In recent years, telemedicine services have significantly improved healthcare quality and access for residents. However, challenges remain, including limited awareness and reluctance to adopt such services, especially among middle-aged and older patients with chronic conditions in remote areas.

This study aimed to investigate the key factors influencing the adoption of telemedicine services among middle-aged and older patients with chronic conditions in rural China.

A multicentre cross-sectional study was conducted between November 2023 and February 2024 at five hospitals in western China, targeting middle-aged and older patients (aged ≥ 45 years) with chronic conditions in rural areas. Participants completed the survey in person via QuestionnaireStar, with assistance provided as needed. The questionnaire was structured based on the Health Information Technology Adoption Model (HITAM) and included constructs such as Social Influence (SI), Perceived Usefulness (PU), Perceived Ease of Use (PEU), Perceived Reliability (PR), Perceived Health Threat (PHT), Self-efficacy (SE), Attitude Toward Using (ATU), Adoption Intention (AI) and Usage Behavior (UB). Data analysis was performed using structural equation modeling (SEM) to examine the relationships between these variables and to assess their impact on patients’ intention to adopt telemedicine services. Additionally, to further explore the factors influencing telemedicine services adoption, a generalized linear model (GLM) analysis was performed, incorporating variables such as age, gender, education level, monthly income, and health status.

A total of 880 middle-aged and older patients with chronic conditions in rural areas were included in this study, with an average adoption intention score for telemedicine services of 3.94 (standard deviation 1.02). SEM analysis revealed that SI, SE, PR, and PHT were peripheral variables, while PEU, PU and ATU acted as core mediating variables that positively influenced patients’ intention to adopt telemedicine services, which in turn impacted their subsequent usage behavior. SI (z = 4.767, P<0.001), PU (z = 2.894, P = 0.004), and ATU (z = 4.545, P<0.001) showed significant direct positive effects on patients’ intention to adopt telemedicine services. GLM analysis revealed that education level (β = 0.032, P < 0.001), monthly income (β = 0.009, P = 0.049), required care (β = 0.034, P = 0.007), and cancer diagnosis (β = − 0.147, P < 0.001) were significant predictors of telemedicine adoption intention.

This study identified key determinants influencing telemedicine services adoption among rural middle-aged and older patients with chronic conditions. The findings offer valuable insights for telemedicine services providers to enhance service design and implementation strategies, thereby promoting greater adoption and sustained use among this target population.

Telemedicine has experienced rapid growth in recent years, driven by the widespread adoption of information technology, seamless communication, wireless sensing technologies, and mobile internet connectivity [1, 2]. As a crucial component of the World Health Organization (WHO) digital health interventions, telemedicine delivers healthcare services remotely through information and communication technologies (ICT), aiming to enhance the accessibility, quality, and efficiency of medical services. Telemedicine encompasses not only remote access to healthcare services by patients via telemedicine devices, but also remote consultations and diagnoses provided by healthcare professionals. This enables patients to access high-quality medical resources regardless of geographic limitations [3].

Chronic non-communicable diseases (NCDs), such as cardiovascular diseases, cancer, and diabetes, have emerged as major global public health challenges. They not only impose a substantial burden on healthcare systems but also undermine the sustainable development of societies and economies [4, 5]. In China, NCDs account for approximately 88.5% of all deaths, with cardiovascular diseases, cancer, and chronic respiratory diseases accounting for over 80% of these cases [6, 7]. These conditions are typically characterised by long-term progression and require continuous, standardised management and monitoring. However, the uneven distribution of healthcare resources has made it difficult for many patients with chronic conditions to access timely and effective medical care, particularly in rural areas where healthcare infrastructure is still limited.

Telemedicine has opened new pathways for the management of chronic conditions by offering services such as remote monitoring of blood pressure and blood glucose levels, online health consultations, and disease management interventions. Such approaches support long-term disease management and contribute to more efficient allocation of healthcare resources [8]. Existing studies have demonstrated that telemedicine can reduce short-term hospitalization and mortality rates among patients with heart failure, while also improving the physical and psychological well-being of individuals with chronic conditions such as cancer [9, 10]. However, the current body of research on the actual effectiveness and patient acceptance of telemedicine remains inconclusive [11,12,13], particularly among middle-aged and older patients with chronic conditions in rural areas, where relevant studies are still relatively scarce. Although the COVID-19 pandemic has accelerated the adoption of telemedicine, patient acceptance remains a critical factor determining its widespread implementation [14,15,16,17]. Moreover, most existing studies have primarily focused on mobile health platforms, with limited exploration of telemedicine acceptance among populations with diverse chronic conditions. Furthermore, prior studies have mainly considered personal characteristics, such as age and gender, as influencing factors, with limited attention to disease-specific factors influencing adoption [18,19,20].

