Natalia M. Rodriguez
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Natalia M. Rodriguez has a total of 84 co-authors. Here are the top co-authors by number of shared publications:
Yumary Ruiz
Department of Public Health, College of Health and Human Sciences, Purdue University, West Lafayette, USA
7 shared publicationsCealia Tolliver
Affiliation Department of Public Health, College of Health and Human Sciences, Purdue University, West Lafayette, Indiana, United States of America
5 shared publicationsLara Balian
Purdue University
5 shared publicationsRebecca G. Martinez
Department of Anthropology, College of Liberal Arts, Purdue University, West Lafayette, USA
5 shared publicationsJacqueline C. Linnes
Purdue University
4 shared publicationsLayla Claure
Department of Public Health, Purdue University, West Lafayette, IN, USA
4 shared publications
Latest Publications
- Open Access
- 18 February 2026
- Published by JMIR Publications Inc. in JMIR Public Health and Surveillance
- Vol. 12, e79720
Examining Artificial Intelligence Chatbots’ Responses in Providing Human Papillomavirus Vaccine Information for Young Adults: Qualitative Content Analysis
Background: The growing use of artificial intelligence (AI) chatbots for seeking health-related information is concerning, as they were not originally developed for delivering medical guidance. The quality of AI chatbots’ responses relies heavily on their training data and is often limited in medical contexts due to their lack of specific training data in medical literature. Findings on the quality of AI chatbot responses related to health are mixed. Some studies showed the quality surpassed physicians’ responses, while others revealed occasional major errors and low readability. This study addresses a critical gap by examining the performance of various AI chatbots in a complex, misinformation-rich environment. Objective: This study examined AI chatbots' responses to human papillomavirus (HPV)–related questions by analyzing structure, linguistic features, information accuracy and currency, and vaccination stance. Methods: We conducted a qualitative content analysis following the approach outlined by Schreier to examine 4 selected AI chatbots’ (ChatGPT 4, Claude 3.7 Sonnet, DeepSeek V3, and Docus [General AI Doctor]) responses to HPV vaccine questions. These questions, simulated by young adults, were adapted from items on the Vaccine Conspiracy Beliefs Scale and Google Trends. The selection criteria for AI chatbots included popularity, accessibility, countries of origin, response update methods, and intended use. Two researchers, simulating a 22-year-old man or woman, collected 8 conversations between February 22 and 28, 2025. We used a deductive approach to develop initial code groups, then an inductive approach to generate codes. The responses were analyzed based on a comprehensive codebook, with codes examining response structure, linguistic features, information accuracy and currency, and vaccination stance. We also assessed readability using the Flesch-Kincaid Grade Level and Reading Ease Score. Results: All AI chatbots cited evidence-based sources from reputable health organizations. We found no fabricated information or inaccuracies in numerical data. For complex questions, all AI chatbots appropriately deferred to health care professionals’ suggestions. All AI chatbots maintained a neutral or provaccine stance, corresponding with scientific consensus. The mean and range of response lengths varied [word count; ChatGPT: 436.4 (218-954); Claude: 188.0 (138-255); DeepSeek: 510.0 (325-735); and Docus: 159.4 (61-200)], as did readability [Flesch-Kincaid Grade Level; ChatGPT: 10.7 (6.0-14.9); Claude: 13.2 (7.7-17.8); DeepSeek: 11.3 (7.0-14.7); and Docus: 12.2 (8.9-15.5); and Flesch-Kincaid Reading Ease Score; ChatGPT: 46.8 (25.4-72.2); Claude: 32.5 (6.3-67.3); DeepSeek: 43.7 (22.8-67.4); and Docus: 40.5 (19.6-58.2)]. ChatGPT and Claude offered personalized responses, while DeepSeek and Docus lacked this. Occasionally, some responses included broken or irrelevant links and medical jargon. Conclusions: Amidst an online environment saturated with misinformation, AI chatbots have the potential to serve as an alternative source of accurate HPV-related information to conventional online platforms (websites and social media). Improvements in readability, personalization, and link accuracy are still needed. Furthermore, we recommend that users treat AI chatbots as complements, not replacements, to health care professionals’ guidance on clinical settings.
