Sage Journals: Discover world-class research

Abstract

Background

Dietitians report insufficient time and low-quality educational materials limit patient-centered care for conditions like diabetes. In other clinical practice areas, technology facilitates patient-centered care, but few practice-based technologies have been developed for dietetics. Thus, little is known about what dietitians need to incorporate technology-based solutions into practice.

Objective

The objective of this study was to understand dietitians’ design needs for a novel education material clinical decision support system (CDSS) to customize materials for improved patient-centered care.

Design

Design needs were evaluated with qualitative user interviews during which participants responded to mockups of the proposed technology. Interviewers coded transcripts to identify themes around three key constructs known to drive technology adoption: perceived usefulness, usability, and workflow integration.

Participants/setting

Dietitians (n = 9) who were licensed in Texas and provided nutrition education to patients as part of their practice participated in the interviews.

Results

Customization drove perceived usefulness, and efficiency drove perceived usability of the CDSS. Participants indicated that workflow integration would depend on if the CDSS met both customization and efficiency needs. Some participants indicated that efficiency was more important than patient-level customization; these participants suggested the possibility of customizing handouts for their typical patient population instead of individual patients.

Conclusions

Dietitians are interested in a CDSS that customizes education materials, with efficiency being essential to adoption. Incorporating these insights into future CDSS design could improve the successful adoption of technology for supporting patient-centered dietetic care.

Keywords

Clinical decision support user-centered design usability nutrition education dietitians patient education materials diabetes

Introduction

Clinical practice guidelines for diabetes and other diet-related chronic diseases emphasize the need for patient-centered care, that is, care that is respectful and responsive to patient needs.^1–3 Dietitians, however, have limited time to individualize care, citing a lack of high-quality materials to effectively deliver nutrition education.^4–6 As early as 2001, the Institute of Medicine (now the National Academy of Science, Engineering, and Medicine) emphasized the important role that technology-based systems could play in supporting patient-centered care by facilitating individualization and increasing efficiency.⁷ Leaders in dietetics have highlighted how technology support could stretch across all steps of dietitians’ care process.^8–10

In particular, clinical decision support systems (CDSS) are particularly promising for patient-centered care. CDSS process patient data (e.g. via guideline-based heuristics or artificial intelligence models) to produce recommendations at the point of care. Through individualization, these systems have increased care quality and improved patient outcomes across clinical settings.^11–13 While a few CDSS have been designed to individualize dietary recommendations (i.e. what nutrients or foods patients should consume), few focus specifically on the needs of dietitians for delivering those recommendations in a patient-centered manner.^14–16 None, to our knowledge, address the lack of high-quality education materials that dietitians’ cite as a barrier to patient-centered care provision.

Although a CDSS could support individualized education material creation, these systems require thoughtful, user-centered design to increase the likelihood of adoption. Usability studies are often used to determine the critical design factors that should be considered when developing new technology. Yet, it is still unclear what these factors are for dietitians, as only a handful of dietitian-specific CDSS have been developed, with few usability studies to identify their design needs.¹⁷ Previous research on increasing technology adoption among other healthcare practitioners suggests that workflow integration, usefulness, and usability are important drivers of CDSS adoption.^18,19

The ability to create customized education materials via CDSS may significantly improve dietitians’ ability to deliver patient-centered care, yet current CDSS for dietetics primarily act as dietary recommendations systems. Thus, the proposed CDSS would be a novel contribution to dietetics practice. Given this novelty and the lack of dietitian-focused usability studies, development must begin with preliminary user experience research to understand the design requirements of dietitians for a CDSS that creates customized education materials. User interviews with a user-informed CDSS mockup were conducted to determine what drives perceived usefulness, usability, and workflow integration.

Methods

Participants and procedure

Participants were recruited during an in-person dietetics educational event hosted by the nutrition department at the University of Texas at Austin in October 2024. Interested attendees consented into the study either virtually during event registration or in-person on the day of the event. Consent was provided after reviewing a study information sheet provided on the first page of a digital survey, after which attendees could select the “Submit” button to indicate their consent and continue to enroll and provide demographic information via the survey or close out of the webpage to not enroll. Eligible dietitians were licensed in Texas and provided nutrition education as part of their dietetic role. Minimum target sample size was set at eight participants given previous user-centered design studies indicating 8 to 12 participants are sufficient to identify the majority of user concerns.²⁰ This was confirmed during our analysis process, as the final two virtual interviews did not add or significantly alter the developing themes found after the initial seven in-person interviews.

