Channing Bice
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Channing Bice has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
Ashley A. Anderson
Colorado State University
2 shared publications- Daniel T. Burns
University of Wyoming Department of Mathematics and Statistics Laramie, Wyoming 82071
2 shared publications Katie M. Abrams
Colorado State University
2 shared publications- Nicholas Chia
Argonne National Laboratory
2 shared publications - Paul Johnson
Department of Otolaryngology- Head and Neck Surgery, University of Washington, Seattle, WA, United States
2 shared publications - Timothy J. Robinson
Department of Mathematics & Statistics, University of Wyoming, Laramie, WY, United States
2 shared publications
Latest Publications
- Open Access
- 6 March 2026
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 28, e85516
Comparison of Emotional Content in Text Responses From Physicians and AI Chatbots to Patient Health Queries: Cross-Sectional Study
Background: Surveys show that many people are willing to use generative artificial intelligence (AI) for health questions. Prior research has largely focused on chatbot accuracy, with some studies finding that both physicians and consumers overwhelmingly prefer chatbot-generated text over physician responses. Objective: This study aimed to characterize and compare the emotional content of responses from physicians and 2 AI chatbots (OpenAI’s ChatGPT and Google’s Gemini) and to assess differences in reading level and use of medical disclaimers. Methods: A public, patient-deidentified telehealth website was used to compile 100 physician-answered questions. The same questions were posed to both chatbots between May 18 and 19, 2025. Two coders classified the emotional content of each sentence using a predefined codebook and reviewed for agreement. Emotions were ranked as primary, secondary, and tertiary by the proportion of sentences classified as each emotion per response. Multinomial logistic regression compared emotional rankings using physician responses as the reference. Word count, Flesch Reading Ease, and Flesch-Kincaid Grade Level were analyzed via ANOVA with the Tukey honestly significant difference test. Disclaimer use was compared between chatbots using a χ2 test. Results: Primary emotions were overwhelmingly neutral, except for one response from each chatbot in which anger was primary. For secondary emotions, the odds ratio of hope was 80.28% (95% CI 37.71%-93.76%) lower for ChatGPT, while the odds ratio of fear was 3.29 (95% CI 1.44-7.49) times higher for Gemini. For tertiary emotions, the odds ratio of compassion was 1.94 (95% CI 1.06-3.54) times higher, and the odds ratio of having no tertiary emotion was 84.33% (95% CI 64.72%-93.04%) lower for Gemini. Gemini responses averaged 889.1 (SD 305.7) words, ChatGPT 476.5 (SD 109.5), and physicians 193.5 (SD 113.6). Gemini had the lowest average Flesch Reading Ease score at 39.9 (SD 8.8), followed by ChatGPT at 45.8 (SD 12.8), while physicians had the highest at 51.9 (SD 13.6). Gemini had the highest average Flesch-Kincaid Grade Level at 11.3 (SD 1.5), followed by ChatGPT at 9.9 (SD 1.9), and physicians at 9.2 (SD 2.4). Gemini was significantly more likely to include a disclaimer than ChatGPT (χ21=49.2; P<.001). Conclusions: Chatbot responses were significantly (P<.001) longer and more difficult to read than physician responses and were more likely to contain a wider range of emotions. Qualitatively, chatbot responses were more varied in their presentation as well as in the breadth of the emotions themselves. The findings of this study could be used to inform more emotionally connected physician responses to patient message queries.