- 18 February 2026
- Published by JMIR Publications Inc. in JMIR Public Health and Surveillance
- Vol. 12, e79720
Open AccessExamining Artificial Intelligence Chatbots’ Responses in Providing Human Papillomavirus Vaccine Information for Young Adults: Qualitative Content Analysis
Background: The growing use of artificial intelligence (AI) chatbots for seeking health-related information is concerning, as they were not originally developed for delivering medical guidance. The quality of AI chatbots’ responses relies heavily on their training data and is often limited in medical contexts due to their lack of specific training data in medical literature. Findings on the quality of AI chatbot responses related to health are mixed. Some studies showed the quality surpassed physicians’ responses, while others revealed occasional major errors and low readability. This study addresses a critical gap by examining the performance of various AI chatbots in a complex, misinformation-rich environment. Objective: This study examined AI chatbots' responses to human papillomavirus (HPV)–related questions by analyzing structure, linguistic features, information accuracy and currency, and vaccination stance. Methods: We conducted a qualitative content analysis following the approach outlined by Schreier to examine 4 selected AI chatbots’ (ChatGPT 4, Claude 3.7 Sonnet, DeepSeek V3, and Docus [General AI Doctor]) responses to HPV vaccine questions. These questions, simulated by young adults, were adapted from items on the Vaccine Conspiracy Beliefs Scale and Google Trends. The selection criteria for AI chatbots included popularity, accessibility, countries of origin, response update methods, and intended use. Two researchers, simulating a 22-year-old man or woman, collected 8 conversations between February 22 and 28, 2025. We used a deductive approach to develop initial code groups, then an inductive approach to generate codes. The responses were analyzed based on a comprehensive codebook, with codes examining response structure, linguistic features, information accuracy and currency, and vaccination stance. We also assessed readability using the Flesch-Kincaid Grade Level and Reading Ease Score. Results: All AI chatbots cited evidence-based sources from reputable health organizations. We found no fabricated information or inaccuracies in numerical data. For complex questions, all AI chatbots appropriately deferred to health care professionals’ suggestions. All AI chatbots maintained a neutral or provaccine stance, corresponding with scientific consensus. The mean and range of response lengths varied [word count; ChatGPT: 436.4 (218-954); Claude: 188.0 (138-255); DeepSeek: 510.0 (325-735); and Docus: 159.4 (61-200)], as did readability [Flesch-Kincaid Grade Level; ChatGPT: 10.7 (6.0-14.9); Claude: 13.2 (7.7-17.8); DeepSeek: 11.3 (7.0-14.7); and Docus: 12.2 (8.9-15.5); and Flesch-Kincaid Reading Ease Score; ChatGPT: 46.8 (25.4-72.2); Claude: 32.5 (6.3-67.3); DeepSeek: 43.7 (22.8-67.4); and Docus: 40.5 (19.6-58.2)]. ChatGPT and Claude offered personalized responses, while DeepSeek and Docus lacked this. Occasionally, some responses included broken or irrelevant links and medical jargon. Conclusions: Amidst an online environment saturated with misinformation, AI chatbots have the potential to serve as an alternative source of accurate HPV-related information to conventional online platforms (websites and social media). Improvements in readability, personalization, and link accuracy are still needed. Furthermore, we recommend that users treat AI chatbots as complements, not replacements, to health care professionals’ guidance on clinical settings.
- Open Access
- 30 January 2026
- Published by Springer Nature in Research Square
Cervical Cancer Screening Rates and Associated Sociodemographic and Behavioral Factors among People Experiencing Homelessness in Indiana
Objective To estimate cervical cancer screening rates, prevalence of risk factors, and factors associated with being overdue for screening among people experiencing homelessness in two Indiana cities. Methods Rapid assessment surveys were conducted at two large homelessness service agencies in Indianapolis and Lafayette, Indiana (November 2023 to November 2024). Participants were aged 21–69 years, assigned female at birth, and currently experiencing homelessness. Screening status was categorized as up-to-date (screened within 5 years) or overdue (more than 5 years or never screened). Descriptive statistics and multivariable logistic regression examined predictors of being overdue. Results Among n = 212 participants, 35% were overdue and 49% had not been screened within 3 years. Prevalence of risk factors was high, including smoking (74%), sexual debut before age 18 (73%), and no HPV vaccination (75%). Older age and having experienced homelessness for 5 years or more were associated with higher odds of being overdue. Conclusions Longer duration of homelessness significantly increased the likelihood of being overdue for cervical cancer screening, underscoring the cumulative disadvantage of chronic housing instability. Cervical cancer prevention is a critical unmet need among women experiencing homelessness, and findings highlight the importance of developing targeted interventions to improve screening access in this population.
