From “Rational Agent” to “Bounded Rationality”: A Qualitative Study of Decision-Making Biases in Exercise Behavior Among Post-PCI Patients with Coronary Heart Disease Authors: Tiantian Xia, Hongyuan Li, Lei Liu, Runyang Jin, Yunqiao Xie, Guangying Zhang, Yutong Han, Haiyang Xing, Yingjie Wang, Jiayu Wang Published: 2026 ABSTRACT AbstractBackgroundThe participation rate of exercise training among post-PCI patients with CHD is low, and their exercise behavior decisions are often influenced by bounded rationality, leading to decision-making biases. However, existing studies lack in-depth exploration. Qualitative research can provide an in-depth understanding of patients’ descriptions of their exercise behaviors and their actual experiences, thereby helping to systematically identify these decision-making biases.ObjectiveTo explore the descriptions and related experiences of post-PCI patients with coronary heart disease regarding their own exercise behaviors, and to systematically identify the types of decision-making biases.DesignA qualitative descriptive study design.MethodsFrom December 2025 to February 2026, semi-structured face-to-face interviews were conducted with hospitalized post-PCI patients, and thematic analysis was used for coding and theme extraction.ResultsThis study enrolled 20 patients. Three major bias types (information perception bias, risk perception bias, and self-regulation bias) and eight specific manifestations. These included individual cognitive fixation, authoritative information transformation, autonomous information compensation, somatic state perception, negative reference constraint, exercise risk generalization, subjective will determination, and daily activity replacement.ConclusionDecision-making biases have a considerable impact on exercise behaviour in post-PCI patients with CHD. This process is profoundly influenced by immediate somatic sensations, reference to others’ experiences, vague concerns about exercise risk, and the replacement of structured exercise with daily activities, leading patients to prefer decisions that afford immediate psychological safety or convenience.Keywords: coronary heart disease, PCI, exercise, behavior decision-making, decision-making bias Background With changes in lifestyle and an ageing population, coronary atherosclerotic heart disease (CHD) has become a major global public health concern. According to the latest statistics from the American Heart Association (AHA) in 2026,1 the total number of adults aged 20 years and older living with CHD in the United States during 2021–2023 was 15.9 million, and deaths attributed to CHD accounted for 38.2% of all cardiovascular disease deaths in 2023. The most recent data from China indicate that the number of people with CHD has reached 11.39 million, and this disease significantly affects the long-term health outcomes of the population.2 Percutaneous coronary intervention (PCI) can quickly open up narrowed or blocked coronary arteries, restoring myocardial blood flow and perfusion, thereby alleviating angina symptoms.3,4 However, some studies have pointed out that PCI can cause coronary inflammatory reactions and acute myocardial injury, which in turn affect cardiac contraction and conduction.5 In addition, patients still face a high risk of adverse cardiovascular events after surgery. Shah et al6 reported that the overall incidence of major adverse cardiovascular events (MACE) within one year after PCI was 10.8%, and among patients who experienced MACE, 26.4% of events occurred within 6 months after discharge, while 73.6% occurred between 6 months and 1 year post-procedure. Huang et al7 similarly indicated that in 151 patients with acute ST-segment elevation myocardial infarction undergoing primary PCI, the in-hospital MACE rate was 47% (7 cases). Therefore, although PCI technology has greatly improved the short-term clinical outcomes of CHD patients, the risk of post-procedural adverse cardiovascular events remains severe, highlighting the urgency and importance of strengthening long-term post-operative management and secondary prevention. Systematic and long-term management after PCI, namely cardiac rehabilitation and secondary prevention, is crucial for improving patients’ long-term prognosis and quality of life. Exercise, as a cost-effective, easy-to-implement, and highly accessible intervention, plays a key role in disease prevention and management.8 Relevant expert consensus and guidelines recommend that exercise-based cardiac rehabilitation can reduce the incidence of adverse cardiovascular events and improve prognosis in patients after PCI for CHD. Medically supervised exercise programs (cardiac rehabilitation) should be recommended to patients after PCI (Class I; Level of Evidence: A).9,10 Although exercise-based cardiac rehabilitation plays an important role post-PCI, numerous studies have shown that the participation rate and long-term adherence to exercise training in post-PCI patients remain low. Surveys show that globally,11 the participation rate in exercise-based cardiac rehabilitation is below 35%, with a dropout rate up to 39% within three months. De Bacquer et al12 conducted a follow-up visit and interview of 16259 patients with CHD at least 6 months after discharge, and the results showed that only 40% of patients achieved the physical activity targets recommended by international guidelines, and those who had not participated in cardiac rehabilitation programs had poorer exercise adherence. However, in China, the proportion of patients receiving standardized cardiac rehabilitation after PCI is less than 10%,13 and this situation severely restricts the actual effectiveness of exercise-based cardiac rehabilitation in the secondary prevention of CHD. Existing research shows that decision-making plays a key role in the process of individual behavior change as well as in the choice and maintenance of health behaviors. Behavioral economics, by revealing the complexity and bounded rationality of human decision-making, provides a richer perspective for understanding health behaviors.14,15 This field points out that behavior originates from decisions, and exploring “how people actually make decisions” and the logic behind them is central to explaining behavioral patterns.16 However, current research on cardiac rehabilitation exercise for patients with CHD mainly focuses on the influence of patients’ personal characteristics, social environment, and healthcare systems. Although