In rural areas with scarce medical resources, telemedicine is crucial for improving healthcare accessibility and promoting health equity. Compared to urban residents, rural inhabitants face challenges in accessing high-quality healthcare services due to the shortage of medical facilities and healthcare professionals, particularly for those with chronic diseases [21, 22]. For example, rural residents tend to have lower digital literacy and health literacy, and are more influenced by traditional medical beliefs, all of which can affect their acceptance of telemedicine. Therefore, studying the attitudes of rural patients with chronic diseases toward telemedicine and the factors influencing these attitudes is essential for advancing digital healthcare and improving the quality of rural healthcare services [23]. Guizhou Province is located in the southwestern part of China, with a rural population accounting for 46.85%. The province’s economic development has been relatively slow, and the distribution of medical resources is uneven, particularly the lack of high-quality healthcare services, which makes it difficult for rural residents to access medical care [24]. Although Guizhou has gradually established a four-tier telemedicine service system through big data policies and 5G technology in recent years to alleviate the shortage of medical resources [25], the low intention to adopt telemedicine remains a major barrier to its promotion [26].

Several theoretical models have been developed to explain the adoption and use of emerging information technologies, including the Theory of Reasoned Action (TRA), the Theory of Planned Behavior (TPB), the Technology Acceptance Model (TAM), the Health Information Technology Acceptance Model (HITAM), the Unified Theory of Acceptance and Use of Technology (UTAUT), the extended UTAUT2 model, and the Value-Attitude-Behaviour (VAB) model [27]. This study adopts the HITAM model as its theoretical foundation, primarily due to its specific applicability to the healthcare domain [28]. Compared with general models such as TAM or UTAUT, HITAM offers greater explanatory power regarding technology acceptance behaviours of patients and healthcare professionals. This is because it integrates constructs from the Health Belief Model (HBM) by incorporating health-related variables [29], thereby enhancing its predictive ability for technology adoption in medical contexts. Moreover, the chronic disease prevalence is higher among middle-aged groups (45 years and older), due to aging population trends and the increasing burden of chronic conditions. Therefore, studying chronic disease management and the adoption of telemedicine among the middle-aged and elderly population is crucial [30, 31]. In summary, this study employs a multicentre cross-sectional survey design to investigate the intention to adopt telemedicine services and the influencing factors among middle-aged and older patients with chronic conditions in rural areas of Guizhou Province.

This study was conducted using a multicentre cross-sectional design in Guizhou Province, a representative region in western China. Five hospitals across six regions were selected using convenience sampling. These included two tertiary hospitals, one municipal hospital, and two county-level hospitals. The study participants were from the Respiratory Medicine and Cardiology Departments of these hospitals. Eligible participants were those aged 45 years or older, diagnosed with at least one chronic condition, with registered and current residence in a rural area, clear consciousness, no cognitive impairment, and who voluntarily agreed to participate. Exclusion criteria included the presence of a diagnosed mental disorder. Sample size was estimated using the root mean square error of approximation (RMSEA) method [32], a commonly used index of model fit. The formula for the RMSEA calculation is as follows.

In this study, the null hypothesis for RMSEA (RMSEA H0) was set at 0.05, and the alternative hypothesis (RMSEA H1) was set to 0.08. The model included 16 degrees of freedom with a significance level of 0.05. A statistical power of 0.95 was selected to ensure sufficient sensitivity for modeling patients’ adoption intention of telemedicine services. The minimum required sample size of 854 was determined using the power4SEM tool (https://sjak.shinyapps.io/power4SEM/). With this sample size, the power to reject an RMSEA of 0.05 was 0.95, indicating 95% confidence in the adequacy of the model fit.