- 30 January 2026
- Published by Springer Nature in Research Square
Open AccessCervical Cancer Screening Rates and Associated Sociodemographic and Behavioral Factors among People Experiencing Homelessness in Indiana
Objective To estimate cervical cancer screening rates, prevalence of risk factors, and factors associated with being overdue for screening among people experiencing homelessness in two Indiana cities. Methods Rapid assessment surveys were conducted at two large homelessness service agencies in Indianapolis and Lafayette, Indiana (November 2023 to November 2024). Participants were aged 21–69 years, assigned female at birth, and currently experiencing homelessness. Screening status was categorized as up-to-date (screened within 5 years) or overdue (more than 5 years or never screened). Descriptive statistics and multivariable logistic regression examined predictors of being overdue. Results Among n = 212 participants, 35% were overdue and 49% had not been screened within 3 years. Prevalence of risk factors was high, including smoking (74%), sexual debut before age 18 (73%), and no HPV vaccination (75%). Older age and having experienced homelessness for 5 years or more were associated with higher odds of being overdue. Conclusions Longer duration of homelessness significantly increased the likelihood of being overdue for cervical cancer screening, underscoring the cumulative disadvantage of chronic housing instability. Cervical cancer prevention is a critical unmet need among women experiencing homelessness, and findings highlight the importance of developing targeted interventions to improve screening access in this population.
- Open Access
- 19 January 2026
- Published by SAGE Publications in International Journal of Qualitative Methods
- Vol. 25
Tangible, Timely, and Thorough: A Case Study on Using Rapid Qualitative Analysis to Inform Human-Centered Intervention Design
Traditional qualitative analysis, while valuable in providing rich understandings of people’s lived experience, can be resource and time-intensive, which is challenging for time-sensitive and action-oriented research such as that required in community-engaged and human-centered design (HCD) projects. Rapid qualitative analysis (RQA) has emerged as a valuable approach to balance timeliness with rigor, enabling real-time intervention development. However, limited literature discusses how RQA has been applied to community-based and HCD approaches in public health interventions. This paper presents a methodological case study illustrating how RQA was integrated into an HCD co-design process as part of a larger community-based participatory research project addressing cervical cancer among people experiencing homelessness (PEH) in Indiana. The study aimed to co-design novel cervical cancer screening strategies with and for PEH. A RQA of qualitative interviews with PEH guided iterative intervention development by streamlining data analysis without sacrificing methodological rigor. The RQA process involved (1) establishing key domains of interest based on study aims and the interview guide, (2) creating and testing a summary template for extracting data, and (3) summarizing interview audio recordings using the template and reaching consensus among team members to ensure consistency. This process resulted in a matrix of summarized interview data (30 participants across 15 template domains) synthesized into major themes and compiled into reports that were shared in real time with community partners to inform ongoing intervention refinements. The use of RQA allowed the research team to deliver timely, actionable findings that supported iterative intervention development, and deepened community engagement. Our experience, presented here as a replicable model, demonstrates that RQA is a valuable methodological tool for community-engaged, time-sensitive public health research and should be considered in future HCD projects.
- 19 January 2026
- Published by SAGE Publications in International Journal of Qualitative Methods
- Vol. 25
Open AccessTangible, Timely, and Thorough: A Case Study on Using Rapid Qualitative Analysis to Inform Human-Centered Intervention Design
Traditional qualitative analysis, while valuable in providing rich understandings of people’s lived experience, can be resource and time-intensive, which is challenging for time-sensitive and action-oriented research such as that required in community-engaged and human-centered design (HCD) projects. Rapid qualitative analysis (RQA) has emerged as a valuable approach to balance timeliness with rigor, enabling real-time intervention development. However, limited literature discusses how RQA has been applied to community-based and HCD approaches in public health interventions. This paper presents a methodological case study illustrating how RQA was integrated into an HCD co-design process as part of a larger community-based participatory research project addressing cervical cancer among people experiencing homelessness (PEH) in Indiana. The study aimed to co-design novel cervical cancer screening strategies with and for PEH. A RQA of qualitative interviews with PEH guided iterative intervention development by streamlining data analysis without sacrificing methodological rigor. The RQA process involved (1) establishing key domains of interest based on study aims and the interview guide, (2) creating and testing a summary template for extracting data, and (3) summarizing interview audio recordings using the template and reaching consensus among team members to ensure consistency. This process resulted in a matrix of summarized interview data (30 participants across 15 template domains) synthesized into major themes and compiled into reports that were shared in real time with community partners to inform ongoing intervention refinements. The use of RQA allowed the research team to deliver timely, actionable findings that supported iterative intervention development, and deepened community engagement. Our experience, presented here as a replicable model, demonstrates that RQA is a valuable methodological tool for community-engaged, time-sensitive public health research and should be considered in future HCD projects.
