The acceptability and effectiveness of artificial intelligence-based chatbot for hypertensive patients in community: protocol for a mixed-methods study

The acceptability and effectiveness of artificial intelligence-based chatbot for hypertensive patients in community: protocol for a mixed-methods study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Study protocol The acceptability and effectiveness of artificial intelligence-based chatbot for hypertensive patients in community: protocol for a mixed-methods study Ping chen, Yi Li, Xuxi Zhang, Xinglin Feng, Xinying Sun This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3903126/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 21 Aug, 2024 Read the published version in BMC Public Health → Version 1 posted 5 You are reading this latest preprint version Abstract Background Chatbots can provide immediate assistance tailored to patients' needs, making them suitable for sustained accompanying interventions. Nevertheless, there is currently no evidence regarding their acceptability by hypertensive patients and the factors influencing the acceptability in the real-world. Existing evaluation scales often focus solely on the technology itself, overlooking the patients' perspective. Utilizing mixed methods can offer a more comprehensive exploration of influencing factors, laying the groundwork for the future integration of artificial intelligence in chronic disease management practices. Methods The mixed methods will provide a holistic view to understand the effectiveness and acceptability of the intervention. Participants will either receive the standard primary health care or obtain a chatbot speaker. The speaker can provide timely reminders, on-demand consultations, personalized data recording, knowledge broadcasts, as well as entertainment features such as telling jokes. The quantitative part will be conducted as a quasi-randomized controlled trial in community in Beijing. And the convergent design will be adopted. When patients use the speaker for 1 month, scales will be used to measure patients' intention to use the speaker. At the same time, semi-structured interviews will be conducted to explore patients' feelings and influencing factors of using speakers. Data on socio-demography, physical examination, blood pressure, acceptability and self-management behavior will be collected at baseline, as well as 1,3,6, and 12 months later. Furthermore, the cloud database will continuously collect patients’ interactions with the speaker. The primary outcome is the efficacy of the chatbot on blood pressure control. The secondary outcome includes the acceptability of the chatbot speaker and the changes of self-management behavior. Discussion Artificial intelligence-based chatbot speaker not only caters to patients' self-management needs at home but also effectively organizes intricate and detailed knowledge system for patients with hypertension through a knowledge graph. Patients can promptly access information that aligns with their specific requirements, promoting proactive self-management and playing a crucial role in disease management. This study will serve as a foundation for the application of artificial intelligence technology in chronic disease management, paving the way for further exploration on enhancing the communicative impact of artificial intelligence technology. Trial registration: Biomedical Ethics Committee of Peking University: IRB00001052-21106, 2021/10/14; Clinical Trials: ChiCTR2100050578,2021/08/29. Hypertension Artificial intelligence voice-based chatbot Mixed-methods Figures Figure 1 Figure 2 Background Global Burden of disease indicates that the prevalence of hypertension has doubled in 2019 compared to 1990, with nearly 1.3 billion people aged 30–79 living with hypertension worldwide [ 1 ], and more than 10 million patients dying from high systolic blood pressure, which accounts for up to 20% of all-cause deaths. In China, the prevalence of hypertension in adults has reached 27.5%[ 2 ], but the control rate is much lower, only 7.2% in patients aged 35 to 75 in China[ 3 ]. In addition, this trend will continue to intensify with the aging of the population and the complications will also lead to a growing medical burden. A good lifestyle is the cornerstone of hypertension treatment[ 4 ]. A British study pointed out that among all genetic risk groups for hypertension, the systolic blood pressure of those with a good lifestyle was significantly lower than that of those with a bad lifestyle [ 5 ]. For every 5mmHg reduction in systolic blood pressure, the risk of cardiovascular events was reduced by 10%[ 6 ]. The guidelines emphasized low-sodium diet, regular exercise, weight loss, limited alcohol consumption and adequate potassium intake[ 7 ]. Many studies have confirmed that changing exposure to these factors will effectively improve blood pressure control. For example, reducing sodium intake can reduce systolic blood pressure by 7.7 mmHg and diastolic blood pressure by 3.0 mmHg[ 8 ], and it has been confirmed that it can effectively reduce the risk of cardiovascular events [ 9 ], which has a dose-response relationship with blood pressure reduction[ 8 ]. It also significantly reduces the risk of stroke, heart disease, and all-cause death[ 10 , 11 ]. Desh diet is the most effective dietary pattern for hypertension management, and there is evidence that the intervention of Desh diet has a lasting effect [ 12 ]. There are studies that have proved the Chinese style-Desh diet can achieve good control of blood pressure in Chinese people[ 5 ]. In addition, moderate-intensity exercise for more than 30 minutes at least 3 times per week in hypertensive patients can effectively lower blood pressure and reduce the risk of cardiovascular death and all-cause death [ 13 ]. Therefore, improved lifestyle is effective in controlling blood pressure and reducing the risk of death. However, the content of hypertension self-management is abundant and the details are complex, and a systematic intervention framework has not yet been formed. Traditional methods, such as health education lecture in person and distributing promotional materials, are unsustainable, limited and difficult to popularize[ 14 ]. With the rapid development of artificial intelligence, the new approach based on digital technology have emerged in recent years, which is expected to solve these traditional problems. Chatbots have been found extensive applications amid the COVID-19 pandemic for enhancing awareness, owing to their cost-effectiveness, capacity to reach a broad audience, and compatibility with multiple devices[ 15 ]. And nowadays, chatbots are already widely used in areas such as weight-loss related behaviors (dietary intake and physical activity) and psycho-emotional support, but there are limited studies on hypertension. A study showed that chatbot facilitating home blood pressure monitoring did not increase adherence to monitoring blood pressure compared to traditional methods, but it did improve knowledge and skills of patients[ 12 ] Moreover, the majority of existing chatbots primarily rely on text, with only a limited number being voice-based. In China, where the aging population is increasing, learning applications and typing pose challenges for the elderly. Overall, there is a shortage of research evidence regarding these artificial intelligence-based products in the self-management of hypertension. A review highlighted that nearly half of the research has been conducted in the United States, along with participation from the United Kingdom, Japan, Australia, Sweden, and other countries. However, there is a notable absence of research on chatbots in China[ 16 ]. The majority of studies focus on users' mental health[ 17 ]or aim to support individuals in maintaining mental well-being[ 18 ]. A smaller number of studies involve lifestyle interventions, such as demonstrating instructional brushing videos[ 19 ], providing breastfeeding support for women's health, or aiding in dietary improvements[ 20 , 21 ]. Additionally, a few of studies have indicated the potential of chatbots in chronic disease management, including applications in pain and diabetes management[ 22 , 23 ]. Technologically advanced products have progressively taken center stage. Consequently, while evaluating the effectiveness of intervention, the acceptance and use of the tool by patients have become fundamental for implementation. Ever since Davis introduced the technology acceptance model, numerous theoretical models concerning technology acceptance have surfaced. These include the technology acceptance model 2[ 24 ]and technology acceptance model 3[ 25 ]. Furthermore, theories such as rational behavior in social psychology[ 26 ], the theory of planned behavior and social cognition and the theory of innovation diffusion in communication have been incorporated into the realm of management information systems to elucidate users' acceptance of technology. In 2003, Venkatesh et al. proposed a united technology acceptance and use theory (UTAUT) after integrating rational behavior theory, planned behavior theory, technology acceptance model, social cognition theory, innovation diffusion theory, complex planned behavior theory, technology acceptance model and motivation model[ 27 ]. The UTAUT is extensively employed in studies on the acceptance of emerging technologies, with a remarkable explanatory power of individuals' adoption of new technologies reaching as high as 69%[ 28 ], surpassing the performance of any preceding models. UTAUT has emerged as one of the most widely employed models in the realm of information technology acceptance in the contemporary era[ 29 ]. It has been applied to validate the adoption intentions of diverse technologies across various environments [ 30 ], encompassing areas such as e-government[ 31 ], mobile banking[ 32 ], mobile payment[ 33 ], mobile commerce[ 34 ], and more. In the domain of health management, the UTAUT has also been found extensive application in hospital electronic information management systems[ 35 ], telemedicine[ 36 ], mobile health[ 37 , 38 ], and so on. However, there is a paucity of studies examining the influence of usage behavior on the intervention effectiveness in chronic disease management. Whether it involves mobile health, digital health, or artificial intelligence-based interventions, patients' acceptability should be a pivotal consideration in assessing the intervention's effectiveness. In the model, intention refers to the user's willingness to adopt a new technology. It is the motivating factor that influences the behavior of an individual. Behavior refers to the user's actual operation behavior of a new technology. As shown in Fig. 1 , intention and behavior are influenced by four key determinants: performance expectations, effort expectations, social influences, and facilitators. Performance expectation refers to the degree to which an individual feels the adoption of innovation is helpful, and effort expectation is derived from variables such as perceived ease of use and complexity, indicating