Digital health readiness, technology acceptance, and organisational readiness for nurse-led endocrine care in China Authors: Jilin Ren, Xiaoyun Wang, Yaoji Liu, Jinfeng Su Published: 2026 ABSTRACT AbstractBackgroundDigital health is becoming central to diabetes care, yet the preparedness of the nursing workforce remains uneven. This study examined digital health readiness, technology acceptance, and perceived organisational readiness for nurse-led telehealth and remote glucose monitoring among endocrinology nurses in China. It also explored individual and organisational factors linked to readiness and behavioural intention to use these tools.MethodsWe conducted a multi-centre cross-sectional study between October 2024 and January 2025 in five public general hospitals in Shanxi Province, China. Eligible endocrinology and diabetes nurses completed a structured questionnaire on demographic and professional characteristics, exposure to digital tools, digital health readiness, Technology Acceptance Model constructs, and organisational readiness. We also undertook pre-specified exploratory subgroup comparisons by recent training, age, and implementation setting. Quantitative data were analysed using descriptive statistics, correlations, subgroup comparisons, and multiple linear regression. Open-ended responses and semi-structured interviews were analysed thematically to explain the quantitative findings.ResultsA total of 230 nurses completed the survey. Overall digital health readiness was moderate to high. Nurses reported strong openness and motivation, but lower perceived organisational support and resources. Telephone follow-up and WeChat-based communication were common, whereas video consultations and structured remote monitoring were less common. Nurses who had received recent digital health training reported broader exposure to digital tools and more favourable readiness and intention patterns than those without training. Perceived usefulness was the strongest correlate of behavioural intention, followed by perceived ease of use, digital readiness, and organisational readiness. Qualitative findings highlighted limited structured training, the “invisible” nature of digital workload, and the importance of leadership and a shared implementation vision.ConclusionsIn this sample, endocrinology nurses in Chinese general hospitals showed generally positive attitudes towards nurse-led digital care, but wider adoption appeared to be constrained by uneven exposure to advanced tools and only moderate organisational readiness. Efforts to strengthen digital endocrine care should focus on practical training, usable systems, and organisational support that recognises digital work as part of routine nursing care.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12913-026-14779-6.Keywords: Digital readiness, Digital health, Telehealth, Remote patient monitoring, Diabetes services, Nursing workforce, Endocrinology nursing, Technology acceptance, Organisational readiness, China Introduction Diabetes has become one of the most pressing chronic health challenges globally. Recent estimates suggest that more than 10% of adults aged 20–79 years—over half a billion people—are living with diabetes worldwide, with numbers projected to rise further by 2045 [1]. The International Diabetes Federation’s 2025 Diabetes Atlas reports that more than half a billion adults live with diabetes worldwide and that diabetes already accounts for around USD 1 trillion in annual health expenditure [2]. China carries the heaviest national burden: an estimated 117 million people had diabetes in 2021, accounting for about 22% of the global diabetes population, and adult prevalence has been reported at roughly 12–13% based on representative survey data [3]. This epidemiological profile places substantial and growing pressure on hospital‑based and outpatient endocrinology services in China, where clinicians and nurses must support long‑term self‑management, frequent treatment adjustments and early detection of complications for large numbers of adults with type 2 diabetes and other endocrine disorders. In this context, telemedicine and broader digital health innovations are increasingly promoted as strategies to sustain and improve diabetes care. Narrative and guideline reviews emphasise that telemedicine modalities—including telephone and video consultations, asynchronous messaging, and remote transmission of blood glucose or continuous glucose monitoring (CGM) data—can improve glycaemic control, increase patient satisfaction and reduce the need for in‑person visits, particularly for people using insulin or living in remote areas [4]. Evidence from controlled trials and implementation studies indicates that remote monitoring of glucose data and virtual clinic models are associated with clinically meaningful reductions in HbA1c and better risk‑factor control compared with usual care [4]. At the same time, emerging literature on digital health in diabetes highlights considerable variation in real‑world uptake across settings and populations, and emphasises that digital interventions often require re‑design of workflows, new competencies for health professionals and supportive organisational infrastructure to achieve sustainable impact [5]. Nurses are central to the success of these digital models of care. In diabetes and endocrinology services, nurses typically lead patient education, support medication titration, interpret SMBG or CGM trends and provide ongoing behavioural and psychological support. Reviews and primary studies of nurse‑led or telenursing interventions show that when nurses use structured telemonitoring, messaging platforms or mobile health (mHealth) tools to provide education and follow‑up, patients often achieve better glycaemic outcomes, improved self‑management and higher satisfaction than with traditional clinic‑based care alone [6]. An umbrella review of telenursing in chronic disease management concluded that nurse‑led telehealth is a “key ally” in improving clinical and patient‑reported outcomes in diabetes, while also noting that benefits are contingent on adequate training, staffing and integration into existing systems [7]. A recent systematic review of nursing‑led digital health interventions for diabetes similarly found consistent