Understanding Uptake of Digital Health Products: Methodology Tutorial for a Discrete Choice Experiment Using the Bayesian Efficient Design

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

BACKGROUND

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.

OBJECTIVES

This study aimed to identify key attributes and elicit healthcare consumer preferences for a healthcare appointment reminder system.

METHODS

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.

RESULTS

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%.

CONCLUSIONS

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.

Benefit-Risk Trade-offs and Patient Preferences for Therapy Selection in Ulcerative Colitis: a Multicountry Preference Study

BACKGROUND

To help navigate the complex treatment landscape of ulcerative colitis (UC), we quantified the benefit-risk trade-offs that patients were willing to make when choosing treatment.

METHODS

Patients completed an online discrete choice experiment. Eligible patients had a UC diagnosis for ≥6 months, were aged ≥18 years, and resided in France, Germany, Italy, Spain, or the UK. Patients chose between 2 hypothetical treatments set up to ensure trade-offs were made. Clinical trial data, literature review, and patient interviews identified treatment attributes. Relative attribute importance (RAI) scores and maximum acceptable risks were generated. A patient-centric benefit-risk assessment of 200 mg of filgotinib was conducted as an example to show how measured trade-offs can be used.

RESULTS

Overall, 631 patients participated; patients had a mean age of 42.2 years and were predominantly male (75.3%). Achieving and maintaining clinical remission was the most important factor for patients (RAI 32.4%); to achieve this, patients were willing to accept slightly higher risks of blood clots, serious infections, and malignancies compared with lower risk treatment profiles. Patients also valued the convenience of oral treatments, avoiding steroids, and the ability to attend school/work. The patient-centric benefit-risk assessment suggested patients are significantly more likely to prefer Janus kinase 1 preferential inhibitor filgotinib over placebo.

CONCLUSIONS

Achieving clinical remission was the highest treatment priority for patients. To attain this, patients were willing to accept some slightly higher risk treatment profiles. Patient choices in the benefit-risk assessment suggested patients were significantly more likely to prefer filgotinib over placebo.

A pilot study on preferences from surgeons to deal with an innovative customized and connected knee prosthesis - A discret choice experiment

Background

To address the increasing global demand for Total Knee Arthroplasty and reduce the need for revisions, several technologies combining 3D planning and artificial intelligence have emerged. These innovations aim to enhance customization, improve component positioning accuracy and precision. The integration of these advancements paves the way for the development of personalized and connected knee implant.

Questions/purposes

These groundbreaking advancements may necessitate changes in surgical practices. Hence, it is important to comprehend surgeons' intentions in integrating these technologies into their routine procedures. Our study aims to assess how surgeons' preferences will affect the acceptability of using this new implant and associated technologies within the entire care chain.

Methods

We employed a Discrete Choice Experiment, a predictive technique mirroring real-world healthcare decisions, to assess surgeons' trade-off evaluations and preferences.

Results

A total of 90 experienced surgeons, performing a significant number of procedures annually (mostly over 51) answered. Analysis indicates an affinity for technology but limited interest in integrating digital advancements like preoperative software and robotics. However, they are receptive to practice improvements and considering the adoption of future sensors.

Conclusions

In conclusion, surgeons prefer customized prostheses via augmented reality, accepting extra cost. Embedded sensor technology is deemed premature by them.

Investigating consumer preferences and experiences of telehealth services in Australian allied health practice: a study protocol

INTRODUCTION

Telehealth service provision and uptake has rapidly increased since the COVID-19 pandemic, allowing healthcare to be delivered safely and reducing non-essential face-to-face (F2F) contact. In Australia, the expansion of subsidisation of telehealth during COVID has led to its permanent installation within Australian primary care in 2022. However, little is known about consumer preferences and experiences with these services, particularly in relation to allied health practice (AHP). Previous studies on telehealth services have focused on general practice rather than allied health (AH) and broader primary care. Given that AH professionals make up a large proportion of the Australian healthcare workforce, the purpose of this study is to explore consumer preferences and experiences with telehealth AHP healthcare.

METHODS AND ANALYSIS

This study uses a mixed methods research design that incorporates three independent but interrelated phases. Phase 2 of the study will use a focus group methodology to discuss consumer attitudes and experiences via a semistructured interview format. Phase 3 involves a discrete choice experiment (DCE) involving a large online survey conducted across the general population. The DCE will be informed by the qualitative findings from phases 1 and 2. The experiment aims to elicit consumer preferences in relation to AH services delivered through telehealth or F2F consultations, based on several hypothetical scenarios and preferences over several different dimensions.