The items used in the questionnaire were primarily adapted from established scales in existing literature and revised to suit the context of telemedicine adoption among rural middle-aged and older patients with chronic conditions. The preliminary questionnaire was reviewed by a panel of experts, including health service management specialists, geriatricians, and epidemiologists, combining their professional expertise with insights from previous studies. Constructs included in the survey were drawn from the Health Information Technology Acceptance Model (HITAM) developed by Kim et al. [29], and comprised: Social Influence (SI), Perceived Usefulness (PU), Perceived Ease of Use (PEU), Perceived Reliability (PR), Perceived Health Threat (PHT), Self-efficacy (SE), Attitude Toward Using (ATU), Adoption Intention (AI), and Usage Behaviour (UB).

In the original HITAM, Health Beliefs & Concerns and Health Status function as antecedent variables that influence PU through PHT. However, for patients with chronic conditions, health beliefs and perceived health status are often internalised and inherently shape their perception of health threats [33]. Furthermore, PHT has been shown to be a key predictor of eHealth technology adoption in medical settings [18, 19]. Therefore, rather than treating Health Beliefs and Health Status as separate constructs, this study operationalises PHT as the core health-related variable, measured through a combination of objective indicators (e.g. disease type, disease counts) and subjective responses. This approach aims to enhance measurement validity while minimising multicollinearity. The questionnaire responses were rated on a 5-point Likert scale, ranging from 1 (completely disagree) to 5 (completely agree). Participants were instructed to choose the option that best reflected their actual situation. A pilot test was conducted with 30 patients prior to the formal survey. The questionnaire was revised based on the difficulties observed during the pilot testing, and adjustments were made accordingly. The 28 adoption intention items that demonstrated the highest internal consistency in the pre-test were ultimately retained. More details are provided in Supplementary File 1. Based on the aforementioned variables and theoretical model, a hypothesis framework was constructed (Fig. 1). Table 1 provides a detailed description of the specific meaning of each hypothesis.

Research model

Full size image

The final questionnaire was developed using the QuestionnaireStar platform and made accessible via a QR code generated through the system (https://www.wjx.cn/). Data quality was ensured by including a trap question, requiring participants to select “neutral” to proceed. This question served as an attention check to assess whether participants were answering carefully. Any deviation from the required response was considered indicative of either careless completion by the respondent or misinterpretation by the data collector, and the corresponding entry was excluded from the dataset.

All questions were marked as mandatory to prevent incomplete submissions. Given that the study focused on rural middle-aged and older patients with chronic conditions, questionnaires that were completed in less than 2 min were excluded from the analysis. Research assistants were assigned to each hospital to oversee the data collection process and ensure the quality and consistency of data collection. Additionally, training was provided to data collectors at each of the five hospitals prior to the commencement of the study. The training covered the study objectives, interpretation of questionnaire content, data collection techniques, and communication skills. Participant recruitment was conducted across all five hospitals between November 2023 and February 2024. Each participant was first informed of the study’s purpose and procedures. In particular, to ensure that participants accurately understood the concept of telemedicine used in this study, data collectors provided detailed explanations to each participant during the data collection process. In this study, telemedicine services refer to medical services provided remotely by licensed physicians using internet technologies (such as smartphones, tablets, wearable devices, or video terminals in rural health centers). These services include, but are not limited to, online consultations, remote medication guidance, remote health monitoring, electronic prescriptions, interpretation of examination reports, and personalized health management. In other words, through telemedicine services, patients can receive the medical support and health management they need without physically visiting a hospital or clinic. Oral informed consent was first obtained by the data collector. Participants were allowed sufficient time to complete the questionnaire at their own pace. They were also informed of their right to withdraw at any stage, in order to reduce overreporting or response bias. No gifts or incentives were provided to minimise potential biases, and participation was entirely voluntary. Patients accessed the online questionnaire by scanning a QR code using their own or a family member’s smartphone. The first section of the questionnaire was designed to collect demographic information, while the second focused on the study variables. For those unable to complete the questionnaire independently, a data collector provided assistance by reading and explaining the items.