it is recognized that individuals’ cognitive processing and executive functions play an important role in their willingness to participate in and adhere to cardiac rehabilitation exercise, the psychological decision-making processes driving these cognition and behaviors have not been deeply analyzed.17,18 As exercise behavior decision-making is the key psychological process linking rehabilitation willingness and actual action, it is the core entry point for exploring and addressing the low participation rates and insufficient adherence in exercise-based cardiac rehabilitation. Existing research has shown that behavioral decision-making plays an important role in promoting patients to change behaviors such as exercise,19 diet,20 and smoking cessation.21 However, in real-life situations, individual decisions do not fully follow rational principles, and their final behavioral choices often involve significant decision-making biases.22 Decision-making bias refers to the phenomenon of deviating from the optimal choice when seeking a satisfactory option.23 Especially in complex health decision-making environments, although individuals may have the best behavioral intentions, predictable decision-making biases such as framing effects and loss aversion often prevent them from making rational decisions consistent with their long-term goals.15 For patients who have undergone PCI for CHD, the desire for active rehabilitation coexists with fear of exercise-related risks, constituting a prominent conflict in exercise behavior decision-making. Therefore, an in-depth understanding of the decision-making performance characteristics underlying this contradiction is of critical theoretical and practical significance for establishing an effective support system for exercise training. Decision-making bias, as an individual’s systematic cognitive deviation in the process of information processing and choice of options, profoundly reflects the internal cognition and subjective preference tendencies of patients with CHD after PCI when faced with diverse exercise options. Qualitative research, with its unique advantages in exploring individuals’ deep psychological activities and capturing behavioral motivations in complex situations, can guide patients to clearly express their choices in specific contexts. It can not only restore the internal thought process driving their decisions but also accurately capture the emotional fluctuations and attitudinal tendencies hidden behind those decisions.24 Based on this, the present study aims to explore the experiences and perceptions of exercise behavior decision-making in patients with CHD after PCI, identify their decision-making biases, reveal the bounded rationality in their exercise behavior decision cognition, and ultimately help patients make decisions and take actions that are more aligned with their own health interests. Methods Research Objective This study explored the experiences and accounts of exercise behavior decision‑making among post‑PCI patients with CHD, aiming to identify the types of biases embedded in their exercise behavior decisions. Research Design This study used a qualitative descriptive research design. Qualitative descriptive research describes the expressed, surface-level, and obvious experiences of individuals, helping to understand the experiences and perspectives of patients, family members, and professionals.25 This method aligns with the objective of this study, allowing for the identification of decision-making biases through an in-depth analysis of patients’ experiences and perceptions during the exercise decision-making process. Sample size was determined based on the principle of data saturation, meaning interviews were terminated when no new information or themes emerged from the data.26 In this study, after conducting interviews with 20 patients and analyzing the data, the research team discussed and deliberated, and it was unanimously agreed that data saturation had been achieved when no new codes or themes emerged, at which point data collection was stopped. This study adhered to the Consolidated Criteria for Reporting Qualitative Research (COREQ) 32-item checklist.27 Research Setting and Study Subjects Research Setting This study used purposive sampling, selecting patients hospitalized in the Cardiology Department of a tertiary hospital in Shenyang, Liaoning Province, China, from December 2025 to February 2026 as study subjects. The study recruited potential participants with the assistance of head nurses from four cardiology wards. Recruitment materials clearly explained the rights and responsibilities of the participants and provided contact information for the researchers. Study Subjects Inclusion Criteria (1) Age ≥18 years, meeting the diagnostic criteria for coronary artery disease according to the “Guidelines for the Diagnosis and Treatment of Stable Coronary Artery Disease”,28 confirmed by coronary angiography; meeting the PCI indications outlined in the “Chinese Guidelines for Percutaneous Coronary Intervention (2025)”,3 receiving PCI treatment and in a stable condition; (2) willing to provide informed consent and voluntarily participate in this study. Exclusion Criteria (1) patients with other major organ dysfunction, including severe renal insufficiency, severe hepatic insufficiency, severe respiratory dysfunction, active malignant tumours, sequelae of severe cerebrovascular disease, or other serious comorbidities that, in the investigator’s judgement, might affect study adherence or outcome assessment; (2) patients with psychiatric disorders or cognitive impairment who were unable to understand the study content or cooperate with the survey; (3) unable to exercise due to physical conditions or disease. Data Collection First, to ensure the sample is representative, this study employed a maximum variation sampling strategy, selecting participants with different characteristics, such as gender, age, educational level, occupation type, and employment status.18 This information is crucial for analyzing the relationship between exercise behavior decisions and social background. Second, interviewers established good interactive relationships with the patients, data were collected through face-to-face interviews, with the interview locations chosen according to the participants’ preferences. Finally, During periods of stable condition, interviewers explained the purpose of the study and invited patients to participate