Participants engaged in either an in-person or virtual semistructured interview (12–27 minutes in length) in which they described how they select education materials for patients. Interviewers walked through the proposed CDSS workflow with the participant, using an exemplar patient to illustrate how the CDSS could adapt the material to the patient's needs. Then participants provided feedback on CDSS mockup screens. For participants who did not complete an in-person interview on the day of the event, virtual interviews were offered following the event. Minor modifications were made to the mockup designs for these later interviews based on the findings from the in-person interviews. The study was designated exempt from full board review by the University of Texas at Austin Institutional Review Board (STUDY00006417). Interviews were recorded on Zoom, which provided automated transcription (Zoom Communications, Inc., San Jose, CA). Participants were not provided incentives.

Interviewers were dietetic interns, a nutrition graduate student, and an undergraduate research assistant. Most interviewers were female (one male dietetic intern was an interviewer). All were trained in qualitative interviewing following an established protocol.²¹ Interviewers reviewed research principles and interviewing best practices, practiced responding appropriately to sample participant statements, and conducted practice interviews in pairs using the semistructured user interview guide. Each interviewer then conducted a mock interview with a dietetic faculty member, after which the trainer provided feedback on the interview recording. These mock interviews were also used to pilot and refine interview questions. See Supplemental materials for copy of participant characteristics survey and final interview guide questions.

Participants did not interact with the interviewers until the interview, which was conducted in a private meeting space separate from the educational event. The interviewers presented themselves as a part of the research team but as someone who was not involved in designing the mockups that were presented. No notes were taken during the interviews, but all were recorded. The lead author maintained a reflexive journal throughout the data collection and analysis process where notes were kept on general study progress as well as personal reflections of the author on the data emerging from interviews and the qualitative analysis. These notes were referred to by the lead author periodically to recall ideas however were not included as a source of data during the analysis. Participants did not provide feedback on transcripts or analysis findings.

Mockup screen development

Mockup screens were developed using a user-centered design approach.²² The purpose of the CDSS was to help a dietitian create a customized educational material for type 2 diabetes nutrition education. We chose one disease condition initially for the sake of initial mockups, however alerted participants that this was an exemplar with the potential for expanding to relevant conditions as development continued. Previous work by the research team with dietitians informed the initial CDSS content and workflow to align with typical practice. First, CDSS users select their desired educational topics and the patient factors that may influence the educational content (e.g. patient literacy level or preferred language). The CDSS then makes recommendations of content for those topics which align with the selected patient factors. Finally, users are then provided an opportunity to rearrange and edit their content to form a single, customized handout for the patient. As mentioned, minor changes were made to the CDSS screen designs between the in-person and virtual rounds of interviews in response to the provided feedback. Primarily, the sample handout on the preview page was altered as initial participants were focused on concerns regarding the specific handout shown instead of providing general feedback about the tool itself.

Analysis

For this deductive thematic analysis, two coders (DL and EA) reviewed each interview transcript and coded excerpts related to the three user-centered design constructs known to be important for CDSS adoption: usability (ease of use), usefulness (benefit to practice), and workflow (compatibility with current work processes).^18,19 Two transcripts were coded initially and then reviewed during a consensus meeting to refine the codebook and ensure consistent coding across coders. All remaining transcripts were coded with the refined codebook by both coders. Given the small number of interviews and the use of frequent consensus meetings, no measures of interrater reliability or quantitative consensus thresholds were used. This is consistent with the study's exploratory aims, which prioritize qualitative interpretation and depth of understanding over quantitative agreement. Coded excerpts were extracted from the Microsoft Word transcripts and organized in Microsoft Excel (Microsoft 365, Microsoft Corporation). Coders identified the primary drivers of adoption within each of the constructs (i.e. themes) from their coded extracts, and then consolidated and refined the themes across multiple consensus meetings with all authors, including peers outside the analysis team.

Results

Fourteen dietitians initially enrolled in the study, with nine ultimately completing an interview. Multiple email attempts were made to schedule interviews with the other five dietitians, however no responses were received to schedule or indicate reason for noncompletion. Of the nine interviews, seven were conducted in-person at the educational event and two were conducted virtually within 2 months of the event. In addition to meeting target sample size, the last two interviews did not provide significantly new or disparate insights from the first seven, suggesting saturation had been reached. Participant characteristics are shown in Table 1. Our themes focus on the primary drivers of adoption within the areas of usefulness (customization), usability (time efficiency), and workflow (customization and efficiency tension), described below.