- Open Access
- 25 November 2025
- Published by Royal Society of Chemistry (RSC) in Analytical Methods
- Vol. 18 (4) , 786-799
Development of an HPV 16 rapid test founded in user-centered design with primary care clinicians
Despite effective screening modalities, cervical cancer remains a leading cause of cancer-related death among women in the United States aged 20 to 39 years old, and incidence is rising in women aged 30–44 years old. Up to 25% of patients who are screened for cervical cancer by testing for human papillomavirus (HPV) do not receive necessary follow-up care with current laboratory-based testing. Applying a user-centered design approach, we surveyed and interviewed practicing clinicians to establish the use case, value proposition, and user requirements of a cervical cancer screening test for use in Indiana, USA. Insights from these stakeholders directly informed design specifications for a point-of-care HPV test capable of providing same-visit results to improve patient follow-up and retention. Guided by these requirements, we designed an isothermal nucleic acid amplification platform suitable for outpatient clinics. The test accepts swabbed endocervical cells, amplifies HPV16 L1 DNA via recombinase polymerase amplification, and provides results within 40 minutes on a lateral flow assay. Further, the test achieves a clinically relevant limit of detection of 1000 HPV 16 copies per reaction and verifies swabbing technique and test operation with a sample adequacy control. The test operation was designed for a minimally-trained user and decreases time-sensitive steps that would interfere with clinical flow. By integrating clinician input to inform development decisions, our device is uniquely tailored to meet the context-specific needs of primary care clinics. This work exemplifies how user-centered design can yield novel diagnostic technologies with greater clinical impact and adoption potential.
- 25 November 2025
- Published by Royal Society of Chemistry (RSC) in Analytical Methods
- Vol. 18 (4) , 786-799
Open AccessDevelopment of an HPV 16 rapid test founded in user-centered design with primary care clinicians
Despite effective screening modalities, cervical cancer remains a leading cause of cancer-related death among women in the United States aged 20 to 39 years old, and incidence is rising in women aged 30–44 years old. Up to 25% of patients who are screened for cervical cancer by testing for human papillomavirus (HPV) do not receive necessary follow-up care with current laboratory-based testing. Applying a user-centered design approach, we surveyed and interviewed practicing clinicians to establish the use case, value proposition, and user requirements of a cervical cancer screening test for use in Indiana, USA. Insights from these stakeholders directly informed design specifications for a point-of-care HPV test capable of providing same-visit results to improve patient follow-up and retention. Guided by these requirements, we designed an isothermal nucleic acid amplification platform suitable for outpatient clinics. The test accepts swabbed endocervical cells, amplifies HPV16 L1 DNA via recombinase polymerase amplification, and provides results within 40 minutes on a lateral flow assay. Further, the test achieves a clinically relevant limit of detection of 1000 HPV 16 copies per reaction and verifies swabbing technique and test operation with a sample adequacy control. The test operation was designed for a minimally-trained user and decreases time-sensitive steps that would interfere with clinical flow. By integrating clinician input to inform development decisions, our device is uniquely tailored to meet the context-specific needs of primary care clinics. This work exemplifies how user-centered design can yield novel diagnostic technologies with greater clinical impact and adoption potential.