in the survey. If they agreed, the study procedures were explained in detail, and the recordings were transcribed into text within 24 hours. Data collection continued until data saturation was reached. Interview Outline Based on the research objectives and combined with the clinical characteristics of patients after coronary heart disease PCI surgery, this study formulated the interview outline through the project team meetings, and it was revised after pre-interviews with two patients to ensure the adaptability and clarity of the outline. Data from the pre-interviews were not included in the analysis. The final interview outline is as follows: (1) What kind of exercise routine do you have? How do you think exercising affects your condition? Where do these views come from? How do you feel during exercise? (2) When you recently decided whether to exercise, what were the main factors you considered? (3) When you hesitate about exercising, how do you usually make your decision? (4) What kind of guidance or support do you think can help you reduce unreasonable choices when deciding on exercise? Research Team The research team comprised four principal investigators and six research assistants, with only one male member among them. Interviews were conducted collaboratively by two master’s-level nursing students (the first and fourth authors). The first author conducted the interviews, while the other was responsible for recording non-verbal behaviors, including tone, intonation, facial expressions and gestures. Prior to the interviews, all researchers had completed systematic training courses in qualitative research methods. In addition, the two interviewers had been engaged in clinical nursing practice at the study site, and established good rapport with all participants before the interviews. The transcription and verification of the interview data were carried out by two independent research assistants separately, with a third research assistant performing cross-checking. Coding was performed independently by two principal investigators, and any disagreements arising during the analysis were resolved through consensus meetings. Data Analysis This study employed inductive thematic analysis,29 with all codes and themes derived progressively from the interview transcripts. Interviews were transcribed verbatim and verified within 24 hours of completion. Two researchers independently read and reviewed the interview transcripts repeatedly using Nvivo 15.0 software to achieve comprehensive familiarisation with the data and generate initial codes that remained close to the original text. The constant comparative method was used to compare codes across different interviews iteratively, and codes with similar meanings were grouped and merged to gradually develop sub-themes and themes. During the coding process, the two researchers reached consensus through discussion and deliberation. Data saturation was confirmed when no new codes or themes emerged. After each interview, the researchers maintained reflective diaries to enhance the auditability of the study. To ensure that the identified codes genuinely reflected decision-making biases rather than clinically prudent risk assessment, we established explicit operational criteria for identification. In this study, a statement was coded as a decision-making bias only when it met both of the following criteria: (a) the patient’s decision systematically over-relied on a single, easily accessible source of information, while systematically neglecting or dismissing contradictory medical evidence or long-term prognostic benefits; (b) the behavioral choice resulting from this decision objectively did not conform to the post-discharge exercise prescription recommended by clinical guidelines.10 Research Rigor This study enhanced rigor through the following methods: (1) Credibility: Representative interviewees were selected using a maximum variation sampling strategy (covering patients of different genders, ages, educational levels, occupational types, and comorbidities) to improve the authenticity and representativeness of the data; triangulation was applied,30 and the results were continuously compared with the original data to enhance the validity of the data and the rationality and logic of analysis and interpretation, thereby improving the credibility of the data; (2) Reliability: The two researchers participating in the interviews were trained in qualitative research and had interview experience. They established good relationships with patients and gained their trust before the interviews. Interview outlines were developed through group discussion and revised by professors in the research team with extensive experience in chronic disease management and qualitative research. Two patients were selected for pre-interviews, and after discussion within the project team, the interview outline was finalized. Pre-interview data were not included in the formal study; (3) Confirmability: Researchers continuously examined the potential influence of their own theoretical presuppositions and personal positions on data collection and interpretation through reflective diary writing, team discussions, and consensus meetings on coding, rather than simply maintaining “neutrality”. The research team acknowledged that complete neutrality is difficult to achieve; however, through the aforementioned transparency strategies, they sought to minimize the interference of subjectivity in the research conclusions; (4) Transferability: Non-verbal behaviors of the interviewees, such as facial expressions, tone, and gestures, were recorded on-site. The real experiences and viewpoints of the interviewees were faithfully transcribed, helping researchers better understand their decision-making biases and allowing readers to judge the transferability of the research results. Ethical Considerations This study was approved by the Ethics Committee of the Affiliated Hospital of Liaoning University of Traditional Chinese Medicine (Approval No.: Y2025334CS(KT)-334-01). The study strictly adhered to the ethical standards of the Declaration of Helsinki, and all participants voluntarily signed informed consent forms prior to participation. Privacy protection measures were implemented throughout the study, and personal information was kept strictly confidential; original materials such as interview recordings and transcripts were used only for research analysis and were not disclosed or used for other