Introduction

Mounting evidence suggests that regular moderate physical activity significantly improves health-related quality of life, promotes community engagement and healthy aging, and may enhance cognition and mental health [,]. It also appears to mitigate the risk of at least 35 chronic diseases and reduce mortality [,-]. Physical activity is currently recognized as a sustainable approach to promoting individual and community health and well-being []. Exercise prescription [], a structured program designed to guide exercise for health promotion and the prevention and management of chronic diseases, has been applied to diverse populations, including healthy individuals, those with sports injuries, perioperative patients, individuals with chronic conditions, and those with disabilities [-]. However, many physicians struggle to provide physical activity guidance and prescribe scientifically effective exercise regimens due to a lack of knowledge, training, awareness, and understanding of exercise rehabilitation []. These challenges, combined with the absence of standardized implementation guidelines [] and other barriers, hinder the promotion and integration of exercise prescriptions.

Most exercise guidelines [-] strongly recommend personalized and well-defined exercise prescriptions that specify the mode, intensity, frequency, and duration of exercise. Traditional exercise prescriptions often struggle to achieve scientific personalization, prompting the development of intelligent, personalized alternatives. By incorporating advanced technologies such as artificial intelligence and big data analysis, these modern approaches provide individuals with more precise, scientifically grounded, and customized exercise plans. For example, Netz et al [,] developed an innovative tool for remotely assessing balance, strength, and flexibility in middle-aged and older individuals, providing personalized exercise plans via smartphones. Lin et al [] and Sun et al [] created and validated distinct systems: a force gauge system that integrates exercise games with the Internet of Things and a cloud-based intelligent personalized exercise prescription system. Both systems are designed to deliver tailored exercise prescriptions for middle-aged and older individuals. Additionally, a multimodal data-driven artificial intelligence system has been developed and validated to generate personalized exercise prescriptions for patients with mental disorders []. These intelligent tools and systems effectively address the limitations of conventional exercise prescription methods. Beyond significantly improving exercise outcomes, reducing exercise-related risks, and enhancing individual health and quality of life, they also alleviate the burden caused by the shortage of professional sports rehabilitation resources, thereby improving the quality and accessibility of sports rehabilitation services [-]. Most research on intelligent personalized exercise prescriptions has focused on system design and development [] and efficacy validation []. However, studies examining the factors that influence individual adherence behaviors and their underlying pathways are currently lacking. Recent systematic reviews and meta-analyses [-] suggest that while smart health interventions hold promise for improving exercise adherence, the specific factors and pathways driving adherence have not been fully explored or systematically explained. This gap makes it challenging for individuals to maintain long-term adherence. Therefore, it is crucial to further investigate and refine the factors, pathways, and theoretical frameworks related to adherence to personalized smart exercise prescriptions.

Motivation is a critical determinant in shaping individuals’ behavioral intentions and driving behavioral changes, with various motivational factors regulating and influencing individual behavior [-]. Although motivation significantly impacts adherence behavior, theoretical limitations, a lack of empirical research, and the complexity of individual motivations contribute to a theoretical “black box” regarding how these motivations influence adherence to exercise prescriptions. The relationship between behavioral motivation and individual adherence behavior remains underexplored [-]. In health promotion, several theoretical models [,] have been used to study individual behavior. Among these, the Transtheoretical Model (TTM) has been widely applied in research on digital health behavior and behavior change [-]. By dividing behavior change into 5 stages—precontemplation, contemplation, preparation, action, and maintenance—the TTM provides a detailed framework for understanding the mechanisms and pathways of individual behavior change []. The TTM has shown promising results in various social studies focused on health behavior change and health promotion [-].

Therefore, this study uses qualitative methods to explore the factors influencing the adherence behavior of community-dwelling middle-aged and older residents who have followed personalized exercise prescriptions issued through an electronic health promotion system for 8 months. By integrating the TTM, the study analyzes behavioral intentions and changes at different stages of behavior change, identifies key factors in the adherence behavior change process, and constructs an implementation path model for adherence to intelligent personalized exercise prescriptions, driven by behavioral motivations and other influencing factors. This study reveals how behavioral motivations drive and sustain adherence to exercise prescriptions, expands the application of the TTM in digital health services, and provides insights for future research on adherence to intelligent personalized exercise prescriptions. Additionally, it offers practical strategies for enhancing and maintaining long-term adherence to these prescriptions. To our knowledge, this is the first qualitative study on adherence behavior to intelligent personalized exercise prescriptions.