positive effects on glycaemic control and self‑care, but highlighted implementation challenges related to digital literacy, workload and organisational support [6]. These findings point to the nursing workforce as both a major opportunity and a potential bottleneck for digital transformation of diabetes care. Workforce digital readiness has therefore become a key focus of contemporary digital health policy and research. Studies in several countries have begun to conceptualise and measure “digital health readiness” among patients, health professionals and organisations, often as a multidimensional construct that includes access, skills, confidence, motivation and support for using digital tools [8]. The Digital Health Readiness Questionnaire and related instruments provide structured ways to assess readiness and identify gaps, although they were originally developed mainly for patients rather than clinicians [8]. A multi‑method study by Dermody and colleagues demonstrated that nurses’ readiness for specific digital health technologies varies substantially across practice settings and roles, and is shaped by factors such as prior exposure, education, attitudes and organisational culture [9]. Complementary work with healthcare leaders suggests that leadership perspectives on digital readiness and change capacity can either accelerate or hinder the implementation of digital health initiatives at hospital level [10]. However, most existing studies examine general hospital nurses or mixed groups of health professionals in high‑income settings; very little is known about the digital readiness of specialised endocrinology nurses in large Chinese general hospitals, where diabetes burden is particularly high and digital health infrastructure is expanding quickly. To explain why clinicians adopt or resist new technologies, many health informatics studies draw on the Technology Acceptance Model (TAM). TAM posits that perceived usefulness and perceived ease of use shape attitudes and behavioural intentions toward technology, which in turn relate to actual use [11]. A systematic review of TAM in health informatics concluded that perceived usefulness and ease of use are consistently strong predictors of clinicians’ and patients’ intentions to use electronic health records, telemedicine systems and other digital tools, including in nursing populations [11]. Nevertheless, TAM focuses primarily on individual‑level beliefs and does not explicitly incorporate organisational or contextual determinants of adoption. Implementation science research emphasises that organisational readiness for change—the shared commitment and collective confidence of organisational members to implement a specific innovation—is an important precondition for successful implementation [12]. Weiner’s theory conceptualises organisational readiness as a shared psychological state characterised by change commitment and change efficacy, while Shea and colleagues developed and validated the Organizational Readiness for Implementing Change (ORIC) measure to operationalise this construct in healthcare settings [13]. A recent systematic review of organisational readiness in healthcare suggests that higher readiness is associated with stronger implementation of new clinical practices, including digital innovations, but also notes that empirical applications in nursing and digital health remain limited [14]. Taken together, current evidence indicates that nurse‑led digital diabetes interventions can be effective but their success depends on both individual nurse‑level factors (skills, attitudes, perceived usefulness and ease of use) and organisational conditions (leadership, infrastructure, collective change commitment and efficacy) [6]. Yet few studies have integrated these perspectives in a single, theory‑driven model, and almost none have focused on endocrinology nurses in China, despite the country’s disproportionate share of the global diabetes burden and rapid expansion of telemedicine and hospital digitalisation [3]. In particular, there is a lack of empirical data on how endocrine nurses in Chinese general hospitals perceive their own digital readiness, how they appraise nurse‑led telehealth and remote glucose monitoring in terms of usefulness and ease of use, and how they view their organisations’ readiness to implement and sustain such services. Without this information, it is difficult for nurse leaders and policymakers to design targeted training and organisational interventions to support safe and scalable digital endocrine care. Guided by TAM and organisational readiness theory, we conceptualised digital adoption as the product of inter‑related individual‑level and organisational‑level factors. We hypothesised that organisational readiness and digital health readiness would be positively associated, and that perceived usefulness, perceived ease of use, digital readiness and organisational readiness would all be positively associated with nurses’ behavioural intention to use telehealth and remote glucose monitoring. The present multi-centre study addresses this gap by examining digital health readiness, technology acceptance and perceived organisational readiness among endocrinology nurses working in five general hospitals in Shanxi Province, China. Specifically, the study addressed the following research questions: What are the levels of exposure to digital tools, digital health readiness, perceived usefulness, perceived ease of use, attitudes and behavioural intention regarding nurse-led telehealth and remote glucose monitoring among endocrinology nurses? What are the levels of exposure to digital tools, digital health readiness, perceived usefulness, perceived ease of use, attitudes and behavioural intention regarding nurse-led telehealth and remote glucose monitoring among endocrinology nurses? What is the level of perceived organisational readiness for implementing nurse-led digital endocrine services? What is the level of perceived organisational readiness for implementing nurse-led digital endocrine services? What individual and organisational factors are associated with digital health readiness and behavioural intention to use telehealth and remote glucose monitoring? What individual and organisational factors are associated with digital health