how much effort an individual needs to pay to adopt innovation. Social influence is the degree to which an individual feels influenced by the group around him or her. Facilitators are distinguished from effort expectations by the degree to which users believe that organizational and technical resources exist to support the adoption of innovation. Effort expectation, performance expectation, social influence and promotion factors are the core factors that affect use intention and use behavior, and gender, age, experience and voluntariness are the four moderating factors[ 39 ]. In summary, the implementation of an intervention for hypertensive patients based on artificial intelligence speakers holds the promise of addressing prevalent challenges in hypertension management in China. However, there has been insufficient research on how people's acceptance of new technologies is influenced, which factors affect this acceptance, and how the degree of acceptance impacts the effectiveness of controlling blood pressure. The objectives of this study are: Examine the acceptability of the chatbot speaker among individuals with hypertension and explore the key factors using the UTAUT model. Investigate different patterns of using the chatbot speaker and figure out users’ characteristics and preferences. Determine the efficacy of the chatbot speaker in improving control of blood pressure and self-management. Methods Study design This study will adopt a mixed method with quantitative survey and qualitative interview. The integration will provide a holistic view to understand the effectiveness and acceptability of the intervention. The quantitative study will be conducted as a quasi-randomized controlled trial in community in Beijing to compare the effectiveness of the intervention on blood pressure control and self-management behaviors (such as medication use, salt reduction, regular exercise and so on). Questionnaires, physical examinations, and personal in-depth interviews will be conducted at baseline, 1, 3, 6 and 12 months to examine changes in blood pressure control, self-management behaviors and acceptability for the intervention. Study setting and randomization The study will be conducted in community health center in Shunyi and Daxing district in Beijing. Patients will be assigned into an intervention group and a control group depending on their willingness. Recruitment ends when the required number of patients is reached. Considering the nature of the intervention, blinding of the participants and researchers cannot be guaranteed. However, they will be blinded to the study aims and hypothesis. The flow chart of patient recruitment and study implementation is shown in Fig. 2 . Population Population in the study are residents aged 18–75 years with hypertension in community health centers. Inclusion criteria are as follows: diagnosed with essential hypertension based on ICD-9 codes 401-401.9; no plans to move for 2 years; can take care of themselves. Exclusion criteria are: having chemotherapy or radiotherapy within the last six months; having mental disorders; refusing to sign informed consent; having disorders that affects normal listening and speaking function. Effect size According to a previous study, the medication adherence of patients with hypertension aged 18 years and above in Shunyi District was about 44%. Assuming an alpha risk of 0.05, a beta risk of 0.10, an increase to 60% of medication adherence and a drop-out rate of 20% will require a total sample of 240 patients in each group. Intervention Control group: Patients in the control group will receive standard health care from community health workers, including regular assessment and follow-up, medication counselling and health education. Intervention group: Patients in the intervention group will receive standard health care plus a chatbot speaker named “hypertensive assistant”. Reminding: chatbot speaker supports the purpose of reminding by setting to-do items or alarm clock. Patients can use this function to remind them to take medication, measure blood pressure or any other they want to set. Take the reminder “take medication” as an example, patients can talk to chatbot “remind me to take medication at 8.am tomorrow morning.” Then, the chatbot speaker will broadcast “you have a reminder: you need to take medicine” until the patients give a response. Counselling: patients are encouraged to interact with the chatbot speaker to obtain accurate and instant knowledge about hypertension (symptoms, risk factors, complications, damage to target organs, related examinations), salt-reduction (purpose, mechanism of salt-reduction, the implementation skills, and how to distinguish foods with high sodium), diet (different grades of food recommended, dietary principles of Desh diet), exercise ( exercise precautions, recommended intensity, time, common exercise (a total of 63) calories consumed and intensity), and blood pressure measurement (precautions, procedure and recommended frequency of blood pressure measurement). A hypertension self-management knowledge graph and a FAQ (Frequently Asked Questions) library support the counselling. Recording: patients are also encouraged to talk with the chatbot speaker to record their blood pressure, their exercise (duration and type) and their dietary. Patients can enter the set conversation flow through the wake word. For example, when the patients say “record blood pressure”, the chatbot speaker will start the record flow asking the relevant questions in turn (time, place and the value), giving immediate feedback comparing with the normal value and last value and finally saving the value to cloud database. Broadcasting: chatbot speaker can go into broadcast mode. Patients can choose the topic from “disease knowledge”, “medication”,“lifestyle” and “blood pressure measurement”. Once they enter this mode and decide the topic, chatbot will broadcast relevant knowledge automatically. Entertainment: chatbot speaker can tell jokes, play music, play novels and play TV series. Patients can call these functions whenever they want. Outcomes Primary outcome: Changes of blood pressure from baseline to 1, 3, 6, and 12-month follow-up. Secondary outcome: Acceptability of the chatbot speaker: patients’ different modes of speaker acceptance and use, what are their characteristics, and which function do the patients prefer. Influencing factors of acceptability: performance expectancy, effort expectancy, social influence, facilitating conditions and personalities. Modifications of self-management behaviors: medication adherence, salt-reduction, dietary, physical activity, blood pressure measurement. Variables measurement Sociodemographic variables will be obtained at baseline. The questionnaire about acceptability of the chatbot speaker will be obtained at 1-month. Physical examinations and other questionnaire surveys will be completed at baseline, 3, 6 and 12-month. All the questionnaire surveys will be completed face-to-face by the trained investigators at the health care center. Interactive data will always be record into the cloud data immediately. The specific schedule is shown in Table 1 . Sociodemographic variables: gender, age, educational level, marital status, income level and occupation will be collected, as well as detailed medical history such as the duration of hypertension, the complications and current hypertensive medications. Blood pressure measurement: the medical staff of the community health care center will be responsible for the blood pressure measurement with the Omron J710 automatic blood pressure meter. Patients will be asked to rest for at least 10 minutes when they arrive at health care center. Th same arm will be measured continuously for 3 times, each interval of 2 min, and the average value of the 3 measurements will be taken as the final blood pressure value. Anthropometric variables: height, weight and waist circumference will be measured twice by certified instruments. Interactive variables: interactions with the chatbot speaker will be recorded into the cloud data instantly. The time, content and key values of the conversation will be recorded. Questionnaires (1) Salt threshold: It consists of three questions: the number when the salt water just was tasted, the number when the salt water just was tasted salty, and the number which was similar to the usual food. Different concentrations of salt water with labels from 1 to 9 are prepared. The patients will be asked to taste the salt water in turn until they completed the three questions. The concentrations are 0.05%, 0.075%, 0.15%, 0.2%, 0.25%, 0.3%, 0.4%, 0.5% and 0.6%[ 40 ]. (2) UTAUT scale: the acceptability of chatbot speaker will be measured by self-designed UTAUT scales referring to previous literatures, including effort expectation (5 items), performance expectation (5 items), social influence (5 items), facilitating conditions (3 items), enjoyment (3 items), habits (3 items) and willingness (5 items). (3) The big-five personality scale: The scale was compiled by Wang et al[ 41 ]. The personality will be measured from neuroticism, openness, agreeableness, conscientiousness and extraversion with 8 items each. The scale has been widely used in studies with good reliability and validity. (4) Medication adherence: The therapeutic adherence scale for hypertensive patients will be used to evaluate the medication adherence with compliance behavior (5 items) and adverse medication-taking behavior (8 items)[ 42 ]. (5) Dietary intake: dietary intake will be assessed by a self-modified simplified version of food frequency questionnaires with reference to the Alternate Healthy Eating Index (AHEI) score[ 43 ]. The questionnaire will assess the intake of eight kinds of foods with the frequency and amount of consumption in the past month, including red meat and processed meat, white meat, fruits, vegetables, nuts and beans, low-fat dairy products, coarse grains, and sugary beverages. (6) Physical activity: A self-modified questionnaire was used to evaluate physical activity weekly in the past month[ 44 ]. The questionnaire includes information on 10 moderate and vigorous activities. Table 1 The schedule of enrolment, interventions, and assessments Study period Recruitment Allocation Intervention and follow-up TIMEPOINT -t1 0 1 3 6 12 ENEOLMENT Eligibility screen √ Informed consent √ Allocation √ INTERVENTIONS Chatbot speaker Control ASSESSMENTS Socio-demographic variables √ Blood pressure √ √ √ √ √ Anthropometric variables √ √ √ √ √ Salt threshold √ √ √ √ √ UTAUT scale √* The big-five personality scale √ Medication adherence √ √* √ √ √ Dietary intake √ √* √ √ √ Physical activity √ √* √ √ √ Cloud data √* √* √* √* √* Personal in-depth interview √* √* √* √* Note: * represents only conducted in intervention group Qualitative analysis Qualitative data will be collected through semi-structured interviews to obtain the better understanding of the acceptability, usability, facilitating factors, hindering factors and actual transformation of the self-management behaviors with the chatbot speaker. We will invite patients