MethodsDesign

This study is part of a longitudinal research project aimed at examining the health impacts of intelligent personalized exercise prescriptions on community-dwelling middle-aged and older residents. It uses a descriptive exploratory qualitative design, based on face-to-face semistructured interviews. The study strictly followed the COREQ (Consolidated Criteria for Reporting Qualitative Research) guidelines [].

Participants

The participants in this study are a subsample from a longitudinal research project conducted by the research team, which aims to investigate the long-term effects of intelligent personalized exercise prescriptions on the health of middle-aged and older individuals. The inclusion criteria for the longitudinal study were age 50 years or older; no severe physical diseases or related complications; no cognitive or mental disorders among community-dwelling middle-aged and older residents; and exclusion of individuals with severe cardiovascular, pulmonary, or renal diseases, severe diabetes or related complications, fasting blood glucose of 13.3 mmol/L or higher with positive urine ketones, postprandial blood glucose of 19.4 mmol/L or higher, resting blood pressure of 180/110 mmHg or higher, or severe cognitive or mental disorders. Participants were initially recruited from the community via telephone and verbal invitations.

To investigate adherence behavior and the factors influencing adherence to intelligent personalized exercise prescriptions, we used purposive sampling to select participants for face-to-face semistructured interviews. The inclusion criteria were (1) community-dwelling middle-aged and older residents who had undergone home-based health checkups between 2021 and 2022 and received personalized exercise prescriptions through an electronic health promotion system administered by community staff; (2) participants who were provided with printed exercise materials and received detailed explanations and guidance from community staff; and (3) individuals whose exercise prescriptions had been active for at least 8 months. To gain a comprehensive understanding of exercise prescription adherence, we specifically recruited participants with neutral or negative attitudes toward intelligent personalized exercise prescriptions, as well as those who self-reported low adherence. The sample size was determined based on theoretical saturation, defined as the point at which no new issues or insights emerge and all relevant conceptual categories have been identified and explored []. Throughout the research process, the team continuously analyzed the dialogues and assessed the saturation level of the interview data at each stage. After the 12th interview, no new information emerged, all major concepts and categories had been thoroughly identified and explored, and theoretical saturation was reached. As a result, 12 eligible community-dwelling middle-aged and older residents participated in the face-to-face semistructured interviews.

Ethics Considerations

This study received ethical approval from the Human Research Ethics Committee of Bengbu Medical University (approval number 2022-103).

Intelligent Personalized Exercise Prescription Program

Details of the intelligent personalized exercise prescription system have been previously described []. The system generally consists of 4 components: user registration and information input, internet-based health monitoring devices, online questionnaire surveys, and a cloud platform.

Before generating a personalized exercise prescription, participants must register with their real names and provide personal information, including basic details (age, gender, height, and weight) and health history (past medical history and exercise habits). Next, participants’ health data are collected using internet-based health monitoring devices. Additionally, participants must complete a series of online health questionnaires to provide the necessary data for generating personalized exercise prescriptions. Finally, all collected health data are uploaded to a cloud platform. The platform uses advanced data analysis techniques and artificial intelligence algorithms to evaluate participants’ health data and generate personalized exercise prescriptions tailored to each individual.

Once the personalized exercise prescription is generated, community health service staff will provide face-to-face health education and guidance based on it. Depending on participants’ needs, the prescription can be delivered in either electronic or paper form, promoting better adherence.

After the exercise prescription is implemented, community health service staff can access user information via the cloud platform and conduct regular follow-up calls. These calls aim to remind participants to adhere to the exercise prescription, answer their questions, and provide necessary support and guidance. illustrates the structure of the intelligent personalized exercise prescription system and the process of generating and applying the prescriptions.