readiness and behavioural intention to use telehealth and remote glucose monitoring? By focusing on a specialised nursing workforce in a high-burden region, and by explicitly combining digital readiness, TAM constructs and organisational readiness in a single analytic model, the study seeks to generate evidence that can inform digital-health training, workforce development and implementation strategies for endocrinology nursing in China and similar settings. Methods Study design This study used an analytical cross-sectional survey design with an explanatory qualitative component. The quantitative phase measured digital health readiness, technology acceptance, and perceived organisational readiness for nurse-led telehealth and remote glucose-monitoring among endocrinology nurses. The qualitative phase comprised semi-structured interviews with a purposive subsample and analysis of open-ended survey responses to help explain the quantitative patterns. The theoretical framework for the study was guided by the Technology Acceptance Model (TAM) and Weiner’s theory of organisational readiness for change [11, 12]. Digital health readiness was conceptualised as a multi‑dimensional construct encompassing nurses’ confidence, technical skills, prior experience, and motivation to engage with digital tools, a definition informed by recent conceptual work on readiness frameworks [15]. Consequently, the study design integrated individual‑level technology acceptance, personal digital readiness, and organisational readiness into a comprehensive explanatory model for the adoption of nurse‑led digital endocrine care. As summarised in Fig. 1, the overall mixed‑methods design, setting and analytic approach are shown for clarity. Study design and data collection overview for the multi‑centre survey Setting The study was conducted across five large public general hospitals in Shanxi Province, northern China. These facilities are tertiary or high-level secondary hospitals providing comprehensive inpatient and outpatient services, including specialised endocrinology and diabetes care. Hospitals were selected purposively to meet four criteria: (1) they provided dedicated endocrinology inpatient and outpatient services, (2) they had established electronic medical records and at least basic teleconsultation capacity, (3) local nursing managers agreed to support recruitment and coordination, and (4) the group of hospitals reflected different stages of digital service implementation. Each participating hospital operated at least one endocrinology ward and a specialist clinic. At the time of data collection, three hospitals had established formal telehealth services for diabetes follow-up, while the remaining two were piloting telephone- or video-based consultations and remote blood glucose review. This variation in digital maturity provided a useful context for examining readiness and technology acceptance within a single provincial health system. Participants and sampling The target population consisted of registered nurses working in endocrinology-related clinical areas across the five hospitals, including inpatient endocrinology wards, diabetes clinics, and general medical wards with a high volume of endocrine patients. Eligibility criteria required nurses to be registered with the Chinese Nursing Association, to have completed at least six months of continuous employment at their current hospital, and to spend at least 50% of their working time providing direct care or education to adult patients with endocrine conditions, mainly diabetes. Nurses were excluded if they held purely administrative posts, were on long-term leave during the study period, or were rotating through the department for short-term training placements. These criteria were intended to ensure that participants had current and sustained exposure to endocrine nursing workflows, including bedside education, follow-up communication, and the local digital systems used in practice. A census sampling strategy was employed to invite all 318 nurses who met the eligibility criteria based on hospital human resources records. This approach was chosen to maximise statistical power. Based on standard guidelines for multiple regression analysis (requiring approximately 10–15 participants per predictor variable), a minimum sample of 120 nurses was deemed necessary to support models with up to ten predictors [13]. Data collection was conducted between October 2024 and January 2025. Response rates and exclusions are reported in the Results section. For the qualitative component, survey respondents indicated their willingness to participate in a follow‑up interview. From this pool, purposive sampling was used to select 18 participants to ensure diversity regarding hospital site, years of nursing experience, digital readiness scores (selecting from both upper and lower quartiles), and extent of exposure to telehealth. This facilitated the collection of rich narrative data to complement statistical findings. Measures Demographic and professional characteristics The questionnaire assessed demographic variables including age, gender, highest nursing qualification, total years of nursing experience, and years of specialist experience in endocrinology or diabetes care. Professional characteristics collected included current position (staff nurse, senior nurse, specialist nurse, or nurse manager), main clinical setting (inpatient, outpatient, or mixed), employment status, and typical weekly workload (patients per shift and shifts per week). Additionally, participants reported whether they had completed any formal training related to digital health, telemedicine, or health information technology in the preceding two years. Exposure to digital tools in endocrine care To quantify the digital context of practice, nurses reported their frequency of use for six specific digital tools relevant to diabetes nursing: (i) telephone consultations, (ii) video consultations, (iii) remote review platforms for continuous glucose monitoring (CGM) or connected glucometers, (iv) secure patient portals or messaging systems (e.g., WeChat), (v) mobile apps recommended for patient self‑management, and (vi) structured remote patient‑monitoring programmes. Responses were recorded on a five‑point Likert scale ranging from “never” to “very often”. These items were informed by existing literature on digital models of endocrine care [10]. For analysis, a digital exposure index (range 0–6) was calculated by summing the number of distinct tools each nurse reported using at least “sometimes”. Higher scores therefore represented broader routine exposure to digital care modalities rather than greater intensity of use of a single tool. Individual digital health readiness Digital health readiness was measured using a 21-item scale developed specifically for the Chinese nursing workforce. The scale development was informed by Dermody and colleagues’ evaluation of nursing readiness and existing digital health readiness questionnaires [9]. The scale assesses three conceptual domains: digital skills and confidence, openness and motivation to learn new technologies, and perceived support and resources. Items were drafted in English, translated into simplified Chinese, and tailored to the context of nurse‑led telehealth (e.g., “I feel confident accessing and interpreting continuous glucose monitoring data through digital platforms”). Each item was rated on a five-point Likert scale from 1 (strongly disagree) to 5 (strongly agree). Domain scores were calculated as the mean of items within each domain, and the overall readiness score was the mean of all 21 items, with higher scores indicating greater readiness. Exploratory factor analysis (principal axis factoring with oblique rotation) supported a three-factor structure with factor loadings > 0.50. The scale demonstrated high internal consistency (Cronbach’s alpha = 0.93), and the total mean score was used as the primary indicator of readiness in regression analyses. Technology acceptance constructs Constructs from the Technology Acceptance Model (TAM) were assessed using items adapted from previous studies on nursing acceptance of telemonitoring systems [16]. Subscales included perceived usefulness (6 items), perceived ease of use (6 items), attitude towards technology (3 items), and behavioural intention to use nurse‑led telehealth (3 items). Items were adapted to specifically refer to “telehealth and remote glucose‑monitoring tools”. Ratings were provided on a five‑point Likert scale, with higher scores indicating more positive perceptions or stronger intentions. Subscale scores were calculated as item means so that all TAM domains could be interpreted on the same 1–5 metric. Organisational readiness Perceived organisational readiness was measured using the 12‑item Organizational Readiness for Implementing Change (ORIC) instrument [13]. This validated tool comprises two subscales: change commitment (5 items) and change efficacy (7 items). The referent for the scale was defined as “implementing nurse‑led telehealth and remote glucose‑monitoring services”. ORIC items were rated on the same five-point Likert format and averaged to produce commitment, efficacy, and total scores. Two additional items were included to assess perceptions of local IT support and hardware availability; these items were analysed descriptively as contextual indicators and were not included in the ORIC total score [10]. Instrument validity and translation The initial English questionnaire was drafted by a multidisciplinary team comprising endocrinology nurse researchers and a health informatics specialist. Content validity was established by a panel of five experts (two senior endocrinology nurses, an endocrinologist, a nursing academic, and a medical informatics researcher). The item-level content validity index (I-CVI) exceeded 0.80 for all retained items, consistent with the recommended threshold for acceptable content validity [28]. The instrument underwent a rigorous translation process. It was translated into simplified Chinese by a bilingual nursing academic and back‑translated into English by an independent translator blinded to the original text. Discrepancies were resolved through consensus to ensure conceptual equivalence [17]. A pilot test was conducted with a convenience sample of 12 endocrinology nurses from a non‑participating hospital to verify item clarity, length, and readability. Minor adjustments were made to the wording of technical items based on pilot feedback. Data collection procedures Questionnaire administration was facilitated by a designated nurse coordinator at each hospital. Eligible nurses received an invitation via hospital email or internal messaging apps containing a participant information sheet and a link (or QR code) to the secure online survey platform. To ensure inclusivity, paper questionnaires were distributed in sealed envelopes to nurses with limited access to digital devices during working hours. Completion time was approximately 20–25 min. Participation was voluntary and anonymous. Two reminder notifications were sent at two‑week intervals to encourage response. For the qualitative phase, interviews were scheduled with selected participants at their convenience, either in a private room at the hospital or via an encrypted video‑conferencing platform. Interviews lasted 30–60 min, were conducted in Mandarin Chinese, and were audio‑recorded with written informed consent. Data analysis Quantitative data were processed using IBM SPSS Statistics (version 29). The dataset was screened for accuracy; missing values (< 5% at item level) were handled using mean substitution for participants completing at least 80% of a scale. This approach was selected because the proportion of missing data at item level was very low (< 5\%), a condition under which mean substitution produces estimates comparable to more complex methods and is widely accepted in survey-based nursing research [29]. Although mean substitution can attenuate variance and correlations, the minimal amount of missingness in the present dataset limits any such bias. Descriptive statistics (means, standard deviations, frequencies) were used to summarise demographic characteristics, digital exposure, and scale scores. Internal consistency was evaluated using Cronbach’s alpha. Bivariate relationships were examined using Pearson correlation coefficients. In addition, we conducted pre-specified exploratory subgroup comparisons to aid interpretation of the main findings. Training status (recent digital health training: yes/no), age group (< 