studies without the consent of the participants. When presenting the research results, participants’ identifying information was anonymized to effectively protect their privacy and rights. Results Demographic Characteristics of the Study Subjects A total of 20 post-PCI patients with CHD were included in this group, comprising 12 males and 8 females, with ages from 42 to 78 years (mean 62 years). Among them, 35% were employed, 85% lived in urban areas, and 90% had an exercise habit (mostly walking). The disease duration ranged from 1 month to 30 years, averaging 6 years, and the comorbidity rate was 75%. Detailed demographic characteristics of the study participants are presented in Table 1. Table 1Demographic Characteristics of the Study ParticipantsNo.GenderAge (Years)Education LevelEmployment StatusPlace of ResidencePrevious Exercise HabitsDisease DurationTime Since Last PCINYHA ClassComorbiditiesStents (n)P1Male42Bachelor’s degreeEmployedUrbanNone1 month1 month post-PCIClass IINone4P2Male46 High schoolEmployedUrbanWalking3 years2 years post-PCIClass IDiabetes4P3Female65Primary schoolRetiredRuralTai Chi10 years4 years post-PCIClass IHypertension1P4Male62Junior high schoolRetiredUrbanWalking4.5 years6 months post-PCIClass IHypertension4P5Male48Associate degreeEmployedUrbanWalking, dumbbell exercises3.5 years3.5 years post-PCIClass IHypertension, Hyperlipidemia3P6Male72Bachelor’s degreeRetiredUrbanWalking2 years1 year post-PCIClass IIHypertension2P7Female65Junior high schoolRetiredUrbanWalking1 year1 year post-PCIClass IHypertension, Hyperlipidemia3P8Female71Primary schoolRetiredRuralWalking30 years25 years post-PCIClass IIIHypertension2P9Male59Primary schoolRetiredRuralWalking2 years9 months post-PCIClass IHyperlipidemia2P10Male64Bachelor’s degreeRetiredUrbanWalking, running, horizontal/parallel bars10 years8 months post-PCIClass IHypertension, Hyperlipidemia, Diabetes1P11Male45Doctoral degreeEmployedUrbanGolf, slow ball, swimming, jogging2 years1.5 years post-PCIClass INone2P12Male66Junior high schoolRetiredUrbanWalking2 years5 months post-PCIClass IIHypertension, Diabetes3P13Male53Technical secondary schoolEmployedUrbanWalking1 month1 month post-PCIClass INone2P14Female76High schoolRetiredUrbanWalking20 years10 years post-PCIClass IIIHypertension2P15Male72Associate degreeEmployedUrbanWalking5 years3 months post-PCIClass IIDiabetes3P16Female70Junior high schoolRetiredUrbanNone4 years11 months post-PCIClass IINone4P17Female69Associate degreeRetiredUrbanWalking9 years9 years post-PCIClass IHypertension1P18Male53High schoolEmployedUrbanWalking2 years1 month post-PCIClass IINone2P19Female75Primary schoolRetiredUrbanWalking3 years1.5 years post-PCIClass IHypertension, Hyperlipidemia, Diabetes3P20Female78 Bachelor’s degreeRetiredUrbanWalking3 years1 year post-PCIClass IHypertension1 Demographic Characteristics of the Study Participants Research Results A total of 177.37 minutes of interviews were conducted, with individual interview durations ranging from 4.23 to 18.08 minutes (median 7.945 minutes), yielding 17124 words of transcribed textual data. Analysis revealed that no new codes emerged starting from the 18th participant (P18). To verify the reliability of the saturation determination, the research team performed a case audit on three interview transcripts (P18~P20). Two researchers independently re-read the complete transcripts and confirmed that all statements could be fully accommodated within the existing coding framework (17 codes). This study involved a single interview session, no repeat interviews were conducted. Saturation was determined based on coding content rather than interview duration. Through analysis, three themes and eight types of biases were ultimately extracted. The thematic map of exercise behavior decision-making biases is presented in Figure 1, and the coding logic judgement illustrative table is provided in Table 2. The thematic map of exercise behavior decision-making biases. Table 2Schematic Table for Coding Logic JudgmentThemeBias TypeOperational Definition (Coding Criteria)Key Biased Statement ExamplesNon-Biased SituationsInformation perception biasIndividual cognitive fixationPatients tend to actively search for, focus on, and act on information consistent with their own pre-existing beliefs, confining themselves to self-determined types of exercise and rejecting individualized exercise plans that contradict their established views.“I have heart problems, so I can only do these two exercises—running and swimming. It’s because of the stent that I can only do these two.” (P1)After being informed of new medical evidence or professional advice, the patient is able to adjust their previous exercise perceptions and try new forms of exercise, rather than adhering rigidly to a single type.Authoritative information transformationPatients’ exercise decisions rely heavily on authoritative information sources, but they subjectively interpret and conservatively translate ambiguous or contradictory parts of such information, leading to behavioral choices that deviate from the original intent.“The doctor didn’t tell me about exercise. Sometimes he told me, after the stent he told me to walk less.” (P8)The patient accurately understands the full meaning of the physician’s advice on “moderate exercise” and develops a specific exercise plan under professional guidance, rather than interpreting the advice one-sidedly as “prohibition of exercise.”Autonomous information compensationWhen professional medical exercise guidance is insufficiently accessible, some patients spontaneously seek information from other sources to fill the knowledge gap, thereby turning to publicly available information platforms as alternative resources.“I usually learn from the internet and watch videos or something.” (P1)After searching for exercise-related information online, the patient proactively verifies the scientific validity and applicability of that information with healthcare providers, rather than directly adopting non-professional information as the basis for decision-making.Risk perception biasSomatic state perceptionPatients use the most readily available immediate somatic sensations as the core reference for exercise decisions, substituting subjective physical feelings for objective functional assessment, and rigidly interpreting postoperative discomfort as a permanent exercise contraindication.