- Open Access
- 17 November 2025
- Published by Cambridge University Press (CUP) in Journal of Clinical and Translational Science
- Vol. 9 (1) , 1-22
Patient attitudes toward HPV self-sampling and community health worker-delivered cervical cancer screening services in an underserved area
Introduction: In Indiana, Black women had 12% higher cervical cancer incidence, and 21% higher mortality compared to White women during the years 2008–2017. Human Papillomavirus (HPV) self-sampling delivered by community health workers (CHW) has demonstrated potential to increase cervical cancer screening rates among at-risk populations. This study aims to understand patient attitudes toward HPV self-sampling and CHW-delivered screening. Methods: We conducted a cross-sectional online survey among patients from three Planned Parenthood clinics in Lake County, IN, which has one of the highest cervical cancer mortality rates in Indiana. Patients’ willingness to self-sample and to receive health education and services related to cervical cancer screening from CHWs was analyzed using logistic regression. Results: In the final sample ( N = 140), the largest proportions of respondents were below age 30 (54.3%), aware of cervical cancer (78.6%), up to date on cervical cancer screening (84.3%), had no prior self-sampling experience (53.6%) and were aware of CHWs (52.1%). Multivariable logistic regression analyses showed Black patients had higher odds of being willing to: self-sample at home (aOR = 3.749, 95% CI = 1.619–8.681), receive health education from CHWs (aOR = 4.952, 95% CI = 2.124–11.548), and receive health services from CHWs (aOR = 3.305, 95% CI = 1.479–7.388) compared to White patients. Conclusion: Our results showed Black patients had higher willingness for HPV self-sampling and CHW delivery of cervical cancer screening services. Study findings can be used to inform future CHW-led interventions for outreach, education, and delivery of self-sampling interventions to increase cervical cancer screening rates among Black women in underserved areas.
- 17 November 2025
- Published by Cambridge University Press (CUP) in Journal of Clinical and Translational Science
- Vol. 9 (1) , 1-22
Open AccessPatient attitudes toward HPV self-sampling and community health worker-delivered cervical cancer screening services in an underserved area
Introduction: In Indiana, Black women had 12% higher cervical cancer incidence, and 21% higher mortality compared to White women during the years 2008–2017. Human Papillomavirus (HPV) self-sampling delivered by community health workers (CHW) has demonstrated potential to increase cervical cancer screening rates among at-risk populations. This study aims to understand patient attitudes toward HPV self-sampling and CHW-delivered screening. Methods: We conducted a cross-sectional online survey among patients from three Planned Parenthood clinics in Lake County, IN, which has one of the highest cervical cancer mortality rates in Indiana. Patients’ willingness to self-sample and to receive health education and services related to cervical cancer screening from CHWs was analyzed using logistic regression. Results: In the final sample ( N = 140), the largest proportions of respondents were below age 30 (54.3%), aware of cervical cancer (78.6%), up to date on cervical cancer screening (84.3%), had no prior self-sampling experience (53.6%) and were aware of CHWs (52.1%). Multivariable logistic regression analyses showed Black patients had higher odds of being willing to: self-sample at home (aOR = 3.749, 95% CI = 1.619–8.681), receive health education from CHWs (aOR = 4.952, 95% CI = 2.124–11.548), and receive health services from CHWs (aOR = 3.305, 95% CI = 1.479–7.388) compared to White patients. Conclusion: Our results showed Black patients had higher willingness for HPV self-sampling and CHW delivery of cervical cancer screening services. Study findings can be used to inform future CHW-led interventions for outreach, education, and delivery of self-sampling interventions to increase cervical cancer screening rates among Black women in underserved areas.
- Open Access
- 30 October 2025
- Published by Public Library of Science (PLoS) in PLOS Global Public Health
- Vol. 5 (10) , e0005366
Stakeholder considerations on acceptability and implementation of a novel rapid test for acute HIV infection: A qualitative study in Indiana
Acute HIV infection (AHI) is the most infectious stage of HIV, yet existing rapid tests cannot reliably detect HIV in this early phase and require up to 90 days post-exposure for accurate results. Laboratory-based nucleic acid tests can detect AHI but are often inaccessible to high-risk populations with limited healthcare access. Novel rapid AHI tests are in development to bridge this gap by enabling earlier, decentralized detection. This study aims to explore the acceptability and future implementation considerations of such a test through engagement with clients (n = 5) and staff (n = 14) of Indiana-based HIV service organizations. Guided by human-centered design frameworks, semi-structured interviews examined experiences with HIV testing, acceptability of a proposed novel rapid AHI test that could detect infection about one month earlier but would require longer time to results (~60 minutes), and preferences for potential end-users (self, community health workers (CHW) or peers). Transcripts were thematic analyzed. Participants strongly supported earlier detection despite longer wait times, describing the trade-off as worthwhile for reducing transmission, initiating treatment sooner, and alleviating anxiety following high-risk exposures. Clients emphasized the benefits of convenience and peace of mind, particularly for people who use drugs or experience stigma. Staff highlighted the potential to retain individuals who are often lost to follow-up and to strengthen linkage to care. Concerns focused on organizational feasibility of accommodating longer test times and client expectations shaped by existing “rapid” tests. CHW/peer-led testing was preferred due to their ability to build trust, provide education and navigation, while self-testing was viewed as useful for overcoming barriers such as stigma and access, though concerns about accuracy, interpretation, and lack of counseling were noted. Stakeholder insights underscore the importance of accuracy, usability, affordability, and integration of counseling and care navigation in designing and implementing AHI rapid tests to maximize their impact among high-risk populations.