ETHICS AND DISSEMINATION

Ethics approval has been obtained from La Trobe University (approval number HEC23404). Findings will be disseminated as reports, presentations and peer-reviewed journal articles.

Preferences in adolescents and young people's sexual and reproductive health services in Nigeria: a discrete choice experiment

BACKGROUND

Barriers to utilization of sexual and reproductive health (SRH) services by adolescents and young people (AYP) have persisted despite evidence that youth-friendly services have a positive effect on contraceptive use, and patient knowledge and satisfaction.

OBJECTIVE

The objective of this study was to elicit, and derive relative valuations for, attributes of SRH services that adolescents and young people value, and their willingness to pay for these services, in public health facilities.

METHODS

A discrete-choice-experiment (DCE) that was developed using a mixed methods approach was administered to AYP from Ogun State, Southwest Nigeria. The DCE attributes were: the type of staff; physical environment; health worker attitude; cost; waiting time; contraceptive availability; and opening hours. The choice tasks had two unlabeled alternatives and an opt-out option. Panel mixed logit choice model was used to fit the choice data, along with estimation of willingness to pay (WTP). Also, a latent class logit model was used to detect underlying preference heterogeneity among the respondents. Finally, the uptake of the services in health facilities was investigated by estimating the probabilities for selecting hypothetical health facilities under different scenarios.

RESULTS

A total of 859 AYP participated resulting in 6872 choice observations. The physical environment attribute had the highest utility rating relative to the other attributes, followed by preference for a doctor and for a service provider who was open and friendly. The cost and time coefficients were negative, revealing preference for lower cost and shorter waiting time. The latent class model had three classes that varied by their background characteristics. Probability of choosing any of the facility alternatives increased with introduction of more favorable facility characteristics.

CONCLUSION

The pattern of preferences identified are potential targets for service design and delivery optimization that may result in improvements in service acceptability and utilization. These results strengthen the call for involving AYP in decision-making in health interventions for them and developing context-specific SRH programs for AYP in public health facilities.

Decision Trade-Offs in Ecological Momentary Assessments and Digital Wearables Uptake: Protocol for a Discrete Choice Experiment

BACKGROUND

Ecological momentary assessments (EMAs) and digital wearables (DW) are commonly used remote monitoring technologies that capture real-time data in people's natural environments. Real-time data are core to personalized medical care and intensively adaptive health interventions. The utility of such personalized care is contingent on user uptake and continued use of EMA and DW. Consequently, it is critical to understand user preferences that may increase the uptake of EMA and DW.

OBJECTIVE

The study aims to quantify users' preferences of EMA and DW, examine variations in users' preferences across demographic and behavioral subgroups, and assess the association between users' preferences and intentions to use EMA and DW.

METHODS

We will administer 2 discrete choice experiments (DCEs) paired with self-report surveys on the internet to a total of 3260 US adults through Qualtrics. The first DCE will assess participants' EMA preferences using a choice-based conjoint design that will ask participants to compare the relative importance of prompt frequency, number of questions per prompt, prompt type, health topic, and assessment duration. The second DCE will measure participants' DW preferences using a maximum difference scaling design that will quantify the relative importance of device characteristics, effort expectancy, social influence, and facilitating technical, health care, and market factors. Hierarchical Bayesian multinomial logistic regression models will be used to generate subject-specific preference utilities. Preference utilities will be compared across demographic (ie, sex, age, race, and ethnicity) and behavioral (ie, substance use, physical activity, dietary behavior, and sleep duration) subgroups. Regression models will determine whether specific utilities are associated with attitudes toward or intentions to use EMA and DW. Mixture models will determine the associations of attitudes toward and intentions to use EMA and DW with latent profiles of user preferences.

RESULTS

The institutional review board approved the study on December 19, 2022. Data collection started on January 20, 2023, and concluded on May 4, 2023. Data analysis is currently underway.

CONCLUSIONS

The study will provide evidence on users' preferences of EMA and DW features that can improve initial uptake and potentially continued use of these remote monitoring tools. The sample size and composition allow for subgroup analysis by demographics and health behaviors and will provide evidence on associations between users' preferences and intentions to uptake EMA and DW. Limitations include the cross-sectional nature of the study, which limits our ability to measure direct behavior. Rather, we capture behavioral intentions for EMA and DW uptake. The nonprobability sample limits the generalizability of the results and introduces self-selection bias related to the demographic and behavioral characteristics of participants who belong to web-based survey panels.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/47567.