30, 30–39, and years), and implementation setting (sites with established telehealth services vs. pilot sites) were selected because they represented plausible sources of variation in everyday digital practice. These analyses were descriptive and stratified rather than formal tests of interaction, as the study was not powered to estimate multiple interaction terms across site and staff characteristics. They were intended to identify practically meaningful differences and to guide interpretation of the regression results rather than to support causal inference. Two multiple linear regression models were constructed to answer the research questions. Model 1 examined factors associated with digital health readiness. Predictors included age, nursing experience, qualification, digital health training, digital exposure index, and ORIC total score. Preliminary analyses excluded hospital site and endocrinology experience as they did not significantly improve model fit. Model 2 examined factors associated with behavioural intention to use telehealth. Predictors included perceived usefulness, perceived ease of use, digital readiness, ORIC score, training, and the digital exposure index. This model specification was grounded in TAM literature, which posits usefulness and ease of use as primary drivers of intention [11]. Regression assumptions (linearity, homoscedasticity, normality of residuals, and multicollinearity) were verified; variance inflation factors were all < 2.0. Qualitative data from open‑ended survey responses and interview transcripts were analysed using reflexive thematic analysis [18]. Audio recordings were transcribed verbatim. Analysis involved familiarisation, inductive code generation, and iterative theme development using NVivo 14 software. An initial coding framework was developed by the first author (JR) and refined through iterative discussion with two co-authors (XW and JS). Both inductive codes (data-driven) and deductive codes were utilised; the deductive coding was guided by the Technology Acceptance Model (TAM) constructs (e.g., perceived usefulness, perceived ease of use) and Weiner’s organisational readiness theory (e.g., change commitment, change efficacy, leadership support, workload recognition) [11, 12]. The integration of quantitative and qualitative findings followed an explanatory approach: quantitative results established general patterns (e.g., levels of readiness), while qualitative themes provided contextual explanations for these patterns (e.g., describing specific organisational constraints or training gaps). Rigour and trustworthiness Methodological rigour was strengthened by the use of established theoretical models (TAM, ORIC) and validated tools adapted for the specific cultural and clinical context. The multi‑centre design enhanced internal validity. Qualitative credibility was supported by purposive sampling for diverse perspectives and systematic thematic analysis. Dependability was ensured through an audit trail of analytic decisions and regular team debriefing to manage researcher reflexivity. Regarding positionality, the first author (JR) is a practising endocrinology nurse at the lead hospital, which afforded insider knowledge of clinical workflows and digital tool use but may also have introduced assumptions about shared professional experiences. To mitigate this, interview transcripts were independently reviewed by a co-author (JS) who holds a supervisory role outside direct clinical care, and reflective memos were maintained throughout the analytic process to document and interrogate the first author’s pre-understandings and their potential influence on code and theme development. Ethical considerations The study protocol received ethical approval from the Institutional Review Board of the lead hospital in Shanxi Province and the governance bodies of all participating sites. Participation was voluntary. Submission of the anonymous questionnaire constituted implied consent. Written informed consent was obtained prior to all interviews. All digital data were stored on password‑protected servers accessible only to the research team to ensure confidentiality. Results Sample and response Across the five participating general hospitals in Shanxi Province, 318 nurses met the eligibility criteria and were invited to take part in the survey. A total of 242 nurses returned questionnaires, giving a crude response rate of 76.1%. Twelve questionnaires contained more than 20% missing responses and were excluded, leaving 230 completed questionnaires for analysis (final analytic response rate 72.3%). The number of respondents per hospital ranged from 37 to 55, broadly reflecting the underlying size of endocrinology and diabetes nursing teams. Participant characteristics Participants were predominantly women (94.8%, n = 218), with a mean age of 33.7 years (SD 6.4, range 23–52). The mean duration of nursing experience was 11.2 years (SD 7.3), and nurses had worked in endocrinology or diabetes care for an average of 6.4 years (SD 5.1). Most respondents held a bachelor’s degree in nursing (72.2%), worked as staff nurses (62.2%), and were employed full‑time. Almost three fifths (58.3%) reported no formal training in digital health or telemedicine in the preceding two years. Table 1 presents detailed demographic and professional characteristics. Table 1Demographic and professional characteristics of participating nurses (N = 230)CharacteristicCategory n %Mean ± SDAge (years)–––33.7 ± 6.4< 307432.2–30–3910445.2–≥ 405222.6–GenderFemale21894.8–Male125.2–Highest nursing qualificationDiploma4117.8–Bachelor’s degree16672.2–Master’s degree or higher2310.0–Years of nursing experience–––11.2 ± 7.3< 5 years5825.2–5–9 years6227.0–10–14 years4820.9–≥ 15 years6227.0–Years in endocrinology/diabetes–––6.4 ± 5.1< 3 years8336.1–3–7 years7833.9–≥ 8 years6930.0–Current roleStaff nurse14362.2–Senior nurse5523.9–Specialist/diabetes nurse2310.0–Nurse manager93.9–Main clinical settingInpatient endocrinology/medical ward13860.0–Outpatient endocrinology/diabetes clinic6026.1–Mixed inpatient/outpatient3213.9–Employment statusFull‑time21493.0–Part‑time167.0–Digital health training (past 2 years)Yes9641.7–No13458.3– Demographic and professional characteristics of participating nurses (N = 230) Nurses cared for a mean