- 30 October 2025
- Published by Public Library of Science (PLoS) in PLOS Global Public Health
- Vol. 5 (10) , e0005366
Open AccessStakeholder considerations on acceptability and implementation of a novel rapid test for acute HIV infection: A qualitative study in Indiana
Acute HIV infection (AHI) is the most infectious stage of HIV, yet existing rapid tests cannot reliably detect HIV in this early phase and require up to 90 days post-exposure for accurate results. Laboratory-based nucleic acid tests can detect AHI but are often inaccessible to high-risk populations with limited healthcare access. Novel rapid AHI tests are in development to bridge this gap by enabling earlier, decentralized detection. This study aims to explore the acceptability and future implementation considerations of such a test through engagement with clients (n = 5) and staff (n = 14) of Indiana-based HIV service organizations. Guided by human-centered design frameworks, semi-structured interviews examined experiences with HIV testing, acceptability of a proposed novel rapid AHI test that could detect infection about one month earlier but would require longer time to results (~60 minutes), and preferences for potential end-users (self, community health workers (CHW) or peers). Transcripts were thematic analyzed. Participants strongly supported earlier detection despite longer wait times, describing the trade-off as worthwhile for reducing transmission, initiating treatment sooner, and alleviating anxiety following high-risk exposures. Clients emphasized the benefits of convenience and peace of mind, particularly for people who use drugs or experience stigma. Staff highlighted the potential to retain individuals who are often lost to follow-up and to strengthen linkage to care. Concerns focused on organizational feasibility of accommodating longer test times and client expectations shaped by existing “rapid” tests. CHW/peer-led testing was preferred due to their ability to build trust, provide education and navigation, while self-testing was viewed as useful for overcoming barriers such as stigma and access, though concerns about accuracy, interpretation, and lack of counseling were noted. Stakeholder insights underscore the importance of accuracy, usability, affordability, and integration of counseling and care navigation in designing and implementing AHI rapid tests to maximize their impact among high-risk populations.
- Open Access
- 13 September 2025
- Published by Elsevier in Preventive Medicine Reports
- Vol. 58, 103241
Human papillomavirus awareness in a digital world: The intersection of faith and online media
- 13 September 2025
- Published by Elsevier in Preventive Medicine Reports
- Vol. 58, 103241
Open AccessHuman papillomavirus awareness in a digital world: The intersection of faith and online media
- Open Access
- 26 June 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Examining Artificial Intelligence Chatbots’ Responses in Providing Human Papillomavirus Vaccine Information for Young Adults: Qualitative Content Analysis (Preprint)
BACKGROUND The growing use of artificial intelligence (AI) chatbots for seeking health-related information is concerning, as they were not originally developed for delivering medical guidance. The quality of AI chatbots’ responses relies heavily on their training data and is often limited in medical contexts due to their lack of specific training data in medical literature. Findings on the quality of AI chatbot responses related to health are mixed. Some studies showed the quality surpassed physicians’ responses, while others revealed occasional major errors and low readability. This study addresses a critical gap by examining the performance of various AI chatbots in a complex, misinformation-rich environment. OBJECTIVE This study examined AI chatbots' responses to human papillomavirus (HPV)–related questions by analyzing structure, linguistic features, information accuracy and currency, and vaccination stance. METHODS We conducted a qualitative content analysis following the approach outlined by Schreier to examine 4 selected AI chatbots’ (ChatGPT 4, Claude 3.7 Sonnet, DeepSeek V3, and Docus [General AI Doctor]) responses to HPV vaccine questions. These questions, simulated by young adults, were adapted from items on the Vaccine Conspiracy Beliefs Scale and Google Trends. The selection criteria for AI chatbots included popularity, accessibility, countries of origin, response update methods, and intended use. Two researchers, simulating a 22-year-old man or woman, collected 8 conversations between February 22 and 28, 2025. We used a deductive approach to develop initial code groups, then an inductive approach to generate codes. The responses were analyzed based on a comprehensive codebook, with codes examining response structure, linguistic features, information accuracy and currency, and vaccination stance. We also assessed readability using the Flesch-Kincaid Grade Level and Reading Ease Score. RESULTS All AI chatbots cited evidence-based