Patient and Public Acceptance of Digital Technologies in Health Care: Protocol for a Discrete Choice Experiment

BACKGROUND

Strokes pose a particular challenge to the health care system. Although stroke-related mortality has declined in recent decades, the absolute number of new strokes (incidence), stroke deaths, and survivors of stroke has increased. With the increasing need of neurorehabilitation and the decreasing number of professionals, innovations are needed to ensure adequate care. Digital technologies are increasingly used to meet patients' unfilled needs during their patient journey. Patients must adhere to unfamiliar digital technologies to engage in health interventions. Therefore, the acceptance of the benefits and burdens of digital technologies in health interventions is a key factor in implementing these innovations.

OBJECTIVE

This study aims to describe the development of a discrete choice experiment (DCE) to weigh criteria that impact patient and public acceptance. Secondary study objectives are a benefit-burden assessment (estimation of the maximum acceptable burden of technical features and therapy-related characteristics for the patient or individual, eg, no human contact), overall comparison (assessment of the relative importance of attributes for comparing digital technologies), and adherence (identification of key attributes that influence patient adherence). The exploratory objectives include heterogeneity assessment and subgroup analysis. The methodological aims are to investigate the use of DCE.

METHODS

To obtain information on the criteria impacting acceptance, a DCE will be conducted including 7 attributes based on formative qualitative research. Patients with stroke (experimental group) and the general population (control group) are surveyed. The final instrument includes 6 best-best choice tasks in partial design. The experimental design is a fractional-factorial efficient Bayesian design (D-error). A conditional logit regression model and mixed logistic regression models will be used for analysis. To consider the heterogeneity of subgroups, a latent class analysis and an analysis of heteroscedasticity will be performed.

RESULTS

The literature review, qualitative preliminary study, survey development, and pretesting were completed. Data collection and analysis will be completed in the last quarter of 2023.

CONCLUSIONS

Our results will inform decision makers about patients' and publics' acceptance of digital technologies used in innovative interventions. The patient preference information will improve decisions regarding the development, adoption, and pricing of innovative interventions. The behavioral changes in the choice of digital intervention alternatives are observable and can therefore be statistically analyzed. They can be translated into preferences, which define the value. This study will investigate the influences on the acceptance of digital interventions and thus support decisions and future research.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/46056.

Public's preferences for health science popularization short videos in China: a discrete choice experiment

Background

Health science popularization short video disseminates health information to the public in an understandable way about health information.

Objective

To investigate the preferences of Chinese residents for health science popularization short videos and provide suggestions for optimizing the production of short videos.

Methods

An online survey of Chinese people was conducted using a self-administered questionnaire, and a discrete choice experiment (DCE) was used to explore the public's preferences for health science popularization short videos.

Results

A total of 618 respondents were included, of which 306 (45.51%) were male and 312 (50.49%) were female, 271 (43.85%) were aged 18-25, 239 (38.67%) were aged 26-60, and 108 (17.48%) were aged 60 and above. Whether the video is charged or not (46.891%) and the account subject (28.806%) were both considered important. The results of the DCE revealed that the participants considered video free of charge as the most significant attribute of health science popularization short videos (OR 3.433, 95% CI 3.243-3.633). Overall, participants preferred and were more willing to pay for health science popularization short videos with a hospital account subject (OR 1.192, 95% CI 1.116-1.274), with the form of graphic narration (OR 1.062, 95% CI 1.003-1.126), free of charge (OR 3.433, 95% CI 3.243-3.633), with the content that satisfies their needs (very much needed: OR 1.253, 95% CI 95% CI 1.197-1.311; generally needed: OR 1.078, 95% CI 1.029-1.129), with platform certification (OR 1.041, 95% CI 1.011-1.073), without commercial advertisements (OR 1.048, 95% CI 1.018-1.080), with simple-to-understand content (OR 1.071, 95% CI 1.040-1.104), and with video content that evokes fear or dread of illness in the viewer (OR 1.046, 95% CI 1.015-1.078).

Conclusion

Participants favor free health popularization short videos, which are hospital accounts, with content that is illustrated, understandable, meets their needs, and can serve as a warning. In the future, the production of health popularization short videos should focus on improving the diversity and relevance of video content, making it as easy to understand to achieve good science popularization effects.

Keeping an eye on cost: What can eye tracking tell us about attention to cost information in discrete choice experiments?