to a 30-minutes telephone-based or face-to-face personal in-depth interview considering both their sociodemographic variables (gender, age, educational level and duration of hypertension) and the using frequency of chatbot speaker. Patients will be encouraged to talk about the topics around their feelings, informative functions, recommendations for the optimization of the chatbot speaker as well as the facilitating and hindering factors of improving self-management behaviors. The interviews will be conducted at 1, 3, 6-month and the participants will be recruited until the information is saturated each time. Interviews will be transcribed and analyzed by Nvivo (version 12, QRS International, Doncaster, Australia) using the thematic framework. Both quantitative and qualitative data will be integrated to better explain the acceptability and effectiveness of the chatbot speaker for hypertensive patients. Statistical analysis Continuous variables, such as age, blood pressure and scores of scales obeying normal distribution, will be described as mean ± standard deviation. Categorical variables will be reported using n (percentage of sample). Continuous variables will be compared using Student’s t test, one-way ANOVA or Kruskal-Walls test according to the conditions. Chi-square test or Mann-Whitney U test will be selected to compare categorical variables. In addition, multiple linear regression, logistic regression and structural equation model will be conducted to explore the influencing factors of acceptability. Latent profile/class analysis will be conducted to explore different behavior patterns of using chatbot speaker. For longitudinal data, generalized estimation equation or generalized linear mixed model (GLMM) will be used to compare the changes between the two groups and dynamic structural equation model will be used to explore the changes of each influencing factor over time. An alpha risk of 0.05 will be set for two-sided tests. IBM SPSS Statistics 22.0, Mplus 7.4, R studio 4.1.2 will be used for statistical analysis. Study management Prior to the study, all the investigator from different community health care will be trained comprehensively, clarifying the recruitment standards and precautions in the investigation process to ensure consistency and quality among all communities. During data collection, a double-entry checking procedure will be implemented to enhance the quality of data entry. Experts will be consulted for appropriate statistical methods. If finding abnormal data, the original questionnaires or record will be checked for verification before proceeding to the next step of analysis. Furthermore, the qualitative data will be transcribed by two researchers to ensure reliability. Discussion Continuously exposed shortcomings in traditional intervention methods that fail to meet evolving demands, theories and applications on artificial intelligence are becoming increasingly mature, gradually permeating various industries and playing crucial and efficient roles. The chatbot speaker in this study based on a knowledge graph can facilitate artificial intelligence intervention and realize the vision of intervention management for hypertensive patients in the current scenario. Firstly, in the construction of the knowledge graph, it is crucial to rigorously select knowledge sources and undergo expert evaluations until evidence-based requirements are met. Secondly, this chatbot speaker can accommodate a vast amount of hypertension management knowledge, providing real-time clarification for patients, offering comprehensive guidance for self-management, and effectively liberating human resources. Thirdly, patients can access relevant knowledge and adjust their self-management behaviors in a timely manner according to their individual needs, achieving a certain degree of personalization and enhancing self-involvement, thus potentially improving compliance. Lastly, this intervention can accompany patients for the long term and continuously improve through technological means, achieving iterative updates. In summary, artificial intelligence interventions for hypertensive patients is feasible and promising. And the acceptability is the most crucial part when it works. This study focuses on whether new technologies or products can be accepted by patients and how the chatbot speaker play a role in blood pressure control. The findings will provide meaningful insights into refining the artificial intelligence-based chatbot for chronic diseases management. Declarations Ethics approval and consent to participate This study was approved by Biomedical Committee of Peking university (IRB00001052-21106) and this trial was registered in Chinese Clinical Trial Registry (registration number: ChiCTR2100050578). Study participation is voluntary. Participations who provided their written informed consent will be included in the study. All information of the study subjects and the obtained data will be kept strictly confidential. Consent withdrawal is possible at any time without cause. Consent for publication Not applicable. Availability of data and materials Anonymized survey data will be available from corresponding author on reasonable request. Competing interests The authors declare that they have no competing interests. Funding This study was supported by National Key Research and Development Program of China(2022YFC3600900) and Beijing Municipal Natural Science Foundation (7202087). Author’s contributions XYS and PC designed this study. YL and XXZ provided the theoretical and methodological guidance. XLF and XYS are responsible for the study management. XYS is responsible for the recruitment. PC is responsible for the study implementation. All authors provided revisions and approved the final manuscript. Acknowledgements We are grateful to the professionals who provided great support for this study in participant recruitment and data collection, and the patients who spent time on this study. References Zhou B, Carrillo-Larco RM, Danaei G, Riley LM, Paciorek CJ, Stevens GA, Gregg EW, Bennett JE, Solomon B, Singleton RK, et al. Worldwide trends in hypertension prevalence and progress in treatment and control from 1990 to 2019: a pooled analysis of 1201 population-representative studies with 104 million participants. Lancet. 2021;398(10304):957–80. Thomopoulos C, Parati G, Zanchetti A. Effects of blood pressure lowering on outcome incidence in hypertension. 1. Overview, meta-analyses, and meta-regression analyses of randomized trials. J Hypertens. 2014;32(12):2285–95. Mackenzie IS, MacDonald TM. Identifying poor adherence to antihypertensive medications in patients with resistant hypertension. Br J Clin Pharmacol. 2019;85(1):5–7. Carey RM, Wright JT, Taler SJ, Whelton PK. Guideline-Driven Management of Hypertension: An Evidence-Based Update. Circ Res. 2021;128(7):827–46. Pazoki R, Dehghan A, Evangelou E, Warren H, Gao H, Caulfield M, Elliott P, Tzoulaki I. Genetic Predisposition to High Blood Pressure and Lifestyle Factors: Associations With Midlife Blood Pressure Levels and Cardiovascular Events. Circulation. 2018;137(7):653–61. Pharmacological blood pressure lowering for. primary and secondary prevention of cardiovascular disease across different levels of blood pressure: an individual participant-level data meta-analysis. Lancet. 2021;397(10285):1625–36. Bakris G, Ali W, Parati G. ACC/AHA Versus ESC/ESH on Hypertension Guidelines: JACC Guideline Comparison. J Am Coll Cardiol. 2019;73(23):3018–26. Huang L, Trieu K, Yoshimura S, Neal B, Woodward M, Campbell NRC, Li Q, Lackland DT, Leung AA, Anderson CAM, et al. Effect of dose and duration of reduction in dietary sodium on blood pressure levels: systematic review and meta-analysis of randomised trials. BMJ. 2020;368:m315. Siegel D, Lopez J, Meier J. Antihypertensive medication adherence in the Department of Veterans Affairs. Am J Med. 2007;120(1):26–32. Choudhry NK, Kronish IM, Vongpatanasin W, Ferdinand KC, Pavlik VN, Egan BM, Schoenthaler A, Houston Miller N, Hyman DJ. Medication Adherence and Blood Pressure Control: A Scientific Statement From the American Heart Association. Hypertension 2022, 79(1). Nieuwlaat R, Wilczynski N, Navarro T, Hobson N, Jeffery R, Keepanasseril A, Agoritsas T, Mistry N, Iorio A, Jack S, et al. Interventions for enhancing medication adherence. Cochrane Database Syst Rev. 2014;2014(11):CD000011. Bundy JD, Li C, Stuchlik P, Bu X, Kelly TN, Mills KT, He H, Chen J, Whelton PK, He J. Systolic Blood Pressure Reduction and Risk of Cardiovascular Disease and Mortality: A Systematic Review and Network Meta-analysis. JAMA Cardiol. 2017;2(7):775–81. Arteriosclerosis T. Vascular Biology Arterioscler Thromb Vasc Biol. 2015;35(1):1. Ruilope LM, Valenzuela PL, Lucia A. Digital therapeutics and lifestyle: the start of a new era in the management of arterial hypertension? Eur Heart J. 2021;42(40):4123–5. Schubel LC, Wesley DB, Booker E, Lock J, Ratwani RM. Population subgroup differences in the use of a COVID-19 chatbot. NPJ Digit Med. 2021;4(1):30. Martinengo L, Jabir AI, Goh WWT, Lo NYW, Ho M-HR, Kowatsch T, Atun R, Michie S, Tudor Car L. Conversational Agents in Health Care: Scoping Review of Their Behavior Change Techniques and Underpinning Theory. J Med Internet Res. 2022;24(10):e39243. Schick A, Feine J, Morana S, Maedche A, Reininghaus U. Validity of Chatbot Use for Mental Health Assessment: Experimental Study. JMIR Mhealth Uhealth. 2022;10(10):e28082. Jiang Q, Zhang Y, Pian W. Chatbot as an emergency exist: Mediated empathy for resilience via human-AI interaction during the COVID-19 pandemic. Inf Process Manag. 2022;59(6):103074. Pithpornchaiyakul S, Naorungroj S, Pupong K, Hunsrisakhun J. Using a Chatbot as an Alternative Approach for In-Person Toothbrushing Training During the COVID-19 Pandemic: Comparative Study. J Med Internet Res. 2022;24(10):e39218. Friederichs S, Bolman C, Oenema A, Guyaux J, Lechner L. Motivational interviewing in a Web-based physical activity intervention with an avatar: randomized controlled trial. J Med Internet Res. 2014;16(2):e48. Jack B, Bickmore T, Hempstead M, Yinusa-Nyahkoon L, Sadikova E, Mitchell S, Gardiner P, Adigun F, Penti B, Schulman D, et al. Reducing Preconception Risks Among African American Women with Conversational Agent Technology. J Am Board Fam Med. 2015;28(4):441–51. Hunt M, Miguez S, Dukas B, Onwude O, White S. Efficacy of Zemedy, a Mobile Digital Therapeutic for the Self-management of Irritable Bowel Syndrome: Crossover Randomized Controlled Trial. JMIR Mhealth Uhealth. 2021;9(5):e26152. Watson A, Bickmore T, Cange A, Kulshreshtha A, Kvedar J. An internet-based virtual coach to promote physical activity adherence in overweight adults: randomized controlled trial. J Med Internet Res. 2012;14(1):e1. Venkatesh V, Davis FD. A Theoretical Extension of the Technology Acceptance Model: Four Longitudinal Field Studies: A Theoretical Extension of the Technology Acceptance Model. Four Longitudinal Field Studies; 2000. Venkatesh V, Bala HJDS. Technology Acceptance Model 3 and a Research Agenda on Interventions. 