sources from reputable health organizations. We found no fabricated information or inaccuracies in numerical data. For complex questions, all AI chatbots appropriately deferred to health care professionals’ suggestions. All AI chatbots maintained a neutral or provaccine stance, corresponding with scientific consensus. The mean and range of response lengths varied [word count; ChatGPT: 436.4 (218-954); Claude: 188.0 (138-255); DeepSeek: 510.0 (325-735); and Docus: 159.4 (61-200)], as did readability [Flesch-Kincaid Grade Level; ChatGPT: 10.7 (6.0-14.9); Claude: 13.2 (7.7-17.8); DeepSeek: 11.3 (7.0-14.7); and Docus: 12.2 (8.9-15.5); and Flesch-Kincaid Reading Ease Score; ChatGPT: 46.8 (25.4-72.2); Claude: 32.5 (6.3-67.3); DeepSeek: 43.7 (22.8-67.4); and Docus: 40.5 (19.6-58.2)]. ChatGPT and Claude offered personalized responses, while DeepSeek and Docus lacked this. Occasionally, some responses included broken or irrelevant links and medical jargon. CONCLUSIONS Amidst an online environment saturated with misinformation, AI chatbots have the potential to serve as an alternative source of accurate HPV-related information to conventional online platforms (websites and social media). Improvements in readability, personalization, and link accuracy are still needed. Furthermore, we recommend that users treat AI chatbots as complements, not replacements, to health care professionals’ guidance on clinical settings. CLINICALTRIAL
- 26 June 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessExamining Artificial Intelligence Chatbots’ Responses in Providing Human Papillomavirus Vaccine Information for Young Adults: Qualitative Content Analysis (Preprint)
BACKGROUND The growing use of artificial intelligence (AI) chatbots for seeking health-related information is concerning, as they were not originally developed for delivering medical guidance. The quality of AI chatbots’ responses relies heavily on their training data and is often limited in medical contexts due to their lack of specific training data in medical literature. Findings on the quality of AI chatbot responses related to health are mixed. Some studies showed the quality surpassed physicians’ responses, while others revealed occasional major errors and low readability. This study addresses a critical gap by examining the performance of various AI chatbots in a complex, misinformation-rich environment. OBJECTIVE This study examined AI chatbots' responses to human papillomavirus (HPV)–related questions by analyzing structure, linguistic features, information accuracy and currency, and vaccination stance. METHODS We conducted a qualitative content analysis following the approach outlined by Schreier to examine 4 selected AI chatbots’ (ChatGPT 4, Claude 3.7 Sonnet, DeepSeek V3, and Docus [General AI Doctor]) responses to HPV vaccine questions. These questions, simulated by young adults, were adapted from items on the Vaccine Conspiracy Beliefs Scale and Google Trends. The selection criteria for AI chatbots included popularity, accessibility, countries of origin, response update methods, and intended use. Two researchers, simulating a 22-year-old man or woman, collected 8 conversations between February 22 and 28, 2025. We used a deductive approach to develop initial code groups, then an inductive approach to generate codes. The responses were analyzed based on a comprehensive codebook, with codes examining response structure, linguistic features, information accuracy and currency, and vaccination stance. We also assessed readability using the Flesch-Kincaid Grade Level and Reading Ease Score. RESULTS All AI chatbots cited evidence-based sources from reputable health organizations. We found no fabricated information or inaccuracies in numerical data. For complex questions, all AI chatbots appropriately deferred to health care professionals’ suggestions. All AI chatbots maintained a neutral or provaccine stance, corresponding with scientific consensus. The mean and range of response lengths varied [word count; ChatGPT: 436.4 (218-954); Claude: 188.0 (138-255); DeepSeek: 510.0 (325-735); and Docus: 159.4 (61-200)], as did readability [Flesch-Kincaid Grade Level; ChatGPT: 10.7 (6.0-14.9); Claude: 13.2 (7.7-17.8); DeepSeek: 11.3 (7.0-14.7); and Docus: 12.2 (8.9-15.5); and Flesch-Kincaid Reading Ease Score; ChatGPT: 46.8 (25.4-72.2); Claude: 32.5 (6.3-67.3); DeepSeek: 43.7 (22.8-67.4); and Docus: 40.5 (19.6-58.2)]. ChatGPT and Claude offered personalized responses, while DeepSeek and Docus lacked this. Occasionally, some responses included broken or irrelevant links and medical jargon. CONCLUSIONS Amidst an online environment saturated with misinformation, AI chatbots have the potential to serve as an alternative source of accurate HPV-related information to conventional online platforms (websites and social media). Improvements in readability, personalization, and link accuracy are still needed. Furthermore, we recommend that users treat AI chatbots as complements, not replacements, to health care professionals’ guidance on clinical settings. CLINICALTRIAL