Data were analysed using IBM SPSS Statistics version 27.0 (IBM Corp., Chicago, IL, USA) and the R packages lavaan and semTools. The data analysis included the following components: (1) Descriptive statistics covering variables such as patients’ age, gender, education level, monthly income, and type of chronic condition. Measurement data were expressed as means ± standard deviations, and categorical data were reported as frequencies and proportions. The scores of each dimension and item of the questionnaire were reported as the mean ± standard deviation. In addition, we conducted descriptive statistical analyses for each latent variable, including the maximum and minimum values, skewness, and kurtosis. (2) Reliability and validity tests: In reliability and validity testing, the Kaiser-Meyer-Olkin (KMO) measure and Bartlett’s test of sphericity were used to assess the suitability of selected variables for factor analysis. A KMO value above 0.7 and a significant result from Bartlett’s test indicate suitability for factor analysis [40]. Convergent validity, discriminant validity, Cronbach’s alpha, average variance extracted (AVE), and composite reliability (CR) were assessed using R. These tests were applied to the measurement model of the latent constructs. The coefficient values ranged from 0 to 1, with higher values indicating stronger reliability and validity. These results confirmed that the measurement items accurately reflected their corresponding latent variables [41]. (3) Structural equation modeling (SEM) analysis: covariance-based SEM was employed to examine the hypotheses. Model fit was assessed using three types of indices: absolute fit indices, relative fit indices, and parsimonious fit indices. The interrelationships among latent variables were tested. A critical ratio greater than 1.96 and a P-value below 0.05 were considered indicative of statistically significant path effects. (4) Lasso regression (L1 regularization) was used to perform variable selection, with the selected variables including gender, age, education level, monthly income, number of diseases, required care (referring to the need for assistance with daily activities such as eating, dressing, and personal hygiene), and disease types. Subsequently, based on the variables selected through Lasso regression, a generalized linear model (GLM) was employed to analyze the factors influencing patients’ intention to use telemedicine services. This study followed STROBE reporting guidelines (Supplementary File 2).

Ethical approval for the study was obtained from the Ethics Committee of the Affiliated Hospital of Guizhou Medical University (2023-061). Before accessing the questionnaire, participants were informed of the study’s objectives on the first page. They were also notified of their rights regarding anonymity, confidentiality, and voluntary participation. Participants were further assured that all data collected would be used exclusively for academic research. No personal information would be disclosed, and individual identities would not be affected in any published results.

A total of 952 questionnaires were collected, of which 72 were excluded from the final analysis for failing to meet the inclusion criteria. The reasons for exclusion were as follows: response time less than two minutes (5/72,6.9%), incorrect answers to trap questions (25/72,34.7%), and uniform or clearly contradictory responses throughout the questionnaire (42/72,58.4%). Thus, 880 valid responses remained for further analysis, yielding an effective recovery rate of 92.4% (880/952). The research participants took an average of 5.45 (4.58, 7.42) minutes to complete the entire questionnaire. The patients’ adoption intention score for telemedicine services was 3.94 ± 1.02. Approximately 46.3% (407/880) were from rural areas in Guiyang, with a nearly equal distribution of male and female participants (450/880, 51.1% vs. 430/880, 48.9%, respectively). The majority of the participants fell within the age range of 45 to 50 (306/880, 34.8%) and had primary school education or were illiterate (520/880, 59.1%). Over half of the participants reported a monthly income of less than 1,000 yuan (478/880, 54.3%), and nearly two-thirds had one chronic condition (578/880, 65.7%). Additionally, about one-third of the participants (331/880, 37.6%) required care. Further details appear in Table 2. The values for each construct of the HITAM are provided in Supplementary File 3. The descriptive analysis results of the latent variables are presented in Supplementary File 4. The absolute values of skewness for all variables are within 2, and the absolute values of kurtosis are within 4, indicating that the data conform to a normal distribution within acceptable error margins [42].

The results showed a KMO value of 0.978 and a significant Bartlett’s test of sphericity (P < 0.001), confirming the data’s suitability for factor analysis. The reliability of the questionnaire was evaluated using Cronbach’s α, with values ranging from 0.833 to 0.970, demonstrating good reliability of the measurement scale. Confirmatory factor analysis was performed using the lavaan package in R to assess discriminant validity. The results showed an adjusted goodness-of-fit index (AGFI) of 0.856, a goodness-of-fit index (GFI) of 0.889, a comparative fit index (CFI) of 0.967, and an RMSEA of 0.065. CR values were all above 0.8, and AVE values exceeded 0.7. The square root of the AVE was found to be greater than the corresponding correlation, confirming the discriminant validity of the data (see Tables 3 and 4; Fig. 2).