‎Figure 1. Generation and application of intelligent personalized exercise prescription. Procedures and Data Collection

This study strictly adheres to the COREQ []. To gather detailed data and gain a deeper understanding of participants’ subjective experiences and true feelings, semistructured interviews were conducted. All interviews were carried out by 3 trained researchers, including the first author (XX), a graduate nursing student, and an undergraduate nursing student, between February and March 2023. Following purposive sampling to select interviewees, rapport was established between the interviewers and participants. Interviews were conducted in quiet, private spaces to minimize interruptions. Before the interviews, participants were briefed on the research objectives, methods, expected duration, and confidentiality principles. Informed consent was obtained, and participants signed consent forms. The interviews were audio-recorded by the interviewers, with additional on-site notes taken for further analysis. Additionally, interviewers observed and promptly documented participants’ nonverbal behaviors, including tone, gestures, and facial expressions, during the interviews. Each interview lasted 20-30 minutes. Within 24 hours of completion, the recordings were transcribed into written transcripts and verified by the participants for content accuracy.

According to the main principles of the TTM, the occurrence and maintenance of individual behavior change occur across multiple stages. Therefore, before conducting the semistructured interviews, we determined the landmark events for each stage—“intention,” “preparation,” “action,” and “maintenance”—based on the research objectives and the definitions and characteristics of these stages in the TTM. Relevant questions about stage transitions were formulated in the interview outline to ensure the interviews were systematic and scientifically rigorous. The interview outline included questions about the respondents’ basic information; their perceptions and attitudes toward the intelligent personalized exercise prescription; factors that promote or hinder progression through specific stages of behavior change; and their activities, emotional states, needs, and suggestions within each stage. To ensure confidentiality, all participant information and data were anonymized and deidentified to prevent personal identification. Recordings and transcripts were stored in password-protected, secure files on the university server. The interviews concluded when no new information emerged.

Analysis

After the qualitative interviews were completed, the initial transcripts of the interview recordings—aligned with the interview questions corresponding to the 4 behavior change stages (“intention stage,” “preparation stage,” “action stage,” and “maintenance stage”) in the TTM—were imported into the qualitative analysis software NVivo 12 (Lumivero). The transcripts were then subjected to a 3-level coding process () based on grounded theory [,] within the framework of the 4 behavior change stages.

Drawing on the TTM and the coding results, we analyzed the factors influencing the adherence behavior to intelligent personalized exercise prescriptions, proposed hypotheses for the pathways to achieving adherence, and developed an implementation path model for adherence to intelligent personalized exercise prescriptions. We validated the hypotheses through follow-up interviews with the 12 respondents and refined the theoretical model based on team discussions, ultimately finalizing it.

For clearer presentation and data referencing, the 12 participants were labeled P1 through P12. Data analysis and organization were primarily carried out by 5 researchers; 3 researchers (XX, GZ, and YX) sorted and classified the interview recordings for open, axial, and selective coding, while the remaining 2 (TS and HX) verified the phased collation and ensured accuracy and consistency in the 3-level coding process. The hypotheses and model framework for adherence behavior pathways to intelligent personalized exercise prescriptions were formulated by the research team, validated through follow-up interviews conducted by 2 researchers (XX and YX), and refined through group discussions.

Textbox 1. The 3-level coding process.

1. Open coding

This is the most basic level of abstraction in the coding process. Each interview transcript was read thoroughly, and the text was coded as it appeared, without unnecessary abstraction. By analyzing the text line by line, we identified and marked concepts and phenomena relevant to the research questions, ensuring that the coding remained original and authentic.

2. Axial coding

Building on open coding, we focused on the logical connections and relationships between the identified concepts. Through ongoing comparison and analysis, we refined these into higher-level main categories.

3. Selective coding

We further examined the relationships between the main categories, identifying a core category that served as the central focus of the coding system. This core category guided the analysis and discussion of the other categories.

ResultsOverview

Effective interviews were conducted with 12 participants, and their general information is summarized in . The results of the 3-level coding process based on the 4 stages of behavior change in the TTM are as follows: (1) Open coding involves breaking down, reorganizing, and conceptualizing the transcribed interview materials. This process led to the identification of 21 categories (see ), which highlight the direct factors influencing adherence to intelligent personalized exercise prescriptions. (2) In the axial coding stage, we examined the relationships and logical connections among the 21 categories identified during open coding. This led to the formation of 8 main categories (). (3) During selective coding, we integrated and refined the existing categories to extract a “core category.” The categories of “hedonic motivation,” “benefit motivation,” “intrinsic motivation,” “extrinsic motivation,” and “achievement motivation” were consolidated into a single category, “multiple motivations of behavior.” Similarly, “self-regulation” and “optimization solutions” were grouped under “problem-solving strategies,” while “perceptual barriers” were categorized as “obstacle factors” (see ). The open coding derived from the original interview data represents direct experiences and situational factors, forming the first level (bottom level). The 21 categories identified during open coding were organized into axial coding, which represents the preliminary classification of motivation and obstacle factors, making it the second level (middle level). Selective coding, which represents the core influencing factors derived from the systematic analysis of the main categories, forms the third level (top level). In summary, we constructed a 3-level model of factors influencing the adherence behavior to intelligent personalized exercise prescriptions, as illustrated in .

Based on the TTM, Multiple Behavioral Motivation Theory, and the 3-level coding results from interviews on the 4 behavior change stages, we propose the following hypotheses for the path to adherence to intelligent personalized exercise prescriptions:

Hypothesis 1: An individual’s intrinsic motivations (such as disease susceptibility, perceived threat, and a sense of family responsibility) and extrinsic motivations (including family support, doctor’s advice, recommendations from friends, and media publicity) contribute to the intention to adhere, leading to the “adherence intention stage.”Hypothesis 2: Achievement motivation encourages the individual to reach the “adherence preparation stage,” where they prepare for the actual adherence behavior to the exercise prescription.Hypothesis 3: Hedonic and benefit motivations drive the individual to enter the “adherence behavior stage,” resulting in the initiation of actual adherence behaviors.Hypothesis 4: The presence of multiple motivations, coupled with problem-solving strategies, leads the individual to the “adherence maintenance stage,” where they progress from partial to full adherence, achieving complete and sustained adherence to the exercise prescription.

The path hypotheses were verified and validated by 2 researchers (XX and YX) through follow-up interviews with the 12 respondents. Based on these findings, a model for the realization path of adherence behavior to intelligent personalized exercise prescriptions was further developed. This model consists of 4 stages of adherence: intention, preparation, action, and maintenance. In the intention stage, individuals are motivated by intrinsic and extrinsic factors, developing the intention to adhere, but without engaging in actual behavior. Driven by achievement motivation, they progress to the preparation stage, where they ready themselves for action and set goal expectations, though behavior has yet to occur. During the action stage, hedonic and benefit motivations play a key role in prompting individuals to take action and experience benefits, though full adherence may not yet be achieved. Finally, in the maintenance stage, supported by multiple motivations and problem-solving strategies, individuals transition from general adherents to complete adherents, fully adhering to the exercise prescription, as illustrated in .

Table 1. General information on interviewees.NumberAge (years)OccupationGenderEducational levelPresence of chronic diseasesType of exercise prescriptionPrevious exercise habitsPrevious success or failure in health behavior change (eg, smoking cessation or weight loss)P166Shop assistantFemaleJunior high schoolYesScientific fitness modeYesNoP271Public officialFemaleJunior high schoolYesScientific fitness modeYesWeight loss successP368TeacherFemaleVocational schoolNoHealth care modeNoNoP472LaborerMalePrimary schoolYesScientific fitness modeYesNoP563LaborerFemalePrimary schoolNoHealth care modeNoNoP668StaffMaleHigh schoolNoExercise habit–formation modeNoSmoking and alcohol cessation successP776StaffMaleJunior high schoolNoExercise habit–formation modeNoNoP857Pharmacy clerkFemaleHigh schoolYesScientific fitness modeYesNoP960StaffFemaleJunior high schoolNoExercise habit–formation modeNoNoP1069StaffMaleJunior high schoolNoHealth care modeNoNoP1172LaborerFemaleHigh schoolYesScientific fitness modeYesNoP1268LaborerMaleHigh schoolNoHealth care modeYesNoTable 2. Categories formed by open coding of interview content.Behavior change stage, category, and initial conceptRepresentative original sentencesIntention stage

Disease threat


Encountering problemsMy cervical vertebra has some issues. Looking at my phone too much, due to my age, so I exercise to relieve the discomfort in my cervical vertebra. [P5]

Worsening of symptomsMy heart wasn’t very good before. The doctor told me to try to increase the vitality of my heart. The lazier you are, the worse your heart condition gets. [P7]
Susceptibility to disease


Frailty due to agingAs I’ve gotten older, my health has declined. [P8]

Decline in functionNow our physical functions are not as good as those of younger people. If we don\'t exercise more, it won’t work. [P8]

Hazards of prolonged sittingIf you don’t exercise and just lie down for a long time without moving, you\'ll definitely have health problems later on. [P4]
Sense of family responsibility


Reducing children’s worriesAlthough my children are grown up, I am still their support. If I am healthy, they will worry less. [P9]

Promoting family healthIf I learn scientific exercise methods, I want to exercise with my family. Won’t our family all be healthy then? [P10]
Family support


Support from childrenMy children also strongly support me participating in this. [P5]

Companionship of spouseUsually, my spouse and I play badminton together. [P1]
Doctor’s advice


Doctor’s notificationThe doctor told me to try to increase the vitality of my heart. [P7]
Friend’s recommendation


Recommending participationMy friend Yue ** joined first and spoke highly of your place, so she recommended me to join personalized exercises here. [P6]
Media publicity


Broadcasting publicityInitially, it was mainly through some radio broadcasts that I learned about the benefits of exercise. [P5]Preparation stage

Technological awareness


Scientifically expectationsYour personalized exercise program must be well-organized and scientifically based. [P1]

Organizing positiveThis exercise prescription must have benefits for the exercise of a specific body structure or physical fitness. [P1]

Authoritative projectsAccording to my daughter-in-law, she said, “Mom, there is a health program here in collaboration with a medical college and the University of Science and Technology of China. They provide free health checkups and tailor personalized exercise prescription for you. They also guide how to exercise. They are here to promote health. You should go and take a look. It\'s much better than those deceptive health products.” [P2]
Exercise preparation


Preparing sports equipmentIn order to adhere to the exercise prescription properly, I even prepared specific exercise clothing and shoes. [P5]

Peer supportAt the beginning, I arranged to exercise together with a few people. [P1]Action stage

Service hedonism


Problem-solvingIt’s great that you can explain any questions we don’t understand so well. [P3]

Thorough guidanceYou health providers are so thoughtful in explaining the examination reports and guiding us in exercise. [P5]

Careful inspectionYour staff are very meticulous in every examination, both before and after exercise. [P4]

Professionalism of the service providerYou have a lot of knowledge and explain things very well... [P12]

Comprehensive reportingThe reports you provide are very clear and comprehensive. [P8]
Social hedonism


Making friendsDuring exercise, you can also make many friends (especially outdoors). [P6]
Physiological benefits


Symptom controlBefore, I had high blood sugar, but now my blood sugar is also very good. [P6]
I used to have constipation, and it’s still there, but the frequency has decreased... [P5]

Physical improvementThere has been an improvement in my body. I used to have high blood pressure and unstable blood pressure, but now it\'s very stable. [P2]

Increasing physical activityLook, I can raise my arms very high now (demonstrating a posture). [P2]
Psychological benefits


Feeling relaxedAfter exercising, I feel my mood is more relaxed, my body feels better, and I have more energy than before. [P3]

EnergeticNow, both mentally and energetically, I am better than before. [P2]

Experiencing well-being(After being guided on walking posture) When the heels touch the ground, I involuntarily straighten up, walking with chest out and stomach in, it feels particularly good, this kind of feeling... [P6]
Personal benefits


Acquiring knowledgeCompared to before, I have learned how to exercise with proper posture, how to relax after exercising, and how to stretch... [P2]

Free medical checkupsThey also conducted physical examinations for us, checking our bones and bone density, teaching us how to improve. They told me that my bone density had reached the bottom line and couldn\'t decline further, advising us to drink milk and eat eggs daily, as well as recommending specific vegetables. They explained it very thoroughly. [P4]Maintenance stage

Self-management


PerseveranceI can persevere because that’s my personality. Once I set my mind on something, I persist until I succeed, no matter what obstacles I face. [P6]

Forming habitsTo develop a good exercise habit, I have to wake up at a certain time every day. [P2]

Strict self-disciplineI must be strict with myself and not make excuses like “it’s cold today” or “it’s windy” to forgive myself. I need to ensure that I accomplish what needs to be done. [P2]
Flexibility adjustment


Persisting to completionIf I don’t meet the planned exercise volume today, I’ll make up for it tomorrow by sticking to the previous day’s shortfall. [P6]

Adapting flexiblyI follow the exercise routine as advised by you. On rainy days, I use the treadmill at home, ensuring I meet the required exercise volume while exercising in a way I enjoy. [P4]
Fulfillment of needs


Adding videosIt would be great if it could be turned into a video. The report is easy to misplace, and it’s easier not to forget with a video, especially at our age...[P1]

Enriching imagesI have a small suggestion. There are too few exercise images in our medical reports. It would be nice if they could be more varied. Otherwise, everything else is quite good. [P10]
Positive feedback on results


Good results upon retestingAfter exercising and reassessing my health, I found that I’m doing well. My grip strength is still good, and I’m almost reaching eighty kilograms. I’m still going strong. Hehehe (laughs happily). [P4]
Ignition of exercise enthusiasm


Eagerness to exerciseAfter exercising for a while, I feel very enthusiastic about it, as if I’m eager to participate in sports from the bottom of my heart. [P5]

Exercising voluntarilyHaving a workout plan makes me feel psychologically engaged. When it’s time, I feel like I should exercise, and there’s a sense of awareness, a kind of concern, that makes me want to work out. [P6]Pan-stage factors

Personal limitations


Struggling to persistSometimes, doing exercise alone is difficult, so doing group activities helps. [P7]

Lack of exercise interestHowever, later on, my legs felt better. Originally, I didn’t like exercising and didn’t do much exercise. [P3]

Physical discomfortBecause of my bad back, my legs hurt when I walk. So, I’m not good at walking and don’t exercise much. [P3]

Bodily fatigueSometimes when I feel tired, I don’t feel like exercising. [P1]

Prone to forgetfulnessI can’t remember the exercise prescription because I’m old and forgetful. I used to stick to it at the beginning, but later, I did less, and I always forget the movements. [P8]
External constraints


Lack of resourcesThere are no good places near my home. There is a small Friendship Square, but there is no track or facilities. [P2]

Unable to free oneselfThe children need to come here for meals, buy groceries in the morning, and cook. It’s usually fine, but sometimes it’s overwhelming when I’m busy. [P3]Table 3. Main categories formed by interview content axial coding.Behavior change stage, main category, and categoryDefinitionIntention stage

Intrinsic motivation


Disease threatPerception of adverse health consequences resulting from one’s current condition.

Susceptibility to diseaseAdverse physical conditions, declining quality, and unhealthy lifestyle habits are more likely to trigger diseases.

Sense of family responsibilityReduce the burden on the family and family members’ worries and promote the health, harmony, and stable development of the family.
Extrinsic motivation


Family supportMaterial or emotional support provided by family members.

Doctor’s adviceReasonable advice from doctors based on individual conditions to improve individual situations.

Friend’s recommendationVerbal introductions from friends that are hoped to be accepted and adopted.

Media publicityDissemination and transmission of information through information dissemination tools.Preparation stage

Achievement motivation


Technological awarenessUnderstanding and familiarizing oneself with the intelligent personalized exercise prescription system.

Exercise preparationPlanning and preparing for upcoming exercise activities.Action stage

Hedonic motivation


Service hedonismEnjoy and experience high-quality, professional exercise testing and guidance services.

Social hedonismExpand social connections, make friends, and enjoy the pleasure and benefits of social interaction.
Benefit motivation


Physiological benefitsControl of adverse physiological symptoms of diseases and improvement in physical function or capability.

Psychological benefitsImproved mental state and psychological well-being, as well as increased energy.
Personal benefitsPersonal access to free benefits, such as exercise and health knowledge, physical fitness, and health assessments.Maintenance stage

Self-regulation

Self-managementSelf-control and using inner strength for behavior change or habit formation.
Flexibility adjustmentFlexibly adjust according to individual circumstances within a certain range.
Optimization solutions


Fulfilling needsFulfill the requirements arising from individual needs.

Positive feedback on resultsFollowing the positive feedback from intelligent personalized exercise prescription to encourage better adherence.

Ignition of exercise enthusiasmStimulating or enhancing interest and enthusiasm for exercise through various means.Pan-stage factors

Perceptual barriers


Personal limitationLimitations imposed by individual conditions on accomplishing tasks.