- Open Access
- 30 May 2025
- Published by Springer Nature in BMC Health Services Research
- Vol. 25 (1) , 1-13
Supporting menstrual health in homeless services: provider-informed strategies for multilevel change
Background: People experiencing homelessness (PEH) face heightened barriers to menstrual health, including limited access to products, hygiene facilities, and consistent care. Social service providers (SSPs) and healthcare providers (HCPs) are critical in supporting PEH but often work within resource-constrained systems. This study explores SSP and HCP perspectives on the menstrual health needs of PEH in a rural-serving community in the U.S., using the Social-Ecological Model (SEM) to identify multilevel barriers and opportunities for intervention. Methods: We conducted semi-structured interviews with 12 SSPs and HCPs in Tippecanoe County, Indiana. Transcripts were analyzed using thematic analysis to identify key challenges and recommendations related to menstrual health support for PEH. Results: Our analysis revealed complex challenges providers faced in supporting the menstrual health of PEH, including emotional burden, resource scarcity, and limited organizational guidance. They described how stigma, provider discomfort, and systemic gaps in training and infrastructure hindered effective care. Providers also shared that menstrual health was often deprioritized due to competing health needs and structural barriers such as lack of housing, transportation, and product access. Despite these challenges, providers offered actionable recommendations to improve menstrual health support through education, policy change, and more equitable organizational practices. Conclusions: This study highlights the multi-level barriers providers face when supporting the menstrual health needs of PEH. Our findings show that meaningful change requires coordinated efforts across all levels of the SEM. Actionable strategies include provider training, improved intake processes, expanded access to menstrual products and hygiene resources, and policy reforms to address housing and insurance gaps. These insights can inform training programs, shelter protocols, and advocacy efforts to promote menstrual health equity and provider sustainability.
- 30 May 2025
- Published by Springer Nature in BMC Health Services Research
- Vol. 25 (1) , 1-13
Open AccessSupporting menstrual health in homeless services: provider-informed strategies for multilevel change
Background: People experiencing homelessness (PEH) face heightened barriers to menstrual health, including limited access to products, hygiene facilities, and consistent care. Social service providers (SSPs) and healthcare providers (HCPs) are critical in supporting PEH but often work within resource-constrained systems. This study explores SSP and HCP perspectives on the menstrual health needs of PEH in a rural-serving community in the U.S., using the Social-Ecological Model (SEM) to identify multilevel barriers and opportunities for intervention. Methods: We conducted semi-structured interviews with 12 SSPs and HCPs in Tippecanoe County, Indiana. Transcripts were analyzed using thematic analysis to identify key challenges and recommendations related to menstrual health support for PEH. Results: Our analysis revealed complex challenges providers faced in supporting the menstrual health of PEH, including emotional burden, resource scarcity, and limited organizational guidance. They described how stigma, provider discomfort, and systemic gaps in training and infrastructure hindered effective care. Providers also shared that menstrual health was often deprioritized due to competing health needs and structural barriers such as lack of housing, transportation, and product access. Despite these challenges, providers offered actionable recommendations to improve menstrual health support through education, policy change, and more equitable organizational practices. Conclusions: This study highlights the multi-level barriers providers face when supporting the menstrual health needs of PEH. Our findings show that meaningful change requires coordinated efforts across all levels of the SEM. Actionable strategies include provider training, improved intake processes, expanded access to menstrual products and hygiene resources, and policy reforms to address housing and insurance gaps. These insights can inform training programs, shelter protocols, and advocacy efforts to promote menstrual health equity and provider sustainability.
- Open Access
- 13 January 2025
- Published by Elsevier in Preventive Medicine Reports
- Vol. 50, 102975
Association between health insurance coverage and stage of diagnosis for cervical cancer among females in Indiana from 2011 - 2019
- 13 January 2025
- Published by Elsevier in Preventive Medicine Reports
- Vol. 50, 102975
Open AccessAssociation between health insurance coverage and stage of diagnosis for cervical cancer among females in Indiana from 2011 - 2019