Table 1.

Dietitians’ (n = 9) characteristics and practice experience related to educating diabetes patients.

Participant characteristic	Stage 2 participants
Years as a registered dietitian (mean ± SD)	9.4 ± 10.5
Race/ethnicity^a
Asian	2 (22%)
Black or African American	1 (11%)
Hispanic, Latino, or Spanish	2 (22%)
Middle Eastern	0
White	5 (56%)
Practice setting
Outpatient	4 (44%)
Private practice	2 (22%)
Public health or social services organization	1 (11%)
Other	2 (22%)
Percent of patients with type 2 diabetes (mean ± SD)	56.7%±32.6
Certified diabetes care and education specialist	2 (22%)

Two participants did not report percent of patients with type 2 diabetes. Percents were calculated using the full sample size. Continuous variables were provided as means and SD, frequencies were provided as count (percentage). ^aParticipants could select more than one race/ethnicity.

Usefulness: Customization

Participants described customization as the primary factor that would make the CDSS useful. Participants reported that currently available handouts were often too wordy, too many pages, or had content that was not relevant for all patients. “I remember even during my clinicals…giving handouts to clients or patients but [saying], ‘Well, this doesn't apply to you. Ignore this section. I'd cross out part of it. So, I feel like this would really help to individualize and tailor [handouts]” (P03). Given their previous experience with outdated handouts, P07 wanted the ability to update content to align with current evidence, explaining “If I would be able to go in and…delete this, or [if it] let me change the wording…”

Participants wanted to customize materials to meet both practice and patient needs. For P03, the ability to customize materials to align with their weight neutral, nondiet approach was seen as a major advantage of the CDSS. “I think I would have to like the actual informational part…tailored to not being diet culture, like all foods fit, and of course fit my practice.” P08 wanted to be able to change the food examples included within the materials to include foods from the patient's diet, saying, “If the tool already has some examples, I think that's amazing. But then, being able to add in maybe examples that the client has mentioned that they like, or they usually will have. Just to include those on there, so they can see what it would look like.”

Usability: Time efficiency

Time was the major factor influencing the perceived usability of the CDSS. Participants described currently having to spend time searching for materials across a range of web sources, including the Academy of Nutrition and Dietetics’ Nutrition Care Manual and the American Diabetes Association. These sources did not always have materials appropriate for their patients, so participants would then need to create their own materials or adapt existing materials. “Because I couldn't find a handout that I like, I actually created the one that I'm using right now, which took a lot of time. This seems like it would be much easier” (P06).

While several participants perceived the CDSS as quick and efficient to use, others noted that they would need additional experience with the technology to determine how quickly they could create materials. P04 emphasized the need to quickly create a relevant handout, “as long as it's something I would take no more than a minute or so to make.” While participants wanted the ability to adapt wording and information provided in the materials, they were concerned about the potential time required if they had to format the content. “But I would want it to make sure that like, because of time, that, okay, click here. It's gonna move up here. Click this. It's gonna move down there. There's no like, you know…when you… add a picture to Word and you're trying to figure out…is it in line with the text? That type of thing” (P01).

Workflow: Customization and efficiency tension

While customization and time efficiency were important to most participants, participants differed on ways they could envision incorporating the CDSS, including before, during, or after a patient visit. While participants’ practice setting and routines influenced the way they envisioned integrating the CDSS into their workflow, they primarily considered how the CDSS would meet both their customization and time efficiency needs. As P04 stated, “I could see this being very helpful when you're actually talking to a client in your office, and when you find out exactly the things they need. So, you're like, ‘Okay, well, let me just like, make this up real quick and you print it out. Let's talk about this.’ That would be the most helpful. I think it just needs to be really, really fast.” P07 noted that they would likely use the CDSS at the end of sessions, unless they were able to create some “templates” of content that they could use with all their patients with diabetes, stating “so I don't have to go back in every time to select the different things that would be helpful, because that saves some steps right?”

In light of the tension between wanting to customize while remaining efficient, several participants appeared willing to sacrifice some level of customization to integrate the CDSS into their workflow, as described by P07, above. Similarly, P09 stated, “I'll have to like tailor it, at least to what most of the patients I see [need], and I would make a copy and keep it in my folder.”

Discussion

In our examination of dietitians’ design needs for an education material CDSS, we found that participants consistently wanted the ability to customize materials to meet both their practice and patient needs while maintaining time efficiency. Participants’ ideas about integrating the CDSS into practice were more varied, highlighting a significant tension between the desire to customize and the desire to save time.

Customization was the key driver of perceived usefulness among participants. Similar to previous studies, our participants described that currently available materials do not meet patient needs, making it difficult to provide individualized education.^23,24 Our study extends these findings by highlighting how dietitians would prefer to customize materials to meet patient needs, from changing individual words to adding in patient-specific food examples.

Although participants were overwhelmingly positive about the usefulness of the proposed CDSS, they were concerned about the time required to use the software in practice. Previous research has indicated dietitians find it challenging to incorporate technology into their everyday practice behaviors.²⁵ Some participants in our study suggested population-level targeted materials underscores the tension between customization and efficiency. This raises the question if a CDSS that meets dietitians’ usability needs will also lead to effective, patient-centered care.^26,27

Participants’ need for rapid customization suggests the potential for large language model (LLM)-based artificial intelligence (AI) tailoring. Tools such as ChatGPT have already demonstrated their capability to deliver dietary recommendations and individualized meal plans, although practitioner supervision is required.^28,29 While LLMs could rapidly customize educational materials, dietitians would still provide oversight to address known issues such as biases and hallucinations of unsubstantiated information, ensuring care is accurate and ethical.³⁰

To appropriately leverage the strengths of both dietitians and AI technology, more research is needed on the acceptability of AI-based tools in dietetics. There is limited research on dietitians’ perceptions of AI-based tools, with one study suggesting dietitians may not place high value in this type of information technology.³¹ Insights from this present study indicate that dietitians desire useful dietetic practice tools, however there was no specific discussion of whether the method of content development (i.e. created by AI) would influence acceptability. Regardless, with the growing development of AI-based, “virtual dietitians,” dietitians must demonstrate their ability to consistently provide individualized, high-quality care exceeding what can be provided by AI tools alone.³²

Overall, this study had several strengths. First, our interviewers underwent rigorous training following best practice guidelines, which included a mock interview and trainer feedback.²¹ We based our analysis of user requirements on known user-centered design constructs important for healthcare practitioners when considering CDSS adoption.^18,19 Additionally, we used a technology probe, the CDSS mockup screens, to facilitate thoughts and feedback from participants who may not have significant experience with CDSS.^33,34

Despite these strengths, our study had some limitations. Participants were those attending an educational event and may not represent the general population of dietitians. We interviewed dietitians from varied practice settings, which may have obscured more minute design needs required within specific practice settings. However, given the novelty of this tool, getting initial perspectives from this cohort on the software's overall concept ensured that further development was warranted. In-person interviews were conducted immediately following an educational session on the use of behavior change theory in education material selection, which may have influenced participants’ perception of the CDSS's value given its inclusion of behavioral theory in the contents’ organization. However, the inclusion of theory in the CDSS design was not emphasized by interviewers during the software walkthrough. Additionally, from this study alone, we are unable to fully determine the degree of customization required to meet dietitians’ usefulness requirement while also maintaining their need for efficiency, or how this balance leads to different integration decisions. In alignment with other user-centered design studies, this initial study focused on identifying user requirements.³⁵ More advanced user-centered design methods, such as user testing of the CDSS in practice simulations would further elucidate the interplay between these user requirements.

Conclusions

This study highlights the design factors that can drive technology acceptance by dietitians. For successful adoption, these technologies must be able to rapidly customize outputs to match patient and dietitian needs. Meeting these needs may require the use of advanced computing methods, creating a synergistic relationship where technology's rapid processing speed and dietitians’ expertise can ensure every patient receives accurate and individualized nutrition education and care.

Supplemental Material

sj-docx-1-dhj-10.1177_20552076261435843 - Supplemental material for Understanding dietitians’ technology design needs through user-centered design of a clinical decision support system for patient-centered diabetes education materials

Supplemental material, sj-docx-1-dhj-10.1177_20552076261435843 for Understanding dietitians’ technology design needs through user-centered design of a clinical decision support system for patient-centered diabetes education materials by Dagny Larson, Elham Almahmoud, Madalyn Rosenthal, Erin Hudson, Elijah Kelley and Marissa Burgermaster in DIGITAL HEALTH

Supplemental Material

sj-pdf-2-dhj-10.1177_20552076261435843 - Supplemental material for Understanding dietitians’ technology design needs through user-centered design of a clinical decision support system for patient-centered diabetes education materials

Supplemental material, sj-pdf-2-dhj-10.1177_20552076261435843 for Understanding dietitians’ technology design needs through user-centered design of a clinical decision support system for patient-centered diabetes education materials by Dagny Larson, Elham Almahmoud, Madalyn Rosenthal, Erin Hudson, Elijah Kelley and Marissa Burgermaster in DIGITAL HEALTH

Footnotes

ORCID iDs

Dagny Larson

Marissa Burgermaster

Ethical considerations

The study was designated exempt by the University of Texas at Austin Institutional Review Board.

Consent to participate

Not applicable as the study was designated exempt by the presiding IRB. All potential participants did receive a study information sheet prior to enrolling in the study.

Consent for publication

Not applicable.

Contributorship

Dagny Larson: Conceptualization; Data curation; Formal analysis; Funding acquisition; Methodology; Investigation; Project administration; Visualization; Writing—original draft.

Elham Almahmoud: Data curation; Formal analysis; Writing—review and editing.

Madalyn Rosenthal: Conceptualization; Data curation; Formal analysis; Writing—review and editing.

Erin Hudson: Conceptualization; Formal analysis; Writing—review and editing.

Elijah Kelley: Conceptualization; Writing—review and editing.

Marissa Burgermaster: Conceptualization; Formal analysis; Methodology; Project administration; Supervision, Writing—review and editing.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors received grant funding from the Academy of Nutrition and Dietetics Foundation for this study.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

De-identifiable data from the datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Supplemental material

Supplemental material for this article is available online.

References

American Diabetes Association Professional Practice Committee. 1. Improving care and promoting health in populations: standards of care in diabetes-2025. Diabetes Care 2024; 48: S14–S26.

American Diabetes Association Professional Practice Committee. 5. Facilitating positive health behaviors and well-being to improve health outcomes: standards of care in diabetes—2025. Diabetes Care 2024; 48: S86–S127.

Navaneethan

Bansal

Cavanaugh

, et al. KDOQI US commentary on the KDIGO 2024 clinical practice guideline for the evaluation and management of CKD. Am J Kidney Dis 2025; 85: 135–176.

Jones

Eggett

Bellini

, et al. Patient-centered care: dietitians’ perspectives and experiences. Patient Educ Couns 2021; 104: 2724–2731.

Prince

Moosa

Lomer

MCE

, et al. Variable access to quality nutrition information regarding inflammatory bowel disease: a survey of patients and health professionals and objective examination of written information. Health Expect 2015; 18: 2501–2512.

Moutou

England

Gutteridge

, et al. Exploring dietitians’ practice and views of giving advice on dietary patterns to patients with type 2 diabetes mellitus: a qualitative study. J Hum Nutr Diet 2022; 35: 179–190.

Institute of Medicine (US) Committee on Quality of Health Care in America. Crossing the quality chasm: a new health system for the 21st century. Washington (DC): National Academies Press (US), http://www.ncbi.nlm.nih.gov/books/NBK222274/ (2001, accessed 21 November 2022).

Abrahams

Matusheski

. Personalised nutrition technologies: a new paradigm for dietetic practice and training in a digital transformation era. J Hum Nutr Diet 2020; 33: 295–298.

Kelly

Collins

McCamley

, et al.

Digital disruption of dietetics: are we ready?

J Hum Nutr Diet 2021; 34: 134–146.

10.

Chen

Gemming

Hanning

, et al. Smartphone apps and the nutrition care process: current perspectives and future considerations. Patient Educ Couns 2018; 101: 750–757.

11.

Sim

Gorman

Greenes

, et al. Clinical decision support systems for the practice of evidence-based medicine. J Am Med Inform Assoc 2001; 8: 527–534.

12.

Jaspers

MWM

Smeulers

Vermeulen

, et al. Effects of clinical decision-support systems on practitioner performance and patient outcomes: a synthesis of high-quality systematic review findings. J Am Med Inform Assoc 2011; 18: 327–334.

13.

Jia

Zhao

Chen

, et al. Evaluation of clinical decision support systems for diabetes care: an overview of current evidence. J Eval Clin Pract 2019; 25: 66–77.