- 6 March 2026
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 28, e85516
Open AccessComparison of Emotional Content in Text Responses From Physicians and AI Chatbots to Patient Health Queries: Cross-Sectional Study
Background: Surveys show that many people are willing to use generative artificial intelligence (AI) for health questions. Prior research has largely focused on chatbot accuracy, with some studies finding that both physicians and consumers overwhelmingly prefer chatbot-generated text over physician responses. Objective: This study aimed to characterize and compare the emotional content of responses from physicians and 2 AI chatbots (OpenAI’s ChatGPT and Google’s Gemini) and to assess differences in reading level and use of medical disclaimers. Methods: A public, patient-deidentified telehealth website was used to compile 100 physician-answered questions. The same questions were posed to both chatbots between May 18 and 19, 2025. Two coders classified the emotional content of each sentence using a predefined codebook and reviewed for agreement. Emotions were ranked as primary, secondary, and tertiary by the proportion of sentences classified as each emotion per response. Multinomial logistic regression compared emotional rankings using physician responses as the reference. Word count, Flesch Reading Ease, and Flesch-Kincaid Grade Level were analyzed via ANOVA with the Tukey honestly significant difference test. Disclaimer use was compared between chatbots using a χ2 test. Results: Primary emotions were overwhelmingly neutral, except for one response from each chatbot in which anger was primary. For secondary emotions, the odds ratio of hope was 80.28% (95% CI 37.71%-93.76%) lower for ChatGPT, while the odds ratio of fear was 3.29 (95% CI 1.44-7.49) times higher for Gemini. For tertiary emotions, the odds ratio of compassion was 1.94 (95% CI 1.06-3.54) times higher, and the odds ratio of having no tertiary emotion was 84.33% (95% CI 64.72%-93.04%) lower for Gemini. Gemini responses averaged 889.1 (SD 305.7) words, ChatGPT 476.5 (SD 109.5), and physicians 193.5 (SD 113.6). Gemini had the lowest average Flesch Reading Ease score at 39.9 (SD 8.8), followed by ChatGPT at 45.8 (SD 12.8), while physicians had the highest at 51.9 (SD 13.6). Gemini had the highest average Flesch-Kincaid Grade Level at 11.3 (SD 1.5), followed by ChatGPT at 9.9 (SD 1.9), and physicians at 9.2 (SD 2.4). Gemini was significantly more likely to include a disclaimer than ChatGPT (χ21=49.2; P<.001). Conclusions: Chatbot responses were significantly (P<.001) longer and more difficult to read than physician responses and were more likely to contain a wider range of emotions. Qualitatively, chatbot responses were more varied in their presentation as well as in the breadth of the emotions themselves. The findings of this study could be used to inform more emotionally connected physician responses to patient message queries.
- Open Access
- 9 October 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Comparison of Emotional Content in Text Responses From Physicians and AI Chatbots to Patient Health Queries: Cross-Sectional Study (Preprint)
BACKGROUND Surveys show that many people are willing to use generative artificial intelligence (AI) for health questions. Prior research has largely focused on chatbot accuracy, with some studies finding that both physicians and consumers overwhelmingly prefer chatbot-generated text over physician responses. OBJECTIVE This study aimed to characterize and compare the emotional content of responses from physicians and 2 AI chatbots (OpenAI’s ChatGPT and Google’s Gemini) and to assess differences in reading level and use of medical disclaimers. METHODS A public, patient-deidentified telehealth website was used to compile 100 physician-answered questions. The same questions were posed to both chatbots between May 18 and 19, 2025. Two coders classified the emotional content of each sentence using a predefined codebook and reviewed for agreement. Emotions were ranked as primary, secondary, and tertiary by the proportion of sentences classified as each emotion per response. Multinomial logistic regression compared emotional rankings using physician responses as the reference. Word count, Flesch Reading Ease, and Flesch-Kincaid Grade Level were analyzed via ANOVA with the Tukey honestly significant difference test. Disclaimer use was compared between chatbots using a χ2 test. RESULTS Primary emotions were overwhelmingly neutral, except for one response from each chatbot in which anger was primary. For secondary emotions, the odds ratio of hope was 80.28% (95% CI 37.71%-93.76%) lower for ChatGPT, while the odds ratio of fear was 3.29 (95% CI 1.44-7.49) times higher for Gemini. For tertiary emotions, the odds ratio of compassion was 1.94 (95% CI 1.06-3.54) times higher, and the odds ratio of having no tertiary emotion was 84.33% (95% CI 64.72%-93.04%) lower for Gemini. Gemini responses averaged 889.1 (SD 305.7) words, ChatGPT 476.5 (SD 109.5), and physicians 193.5 (SD 113.6). Gemini had the lowest average Flesch Reading Ease score at 39.9 (SD 8.8), followed by ChatGPT at 45.8 (SD 12.8), while physicians had the highest at 51.9 (SD 13.6). Gemini had the highest average Flesch-Kincaid Grade Level at 11.3 (SD 1.5), followed by ChatGPT at 9.9 (SD 1.9), and physicians at 9.2 (SD 2.4). Gemini was significantly more likely to include a disclaimer than ChatGPT (χ21=49.2; P<.001). CONCLUSIONS Chatbot responses were significantly (P<.001) longer and more difficult to read than physician responses and were more likely to contain a wider range of emotions. Qualitatively, chatbot responses were more varied in their presentation as well as in the breadth of the emotions themselves. The findings of this study could be used to inform more emotionally connected physician responses to patient message queries.