“After I got the stent, I felt chest tightness, couldn’t exercise, and dare not exercise.” (P10)The patient temporarily reduces activity due to acute chest tightness or severe fatigue, but resumes moderate exercise once symptoms subside, and actively seeks professional evaluation when symptoms persist.Negative reference constraintPatients use a single negative case they first encountered as a fixed reference point for assessing their own exercise risk; once this anchor is formed, it becomes difficult to be overturned by contrary safety evidence or successful cases.“There was an old man who didn’t care, also had cerebral thrombosis, and he went out for long walks every day, wandered around for three days, and on that night he vomited blood and died; I saw it with my own eyes.” (P13)After hearing about others’ adverse exercise events, the patient becomes somewhat cautious but can still rationally adjust their exercise decisions after obtaining professional assessment or comprehensive information, rather than rigidly treating a single event as an absolute contraindication.Exercise risk generalizationPatients exhibit avoidance and fear of exercise due to a strong sense of uncertainty regarding the risks of exercise after stent implantation.“I wouldn’t dare to do strenuous exercise either, since I have heart disease and a stent. What if something goes wrong.” (P6)The patient acknowledges uncertainty about postoperative exercise but actively seeks objective functional assessment (eg, cardiopulmonary exercise testing) and professional exercise prescriptions to quantify risks, rather than completely avoiding or ignoring exercise under the pretext of “uncertainty.”Self-regulation biasSubjective will determinationPatients attribute the decision to exercise or not entirely to their personal will, overemphasizing the primacy of autonomous choice while disregarding external reference standards.“I think that exercise is purely a matter of personal will.” (P5)The patient acknowledges that personal will is an important part of exercise decisions, but simultaneously considers objective conditions such as disease suitability and exercise intensity safety, and seeks professional guidance before exercising.Daily activity replacementPatients prioritize alternative daily activities that are immediately feasible and low-resource-consuming, thereby weakening the perceived necessity of medically recommended structured rehabilitation exercise in pursuit of immediate behavioral convenience.“Housework is quite heavy, the walking amount is enough for exercise.” (P16)Due to realistic constraints such as old age, frailty, joint diseases, or geographical inaccessibility of rehabilitation resources, the patient chooses alternative activities within their capacity, and returns to structured exercise programs when conditions improve. Schematic Table for Coding Logic Judgment Information Perception Bias In this study, information perception bias refers to patients’ tendency to selectively filter, interpret, or seek out exercise-related information. It should be emphasised that this tendency is significantly influenced by the accessibility and quality of professional guidance. When the time for exercise guidance provided by healthcare providers is limited, patients naturally turn to social networks and online platforms as alternative sources of information, actively constructing an information environment that supports their own positions. Individual Cognitive Fixation Patients undergoing PCI for CHD tend to actively search for, pay attention to, and adopt information that aligns with their inherent attitudes during the exercise decision-making process, showing a clear cognitive fixation. In choosing exercise programs, some patients are dominated by their existing cognition, limiting themselves to exercise types they personally recognize and refusing to accept more individualized and adaptable options. For example, a participant mentioned, Since my heart has problems, I can only do these two exercises, running and swimming, simply because I have a stent, so those are the only two exercises I can do. (P1) Some patients, based on their own cognition, judge certain types of exercise as risky and directly reject them, If you want to dance, that’s a bit too strenuous, too intense (P3); Especially when it comes to strength training exercises, I still feel very fearful (P11). Athoritative Information Transformation Patients’ exercise behavior decision-making relies heavily on guidance from authoritative information sources, while they also make conservative transformations of the information conveyed by these sources, leading to differentiated behavior choices. In the process of receiving authoritative recommendations, some patients, due to the vagueness of the information transmitted or differences in their own risk perception, make conservative transformations of medical instructions, thereby generating contradictory decision-making tendencies. For example, regarding the core principle of “moderate exercise”, some patients selectively magnify the statements that suggest reducing activity, distorting medical instructions into signals that restrict exercise, The doctor didn’t tell me to exercise; sometimes he told me, after stenting, to walk less (P8). Similarly, another patient further equated the doctor’s advice of “not going out” directly with “not exercising”, thereby limiting the types of exercise and activity space, The doctor told me it’s best not to go out, not being allowed to go out means not being allowed to go out for exercise, just walk at home (P13). Autonomous Information Compensation The study found that when access to professional medical exercise guidance was insufficient, some patients proactively sought information from other sources to fill their knowledge gap. Some patients explicitly mentioned that they had not received adequate exercise guidance in clinical settings, and therefore turned to publicly accessible information platforms as alternative options, I usually learn from the internet and watch videos or something (P1); Doctors are too busy to tell us in detail how to exercise, isn’t everything available online? (P20). It is noteworthy that when referring to online information, patients also screened and adjusted the information based on their own judgments of their physical condition, I’ve watched some videos, not high-intensity exercise, but brisk walking or jogging, but I also consider my own physical condition, I can’t completely follow them. (P2) Risk Perception Bias In this study, risk