2010, 39(2):273–315. Fishbein M, Ajzen IJP, Rhetoric. Belief, Attitude, Intention and Behaviour: An Introduction to Theory and Research. Addison-Wesley, Reading MA. 1977, 41(4):842–844. Maillet É, Mathieu L, Sicotte C. UTAUT Int J Med Inf. 2015;84(1):36–47. Cao LH. Study on users’ adoption of mobile social commerce based on UTAUT2. Dalian University of Technology; 2018. Dwivedi YK, Rana NP, Janssen M, Lal B, Williams MD, Clement M. An empirical validation of a unified model of electronic government adoption (UMEGA). Government Inform Q. 2017;34(2):211–30. Williams MD, Rana NP, Dwivedi YK. The unified theory of acceptance and use of technology (UTAUT): a literature review. J Enterp Inform Manage. 2015;28(3):443–88. Rana NP, Dwivedi YK, Lal B, Williams MD, Clement M. Citizens’ adoption of an electronic government system: towards a unified view. Inform Syst Front. 2015;19(3):549–68. Baabdullah AM, Alalwan AA, Rana NP, Kizgin H, Patil PPJIJIM. Consumer use of mobile banking (M-Banking) in Saudi Arabia: Towards an integrated model. 2019, 44:38–52. Dwivedi YKJTMR. Devising a research model to examine adoption of mobile payments: An extension of UTAUT2. 2014, 14:310–35. Shaw N, Sergueeva K. The non-monetary benefits of mobile commerce: Extending UTAUT2 with perceived value. Int J Inf Manag. 2019;45:44–55. Zhou LL, Owusu-Marfo J, Asante Antwi H, Antwi MO, Kachie ADT, Ampon-Wireko S. Assessment of the social influence and facilitating conditions that support nurses' adoption of hospital electronic information management systems (HEIMS) in Ghana using the unified theory of acceptance and use of technology (UTAUT) model. BMC Med Inf Decis Mak. 2019;19(1):230. Cimperman M, Makovec Brenčič M, Trkman P. Analyzing older users' home telehealth services acceptance behavior-applying an Extended UTAUT model. Int J Med Inf. 2016;90:22–31. Hoque R, Sorwar G. Understanding factors influencing the adoption of mHealth by the elderly: An extension of the UTAUT model. Int J Med Inf. 2017;101:75–84. Dwivedi YK, Shareef MA, Simintiras AC, Lal B, Weerakkody V. A generalised adoption model for services: A cross-country comparison of mobile health (m-health). Government Inform Q. 2016;33(1):174–87. Song T. A research on user behavior on mobile shoppoing app based on UTAUT model. Shandong Univ. 2017. Liu T. Analysis of salt taste/salt intake and the effect of capsaicin on salt taste sensitivity of the elderly in northeast China: a clinical study. China medical university; 2019. Wang MC, Dai XY, Yao SQ. Development of the Chinese big five personality inventory(CBF-PI)ⅲ: psychometric properties of CBF-PI brief version. Tang HY, Zhu JC, He HY, Qian CR, Yang YN. Development and evaluation of a new therapeutic adherence scale for hypertensive patients. J Third Military Med Univ. 2011;33(13):1400–3. Sotos-Prieto M, Bhupathiraju SN, Mattei J, Fung TT, Li Y, Pan A, Willett WC, Rimm EB, Hu FB. Changes in Diet Quality Scores and Risk of Cardiovascular Disease Among US Men and Women. Circulation. 2015;132(23):2212–9. Shan Z, Li Y, Zong G, Guo Y, Li J, Manson JE, Hu FB, Willett WC, Schernhammer ES, Bhupathiraju SN. Rotating night shift work and adherence to unhealthy lifestyle in predicting risk of type 2 diabetes: results from two large US cohorts of female nurses. BMJ. 2018;21:363k4641. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 21 Aug, 2024 Read the published version in BMC Public Health → Version 1 posted Editorial decision: Revision requested 26 Mar, 2024 Editor assigned by journal 01 Mar, 2024 Editor invited by journal 01 Feb, 2024 Submission checks completed at journal 01 Feb, 2024 First submitted to journal 27 Jan, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3903126","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Study protocol","associatedPublications":[],"authors":[{"id":270387621,"identity":"e2a0f7cc-742d-4e7f-a475-6abb7138b8f8","order_by":0,"name":"Ping chen","email":"","orcid":"","institution":"Department of Social Medicine and Health Education, School of Public Health, Peking University","correspondingAuthor":false,"prefix":"","firstName":"Ping","middleName":"","lastName":"chen","suffix":""},{"id":270387622,"identity":"1ad95c8c-f692-4c10-b41a-1c4575aede1b","order_by":1,"name":"Yi Li","email":"","orcid":"","institution":"Center of medical informatics, Peking University","correspondingAuthor":false,"prefix":"","firstName":"Yi","middleName":"","lastName":"Li","suffix":""},{"id":270387623,"identity":"8c3aa0ce-15a7-4b10-b2dd-64de6d7c1545","order_by":2,"name":"Xuxi Zhang","email":"","orcid":"","institution":"Department of Social Medicine and Health Education, School of Public Health, Peking University","correspondingAuthor":false,"prefix":"","firstName":"Xuxi","middleName":"","lastName":"Zhang","suffix":""},{"id":270387624,"identity":"8a4b0f96-2c35-465e-9f14-8a3ed56aea45","order_by":3,"name":"Xinglin Feng","email":"","orcid":"","institution":"Department of Health Policy and Management, School of Public Health, Peking University","correspondingAuthor":false,"prefix":"","firstName":"Xinglin","middleName":"","lastName":"Feng","suffix":""},{"id":270387625,"identity":"ad158dbf-72b7-4d7b-b76e-84ceb6ddad96","order_by":4,"name":"Xinying Sun","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAxUlEQVRIiWNgGAWjYNCCigPMYJqHeC1nSNbC2HaAgXgtBjfSn0kXzrvDrjsjgfHB2zYGeXNCWiRn5JhJz9z2jNnsRgKz4dw2BsOdDQS08EvksEnzbjsM0gJktDEkGBwgoIVNAugw3jlgLey/idLCL5FgJs3bALGFmSgtkj1vjK15jgH9cuZhs+SccxKGGwhpMTie/vA2T82dZLPjyQc/vCmzkSdoC4NAAphKBsZOA5CWIKQeCPghhtoRoXQUjIJRMApGKgAAJ7Y9LB118GgAAAAASUVORK5CYII=","orcid":"","institution":"Department of Social Medicine and Health Education, School of Public Health, Peking University","correspondingAuthor":true,"prefix":"","firstName":"Xinying","middleName":"","lastName":"Sun","suffix":""}],"badges":[],"createdAt":"2024-01-27 13:44:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3903126/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3903126/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12889-024-19667-4","type":"published","date":"2024-08-21T15:57:10+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":50569835,"identity":"b63d958a-18e1-4430-8e9c-7b5cb5292b09","added_by":"auto","created_at":"2024-02-02 15:35:37","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":56996,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe UTAUT model\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3903126/v1/bacbe11a69298b748d94e7aa.jpg"},{"id":50568945,"identity":"1adc9e6e-da61-4712-a877-0ac277e42316","added_by":"auto","created_at":"2024-02-02 15:27:37","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":152370,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFlow chart of patient recruitment and study implementation\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3903126/v1/c113b74b9a9430c3b9ff498b.jpg"},{"id":63300099,"identity":"8e314eac-1c5b-4843-852e-770cff7fd6c8","added_by":"auto","created_at":"2024-08-26 16:11:05","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":724212,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3903126/v1/35905c02-43ab-46e4-8ae9-205ee8265c91.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"The acceptability and effectiveness of artificial intelligence-based chatbot for hypertensive patients in community: protocol for a mixed-methods study","fulltext":[{"header":"Background","content":"\u003cp\u003eGlobal Burden of disease indicates that the prevalence of hypertension has doubled in 2019 compared to 1990, with nearly 1.3\u0026nbsp;billion people aged 30\u0026ndash;79 living with hypertension worldwide [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e], and more than 10\u0026nbsp;million patients dying from high systolic blood pressure, which accounts for up to 20% of all-cause deaths. In China, the prevalence of hypertension in adults has reached 27.5%[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e], but the control rate is much lower, only 7.2% in patients aged 35 to 75 in China[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. In addition, this trend will continue to intensify with the aging of the population and the complications will also lead to a growing medical burden.\u003c/p\u003e \u003cp\u003eA good lifestyle is the cornerstone of hypertension treatment[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. A British study pointed out that among all genetic risk groups for hypertension, the systolic blood pressure of those with a good lifestyle was significantly lower than that of those with a bad lifestyle [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. For every 5mmHg reduction in systolic blood pressure, the risk of cardiovascular events was reduced by 10%[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. The guidelines emphasized low-sodium diet, regular exercise, weight loss, limited alcohol consumption and adequate potassium intake[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Many studies have confirmed that changing exposure to these factors will effectively improve blood pressure control. For example, reducing sodium intake can reduce systolic blood pressure by 7.7 mmHg and diastolic blood pressure by 3.0 mmHg[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], and it has been confirmed that it can effectively reduce the risk of cardiovascular events [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e], which has a dose-response relationship with blood pressure reduction[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. It also significantly reduces the risk of stroke, heart disease, and all-cause death[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eDesh diet is the most effective dietary pattern for hypertension management, and there is evidence that the intervention of Desh diet has a lasting effect [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. There are studies that have proved the Chinese style-Desh diet can achieve good control of blood pressure in Chinese people[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. In addition, moderate-intensity exercise for more than 30 minutes at least 3 times per week in hypertensive patients can effectively lower blood pressure and reduce the risk of cardiovascular death and all-cause death [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Therefore, improved lifestyle is effective in controlling blood pressure and reducing the risk of death. However, the content of hypertension self-management is abundant and the details are complex, and a systematic intervention framework has not yet been formed. Traditional methods, such as health education lecture in person and distributing promotional materials, are unsustainable, limited and difficult to popularize[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. With the rapid development of artificial intelligence, the new approach based on digital technology have emerged in recent years, which is expected to solve these traditional problems.