of 8.9 patients per shift (SD 2.3) and worked an average of 5.3 shifts per week (SD 1.1). There were no statistically significant differences in age or years of endocrinology experience across hospitals (p > 0.10 for all comparisons), suggesting broadly comparable staff profiles between sites. Exposure to digital tools in diabetes and endocrine care Use of digital tools related to diabetes and endocrine care was widespread but uneven across specific technologies (Table 2). Telephone follow‑up and WeChat‑based communication or patient portals were the most frequently used tools, with 88.7% and 86.1% of nurses respectively reporting that they used these at least “sometimes” in their routine practice. Just over half of nurses (51.7%) reported recommending mobile applications to patients for diabetes self‑management at least sometimes. In contrast, only 49.1% used dedicated CGM or glucometer data platforms at least sometimes, and only 21.3% reported involvement in a structured remote patient‑monitoring programme. Table 2Frequency of use of digital tools for diabetes and endocrine care (N = 230)Digital toolNever n (%)Rarely n (%)Sometimes n (%)Often n (%)Very often n (%)≥Sometimes n (%)Telephone follow‑up with adults with diabetes5 (2.2)21 (9.1)84 (36.5)88 (38.3)32 (13.9)204 (88.7)Video consultation96 (41.7)68 (29.6)42 (18.3)18 (7.8)6 (2.6)66 (28.7)CGM / glucometer web portal54 (23.5)63 (27.4)62 (27.0)37 (16.1)14 (6.1)113 (49.1)WeChat / patient portal for messaging or results10 (4.3)22 (9.6)67 (29.1)86 (37.4)45 (19.6)198 (86.1)Recommending mobile apps for self‑management37 (16.1)74 (32.2)70 (30.4)35 (15.2)14 (6.1)119 (51.7)Participation in remote patient monitoring programme134 (58.3)47 (20.4)29 (12.6)15 (6.5)5 (2.2)49 (21.3) Frequency of use of digital tools for diabetes and endocrine care (N = 230) The mean digital exposure index, representing the number of different digital tools used at least sometimes (out of six), was 3.0 (SD 1.3; range 0–6). Nurses who had received recent digital health training had a significantly higher exposure index than those without training (mean 3.5 vs. 2.6, t = 5.15, p < 0.001). The distribution of frequency of digital tool use is illustrated in Fig. 2. Frequency of digital tool use for diabetes and endocrine care among nurses (stacked percentages by tool and frequency) Descriptive statistics and reliability of scales All multi‑item scales showed good to excellent internal consistency (Table 3). The total digital health readiness score had a mean of 3.74 (SD 0.52, range 2.24–4.90) on a 1–5 scale, with Cronbach’s alpha of 0.93. Subscale means suggested that nurses reported particularly high openness and motivation to use digital tools (mean 4.10, SD 0.57), but somewhat lower perceived organisational support and resources for digital work (mean 3.21, SD 0.74). Table 3Descriptive statistics and internal consistency of main scales (N = 230)Scale / subscaleItems (n)Possible rangeObserved meanSDCronbach’s αDigital health readiness – total211–53.740.520.93Digital skills and confidence81–53.630.630.91Openness and motivation71–54.100.570.88Perceived support and resources61–53.210.740.84Perceived usefulness (PU)61–54.010.560.92Perceived ease of use (PEOU)61–53.480.640.89Attitude toward telehealth/RPM31–53.880.680.86Behavioural intention to use telehealth/RPM31–53.790.770.90ORIC – commitment51–53.630.610.88ORIC – efficacy71–53.420.660.90ORIC – total121–53.510.590.92 Descriptive statistics and internal consistency of main scales (N = 230) Perceived usefulness of nurse‑led telehealth and remote glucose monitoring was generally high (mean 4.01, SD 0.56, α = 0.92), while perceived ease of use was more moderate (mean 3.48, SD 0.64, α = 0.89). Attitudes toward using these tools were favourable (mean 3.88, SD 0.68, α = 0.86), and the mean behavioural intention score was 3.79 (SD 0.77, α = 0.90). For organisational readiness, the ORIC commitment subscale had a mean of 3.63 (SD 0.61) and efficacy 3.42 (SD 0.66), with an overall ORIC score of 3.51 (SD 0.59) and α = 0.92. These values indicate moderate organisational readiness for implementing nurse‑led telehealth and remote monitoring, with somewhat stronger shared commitment than perceived capability. Exploratory factor analyses of the digital readiness and TAM items (not shown) supported the intended factor structure, with factor loadings above 0.50 on their hypothesised factors and minimal cross‑loadings, providing evidence of construct validity in this sample. Bivariate associations Correlation analyses demonstrated substantial and statistically significant associations among digital readiness, technology acceptance constructs, organisational readiness and exposure to digital tools (Table 4). Higher digital health readiness was strongly correlated with higher perceived ease of use (r = 0.60, p < 0.001) and perceived usefulness (r = 0.52, p < 0.001), and moderately correlated with organisational readiness (r = 0.45, p < 0.001) and greater digital exposure (r = 0.43, p < 0.001). Nurses who had received recent digital health training had higher digital readiness scores (r = 0.34, p < 0.001, point‑biserial). Table 4Correlations among key variables (N = 230)Variable1. Digital readiness2. PU3. PEOU4. Intention5. ORIC total6. Exposure index7. Training (0/1)8. Age1. Digital readiness––––––––2. Perceived usefulness (PU)0.52***–––––––3. Perceived ease of use (PEOU)0.60***0.49***––––––4. Behavioural intention0.57***0.68***0.55***–––––5. ORIC total0.45***0.49***0.41***0.46***––––6. Exposure index0.43***0.29***0.37***0.40***0.33***–––7. Digital health training (0 = no, 1 = yes)0.34***0.18**0.24***0.28***0.20**0.31***––8. Age (years)−0.14*−0.07−0.21**−0.10−0.05−0.08−0.25***–Note. PU = perceived usefulness; PEOU = perceived ease of use; ORIC = Organizational Readiness for Implementing Change; Exposure index = number of tools used at least sometimes. * p < 0.05; ** p < 0.01; *** p < 0.001 Correlations among key variables (N = 230) Note. PU = perceived usefulness; PEOU = perceived ease of use; ORIC = Organizational Readiness for Implementing Change; Exposure index = number of tools used at least sometimes. * p < 0.05; ** p < 0.01; *** p < 0.001 Behavioural intention to use telehealth and remote monitoring showed the strongest correlations with perceived usefulness (r = 