14.

Abu-Saad

Murad

Barid

, et al. Development and efficacy of an electronic, culturally adapted lifestyle counseling tool for improving diabetes-related dietary knowledge: randomized controlled trial among ethnic minority adults with type 2 diabetes mellitus. J Med Internet Res 2019; 21: e13674.

15.

Paulsen

Paur

Gjestland

, et al. Effects of using the MyFood decision support system on hospitalized patients’ nutritional status and treatment: a randomized controlled trial. Clin Nutr 2020; 39: 3607–3617.

16.

Hsu

C-Y

Huang

L-C

Chen

, et al. A web-based decision support system for dietary analysis and recommendations. Telemed e-Health 2011; 17: 68–75.

17.

Barnett

Harricharan

Fletcher

, et al. Mypace: an integrative health platform for supporting weight loss and maintenance behaviors. IEEE J Biomed Health Inform 2015; 19: 109–116.

18.

Horsky

Schiff

Johnston

, et al. Interface design principles for usable decision support: a targeted review of best practices for clinical prescribing interventions. J Biomed Inform 2012; 45: 1202–1216.

19.

Meunier

P-Y

Raynaud

Guimaraes

, et al. Barriers and facilitators to the use of clinical decision support systems in primary care: a mixed-methods systematic review. Ann Fam Med 2023; 21: 57–69.

20.

Kushniruk

Patel

. Cognitive and usability engineering methods for the evaluation of clinical information systems. J Biomed Inform 2004; 37: 56–76.

21.

Goodell

Stage

Cooke

. Practical qualitative research strategies: training interviewers and coders. J Nutr Educ Behav 2016; 48: 578–585.e1.

22.

Norman

Draper

. User centered system design; new perspectives on human-computer interaction. Hillsdale, NJ, USA: L. Erlbaum Associates Inc., 1986.

23.

Tagtow

Amos

. Extent to which dietitians evaluate nutrition education materials. J Nutr Educ 2000; 32: 161–168.

24.

Levey

Ball

Chaboyer

, et al. Dietitians’ perspectives of the barriers and enablers to delivering patient-centred care. J Hum Nutr Diet 2020; 33: 106–114.

25.

Jones

Mitchell

O’Connor

, et al. Investigating the perceptions of primary care dietitians on the potential for information technology in the workplace: qualitative study. J Med Internet Res 2018; 20. DOI: 10.2196/jmir.9568

26.

Kreuter

Wray

. Tailored and targeted health communication: strategies for enhancing information relevance. Am J Health Behav 2003; 27: S227–S232.

27.

Kreuter

Oswald

Bull

, et al.

Are tailored health education materials always more effective than non-tailored materials?

Health Educ Res 2000; 15: 305–315.

28.

Chatelan

Clerc

Fonta

P-A

. ChatGPT and future artificial intelligence chatbots: what may be the influence on credentialed nutrition and dietetics practitioners? J Acad Nutr Diet 2023; 123: 1525–1531.

29.

Guo

Liu

Xiang

, et al. From AI to the table: a systematic review of ChatGPT’s potential and performance in meal planning and dietary recommendations. Dietetics 2025; 4: 7.

30.

Bailey

MacFarlane

Field

, et al. Artificial intelligence in food and nutrition evidence: the challenges and opportunities. PNAS Nexus 2024; 3: pgae461.

31.

Abrahams

Frewer

Bryant

, et al. Personalised nutrition technologies and innovations: a cross-national survey of registered dietitians. PHG 2019; 22: 119–131.

32.

Buchan

Goel

Schneider

, et al. National implementation of an artificial intelligence-based virtual dietitian for patients with cancer. JCO Clin Cancer Inform 2024: e2400085. doi:10.1200/CCI.24.00085

33.

Gaver

Dunne

Pacenti

. Design: cultural probes. Interactions 1999; 6: 21–29.

34.

Kulp

Sarcevic

Farneth

, et al.

Exploring design opportunities for a context-adaptive medical checklist through technology probe approach

Proceedings of the 2017 conference on designing interactive systems. Edinburgh, UK: ACM, 2017, pp.57–68.

35.

Tyo

Desroches

. Content analysis of qualitative interviews for user-centered design of a prototype mobile health app for direct support professionals’ resilience. Disabil Health J 2024; 17: 101544.