Concern has been expressed about including a cost attribute within discrete choice experiments (DCEs) when individuals do not have to pay at the point of consumption. We use eye tracking to investigate attention to cost when valuing publicly financed health care. One-hundred and four individuals completed a DCE concerned with preferences for UK general practitioner appointments: 51 responded to a DCE with cost included and 53 to the same DCE without cost. Eye-movements were tracked whilst respondents completed the DCE. We assessed if respondents pay attention to cost. We then compare fixation time (FT) on attributes, eye movement patterns and mental effort across the experimental groups. Results are encouraging for the inclusion of cost in DCEs valuing publicly provided healthcare. Most respondents gave visual attention to the cost attribute most of the time. Average FT on multi-attribute tasks increased by 44% in the cost DCE, with attention to non-monetary attributes increasing by 22%. Including cost led to more structured decision-making and did not increase mental effort. Acceptability of the cost attribute and difficulty of choice tasks were predictors of cost information processing, highlighting the importance of both motivating the cost attribute and considering difficulty of the tasks when developing DCEs.

Tradeoffs in Triage of Motor Vehicle Trauma by Rural 911 Emergency Medical Services Practitioners

PURPOSE

Identification and triage of severely injured patients to trauma centers is paramount to survival. Many patients are undertriaged in rural areas and do not receive proper care. The decision-making processes involved in triage are not well understood and should be assessed to improve the triage process and outcomes.

METHODS

Triage decision-making processes were explored through emergency medical services (EMS) practitioner focus groups and a discrete choice experiment (DCE). Attributes of trauma determined from focus groups and the literature included patient demography, injury mechanism, and trauma center distance. DCE data were analyzed using mixed logit models.

RESULTS

High-risk mechanism, decreased age, multiple comorbidities, and pregnancy were found to increase the preference for triage. Greater trauma center distance was found to decrease preference for triage, but practitioners were willing to trade off up to 2 h of travel time to transport a third-trimester pregnancy and 48 min of travel time to transport a 25-y-old than they would a 50-y-old with the same comorbidities, injuries, and stability.

CONCLUSIONS

Our findings suggest that current forms of EMS protocols may not be appropriately tailored to support the mechanisms underlying practitioner decision making. Public health professionals and researchers should consider using DCEs to better understand EMS practitioner decision making and identify structures and incentives that may improve patient outcomes and optimally guide appropriate triage decisions.

HIGHLIGHTS

Discrete choice experiments are an effective method to elicit prehospital practitioners' preferences around transport of the traumatized patient.Practitioner biases observed in EMS transport data are recovered in stated preference models incorporating individual preference heterogeneity.There is a discrepancy between the triage priorities recommended by protocol and those measured from prehospital practitioners' decisions-this may have implications in over- and undertriage rates and prehospital protocol design.

The heterogeneity of public preferences for the first healthcare visit: A discrete choice experiment in the context of Vietnam

Primary healthcare is critical in addressing the main health problems of communities. In Vietnam, the increasing healthcare demands cause major challenges, especially overcrowding. This study identified public preferences regarding the selection of healthcare facilities for first visit. A discrete choice online survey was generated from five attributes including visit duration, travel time, personal connection with medical staff, doctors' experience, and health insurance. A D_z -efficient design constructed 36 choice sets, divided into three blocks of 12 choice sets. Each block formed one version of the questionnaire, which was randomly distributed to the participants. Heterogeneity in participant preferences was analysed by a latent class model with socio demographic characteristics and experiences of the last visit. 822 participants valued doctors' experience for both minor and severe symptoms. Preference heterogeneity for minor symptoms was quick service provision, highly experienced doctors, and payment through health insurance for the first (44.18%), second (32.17%), and third classes (23.66%), respectively. Regarding severe symptoms, they favoured all five attributes, quick health service, and reduced travel time for the first, second, and third classes, respectively (heterogeneities of 58.16%, 27.79%, and 14.05%, respectively). Predictions of choice from the worst to optimal healthcare facility scenario were 8.91%-61.91% and 10.16%-69.83% for minor and severe symptoms, respectively. Knowledge regarding public preference heterogeneity supports policymakers increase public acceptance in choosing primary healthcare facilities. Visit duration and doctors' experience should be considered a priority in decision making.

Societal preferences for funding orphan drugs in China: An application of the discrete choice experiment method

Objectives

To explore whether a societal preference for orphan drugs exists in Chinese general public and to quantitatively measure the personal trade-off between essential attributes of orphan drugs through a discrete choice experiment.