Comprehensive display of item standard loadings, construct reliability, and validity metrics. Note: SI = Social Influence; AI = Adoption Intention; UB = Usage Behaviuor; PR = Perceived Reliability; SE = Self-efficacy; PHT = Perceived Health Threat; PU = Perceived Usefulness; PEU = Perceived Ease of Use; ATU = Attitude Toward Using

Full size image

The results of the SEM are depicted in Fig. 3. Model validation involves analyzing the parameters of the initial hypothesis model to understand the relationships between variables and their influencing mechanisms. The model path coefficient analysis revealed that hypotheses H1a, H1b, H2b, H2c, H3c, H4b, H5a, H6a, H6b, and H8a were supported, while hypotheses H2a, H3a, H3b, H4a, H7a, H7b and H7c were not supported. Detailed results are provided in Supplementary File 5. Additionally, the conceptual model demonstrated a strong predictive ability for adoption intention, with an R² value of 0.850. High R² values were also observed for UB (0.755), PU (0.904), PEU (0.887), and ATU (0.820), indicating the predictive validity of the study model. The detailed results of indirect effects and their significance levels are presented in Supplementary File 6. The comparative analysis of indirect effects, direct effects, and total effects in the model is presented in the form of bar charts in Supplementary File 7.

Results of the structural equation model analysis

Full size image

We initially conducted a univariate analysis to examine the relationship between patient characteristics and their intention to adopt telemedicine services. The findings revealed that having cancer or chronic respiratory diseases (CRD) had a significant impact on intention: cancer negatively influenced adoption(t=-5.377, P<0.001), whereas CRD had a positive effect(t = 2.880, P = 0.004) (Supplementary File 8). Patients with chronic conditions displayed varying levels of intention to adopt telemedicine services based on their educational level(F = 15.094, P<0.001) and monthly income(F = 7.472, P<0.001). Using Lasso regression, nine key variables were identified: ‘hypertension’, ‘coronary heart disease’, ‘cancer’, ‘osteoporosis’, ‘chronic respiratory disease’, ‘other chronic diseases’, ‘education level’, ‘monthly income’, and ‘required care’. Further analysis with a generalized linear regression model revealed that a diagnosis of cancer had a significant negative effect on the adoption of telemedicine services (β = -0.147, P < 0.001), indicating that patients with cancer demonstrated lower adoption intentions. In contrast, education level (β = 0.032, P < 0.001) and monthly income (β = 0.009, P = 0.049) were significantly positively associated with adoption intentions, suggesting that patients with higher educational attainment and income levels were more likely to utilize telemedicine services. Additionally, patients who required caregiving exhibited significantly higher willingness to adopt telemedicine (β = 0.034, P = 0.007). Detailed results are presented in Table 5.

This study aims to explore the key factors influencing the adoption of telemedicine services among middle-aged and older patients with chronic conditions in rural areas of China. The results indicate that social influence, attitude toward use, and perceived usefulness are the most significant factors, all of which directly or indirectly affect patients’ intention to adopt telemedicine. In particular, SI, PU, and ATU exerted significant positive effects on adoption intention. The study also found that PEU did not have a direct impact on adoption intention but played an indirect role through perceived usefulness and attitude toward use. In addition, individual factors such as educational level, monthly income, caregiving needs, and the presence of a cancer diagnosis were also found to significantly influence patients’ intention to adopt telemedicine services. These findings suggest that the adoption of telemedicine among rural middle-aged and older patients with chronic conditions is affected by multiple factors, particularly those related to the social environment, technology perceptions, and personal characteristics.