\u003c/p\u003e \u003cp\u003eChatbots have been found extensive applications amid the COVID-19 pandemic for enhancing awareness, owing to their cost-effectiveness, capacity to reach a broad audience, and compatibility with multiple devices[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. And nowadays, chatbots are already widely used in areas such as weight-loss related behaviors (dietary intake and physical activity) and psycho-emotional support, but there are limited studies on hypertension. A study showed that chatbot facilitating home blood pressure monitoring did not increase adherence to monitoring blood pressure compared to traditional methods, but it did improve knowledge and skills of patients[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e] Moreover, the majority of existing chatbots primarily rely on text, with only a limited number being voice-based. In China, where the aging population is increasing, learning applications and typing pose challenges for the elderly. Overall, there is a shortage of research evidence regarding these artificial intelligence-based products in the self-management of hypertension.\u003c/p\u003e \u003cp\u003eA review highlighted that nearly half of the research has been conducted in the United States, along with participation from the United Kingdom, Japan, Australia, Sweden, and other countries. However, there is a notable absence of research on chatbots in China[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. The majority of studies focus on users' mental health[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]or aim to support individuals in maintaining mental well-being[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. A smaller number of studies involve lifestyle interventions, such as demonstrating instructional brushing videos[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e], providing breastfeeding support for women's health, or aiding in dietary improvements[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Additionally, a few of studies have indicated the potential of chatbots in chronic disease management, including applications in pain and diabetes management[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eTechnologically advanced products have progressively taken center stage. Consequently, while evaluating the effectiveness of intervention, the acceptance and use of the tool by patients have become fundamental for implementation. Ever since Davis introduced the technology acceptance model, numerous theoretical models concerning technology acceptance have surfaced. These include the technology acceptance model 2[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]and technology acceptance model 3[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. Furthermore, theories such as rational behavior in social psychology[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e], the theory of planned behavior and social cognition and the theory of innovation diffusion in communication have been incorporated into the realm of management information systems to elucidate users' acceptance of technology. In 2003, Venkatesh et al. proposed a united technology acceptance and use theory (UTAUT) after integrating rational behavior theory, planned behavior theory, technology acceptance model, social cognition theory, innovation diffusion theory, complex planned behavior theory, technology acceptance model and motivation model[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. The UTAUT is extensively employed in studies on the acceptance of emerging technologies, with a remarkable explanatory power of individuals' adoption of new technologies reaching as high as 69%[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e], surpassing the performance of any preceding models.\u003c/p\u003e \u003cp\u003eUTAUT has emerged as one of the most widely employed models in the realm of information technology acceptance in the contemporary era[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. It has been applied to validate the adoption intentions of diverse technologies across various environments [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e], encompassing areas such as e-government[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e], mobile banking[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e], mobile payment[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e], mobile commerce[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e], and more. In the domain of health management, the UTAUT has also been found extensive application in hospital electronic information management systems[\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e], telemedicine[\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e], mobile health[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e], and so on. However, there is a paucity of studies examining the influence of usage behavior on the intervention effectiveness in chronic disease management. Whether it involves mobile health, digital health, or artificial intelligence-based interventions, patients' acceptability should be a pivotal consideration in assessing the intervention's effectiveness.\u003c/p\u003e \u003cp\u003eIn the model, intention refers to the user's willingness to adopt a new technology. It is the motivating factor that influences the behavior of an individual. Behavior refers to the user's actual operation behavior of a new technology. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, intention and behavior are influenced by four key determinants: performance expectations, effort expectations, social influences, and facilitators. Performance expectation refers to the degree to which an individual feels the adoption of innovation is helpful, and effort expectation is derived from variables such as perceived ease of use and complexity, indicating how much effort an individual needs to pay to adopt innovation. Social influence is the degree to which an individual feels influenced by the group around him or her. Facilitators are distinguished from effort expectations by the degree to which users believe that organizational and technical resources exist to support the adoption of innovation. Effort expectation, performance expectation, social influence and promotion factors are the core factors that affect use intention and use behavior, and gender, age, experience and voluntariness are the four moderating factors[\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn summary, the implementation of an intervention for hypertensive patients based on artificial intelligence speakers holds the promise of addressing prevalent challenges in hypertension management in China. However, there has been insufficient research on how people's acceptance of new technologies is influenced, which factors affect this acceptance, and how the degree of acceptance impacts the effectiveness of controlling blood pressure. The objectives of this study are:\u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eExamine the acceptability of the chatbot speaker among individuals with hypertension and explore the key factors using the UTAUT model.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eInvestigate different patterns of using the chatbot speaker and figure out users\u0026rsquo; characteristics and preferences.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eDetermine the efficacy of the chatbot speaker in improving control of blood pressure and self-management.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy design\u003c/h2\u003e \u003cp\u003eThis study will adopt a mixed method with quantitative survey and qualitative interview. The integration will provide a holistic view to understand the effectiveness and acceptability of the intervention. The quantitative study will be conducted as a quasi-randomized controlled trial in community in Beijing to compare the effectiveness of the intervention on blood pressure control and self-management behaviors (such as medication use, salt reduction, regular exercise and so on). Questionnaires, physical examinations, and personal in-depth interviews will be conducted at baseline, 1, 3, 6 and 12 months to examine changes in blood pressure control, self-management behaviors and acceptability for the intervention.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eStudy setting and randomization\u003c/h2\u003e \u003cp\u003eThe study will be conducted in community health center in Shunyi and Daxing district in Beijing. Patients will be assigned into an intervention group and a control group depending on their willingness. Recruitment ends when the required number of patients is reached. Considering the nature of the intervention, blinding of the participants and researchers cannot be guaranteed. However, they will be blinded to the study aims and hypothesis. The flow chart of patient recruitment and study implementation is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003ePopulation\u003c/h2\u003e \u003cp\u003ePopulation in the study are residents aged 18–75 years with hypertension in community health centers. Inclusion criteria are as follows: diagnosed with essential hypertension based on ICD-9 codes 401-401.9; no plans to move for 2 years; can take care of themselves. Exclusion criteria are: having chemotherapy or radiotherapy within the last six months; having mental disorders; refusing to sign informed consent; having disorders that affects normal listening and speaking function.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eEffect size\u003c/h2\u003e \u003cp\u003eAccording to a previous study, the medication adherence of patients with hypertension aged 18 years and above in Shunyi District was about 44%. Assuming an alpha risk of 0.05, a beta risk of 0.10, an increase to 60% of medication adherence and a drop-out rate of 20% will require a total sample of 240 patients in each group.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eIntervention\u003c/h2\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003eControl group:\u003c/h2\u003e \u003cp\u003ePatients in the control group will receive standard health care from community health workers, including regular assessment and follow-up, medication counselling and health education.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eIntervention group:\u003c/h2\u003e \u003cp\u003ePatients in the intervention group will receive standard health care plus a chatbot speaker named “hypertensive assistant”.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eReminding: chatbot speaker supports the purpose of reminding by setting to-do items or alarm clock. Patients can use this function to remind them to take medication, measure blood pressure or any other they want to set. Take the reminder “take medication” as an example, patients can talk to chatbot “remind me to take medication at 8.am tomorrow morning.” Then, the chatbot speaker will broadcast “you have a reminder: you need to take medicine” until the patients give a response.