Methods

A labeled discrete choice experiment was employed to measure public preference. Six attributes (impact of diseases on life-years, impact of diseases on quality of life, availability of alternative drug treatments, annual cost per patient paid by medical insurance, expected increases in life-expectancy, and improvements to the quality of life) were identified through a literature review, experts' suggestions, and stakeholders' semi-structured interviews, then refined through a pre-survey. The current study used a D-efficient design to yield 27 choice sets divided into three blocks with nine questions containing the labeled treatment (either orphan drugs or common drugs). Information on sociodemographic characteristics and individual preferences was collected through a web-based questionnaire using convenience sampling. A mixed logit model was used to test societal preferences for orphan drugs over common drugs, while a binary logit model was used to measure the relative importance of each attribute in orphan drug access for the National Reimbursement Drug List and its willingness to pay.

Results

A total of 323 persons participated in this study. Respondents largely had indifferent attitudes toward orphan drugs and common drugs. The binary logit model results showed that 5 of the 6 attributes were significant, except for the availability of alternative drug treatments. The most impacted factor was the annual cost per patient paid by medical insurance (β = -1.734, odds ratio [OR] = 0.177). Among non-economic attributes, the impact of diseases on life-years-with no treatment, the patient will die in the prime of life (β = 0.523, OR = 1.688, willingness to pay = 301,895)-was most concerning, followed by significant improvements to the quality of life (β = 0.516, OR = 1.676, willingness to pay = 297,773).

Conclusion

The general public in China does not value rarity as a sufficient reason to justify special consideration in funding orphan drugs. When making orphan drug coverage decisions, the public prioritized the annual cost, disease severity, and drug effects.

Developing mHealth to the Context and Valuation of Injured Patients and Professionals in Hospital Trauma Care: Qualitative and Quantitative Formative Evaluations (Preprint)

Background

Trauma care faces challenges to innovating their services, such as with mobile health (mHealth) app, to improve the quality of care and patients’ health experience. Systematic needs inquiries and collaborations with professional and patient end users are highly recommended to develop and prepare future implementations of such innovations.

Objective

This study aimed to develop a trauma mHealth app for patient information and support in accordance with the Center for eHealth Research and Disease Management road map and describe experiences of unmet information and support needs among injured patients with trauma, barriers to and facilitators of the provision of information and support among trauma care professionals, and drivers of value of an mHealth app in patients with trauma and trauma care professionals.

Methods

Formative evaluations were conducted using quantitative and qualitative methods. Ten semistructured interviews with patients with trauma and a focus group with 4 trauma care professionals were conducted for contextual inquiry and value specification. User requirements and value drivers were applied in prototyping. Furthermore, a complementary quantitative discrete choice experiment (DCE) was conducted with 109 Dutch trauma surgeons, which enabled triangulation on value specification results. In the DCE, preferences were stated for hypothetical mHealth products with various attributes. Panel data from the DCE were analyzed using conditional and mixed logit models.

Results

Patients disclosed a need for more psychosocial support and easy access to more extensive information on their injury, its consequences, and future prospects. Health care professionals designated workload as an essential issue; a digital solution should not require additional time. The conditional logit model of DCE results suggested that access to patient app data through electronic medical record integration (odds ratio [OR] 3.3, 95% CI 2.55-4.34; P<.001) or a web viewer (OR 2.3, 95% CI 1.64-3.31; P<.001) was considered the most important for an mHealth solution by surgeons, followed by the inclusion of periodic self-measurements (OR 2, 95% CI 1.64-2.46; P<.001), the local adjustment of patient information (OR 1.8, 95% CI 1.42-2.33; P<.001), local hospital identification (OR 1.7, 95% CI 1.31-2.10; P<.001), complication detection (OR 1.5, 95% CI 1.21-1.84; P<.001), and the personalization of rehabilitation through artificial intelligence (OR 1.4, 95% CI 1.13-1.62; P=.001).

Conclusions

In the context of trauma care, end users have many requirements for an mHealth solution that addresses psychosocial functioning; dependable information; and, possibly, a prediction of how a patient’s recovery trajectory is evolving. A structured development approach provided insights into value drivers and facilitated mHealth prototype enhancement. The findings imply that iterative development should move on from simple and easily implementable mHealth solutions to those that are suitable for broader innovations of care pathways that most—but plausibly not yet all—end users in trauma care will value. This study could inspire the trauma care community.