Our research found that the HITAM model (SI, ATU and PU) was significantly associated with the behavioral intention to adopt telemedicine compared to other constructs. This finding aligns with previous studies on the intention to adopt other e-health technologies [18,19,20, 43, 44]. Furthermore, this study found that PEU did not have a direct and significant impact on patients’ intention to adopt. Instead, PEU influenced adoption intention indirectly by positively affecting perceived usefulness and attitude toward use, aligning with Kim’s original findings during the development of the HITAM model [29]. This may be attributed to the unique characteristics of the target population: middle-aged and older patients in rural areas generally lack experience with information technology [45, 46]. Specifically, the lack of experience with information technology makes it difficult for these patients to independently assess the technical details and functional advantages of telemedicine. As a result, they tend to rely more heavily on the opinions and recommendations of others within their social environment, such as family members, friends, and community healthcare providers, during their decision-making process. This explains the significant influence of SI on their intention to adopt telemedicine. Furthermore, due to their limited capacity for self-exploration and mastery of new technologies, these individuals are more inclined to judge the value of telemedicine based on its PU, rather than on specific system functionalities. In other words, they are primarily concerned with whether telemedicine can effectively improve their health conditions. Additionally, when confronted with the uncertainties and potential difficulties associated with new technologies, these patients tend to develop relatively stable ATU, which substantially determines their ultimate behavioral intentions. In addition, this suggests that while operational simplicity remains important for middle-aged and older patients in rural areas, its influence is primarily realized through enhancing perceived usefulness and fostering a positive attitude toward technology adoption. In the context of telemedicine services, PEU emphasizes users’ experiences in learning and utilizing the service [47]. Therefore, service providers should focus on optimizing the user interface and interaction design. This includes simplifying sign-up and login procedures, providing clear navigation menus, and ensuring quick access to commonly used features. Easily accessible online assistance, such as Frequently Asked Questions (FAQs) and video tutorials, can help users quickly resolve common issues. Regular user training sessions not only enable patients to use telemedicine services more effectively but also provide valuable feedback for service refinement [48]. This bidirectional interaction contributes to the continuous improvement of telemedicine services, aligning them more closely with users’ needs and expectations, and ultimately enhancing the overall user experience and satisfaction. Equally important is fostering a positive attitude toward telemedicine through targeted promotional efforts and user education, which plays a crucial role in increasing users’ intention to adopt such services. At the same time, it is recommended that telemedicine services be closely integrated with existing rural healthcare infrastructure, such as township health centers and village clinics. Involving community doctors in the promotion and implementation of telemedicine can facilitate broader adoption and ensure the long-term sustainability of these services in rural areas.

Unexpectedly, this study found that among rural middle-aged and older patients with chronic conditions utilizing telemedicine services, PR did not exert a positive effect on PU, but rather showed a slight negative correlation (z = -0.051). This result contrasts with existing reports and empirical findings [18, 29]. However, given the small absolute value of the coefficient, the negative correlation between PR and PU is relatively weak, indicating that its actual impact is limited. The possible explanation for this discrepancy lies in the distinct composition of the study participants compared to previous studies [18, 29]. The participants in this study had chronic conditions, and may perceive telemedicine services as technically reliable, yet feel that the content of the services may not fully address their specific health needs. Given that telemedicine services currently lack sufficient personalization and precision, this may influence their perception of the technology’s utility. Strategies to enhance the perceived usefulness of telemedicine services for this particular demographic could focus on improving service personalization, offering technical training and support, empowering users with a sense of control and privacy protection, and enhancing their comprehension and trust in telemedicine services through education [49]. These measures not only promote patient acceptance of telemedicine services but also ensure that the technology effectively meets the health management needs of middle-aged and older patients with chronic conditions in rural areas. It can be seen that a comprehensive set of strategies must be applied to enhance the intention to adopt telemedicine services among middle-aged and older rural patients with chronic conditions. SI, SE, PR, and PHT were the key peripheral variables in the telemedicine adoption intention model for this demographic. Strategies can focus on boosting social influence by involving community leaders and experienced users in promoting telemedicine benefits [50, 51]. Improving SE can be achieved through tailored technical training and peer support [52]. Enhancing PR involves building trust through transparent service standards and professional certification [53]. Last, leveraging health education and personalized risk assessment can enhance the perception of health threats, thus increasing telemedicine adoption intention [54]. These strategies should not only address technical aspects but also consider the psychosocial needs of patients to ensure that telemedicine services effectively meet the health management requirements of rural middle-aged and older individuals with chronic illnesses [46]. Moreover, PU directly influences patients’ perceptions of the value of adopting telemedicine technology and their attitude toward using it. To enhance patients’ perceived usefulness of telemedicine services, it is important to clearly demonstrate how these services address their actual health needs and improve their quality of life. Providers can use real patient stories and data to showcase the specific benefits of telehealth services, such as lower rates of acute events and better management of chronic conditions [8, 55, 56].