0.68, p < 0.001), followed by perceived ease of use (r = 0.55, p < 0.001), digital readiness (r = 0.57, p < 0.001) and organisational readiness (r = 0.46, p < 0.001). Age showed small negative correlations with digital readiness (r = − 0.14, p = 0.03) and perceived ease of use (r = − 0.21, p = 0.002), suggesting slightly lower readiness and ease‑of‑use perceptions among older nurses, although these effects were modest in magnitude. Mean digital readiness and behavioural intention by training status are shown in Fig. 3. Digital readiness and behavioural intention by training status (means with 95% CIs) Exploratory subgroup patterns Exploratory subgroup analyses suggested that recent digital health training was the clearest differentiating factor across the sample. Compared with nurses without recent training, trained nurses reported broader exposure to digital tools (mean exposure index 3.5 vs. ) and consistently more favourable readiness and behavioural-intention patterns in Fig. 3. This pattern was not confined to a single measure; it aligned with more positive views of usefulness and ease of use, suggesting that training may strengthen both practical familiarity and confidence in the value of digital care. Age-related differences were smaller and more selective. Older nurses reported slightly lower perceived ease of use and modestly lower readiness, but age was not an independent correlate of readiness once training, exposure, and organisational readiness were taken into account. This finding suggests that apparent age differences may largely reflect unequal access to training opportunities and routine digital practice rather than age itself. Implementation setting also helped explain subgroup variation. Nurses working in hospitals where telehealth services were still being piloted more often described uncertainty about protocols, platform access, and role expectations, whereas nurses in more established services described clearer workflows and smoother integration of digital tasks into routine care. Together, these findings indicate that subgroup differences in this study were shaped more by modifiable implementation conditions than by fixed staff characteristics. The summary below synthesises the subgroup patterns to clarify their practical meaning. Subgroup contrastMain quantitative signalQualitative/contextual interpretationPractical implicationRecent digital health training vs. no recent trainingHigher exposure index (3.5 vs. 2.6) and more favourable readiness and intention patternsTraining appears to increase familiarity, confidence, and perceived value of digital carePrioritise repeated, role-specific digital health trainingYounger vs. older nursesOlder nurses reported slightly lower ease of use and modestly lower readiness, but age was not independently associated with readiness after adjustmentDifferences may reflect unequal access to training and routine practice rather than age aloneAvoid age-based assumptions; focus on support and opportunitiesEstablished telehealth services vs. pilot sitesMore uncertainty and less smooth workflow integration were reported in pilot settingsWorkflow maturity, local protocols, and leadership support shape daily digital practiceAssess local readiness and workflow fit before scaling services Factors associated with digital health readiness In the first multiple linear regression model, the outcome variable was the total digital health readiness score (Table 5, Model 1). Predictor variables included age, years of nursing experience, highest qualification (diploma vs. bachelor’s degree or higher), digital health training (yes/no), the digital exposure index and ORIC total score. Hospital site and years of endocrinology experience were explored in preliminary models but were not retained, as they were not significantly associated with readiness and did not improve model fit. Table 5Multiple regression models predicting digital readiness and behavioural intention (N = 230)PredictorModel 1: Digital readiness (1–5)Model 2: Behavioural intention (1–5)B95% CI p B95% CI p Constant1.470.94 to 2.00< 0.0010.28−0.26 to 0.820.31Age (years)−0.01−0.03 to 0.010.23–––Years of nursing experience0.01−0.00 to 0.020.19–––Higher qualification (≥ bachelor’s vs. diploma)0.09−0.01 to 0.190.07–––Digital health training (yes vs. no)0.270.16 to 0.38< 0.0010.140.00 to 0.280.047Exposure index (0–6)0.080.05 to 0.11< 0.0010.060.02 to 0.100.004ORIC total score (1–5)0.230.14 to 0.32< 0.0010.160.04 to 0.280.010Perceived usefulness (PU)–––0.470.34 to 0.60< 0.001Perceived ease of use (PEOU)–––0.180.06 to 0.300.004Digital readiness (total)–––0.230.09 to 0.370.001Model statisticsAdj. R² = 0.43F(6, 223) = 29.4< 0.001Adj. R² = 0.62F(6, 223) = 63.2< 0.001Note. A dash indicates the variable was not entered in that model. Years of nursing experience and qualification were included only in Model 1; TAM constructs were included only in Model 2 Multiple regression models predicting digital readiness and behavioural intention (N = 230) Note. A dash indicates the variable was not entered in that model. Years of nursing experience and qualification were included only in Model 1; TAM constructs were included only in Model 2 The overall model was statistically significant (F(6, 223) = 29.4, p < 0.001) and explained 43% of the variance in digital readiness (adjusted R² = 0.43). After adjustment for other factors, higher organisational readiness (B = 0.23, 95% CI 0.14–0.32, p < 0.001), greater exposure to digital tools (B = 0.08 per additional tool, 95% CI 0.05–0.11, p < 0.001), and having received digital health training (B = 0.27, 95% CI 0.16–0.38, p < 0.001) remained significant correlates of higher digital readiness. Highest qualification showed a small, non‑significant positive association with readiness (B = 0.09, 95% CI − 0.01–0.19, p = 0.07). Age and years of nursing experience were not independently associated with digital readiness after accounting for other variables. Variance inflation factors were all below 2.0, indicating no problematic multicollinearity. Factors associated with behavioural intention to use telehealth and remote monitoring In the second regression model, behavioural intention to use nurse‑led telehealth and remote glucose‑monitoring was the outcome. Predictors were perceived usefulness, perceived