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eCounselling: patients are encouraged to interact with the chatbot speaker to obtain accurate and instant knowledge about hypertension (symptoms, risk factors, complications, damage to target organs, related examinations), salt-reduction (purpose, mechanism of salt-reduction, the implementation skills, and how to distinguish foods with high sodium), diet (different grades of food recommended, dietary principles of Desh diet), exercise ( exercise precautions, recommended intensity, time, common exercise (a total of 63) calories consumed and intensity), and blood pressure measurement (precautions, procedure and recommended frequency of blood pressure measurement). A hypertension self-management knowledge graph and a FAQ (Frequently Asked Questions) library support the counselling.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eRecording: patients are also encouraged to talk with the chatbot speaker to record their blood pressure, their exercise (duration and type) and their dietary. Patients can enter the set conversation flow through the wake word. For example, when the patients say “record blood pressure”, the chatbot speaker will start the record flow asking the relevant questions in turn (time, place and the value), giving immediate feedback comparing with the normal value and last value and finally saving the value to cloud database.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eBroadcasting: chatbot speaker can go into broadcast mode. Patients can choose the topic from “disease knowledge”, “medication”,“lifestyle” and “blood pressure measurement”. Once they enter this mode and decide the topic, chatbot will broadcast relevant knowledge automatically.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eEntertainment: chatbot speaker can tell jokes, play music, play novels and play TV series. Patients can call these functions whenever they want.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003cp\u003e\u003c/p\u003e \u003c/div\u003e "},{"header":"Outcomes","content":"\u003cp\u003ePrimary outcome: Changes of blood pressure from baseline to 1, 3, 6, and 12-month follow-up.\u003c/p\u003e\u003cp\u003eSecondary outcome:\u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eAcceptability of the chatbot speaker: patients’ different modes of speaker acceptance and use, what are their characteristics, and which function do the patients prefer.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eInfluencing factors of acceptability: performance expectancy, effort expectancy, social influence, facilitating conditions and personalities.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eModifications of self-management behaviors: medication adherence, salt-reduction, dietary, physical activity, blood pressure measurement.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e\u003cp\u003e\u003c/p\u003e\u003ch2\u003eVariables measurement\u003c/h2\u003e\u003cp\u003eSociodemographic variables will be obtained at baseline. The questionnaire about acceptability of the chatbot speaker will be obtained at 1-month. Physical examinations and other questionnaire surveys will be completed at baseline, 3, 6 and 12-month. All the questionnaire surveys will be completed face-to-face by the trained investigators at the health care center. Interactive data will always be record into the cloud data immediately. The specific schedule is shown in Table \u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003eSociodemographic variables: gender, age, educational level, marital status, income level and occupation will be collected, as well as detailed medical history such as the duration of hypertension, the complications and current hypertensive medications.\u003c/p\u003e\u003cp\u003eBlood pressure measurement: the medical staff of the community health care center will be responsible for the blood pressure measurement with the Omron J710 automatic blood pressure meter. Patients will be asked to rest for at least 10 minutes when they arrive at health care center. Th same arm will be measured continuously for 3 times, each interval of 2 min, and the average value of the 3 measurements will be taken as the final blood pressure value.\u003c/p\u003e\u003cp\u003eAnthropometric variables: height, weight and waist circumference will be measured twice by certified instruments.\u003c/p\u003e\u003cp\u003eInteractive variables: interactions with the chatbot speaker will be recorded into the cloud data instantly. The time, content and key values of the conversation will be recorded.\u003c/p\u003e\u003ch2\u003eQuestionnaires\u003c/h2\u003e\u003cp\u003e(1) Salt threshold: It consists of three questions: the number when the salt water just was tasted, the number when the salt water just was tasted salty, and the number which was similar to the usual food. Different concentrations of salt water with labels from 1 to 9 are prepared. The patients will be asked to taste the salt water in turn until they completed the three questions. The concentrations are 0.05%, 0.075%, 0.15%, 0.2%, 0.25%, 0.3%, 0.4%, 0.5% and 0.6%[\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e].\u003c/p\u003e\u003cp\u003e(2) UTAUT scale: the acceptability of chatbot speaker will be measured by self-designed UTAUT scales referring to previous literatures, including effort expectation (5 items), performance expectation (5 items), social influence (5 items), facilitating conditions (3 items), enjoyment (3 items), habits (3 items) and willingness (5 items).\u003c/p\u003e\u003cp\u003e(3) The big-five personality scale: The scale was compiled by Wang et al[\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. The personality will be measured from neuroticism, openness, agreeableness, conscientiousness and extraversion with 8 items each. The scale has been widely used in studies with good reliability and validity.\u003c/p\u003e\u003cp\u003e(4) Medication adherence: The therapeutic adherence scale for hypertensive patients will be used to evaluate the medication adherence with compliance behavior (5 items) and adverse medication-taking behavior (8 items)[\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e].\u003c/p\u003e\u003cp\u003e(5) Dietary intake: dietary intake will be assessed by a self-modified simplified version of food frequency questionnaires with reference to the Alternate Healthy Eating Index (AHEI) score[\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. The questionnaire will assess the intake of eight kinds of foods with the frequency and amount of consumption in the past month, including red meat and processed meat, white meat, fruits, vegetables, nuts and beans, low-fat dairy products, coarse grains, and sugary beverages.\u003c/p\u003e\u003cp\u003e(6) Physical activity: A self-modified questionnaire was used to evaluate physical activity weekly in the past month[\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. The questionnaire includes information on 10 moderate and vigorous activities.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eTable 1 The schedule of enrolment, interventions, and assessments\u003c/strong\u003e\u003c/p\u003e\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"557\"\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd width=\"40.57450628366248%\" valign=\"top\"\u003e\n \u003cp\u003eStudy period\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.26032315978456%\" valign=\"top\"\u003e\n \u003cp\u003eRecruitment\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.644524236983843%\" valign=\"top\"\u003e\n \u003cp\u003eAllocation\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"30.52064631956912%\" colspan=\"4\" valign=\"top\"\u003e\n \u003cp\u003eIntervention and follow-up\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eTIMEPOINT\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e-t1\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e12\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eENEOLMENT\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eEligibility screen\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eInformed consent\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eAllocation\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eINTERVENTIONS\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eChatbot speaker\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\" colspan=\"5\" style=\"width: 44.1652%;\"\u003e\u003cimg src=\"data:image/png;base64,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\" style=\"width: 272px;\"\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eControl\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\" colspan=\"5\" style=\"width: 44.1652%;\"\u003e\u003cimg src=\"data:image/png;base64,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\" style=\"width: 274px;\"\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eASSESSMENTS\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eSocio-demographic variables\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eBlood pressure\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eAnthropometric variables\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eSalt threshold\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eUTAUT scale\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√*\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eThe big-five personality scale\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eMedication adherence\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√*\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eDietary intake\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√*\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003ePhysical activity\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√*\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e√\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003eCloud data\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e√*\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√*\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√*\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√*\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e√*\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd width=\"40.50179211469534%\" valign=\"top\"\u003e\n \u003cp\u003ePersonal in-depth interview\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"15.232974910394265%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"13.620071684587813%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√*\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√*\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"6.810035842293907%\" valign=\"top\"\u003e\n \u003cp\u003e√*\u003c/p\u003e\n \u003c/td\u003e\u003ctd width=\"10.21505376344086%\" valign=\"top\"\u003e\n \u003cp\u003e√*\u003c/p\u003e\n \u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\u003cp\u003eNote: * represents only conducted in intervention group\u003c/p\u003e\u003ch2\u003eQualitative analysis\u003c/h2\u003e\u003cp\u003eQualitative data will be collected through semi-structured interviews to obtain the better understanding of the acceptability, usability, facilitating factors, hindering factors and actual transformation of the self-management behaviors with the chatbot speaker. We will invite patients to a 30-minutes telephone-based or face-to-face personal in-depth interview considering both their sociodemographic variables (gender, age, educational level and duration of hypertension) and the using frequency of chatbot speaker. Patients will be encouraged to talk about the topics around their feelings, informative functions, recommendations for the optimization of the chatbot speaker as well as the facilitating and hindering factors of improving self-management behaviors. The interviews will be conducted at 1, 3, 6-month and the participants will be recruited until the information is saturated each time. Interviews will be transcribed and analyzed by Nvivo (version 12, QRS International, Doncaster, Australia) using the thematic framework. Both quantitative and qualitative data will be integrated to better explain the acceptability and effectiveness of the chatbot speaker for hypertensive patients.