Personal factors, including education level and monthly income, have been identified as significant influences on patients’ intentions to use telemedicine services, aligning with findings from prior studies [45]. Higher education levels may enhance individuals’ abilities to process and evaluate health-related information, thereby increasing their propensity to explore and adopt innovative healthcare delivery models. For patients with lower educational attainment, local governments, in collaboration with healthcare institutions, could implement “Digital Health Assistance Programs.” These programs may include distributing user-friendly devices at no cost and providing personalized guidance from village-based coordinators, which could help mitigate the adverse impact of low education levels on technology adoption. Similarly, patients with higher income levels typically enjoy greater financial security, making them more inclined to invest in health-related innovations. In contrast, individuals with lower incomes may be reluctant to use telemedicine services due to concerns regarding out-of-pocket expenses. This indicates that government subsidies could serve a moderating role by alleviating financial burdens and enhancing adoption rates among low-income populations. For instance, expanding the reimbursement scope of the New Rural Cooperative Medical Scheme (NRCMS) to include telemedicine services, such as remote consultation fees, could significantly lower entry barriers for economically disadvantaged groups [57].

Furthermore, significant differences were observed in patients’ intention to adopt telemedicine services based on the type of chronic conditions. The results of the GLM analysis indicated that a diagnosis of cancer was negatively associated with the intention to adopt telemedicine, which may be attributed to the complexity of cancer treatment and a stronger preference for in-person consultations to receive more comprehensive and accurate medical care. Additionally, cancer patients may place greater value on direct communication and emotional support from healthcare providers [57], which are key strengths of traditional face-to-face medical encounters. However, only 54 cancer patients were included in this study, and further research is needed to validate these findings. In contrast, univariate analysis revealed that patients with chronic respiratory diseases showed a higher intention to adopt telemedicine. This may be due to their greater need for regular self-monitoring and the ability of telemedicine to provide convenient tools for daily health tracking [59, 60]. Moreover, patients in need of care showed higher adoption intentions of telemedicine services, emphasizing the convenience and reduced burden that such services offer. These patients, restricted by health limitations from easily accessing distant senior care facilities, value the continuity of care and monitoring that telemedicine provides. For this group of patients, telemedicine service providers should focus on offering customized telehealth monitoring and management plans. Thus, enhancing the acceptance of telemedicine services among middle-aged and older rural patients with chronic conditions necessitates a thorough evaluation of factors including education level, economic status, types of chronic conditions, and care requirements. Our study findings indicated that specific patient characteristics could impact an individual’s intention to engage with telemedicine. Further investigation is warranted to delve into these associations more extensively.

The strength of this study lies in its multicenter design, which ensured a broad representation of the sample. The research focused on the intention of rural middle-aged and older patients with chronic conditions to adopt telemedicine services, examining the influencing factors with particular attention to the role of demographic characteristics in shaping adoption intentions. This provides a theoretical foundation for exploring the potential of telemedicine to enhance healthcare accessibility in remote rural areas. Furthermore, as a post-pandemic study, this research offers valuable empirical data and policy insight for the promotion and implementation of telemedicine services, which is of significant practical relevance.

Despite its comprehensive consideration of various factors, this study had several limitations. First, while the sample encompassed multiple regions in Guizhou Province, the geographical scope was confined to western China, which may not fully represent the situation of rural middle-aged and older patients with chronic conditions in other regions. Future research would benefit from conducting comparative studies across broader areas to assess whether the identified factors have universal applicability. Second, the cross-sectional design of this study, which collected data at a single time point, limited causal inference. Although SEM provides a thorough analysis of relationships between variables, it cannot establish the temporal sequence of these variables. Therefore, future research could adopt a longitudinal design to track changes in patient attitudes and behaviors toward telemedicine, thereby exploring the long-term effects of various influencing factors. Additionally, this study did not collect information on whether participants had used telemedicine services. Therefore, the potential impact of this variable on the regression model was not considered. Future research could consider collecting such information and incorporating it into the analysis to more comprehensively assess the factors influencing the intention to adopt telemedicine.

Download references

L.Y. conceptualized the study and contributed to data collection, analysis, and writing of the original draft. Qiao.L. was responsible for data analysis, project administration, and supervision, and contributed to writing, review, and editing of the manuscript. Qin.L. contributed to writing, review, and editing of the manuscript. T.W. participated in data collection and analysis. S.P., X.F., W.W., C.Y., L.Z. and M.C. all contributed to data collection. All authors contributed to the critical revision of the manuscript and approved the final version.

Correspondence to Qiaoxing Li.

The authors declare no competing interests.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions