancer Patient Preferences for First-Line Treatment: A Discrete Choice Experiment

2020 ◽  
Vol 5 (1) ◽  
pp. 238146832092220
Author(s):  
Joanna P. MacEwan ◽  
Komal Gupte-Singh ◽  
Lauren M. Zhao ◽  
Karen L. Reckamp
Keyword(s):  
Lung Cancer ◽  
Discrete Choice ◽  
Choice Experiment ◽  
Latent Class ◽  
Latent Class Model ◽  
Small Cell ◽  
Small Cell Lung ◽  
First Line ◽  
Class Model ◽  
Treatment Delays

Background. There has been much innovation in the treatment of non–small cell lung cancer (NSCLC) in recent years. In particular, use of immuno-oncology (IO) therapies has been growing. Methods. Patients with NSCLC in the United States were surveyed online using a discrete choice experiment to elicit first-line (1L) treatment preferences across six treatment attributes: survival, adverse events (AEs), mechanism of action (MOA), subsequent treatment options (STOs), genetic testing treatment delay, and out-of-pocket cost (OOPC). Preferences were estimated using a latent-class model. Preference shares were estimated for IO-IO, IO-chemo, and chemo-like regimens. Results. Of the 199 patients who completed the survey, 55% were male, 76% were white, 19% had not begun or were on 1L treatment, and the median age was 43 years. Based on a latent-class model with 3 preference classes, 53.0% of patients considered survival and OOPC alone and were less likely to choose an option with a higher OOPC and lower survival, 12.7% of patients were likely to choose the more expensive option, and for 34.3% of patients, survival, AE risk, and treatment delays all significantly influenced choices. MOA and STOs did not significantly influence treatment choices in any preference class. Approximately 53%, 27%, and 20% of patients preferred IO-IO-like, IO-chemo-like, and chemo-like regimens in 1L, respectively. Respondents were younger, more likely to be Caucasian, and more likely to speak English than the general NSCLC patient population. Conclusions. OOPC, effectiveness, treatment delays, and safety influenced NSCLC patients’ 1L treatment decisions, and most patients preferred an IO-IO followed by IO-chemo-like regimen in 1L. Cancer treatment decisions are complex and patient preferences are unique; therefore, patients’ treatment objectives should be discussed in shared treatment decision making.

Is “Localness” about Distance or Relationships? Evidence from Hard Cider

2019 ◽  
Vol 14 (3) ◽  
pp. 252-273 ◽  
Author(s):  
Jarrad Farris ◽  
Trey Malone ◽  
Lindon J. Robison ◽  
Nikki L. Rothwell
Keyword(s):  
Discrete Choice Experiment ◽  
Discrete Choice ◽  
Choice Experiment ◽  
Latent Class ◽  
Consumer Preferences ◽  
Geographic Distance ◽  
Social Capital Theory ◽  
The State ◽  
Travel Distance ◽  
Local Foods

AbstractWhile many studies have evaluated consumer demand for local foods, fewer studies have focused on the mechanism that has created the positive willingness-to-pay for local foods. This article compares the role of geographic distance and attachment value in consumer preferences for locally produced hard cider. Consumer valuations are estimated via a “branded” discrete choice experiment where the respondents chose between an in-state hard cider, an out-of-state hard cider, and a no buy option. Our measure of travel distance is based on the optimal driving route between each consumer's GPS location and the locations of the cideries while our attachment value measure is based on social capital theory. This allows us to analyze individual-specific travel distance heterogeneity in consumer choice as it relates to attachment value. Based on a latent class logit model estimated from a discrete choice experiment with 441 participants, we show that attachment value is higher for a cider produced within the state than for a cider produced outside the state. Furthermore, we show that increases in attachment value increase demand for locally produced hard cider more than an equal increase in attachment value for non-locally produced hard cider. Our findings are consistent with “local” preferences based on geopolitical boundaries (e.g., the state of Michigan) and not distance. (JEL Classifications: B55, M3, Q13, C83)

scholarly journals POSITIONING VIETNAM’S PANGASIUS CATFISH IN THE FRENCH MARKET USING DISCRETE CHOICE MODEL

2014 ◽  
Vol 17 (4) ◽  
pp. 96-111
Author(s):  
Thong Tien Nguyen ◽  
Hung Manh Nguyen
Keyword(s):  
Discrete Choice ◽  
Low Income ◽  
Latent Class ◽  
Choice Model ◽  
Latent Class Model ◽  
Random Parameter ◽  
Discrete Choice Model ◽  
Class Model ◽  
Additional Information ◽  
Random Parameter Model

The study used discrete choice model to investigate the position of Vietnam’s Pangasius catfish in the French market. Data was collected via a choice experiment designed for 12 aquaculture species familiar to French consumers. The random parameter model was estimated and used to calculate the share elasticity. The market position of the aquaculture products in this study was calculated based on the competitive clout, vulnerability scores, and ranked-order implicit values. The results show that Vietnam’s Pangasius has a low competitive clout, high vulnarability score, and low ranked-order implicit value. A latent class model was also estimated for comparison and acquisition of additional information. A strong segment of Pangasius (11.9%) is described by low income and education consumers, women at mid-age dominated, and family with children. To improve the Pangasius position and image in the international market, Vietnam needs promotional and marketing campaigns at global level for the product.

scholarly journals Incentivising doctor attendance in rural Bangladesh: a latent class analysis of a discrete choice experiment

2021 ◽  
Vol 6 (7) ◽  
pp. e006001
Author(s):  
Blake Angell ◽  
Mushtaq Khan ◽  
Raihanul Islam ◽  
Kate Mandeville ◽  
Nahitun Naher ◽  
...  
Keyword(s):  
Latent Class Analysis ◽  
Discrete Choice Experiment ◽  
Discrete Choice ◽  
Choice Experiment ◽  
Latent Class ◽  
Latent Class Models ◽  
Class Analysis ◽  
Sustainable Solutions ◽  
Rural Bangladesh ◽  
Incentive Payments

ObjectiveDoctor absenteeism is widespread in Bangladesh, and the perspectives of the actors involved are insufficiently understood. This paper sought to elicit preferences of doctors over aspects of jobs in rural areas in Bangladesh that can help to inform the development of packages of policy interventions that may persuade them to stay at their posts.MethodsWe conducted a discrete choice experiment with 308 doctors across four hospitals in Dhaka, Bangladesh. Four attributes of rural postings were included based on a literature review, qualitative research and a consensus-building workshop with policymakers and key health-system stakeholders: relationship with the community, security measures, attendance-based policies and incentive payments. Respondents’ choices were analysed with mixed multinomial logistic and latent class models and were used to simulate the likely uptake of jobs under different policy packages.ResultsAll attributes significantly impacted doctor choices (p<0.01). Doctors strongly preferred jobs at rural facilities where there was a supportive relationship with the community (β=0.93), considered good attendance in education and training (0.77) or promotion decisions (0.67), with functional security (0.67) and higher incentive payments (0.5 per 10% increase of base salary). Jobs with disciplinary action for poor attendance were disliked by respondents (−0.63). Latent class analysis identified three groups of doctors who differed in their uptake of jobs. Scenario modelling identified intervention packages that differentially impacted doctor behaviour and combinations that could feasibly improve doctors’ attendance.ConclusionBangladeshi doctors have strong but varied preferences over interventions to overcome absenteeism. We generated evidence suggesting that interventions considering the perspective of the doctors themselves could result in substantial reductions in absenteeism. Designing policy packages that take account of the different situations facing doctors could begin to improve their ability and motivation to be present at their job and generate sustainable solutions to absenteeism in rural Bangladesh.

scholarly journals Public Preferences for Social Distancing Behaviors to Mitigate the Spread of COVID-19: A Discrete Choice Experiment

2020 ◽  
Author(s):  
Ingrid Eshun Wilson ◽  
Aaloke Mody ◽  
Ginger McKay ◽  
Mati Hlatshwayo ◽  
Cory Bradley ◽  
...  
Keyword(s):  
Public Health ◽  
Discrete Choice Experiment ◽  
Discrete Choice ◽  
Choice Experiment ◽  
Latent Class ◽  
Latent Class Models ◽  
Second Phase ◽  
Sampling Weights ◽  
Social Distancing ◽  
The Public

AbstractPolicies to promote social distancing can minimize COVID-19 transmission, but come with substantial social and economic costs. Quantifying relative preferences of the public for such practices can inform policy prioritization and optimize uptake. We used a discrete choice experiment (DCE) to quantify relative “utilities” (preferences) for five COVID-19 pandemic social distances strategies (e.g., closure of restaurants, restriction of large gatherings) against the hypothetical risk of acquiring COVID-19 and anticipated income loss. The survey was distributed in Missouri in May-June, 2020. We applied inverse probability sampling weights to mixed logit and latent class models to generate mean preferences and identify preference classes. Overall (n=2,428), the strongest preference was for the prohibition of large gatherings, followed by preferences to keep outdoor venues, schools, and social and lifestyle venues open, 75% of the population showing probable support for a strategy that prohibited large gatherings and closed lifestyle and social venues. Latent class analysis, however revealed four preference sub-groups in the population - “risk eliminators”, “risk balancers”, “altruistic” and “risk takers”, with men twice as likely as women to belong to the risk-taking group. In this setting, public health policies which as a first phase prohibit large gatherings, as well as close social and lifestyle venues may be acceptable and adhered to by the public. In addition, policy messages that address preference heterogeneity, for example by targeting public health messages at men, could improve adherence to social distancing measures and prevent further COVID-19 transmission prior to vaccine distribution and in the event of future pandemics.Significance StatementPreferences drive behavior – DCE’s are a novel tool in public health that allow examination of preferences for a product, service or policy, identifying how the public prioritizes personal risks and cost in relation to health behaviors. Using this method to establish preferences for COVID-19 mitigation strategies, our results suggest that, firstly, a tiered approach to non-essential business closures where large gatherings are prohibited and social and lifestyle venues are closed as a first phase, would be well aligned with population preferences and may be supported by the public, while school and outdoor venue closures may require more consideration prior to a second phase of restrictions. And secondly, that important distinct preference phenotypes - that are not captured by sociodemographic (e.g., age, sex, race) characteristics - exist, and therefore that messaging should be target at such subgroups to enhance adherence to prevention efforts.

Sign in / Sign up

Export Citation Format

Share Document