- 9 October 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessComparison of Emotional Content in Text Responses From Physicians and AI Chatbots to Patient Health Queries: Cross-Sectional Study (Preprint)
BACKGROUND Surveys show that many people are willing to use generative artificial intelligence (AI) for health questions. Prior research has largely focused on chatbot accuracy, with some studies finding that both physicians and consumers overwhelmingly prefer chatbot-generated text over physician responses. OBJECTIVE This study aimed to characterize and compare the emotional content of responses from physicians and 2 AI chatbots (OpenAI’s ChatGPT and Google’s Gemini) and to assess differences in reading level and use of medical disclaimers. METHODS A public, patient-deidentified telehealth website was used to compile 100 physician-answered questions. The same questions were posed to both chatbots between May 18 and 19, 2025. Two coders classified the emotional content of each sentence using a predefined codebook and reviewed for agreement. Emotions were ranked as primary, secondary, and tertiary by the proportion of sentences classified as each emotion per response. Multinomial logistic regression compared emotional rankings using physician responses as the reference. Word count, Flesch Reading Ease, and Flesch-Kincaid Grade Level were analyzed via ANOVA with the Tukey honestly significant difference test. Disclaimer use was compared between chatbots using a χ2 test. RESULTS Primary emotions were overwhelmingly neutral, except for one response from each chatbot in which anger was primary. For secondary emotions, the odds ratio of hope was 80.28% (95% CI 37.71%-93.76%) lower for ChatGPT, while the odds ratio of fear was 3.29 (95% CI 1.44-7.49) times higher for Gemini. For tertiary emotions, the odds ratio of compassion was 1.94 (95% CI 1.06-3.54) times higher, and the odds ratio of having no tertiary emotion was 84.33% (95% CI 64.72%-93.04%) lower for Gemini. Gemini responses averaged 889.1 (SD 305.7) words, ChatGPT 476.5 (SD 109.5), and physicians 193.5 (SD 113.6). Gemini had the lowest average Flesch Reading Ease score at 39.9 (SD 8.8), followed by ChatGPT at 45.8 (SD 12.8), while physicians had the highest at 51.9 (SD 13.6). Gemini had the highest average Flesch-Kincaid Grade Level at 11.3 (SD 1.5), followed by ChatGPT at 9.9 (SD 1.9), and physicians at 9.2 (SD 2.4). Gemini was significantly more likely to include a disclaimer than ChatGPT (χ21=49.2; P<.001). CONCLUSIONS Chatbot responses were significantly (P<.001) longer and more difficult to read than physician responses and were more likely to contain a wider range of emotions. Qualitatively, chatbot responses were more varied in their presentation as well as in the breadth of the emotions themselves. The findings of this study could be used to inform more emotionally connected physician responses to patient message queries.
- 1 October 2025
- Published by Taylor & Francis in Journal of Applied Communication Research
- Vol. 53 (6) , 561-575
Air Aware: a theory-driven, co-designed air quality communication campaign to promote health protective behaviors among outdoor workers
- 1 October 2025
- Published by Taylor & Francis in Journal of Applied Communication Research
- Vol. 53 (6) , 561-575
Air Aware: a theory-driven, co-designed air quality communication campaign to promote health protective behaviors among outdoor workers
- 21 February 2024
- Published by Taylor & Francis in Journal of Communication in Healthcare
- Vol. 17 (2) , 214-222
Breathing on the job: investigating predictors of air quality protective actions and information seeking among outdoor workers
Air quality issues, exacerbated by wildfire smoke and excessive ozone that is worsened by climate change, pose significant health risks to outdoor workers, who are often overlooked in regulatory protection and communication efforts. This study examined how outdoor worker demographics, risk perceptions, and efficacy beliefs predict air quality protective actions and information seeking. Additionally, it investigates the sources of information that this population relies on for understanding air quality. A survey was conducted with 256 outdoor workers in Colorado, a state regularly affected by wildfire smoke and ozone. Measures included demographics, perceived risk, efficacy beliefs, air quality actions, and information seeking behavior. Both perceived risk and efficacy beliefs influenced health-protective actions during poor air quality events. Interestingly, efficacy beliefs were found to be a more reliable predictor of air quality information seeking than perceived risk. The top sources of air quality information among outdoor workers were local news media, The Weather Channel, mobile apps, state public health authorities, and the National Weather Service. These findings enhance our understanding of how perceived risk and efficacy beliefs promote health-protective behaviors among outdoor workers. They lay the groundwork for future research and initiatives to improve air quality communication and promote health-protective actions for this population group. Promoting the efficacy of health-protective actions and seeking information are important components of air quality communication.