External constraintsLimitations imposed by external factors beyond the individual on accomplishing tasks.Table 4. Core categories formed by interview content selective coding.Core category and main categoryDefinitionMultiple behavioral motivations

Hedonic motivationBehavioral motivation generated from the pursuit of the pleasant feelings brought by exercise.Benefit motivationBehavioral motivation arising from the actual benefits that can be obtained by following the exercise prescription.Intrinsic motivationBehavioral motivation originating from the individual’s internal needs and perceptions.Extrinsic motivationBehavioral motivation resulting from the support and incentives of external factors.Achievement motivationBehavioral motivation arising from the expectation of achieving good performance or reaching specific goals by adhering to the exercise prescription.Problem-solving strategies

Self-regulationSelf-management and adjustment of an individual’s behavior or thinking for better compliance with the exercise prescription.Optimization solutionsImprovement measures and solutions for enhancing the compliance effect and adherence to the exercise prescription.Obstacle factors

Perceptual barriersDifficulties and obstacles that an individual subjectively perceives as restricting their exercise in accordance with the exercise prescription.‎Figure 2. The 3-level model of factors influencing the adherence behavior to intelligent personalized exercise prescription. ‎Figure 3. The implementation path model of adherence behavior to intelligent personalized exercise prescription based on the Transtheoretical Model and Multiple Behavioral Motivation Theory. Factors Influencing Adherence Behavior With Intelligent Personalized Exercise Prescription and Pathways for ImplementationMultiple Behavioral Motivational Factors

Motivation is the central factor driving behavior change and serves as the intrinsic force behind individuals’ actions [,]. This study explores how various motivations—hedonic, benefit, intrinsic, extrinsic, and achievement—affect participants’ adherence to intelligent personalized exercise prescriptions. These motivations directly influence adherence behaviors []. (1) Intrinsic motivations (eg, perceived disease threat and susceptibility, sense of family responsibility) and extrinsic motivations (eg, family support, doctor recommendations, friend referrals, and media publicity) drive participants to develop an intention to adhere to the intelligent personalized exercise prescription. This phase, referred to as the “adherence intention stage,” involves participants forming the intention to adhere, but without yet engaging in actual adherence behaviors. (2) After forming the intention to adhere, participants are driven by achievement motivation, aiming to reach self-set goals through adherence to the exercise prescription. At this stage, they begin preparing for adherence by seeking information about the exercise prescription and acquiring the necessary equipment. This phase, known as the “adherence preparation stage,” builds upon the previous stage but does not yet involve actual adherence behavior. (3) When participants have a strong intention to adhere and are well-prepared, they enter the “adherence action stage” [], where actual adherence behavior takes place. During this stage, high-quality health guidance, positive service experiences, and expanded social connections foster hedonic motivation. Additionally, the physiological and psychological benefits of adherence, increased health knowledge, and free health checkups provided by health care providers contribute to benefit motivation. Furthermore, hedonic and benefit motivations play a key role in sustaining and further developing adherence behavior during this phase. Therefore, the various motivations influencing participants in the context of intelligent personalized exercise prescriptions are crucial in the first 3 stages and serve as the primary drivers of adherence behavior.

Obstacles

“Perceptual barriers,” or obstacles, are critical factors affecting adherence behavior. These barriers can undermine an individual’s confidence and motivation, thereby hindering the sustainability of adherence. They negatively impact both the initiation and maintenance of adherence []. The findings by Kilgour et al [] suggest that these obstacles stem from both individual limitations and external constraints. Individual limitations include disliking exercise, difficulty adhering due to personal traits, memory decline that causes the exercise prescription to be forgotten, and physical ailments or fatigue that prevent adherence. External constraints include insufficient exercise facilities, lack of necessary equipment, and a busy family life that interferes with following the exercise prescription. These factors impede the sustainability of adherence behavior and may lead participants to discontinue or abandon it. As obstacle factors do not occur exclusively in a specific stage of adherence behavior transformation but rather persist throughout each stage, they are classified as “pan-stage factors” during the coding process and the division of behavior change stages. These factors are independent of the 4 distinct stages of behavior change.

Solution Strategies

Fro