\u003c/p\u003e\u003ch2\u003eStatistical analysis\u003c/h2\u003e\u003cp\u003eContinuous variables, such as age, blood pressure and scores of scales obeying normal distribution, will be described as mean ± standard deviation. Categorical variables will be reported using n (percentage of sample). Continuous variables will be compared using Student’s t test, one-way ANOVA or Kruskal-Walls test according to the conditions. Chi-square test or Mann-Whitney U test will be selected to compare categorical variables. In addition, multiple linear regression, logistic regression and structural equation model will be conducted to explore the influencing factors of acceptability. Latent profile/class analysis will be conducted to explore different behavior patterns of using chatbot speaker. For longitudinal data, generalized estimation equation or generalized linear mixed model (GLMM) will be used to compare the changes between the two groups and dynamic structural equation model will be used to explore the changes of each influencing factor over time. An alpha risk of 0.05 will be set for two-sided tests. IBM SPSS Statistics 22.0, Mplus 7.4, R studio 4.1.2 will be used for statistical analysis.\u003c/p\u003e\u003ch2\u003eStudy management\u003c/h2\u003e\u003cp\u003ePrior to the study, all the investigator from different community health care will be trained comprehensively, clarifying the recruitment standards and precautions in the investigation process to ensure consistency and quality among all communities. During data collection, a double-entry checking procedure will be implemented to enhance the quality of data entry. Experts will be consulted for appropriate statistical methods. If finding abnormal data, the original questionnaires or record will be checked for verification before proceeding to the next step of analysis. Furthermore, the qualitative data will be transcribed by two researchers to ensure reliability.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eContinuously exposed shortcomings in traditional intervention methods that fail to meet evolving demands, theories and applications on artificial intelligence are becoming increasingly mature, gradually permeating various industries and playing crucial and efficient roles.\u003c/p\u003e \u003cp\u003eThe chatbot speaker in this study based on a knowledge graph can facilitate artificial intelligence intervention and realize the vision of intervention management for hypertensive patients in the current scenario. Firstly, in the construction of the knowledge graph, it is crucial to rigorously select knowledge sources and undergo expert evaluations until evidence-based requirements are met. Secondly, this chatbot speaker can accommodate a vast amount of hypertension management knowledge, providing real-time clarification for patients, offering comprehensive guidance for self-management, and effectively liberating human resources. Thirdly, patients can access relevant knowledge and adjust their self-management behaviors in a timely manner according to their individual needs, achieving a certain degree of personalization and enhancing self-involvement, thus potentially improving compliance. Lastly, this intervention can accompany patients for the long term and continuously improve through technological means, achieving iterative updates.\u003c/p\u003e \u003cp\u003eIn summary, artificial intelligence interventions for hypertensive patients is feasible and promising. And the acceptability is the most crucial part when it works. This study focuses on whether new technologies or products can be accepted by patients and how the chatbot speaker play a role in blood pressure control. The findings will provide meaningful insights into refining the artificial intelligence-based chatbot for chronic diseases management.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was approved by Biomedical Committee of Peking university (IRB00001052-21106) and this trial was registered in Chinese Clinical Trial Registry (registration number: ChiCTR2100050578). Study participation is voluntary. Participations who provided their written informed consent will be included in the study. All information of the study subjects and the obtained data will be kept strictly confidential. Consent withdrawal is possible at any time without cause.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAnonymized survey data will be available from corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was supported by National Key Research and Development Program of China(2022YFC3600900) and Beijing Municipal Natural Science Foundation (7202087).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor\u0026rsquo;s contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eXYS and PC designed this study. YL and XXZ provided the theoretical and methodological guidance. XLF and XYS are responsible for the study management. XYS is responsible for the recruitment. PC is responsible for the study implementation. All authors provided revisions and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe are grateful to the professionals who provided great support for this study in participant recruitment and data collection, and the patients who spent time on this study.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eZhou B, Carrillo-Larco RM, Danaei G, Riley LM, Paciorek CJ, Stevens GA, Gregg EW, Bennett JE, Solomon B, Singleton RK, et al. Worldwide trends in hypertension prevalence and progress in treatment and control from 1990 to 2019: a pooled analysis of 1201 population-representative studies with 104 million participants. Lancet. 2021;398(10304):957\u0026ndash;80.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThomopoulos C, Parati G, Zanchetti A. Effects of blood pressure lowering on outcome incidence in hypertension. 1. Overview, meta-analyses, and meta-regression analyses of randomized trials. J Hypertens. 2014;32(12):2285\u0026ndash;95.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMackenzie IS, MacDonald TM. Identifying poor adherence to antihypertensive medications in patients with resistant hypertension. Br J Clin Pharmacol. 2019;85(1):5\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarey RM, Wright JT, Taler SJ, Whelton PK. Guideline-Driven Management of Hypertension: An Evidence-Based Update. Circ Res. 2021;128(7):827\u0026ndash;46.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePazoki R, Dehghan A, Evangelou E, Warren H, Gao H, Caulfield M, Elliott P, Tzoulaki I. Genetic Predisposition to High Blood Pressure and Lifestyle Factors: Associations With Midlife Blood Pressure Levels and Cardiovascular Events. Circulation. 2018;137(7):653\u0026ndash;61.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePharmacological blood pressure lowering for. primary and secondary prevention of cardiovascular disease across different levels of blood pressure: an individual participant-level data meta-analysis. Lancet. 2021;397(10285):1625\u0026ndash;36.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBakris G, Ali W, Parati G. ACC/AHA Versus ESC/ESH on Hypertension Guidelines: JACC Guideline Comparison. J Am Coll Cardiol. 2019;73(23):3018\u0026ndash;26.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang L, Trieu K, Yoshimura S, Neal B, Woodward M, Campbell NRC, Li Q, Lackland DT, Leung AA, Anderson CAM, et al. Effect of dose and duration of reduction in dietary sodium on blood pressure levels: systematic review and meta-analysis of randomised trials. BMJ. 2020;368:m315.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSiegel D, Lopez J, Meier J. Antihypertensive medication adherence in the Department of Veterans Affairs. Am J Med. 2007;120(1):26\u0026ndash;32.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChoudhry NK, Kronish IM, Vongpatanasin W, Ferdinand KC, Pavlik VN, Egan BM, Schoenthaler A, Houston Miller N, Hyman DJ. Medication Adherence and Blood Pressure Control: A Scientific Statement From the American Heart Association. Hypertension 2022, 79(1).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNieuwlaat R, Wilczynski N, Navarro T, Hobson N, Jeffery R, Keepanasseril A, Agoritsas T, Mistry N, Iorio A, Jack S, et al. Interventions for enhancing medication adherence. Cochrane Database Syst Rev. 2014;2014(11):CD000011.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBundy JD, Li C, Stuchlik P, Bu X, Kelly TN, Mills KT, He H, Chen J, Whelton PK, He J. Systolic Blood Pressure Reduction and Risk of Cardiovascular Disease and Mortality: A Systematic Review and Network Meta-analysis. JAMA Cardiol. 2017;2(7):775\u0026ndash;81.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArteriosclerosis T. Vascular Biology Arterioscler Thromb Vasc Biol. 2015;35(1):1.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRuilope LM, Valenzuela PL, Lucia A. Digital therapeutics and lifestyle: the start of a new era in the management of arterial hypertension? Eur Heart J. 2021;42(40):4123\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchubel LC, Wesley DB, Booker E, Lock J, Ratwani RM. Population subgroup differences in the use of a COVID-19 chatbot. NPJ Digit Med. 2021;4(1):30.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartinengo L, Jabir AI, Goh WWT, Lo NYW, Ho M-HR, Kowatsch T, Atun R, Michie S, Tudor Car L. Conversational Agents in Health Care: Scoping Review of Their Behavior Change Techniques and Underpinning Theory. J Med Internet Res. 2022;24(10):e39243.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchick A, Feine J, Morana S, Maedche A, Reininghaus U. Validity of Chatbot Use for Mental Health Assessment: Experimental Study. JMIR Mhealth Uhealth. 2022;10(10):e28082.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiang Q, Zhang Y, Pian W. Chatbot as an emergency exist: Mediated empathy for resilience via human-AI interaction during the COVID-19 pandemic. Inf Process Manag. 2022;59(6):103074.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePithpornchaiyakul S, Naorungroj S, Pupong K, Hunsrisakhun J. Using a Chatbot as an Alternative Approach for In-Person Toothbrushing Training During the COVID-19 Pandemic: Comparative Study. J Med Internet Res. 2022;24(10):e39218.