perception bias refers to patients’ tendency to cite immediate, emotionally charged somatic sensations, negative cases, and vague concerns about exercise risk as the primary reasons for avoiding exercise. In this process, patients rarely refer to objective functional assessment indicators or healthcare professionals’ confirmation of the safety of low-to-moderate intensity exercise, instead, they rely on personalised, emotionally coloured subjective judgments as the basis for their decisions on whether to engage in exercise. Somatic State Perception In their exercise behavior decision-making, patients with CHD after PCI often rely on the most readily accessible cue, their own immediate bodily sensations as the basis for their decisions. Some patients regarded subjective physical feelings as the core of their decision-making, Now I can only do mild exercise, just walk a few laps (P2); If I feel lazy, tired, or have done too much housework today, I won’t go out. If I feel I have enough energy, I will go out, just according to my own physical condition. (P17) Notably, persistent or intermittent physical discomfort experienced by some patients post-surgery further reinforced their perception of exercise risk, making it difficult to initiate exercise behaviour, After I got the stent, I felt chest tightness, couldn’t exercise, and dare not exercise (P10). Such immediate bodily sensation information, being easily perceived and retrieved, became a substantial barrier to patients’ initiation of exercise behaviour. Negative Reference Constraint Patients after PCI tend to over rely on a particular reference that they encounter first when making decisions about exercise behaviour. This study found that when deciding whether to engage in exercise, some patients did not base their decisions on their own postoperative recovery status or medical assessments, but rather used negative postoperative cases of others around them as the primary behavior reference, That (stent) he did it and how come he died within a month? (the patient’s friend died one month after stenting) (P6). Similarly, another respondent described a witnessed experience of others, There was an old man who didn’t care, also had cerebral thrombosis, and he went out for long walks every day, wandered around for three days, and on that night he vomited blood and died, I saw it with my own eyes. (P13) Exercise Risk Generalisation Patients with CHD after PCI often experience heightened concerns about the uncertainty of exercise risk following stent implantation when making exercise behavior decisions. Most patients have strong concerns about the unknown risks of post-surgery exercise, showing clear characteristics of ambiguity aversion, specifically manifested as fear and avoidance of exercise. Some patients worry that strenuous exercise may trigger stent-related adverse events, developing a perception that exercise equals risk. I do not even dare to do strenuous exercise anymore, is not it a heart problem? I have a stent, what if something goes wrong. (P6); some patients, even concerned that daily activities might trigger cardiovascular accidents, feel psychological burden even just going downstairs, (After the stent) of course there’s worry, how could I not be worried? Just afraid to go downstairs, what if something happens? (P18). Additionally, patients are susceptible to interference from the remarks and experiences of those around them, showing pronounced psychological concerns. For instance, some participants expressed concerns due to negative comments from others, Some say that these stents are not good, not good for the body, and I’m still worried and scared (P9). Self-Regulation Bias Self-regulation bias refers to patients’ tendency, when translating exercise intentions into actual actions and when interpreting their own behavior outcomes, to preferentially select action plans that are immediately feasible and require low resource consumption, and to attribute the success or failure of their behaviour primarily to subjective judgments of personal will. Subjective Will Determination This study found that, at the levels of initiation, implementation, and evaluation of exercise behaviour, patients with CHD after PCI tend to attribute the decision to exercise or not entirely to their own internal willingness, while rarely referring to external standards, and overemphasising the dominance of autonomous choice. Some participants explicitly defined exercise behavior decisions as purely personal choices, stressing the decisive role of subjective willingness, I walk by myself, if I want to go, I go, and I can exercise anytime, anywhere (P4); I think exercise is purely a personal matter of one’s own willingness (P5); Walk more if you want, don’t if you don’t (P7). Other patients, when assessing their own exercise intensity, similarly replaced objective quantitative criteria with subjective, general judgments, “It’s subjective, just do some activity and it’s about enough” (P16). Daily Activity Replacement Compared with long-term, delayed health benefits, individuals tend to prefer immediate and easily accessible behavior options. At the level of exercise behavior implementation, patients with CHD after PCI tend to replace medically recommended structured rehabilitation exercise with daily unstructured activities. Some participants equated the walking involved in shopping trips with rehabilitation exercise, believing that the walking during grocery shopping already met their exercise needs, while rarely initiating dedicated, regular exercise, Just going grocery shopping, isn’t that going out for a walk and exercise? (P7). Other patients directly equated household chores with rehabilitation exercise, thereby reducing the perceived necessity of structured exercise, Cleaning the house, that also counts as exercise, quite a lot of exercise. you get out of bed, go to the toilet, all these are activities (P16); Housework is quite heavy, the walking amount is enough for exercise (P19). Some patients also regarded their daily workload as physical activity, Going to work and seeing patients is so tiring, forty or fifty patients a day, that workload is enough (P15). Discussion Through in-depth analysis of the interview data, this study described the types of biases in exercise behavior decision-making among patients with CHD who underwent PCI. The findings indicate that patients’ judgments and decisions in this process are influenced by multiple factors, including