- 21 February 2024
- Published by Taylor & Francis in Journal of Communication in Healthcare
- Vol. 17 (2) , 214-222
Breathing on the job: investigating predictors of air quality protective actions and information seeking among outdoor workers
Air quality issues, exacerbated by wildfire smoke and excessive ozone that is worsened by climate change, pose significant health risks to outdoor workers, who are often overlooked in regulatory protection and communication efforts. This study examined how outdoor worker demographics, risk perceptions, and efficacy beliefs predict air quality protective actions and information seeking. Additionally, it investigates the sources of information that this population relies on for understanding air quality. A survey was conducted with 256 outdoor workers in Colorado, a state regularly affected by wildfire smoke and ozone. Measures included demographics, perceived risk, efficacy beliefs, air quality actions, and information seeking behavior. Both perceived risk and efficacy beliefs influenced health-protective actions during poor air quality events. Interestingly, efficacy beliefs were found to be a more reliable predictor of air quality information seeking than perceived risk. The top sources of air quality information among outdoor workers were local news media, The Weather Channel, mobile apps, state public health authorities, and the National Weather Service. These findings enhance our understanding of how perceived risk and efficacy beliefs promote health-protective behaviors among outdoor workers. They lay the groundwork for future research and initiatives to improve air quality communication and promote health-protective actions for this population group. Promoting the efficacy of health-protective actions and seeking information are important components of air quality communication.
- Open Access
- 16 August 2023
- Published by Clemson University in Journal of Extension
- Vol. 61 (2) , 11
After the Disaster Guidebook: Designing a Post-Disaster Communication Intervention for Rural Landowners
Extension is well-positioned to facilitate communication strategies that foster community resilience and disaster recovery, particularly for rural residents. This paper proposes a new approach to post-disaster communication that strengthens rural community capacities in locally and culturally relevant ways. The findings revealed specific post-disaster information needs, preferences for local resources, and communication that encourages resilience through a document analysis and interviews with informants recovering from the 2020 Colorado wildfires. The practical recommendations discussed serve as a starting point for Extension professionals in other areas to consider ways to engage with their communities before, during, and after a disaster.
- 16 August 2023
- Published by Clemson University in Journal of Extension
- Vol. 61 (2) , 11
Open AccessAfter the Disaster Guidebook: Designing a Post-Disaster Communication Intervention for Rural Landowners
Extension is well-positioned to facilitate communication strategies that foster community resilience and disaster recovery, particularly for rural residents. This paper proposes a new approach to post-disaster communication that strengthens rural community capacities in locally and culturally relevant ways. The findings revealed specific post-disaster information needs, preferences for local resources, and communication that encourages resilience through a document analysis and interviews with informants recovering from the 2020 Colorado wildfires. The practical recommendations discussed serve as a starting point for Extension professionals in other areas to consider ways to engage with their communities before, during, and after a disaster.
- Open Access
- 16 October 2021
- Published by MDPI AG in International Journal of Environmental Research and Public Health
- Vol. 18 (20) , 10882
Visualizing the Invisible: Visual-Based Design and Efficacy in Air Quality Messaging
This study examines the effect and efficacy of visual designs for messages about poor air quality. The study utilized a 2 (message efficacy: high vs. low) × 2 (message design: visual vs. text) between-subjects experimental design, of N = 95 students from a large Western university. This experiment assessed the effects of message design and efficacy of language on students’ visual comprehension, source credibility, self-efficacy, and protective behavioral intention. Hypotheses 1 and 2 were partially supported, finding that there were some statistically significant effects for efficacy and message design on students’ comprehension and protective behavioral intention. Future work should focus on strategies for more salient air quality health communication because wildfires will continue to impact the western United States.
- 16 October 2021
- Published by MDPI AG in International Journal of Environmental Research and Public Health
- Vol. 18 (20) , 10882
Open AccessVisualizing the Invisible: Visual-Based Design and Efficacy in Air Quality Messaging
This study examines the effect and efficacy of visual designs for messages about poor air quality. The study utilized a 2 (message efficacy: high vs. low) × 2 (message design: visual vs. text) between-subjects experimental design, of N = 95 students from a large Western university. This experiment assessed the effects of message design and efficacy of language on students’ visual comprehension, source credibility, self-efficacy, and protective behavioral intention. Hypotheses 1 and 2 were partially supported, finding that there were some statistically significant effects for efficacy and message design on students’ comprehension and protective behavioral intention. Future work should focus on strategies for more salient air quality health communication because wildfires will continue to impact the western United States.
- 1 August 2019
The Relative Persuasiveness of Health Infographics