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFriederichs S, Bolman C, Oenema A, Guyaux J, Lechner L. Motivational interviewing in a Web-based physical activity intervention with an avatar: randomized controlled trial. J Med Internet Res. 2014;16(2):e48.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJack B, Bickmore T, Hempstead M, Yinusa-Nyahkoon L, Sadikova E, Mitchell S, Gardiner P, Adigun F, Penti B, Schulman D, et al. Reducing Preconception Risks Among African American Women with Conversational Agent Technology. J Am Board Fam Med. 2015;28(4):441\u0026ndash;51.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHunt M, Miguez S, Dukas B, Onwude O, White S. Efficacy of Zemedy, a Mobile Digital Therapeutic for the Self-management of Irritable Bowel Syndrome: Crossover Randomized Controlled Trial. JMIR Mhealth Uhealth. 2021;9(5):e26152.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWatson A, Bickmore T, Cange A, Kulshreshtha A, Kvedar J. An internet-based virtual coach to promote physical activity adherence in overweight adults: randomized controlled trial. J Med Internet Res. 2012;14(1):e1.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVenkatesh V, Davis FD. A Theoretical Extension of the Technology Acceptance Model: Four Longitudinal Field Studies: A Theoretical Extension of the Technology Acceptance Model. Four Longitudinal Field Studies; 2000.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVenkatesh V, Bala HJDS. Technology Acceptance Model 3 and a Research Agenda on Interventions. 2010, 39(2):273\u0026ndash;315.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFishbein M, Ajzen IJP, Rhetoric. Belief, Attitude, Intention and Behaviour: An Introduction to Theory and Research. Addison-Wesley, Reading MA. 1977, 41(4):842\u0026ndash;844.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaillet \u0026Eacute;, Mathieu L, Sicotte C. UTAUT Int J Med Inf. 2015;84(1):36\u0026ndash;47.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCao LH. Study on users\u0026rsquo; adoption of mobile social commerce based on UTAUT2. Dalian University of Technology; 2018.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDwivedi YK, Rana NP, Janssen M, Lal B, Williams MD, Clement M. An empirical validation of a unified model of electronic government adoption (UMEGA). Government Inform Q. 2017;34(2):211\u0026ndash;30.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWilliams MD, Rana NP, Dwivedi YK. The unified theory of acceptance and use of technology (UTAUT): a literature review. J Enterp Inform Manage. 2015;28(3):443\u0026ndash;88.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRana NP, Dwivedi YK, Lal B, Williams MD, Clement M. Citizens\u0026rsquo; adoption of an electronic government system: towards a unified view. Inform Syst Front. 2015;19(3):549\u0026ndash;68.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaabdullah AM, Alalwan AA, Rana NP, Kizgin H, Patil PPJIJIM. Consumer use of mobile banking (M-Banking) in Saudi Arabia: Towards an integrated model. 2019, 44:38\u0026ndash;52.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDwivedi YKJTMR. Devising a research model to examine adoption of mobile payments: An extension of UTAUT2. 2014, 14:310\u0026ndash;35.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShaw N, Sergueeva K. The non-monetary benefits of mobile commerce: Extending UTAUT2 with perceived value. Int J Inf Manag. 2019;45:44\u0026ndash;55.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou LL, Owusu-Marfo J, Asante Antwi H, Antwi MO, Kachie ADT, Ampon-Wireko S. Assessment of the social influence and facilitating conditions that support nurses' adoption of hospital electronic information management systems (HEIMS) in Ghana using the unified theory of acceptance and use of technology (UTAUT) model. BMC Med Inf Decis Mak. 2019;19(1):230.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCimperman M, Makovec Brenčič M, Trkman P. Analyzing older users' home telehealth services acceptance behavior-applying an Extended UTAUT model. Int J Med Inf. 2016;90:22\u0026ndash;31.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoque R, Sorwar G. Understanding factors influencing the adoption of mHealth by the elderly: An extension of the UTAUT model. Int J Med Inf. 2017;101:75\u0026ndash;84.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDwivedi YK, Shareef MA, Simintiras AC, Lal B, Weerakkody V. A generalised adoption model for services: A cross-country comparison of mobile health (m-health). Government Inform Q. 2016;33(1):174\u0026ndash;87.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSong T. A research on user behavior on mobile shoppoing app based on UTAUT model. Shandong Univ. 2017.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu T. Analysis of salt taste/salt intake and the effect of capsaicin on salt taste sensitivity of the elderly in northeast China: a clinical study. China medical university; 2019.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang MC, Dai XY, Yao SQ. Development of the Chinese big five personality inventory(CBF-PI)ⅲ: psychometric properties of CBF-PI brief version.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTang HY, Zhu JC, He HY, Qian CR, Yang YN. Development and evaluation of a new therapeutic adherence scale for hypertensive patients. J Third Military Med Univ. 2011;33(13):1400\u0026ndash;3.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSotos-Prieto M, Bhupathiraju SN, Mattei J, Fung TT, Li Y, Pan A, Willett WC, Rimm EB, Hu FB. Changes in Diet Quality Scores and Risk of Cardiovascular Disease Among US Men and Women. Circulation. 2015;132(23):2212\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShan Z, Li Y, Zong G, Guo Y, Li J, Manson JE, Hu FB, Willett WC, Schernhammer ES, Bhupathiraju SN. Rotating night shift work and adherence to unhealthy lifestyle in predicting risk of type 2 diabetes: results from two large US cohorts of female nurses. BMJ. 2018;21:363k4641.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-public-health","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pubh","sideBox":"Learn more about [BMC Public Health](http://bmcpublichealth.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/pubh/default.aspx","title":"BMC Public Health","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Hypertension, Artificial intelligence, voice-based chatbot, Mixed-methods","lastPublishedDoi":"10.21203/rs.3.rs-3903126/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3903126/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eChatbots can provide immediate assistance tailored to patients' needs, making them suitable for sustained accompanying interventions. Nevertheless, there is currently no evidence regarding their acceptability by hypertensive patients and the factors influencing the acceptability in the real-world. Existing evaluation scales often focus solely on the technology itself, overlooking the patients' perspective. Utilizing mixed methods can offer a more comprehensive exploration of influencing factors, laying the groundwork for the future integration of artificial intelligence in chronic disease management practices.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eThe mixed methods will provide a holistic view to understand the effectiveness and acceptability of the intervention. Participants will either receive the standard primary health care or obtain a chatbot speaker. The speaker can provide timely reminders, on-demand consultations, personalized data recording, knowledge broadcasts, as well as entertainment features such as telling jokes. The quantitative part will be conducted as a quasi-randomized controlled trial in community in Beijing. And the convergent design will be adopted. When patients use the speaker for 1 month, scales will be used to measure patients' intention to use the speaker. At the same time, semi-structured interviews will be conducted to explore patients' feelings and influencing factors of using speakers. Data on socio-demography, physical examination, blood pressure, acceptability and self-management behavior will be collected at baseline, as well as 1,3,6, and 12 months later. Furthermore, the cloud database will continuously collect patients\u0026rsquo; interactions with the speaker. The primary outcome is the efficacy of the chatbot on blood pressure control. The secondary outcome includes the acceptability of the chatbot speaker and the changes of self-management behavior.\u003c/p\u003e\u003ch2\u003eDiscussion\u003c/h2\u003e \u003cp\u003eArtificial intelligence-based chatbot speaker not only caters to patients' self-management needs at home but also effectively organizes intricate and detailed knowledge system for patients with hypertension through a knowledge graph. Patients can promptly access information that aligns with their specific requirements, promoting proactive self-management and playing a crucial role in disease management. This study will serve as a foundation for the application of artificial intelligence technology in chronic disease management, paving the way for further exploration on enhancing the communicative impact of artificial intelligence technology.\u003c/p\u003e\u003ch2\u003eTrial registration:\u003c/h2\u003e \u003cp\u003e Biomedical Ethics Committee of Peking University: IRB00001052-21106, 2021/10/14; Clinical Trials: ChiCTR2100050578,2021/08/29.\u003c/p\u003e","manuscriptTitle":"The acceptability and effectiveness of artificial intelligence-based chatbot for hypertensive patients in community: protocol for a mixed-methods study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-02-02 15:27:32","doi":"10.21203/rs.3.rs-3903126/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-03-26T10:29:50+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-03-01T09:13:49+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2024-02-01T05:39:45+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-02-01T05:37:28+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Public Health","date":"2024-01-27T13:38:44+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-public-health","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pubh","sideBox":"Learn more about [BMC Public Health](http://bmcpublichealth.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/pubh/default.aspx","title":"BMC Public Health","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"a6890ab2-12df-4277-aa8b-eea66d6cffa6","owner":[],"postedDate":"February 2nd, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2024-08-26T16:00:38+00:00","versionOfRecord":{"articleIdentity":"rs-3903126","link":"https://doi.org/10.1186/s12889-024-19667-4","journal":{"identity":"bmc-public-health","isVorOnly":false,"title":"BMC Public Health"},"publishedOn":"2024-08-21 15:57:10","publishedOnDateReadable":"August 21st, 2024"},"versionCreatedAt":"2024-02-02 15:27:32","video":"","vorDoi":"10.1186/s12889-024-19667-4","vorDoiUrl":"https://doi.org/10.1186/s12889-024-19667-4","workflowStages":[]},"version":"v1","identity":"rs-3903126","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3903126","identity":"rs-3903126","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}