their own somatic sensations, daily experiences, others’ experiences, and information access channels, resulting in diverse personalized decision-making patterns that lead to exercise behavior plans congruent with their own cognition and circumstances. This study may facilitate the establishment of personalized comprehensive control targets and management content tailored to different types of decision-making biases in post-PCI patients, thereby providing a reference for clinical interventions. The results of this study indicated that post-PCI patients exhibit relatively complex information processing patterns in their exercise behavior decision-making. On the one hand, patients’ information acquisition channels are characterized by diversity, and their information choices are influenced by factors such as the accessibility of information, familiarity with information sources, and the congruence between information content and their own experiences. On the other hand, in the process of information use, patients demonstrate individualized patterns of judgment and selection when confronted with information from different sources and with different orientations. Han31 also noted that during the information reception stage, patients tend to interpret the advice of authoritative figures according to their own understanding, thereby exhibiting bounded rationality. Pender,32 in his health promotion model, noted that individuals’ health behaviors are deeply influenced by their cognitive and emotional responses to specific behaviors, with cognitive or perceptual factors playing a central role. This suggests that exercise decision-making is not only based on objective information but is also heavily shaped by subjective cognitive frameworks. Studies have shown that when making decisions, people are often constrained by cognitive abilities and lack of information, highlighting the need to create a healthy decision-making environment and frequently provide feedback to enhance self-management capabilities and maintain healthy behaviors.33 For example, when nurses provide personalized lifestyle intervention plans that include motivational interviewing over the phone to guide patients in weighing the pros and cons of physical exercise and reshaping their decision-making cognitive framework, patients’ activity levels can be improved to some extent.34 In addition, frequently feeding back patients’ exercise data using technologies such as health monitoring apps can enhance their engagement in exercise training and help monitor the appropriateness of exercise programs, significantly improving the quality of their cardiac rehabilitation.35 Therefore, it can be seen that the exercise behavior decision-making of patients after PCI for CHD is a process deeply shaped by information environment. Mere knowledge infusion is often insufficient to promote behavior change, suggesting that future interventions need to shift from providing information to reconstructing decision contexts, proactively guiding patients to overcome cognitive limitations, and establishing a more rational and positive perception of exercise at the early stages of decision-making. The study found that, at the level of risk perception, patients generally tend to prioritize immediate somatic sensations as the basis for exercise decisions. In addition, the uncertainty following stent implantation and postoperative adverse events experienced by others around them trigger fear-avoidance behaviors, which severely hinder the initiation and adherence to exercise programs. Yan et al23 pointed out that patients with type 2 diabetes often rely on experience and whether they exhibit obvious physical symptoms as a basis for rapid decision-making during the disease monitoring stage. To correct such cognitive biases, future interventions could focus on helping patients establish a scientific and positive “reference point” in their decision-making. For example, Pedersen et al36 conducted a 24-week randomized controlled trial in elderly patients with ischaemic heart disease, and demonstrated that support and encouragement provided by peer mentors with successful rehabilitation experience, based on their own personal experience, significantly improved patients’ participation rates in cardiac rehabilitation, with the initial participation rate increasing from 58.6% to 78.0%. In addition, according to the fear-avoidance model, patients affected by catastrophic thinking become extremely sensitive to exercise behavior and avoid physical activity, generating significant psychological stress and leading to various negative health outcomes, including physical deterioration.37 Li et al38 conducted a randomized controlled trial in post-PCI patients based on the fear-avoidance model, in which the intervention group received a stepwise intervention grounded in this model. The results showed that this intervention effectively reduced fear of exercise and enhanced exercise self-efficacy, providing a practical reference for psychological interventions in post-PCI patients. Notably, Guo et al,39 drawing on information framing theory, designed three types of personalized risk communication messages: gain-framed, loss-framed, and unframed; targeting three subgroups with biased cardiovascular disease risk perception (underestimation, overestimation, and accurate perception), to help patients form accurate disease risk perception and thereby promote rational health behaviour decisions. In summary, patients’ cognition of exercise is not a static knowledge gap but a dynamic assessment process shaped in real-time by physical and mental experiences. Therefore, intervention strategies need to consider both the immediate feelings and long-term risk adjustments. This study found that, at the level of exercise implementation, post-PCI patients face barriers to translating cognition into action. On the one hand, the initiation, intensity, and frequency of exercise behaviour are determined entirely by the individual’s immediate subjective willingness, lacking stable external reference standards; on the other hand, patients frequently replace regular structured exercise training with unstructured daily activities, resulting in their exercise behaviour deviating from the optimal rehabilitation pathway. Belizan et al40 from a behavioral economics perspective, found that overconfidence, optimism bias, and present bias can cause patients with hypertension and diabetes to overlook long-term health risks, becoming the most significant obstacles to disease screening and treatment. The nudge theory in behavioral economics provides an intervention approach in this regard. The theory advocates optimizing the “choice architecture” that is, adjusting the way options are presented, their accessibility, or the feedback mechanism so that help individuals shift from boundedly rational decision-making to more rational patterns, or to make decisions beneficial to themselves even under cognitive constraints.40 Research shows that nudge methods using economic incentives can, to some extent, bridge the intention-behavior gap and help patients make decisions consistent with the best long-term prognosis.41 Chokshi et al42 applied loss-framed financial incentives to patients with ischaemic heart disease and conducted the ACTIVE REWARD randomized controlled trial: patients wore wearable devices to track step counts, with $14 deposited into a virtual account each week and $2 deducted for each day the target was not met. The intervention lasted 16 weeks, with an 8-week follow-up. The results showed that the financial incentive group increased their daily step count by 1,501 steps from baseline during the intervention period (compared with an increase of only 264 steps in the control group), confirming that loss-framed financial incentives can effectively promote physical activity in patients. However, it is worth noting that economic incentives are considered more effective in promoting one-time behaviors than in sustaining ongoing behaviors.43 Therefore, in managing exercise behavior for cardiovascular diseases such as CHD, it may be worth attempting to appropriately increase economic rewards at the low points of maintaining a patient’s exercise behavior to encourage continued activity. In summary, even when patients have the intention to exercise, they may still fall into the dilemma of knowing but not acting due to the cognitive tendency for immediate gratification and a lack of executive control. Future research should go beyond simple economic incentives, focusing on individual differences and stage-specific characteristics to achieve a sustainable transition from passive motivation to proactive adherence. Limitations The findings reported in this study are mainly derived from inpatients at a single hospital, and the generalisability is limited by the specific regional context and disease stage, making it difficult to fully capture the decision-making experiences of post-PCI patients with CHD across different geographic regions and healthcare settings. Future research should include patients with CHD after PCI from diverse cultural backgrounds, at different clinical stages, and with multidimensional clinical data, in order to better elucidate their exercise behavior decision-making experiences. Additionally, since patients’ perceptions and experiences of exercise behavior decision-making may differ at various stages of post-PCI rehabilitation, future studies should conduct longitudinal follow-ups with such patients to more clearly explore the dynamic process of exercise behavior decision-making, revealing the long-term evolution of decision biases and their influencing factors. Conclusion Through qualitative interviews, this study identified specific manifestations of decision-making biases in exercise behavior decision-making among post-PCI patients with CHD. The findings revealed that their exercise choices were not primarily based on medical rehabilitation guidelines or assessments of long-term health benefits, but were influenced by multiple factors, including pre-existing exercise beliefs, immediate somatic sensations, reference to others’ experiences, vague concerns about exercise risk, and replacement of structured exercise with daily activities. These factors often led them to choose behavior options that provided immediate satisfaction or reassurance rather than medically optimal ones. To advance research in this field and make findings more generalizable and intervention-specific, a key direction for future studies is to develop and validate a scale measuring decision-making biases in exercise behavior among post-PCI with CHD patients. Healthcare professionals could use this scale to quickly identify the predominant bias types in individual patients, enabling precise, personalized cognitive-behavioral interventions or “nudge” strategies to promote effective exercise management. Acknowledgments We sincerely thank all the master’s nursing students who participated in the survey. Funding Statement This work was supported by 2025 Basic Scientific Research Projects of Colleges and Universities in Humanities and Social Sciences Category of Liaoning Provincial Department of Education (grant number LJ112510162012);2025 Nursing Discipline Research Projects of the Chinese Medical Association Publishing House (grant number CMAPH-NRC2025011). Patient or Public Contribution A pilot test was conducted prior to the formal interviews in this study. Two post-PCI patients who met the inclusion criteria were invited to participate in the pilot interviews. Their feedback was used to refine the wording and phrasing of the semi-structured interview guide, so that the interview tool would better align with patients’ everyday language and comprehension levels. Patients were not involved in the formulation of research questions, data analysis, result interpretation, or dissemination of findings. Abbreviations CHD, Coronary Heart Disease; PCI, Percutaneous coronary intervention. Data Sharing Statement Data supporting the results of this study are available from the first corresponding author (Jiayu Wang) upon request. Ethics Approval and Consent to Participate This study was approved by the Ethics Committee of the Affiliated Hospital of Liaoning University of Traditional Chinese Medicine (Approval No.: Y2025334CS(KT)-334-01). The study strictly adhered to the ethical standards of the Declaration of Helsinki. All procedures in this study were in accordance with ethical guidelines and regulations, and informed consent was obtained from participants before data collection. Author Contributions All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work. Disclosure The authors declare that they have no conflicts of interest in this work. References Associated Data Data Availability Statement Data supporting the results of this study are available from the first corresponding author (Jiayu Wang) upon request.