ease of use, digital readiness, ORIC total score, digital health training and the digital exposure index. The model was statistically significant (F(6, 223) = 63.2, p < 0.001), explaining 62% of the variance in behavioural intention (adjusted R² = 0.62). Perceived usefulness emerged as the strongest correlate of behavioural intention (B = 0.47, 95% CI 0.34–0.60, p < 0.001, standardised β = 0.43). Perceived ease of use (B = 0.18, 95% CI 0.06–0.30, p = 0.004, β = 0.18) and digital readiness (B = 0.23, 95% CI 0.09–0.37, p = 0.001, β = 0.19) were also significant independent correlates. Organisational readiness was positively associated with intention (B = 0.16, 95% CI 0.04–0.28, p = 0.010, β = 0.14), indicating that nurses who perceived their organisation as more ready for change reported stronger intention to use digital tools. Recent digital health training (B = 0.14, 95% CI 0.00–0.28, p = 0.047, β = 0.09) and greater exposure to digital tools (B = 0.06 per additional tool, 95% CI 0.02–0.10, p = 0.004) showed smaller but significant associations with higher behavioural intention. Qualitative findings from open‑ended responses and interviews Open‑ended responses were provided by 186 nurses (80.9% of the sample), and 18 nurses participated in individual semi‑structured interviews. Thematic analysis of the combined qualitative dataset identified three overarching themes that helped explain the quantitative findings: perceived gaps in training and infrastructure, the invisibility of digital workload, and the critical role of leadership and shared vision. The first theme, “Running faster than the system”, captured nurses’ sense that their interest in digital tools often outpaced available training and infrastructure. Nurses described feeling motivated to use telehealth and remote monitoring but lacking structured training on platforms and protocols, particularly for interpreting CGM data and handling technical problems. This theme was more prominent among younger nurses and those in hospitals where telehealth services were still in pilot phases. The second theme, “Digital work as invisible workload”, reflected the perception that reviewing remote data, documenting teleconsultations and troubleshooting technical issues added to nurses’ workload without always being recognised in staffing or workload metrics. Several nurses commented that digital tasks were often expected to be done in addition to, rather than instead of, face‑to‑face care, which contributed to ambivalence despite generally positive attitudes toward telehealth. This theme provides qualitative context for the moderate scores on perceived support and time resources in the quantitative data. The third theme, “Leadership and shared vision enable digital care”, highlighted the enabling effect of supportive nurse managers and hospital leaders. Nurses from departments with clear local protocols, dedicated IT support and visible endorsement from leadership described more confidence and willingness to engage with telehealth and remote monitoring. In contrast, in settings where digital initiatives were perceived as fragmented or driven externally, nurses expressed uncertainty about expectations and long‑term sustainability. Table 6 summarises the main themes and provides illustrative quotes. Table 6Themes from qualitative data on digital endocrine care (open‑ended responses and interviews)ThemeBrief descriptionIllustrative quote (example)Running faster than the systemNurses feel personally motivated to use digital tools but perceive gaps in structured training, platform access and clear procedures.“I want to use CGM platforms more, but I only learned by watching colleagues. There was no formal training or guidelines.” (Interview participant)Digital work as invisible workloadTelehealth and remote monitoring tasks are experienced as extra work that is not fully recognised or resourced.“After clinic I still need to check messages and glucose data at home time. It is care work, but it is not counted in our workload.” (Survey respondent)Leadership and shared vision enable digital careStrong, supportive leadership and a shared departmental plan make it easier for nurses to engage in digital care.“Our head nurse treats telehealth as part of routine care, gives us time and IT support, so everyone is more confident to try these new ways.” (Interview participant) Themes from qualitative data on digital endocrine care (open‑ended responses and interviews) Table 7 summarises the main quantitative and qualitative findings and their practical implications. Table 7Integrated summary of key findings and implicationsDomainMain findingInterpretationPractical implicationDigital readinessOverall readiness was moderate to high, with strongest scores for openness and motivation.Nurses were generally willing to engage with digital care but not equally supported to do so.Training should build on motivation while addressing skills gaps and workflow support.Technology useTelephone and WeChat-based communication were common, whereas video visits and structured RPM were less common.Low-threshold tools appear easier to embed in practice than more structured digital services.Services should move gradually from informal communication to protocol-based digital models.TrainingNurses with recent training had broader digital exposure and more favourable readiness and intention patterns.Training appears to be an important modifiable enabler of adoption.Hospitals should provide role-specific and repeated digital-health education.Organisational readinessORIC scores were moderate and were independently associated with readiness and behavioural intention.Organisational climate influences whether positive attitudes can translate into implementation.Leaders should assess readiness before rollout and address staffing, IT support, and workflow redesign.Qualitative insightsNurses described limited structured training, invisible digital workload, and the importance of leadership.The implementation gap is not only technical; it is also organisational and managerial.Digital nursing work should be formally recognised, resourced, and incorporated into routine service planning. Integrated summary of key findings and implications Discussion This multi-centre study offers a detailed assessment of digital health readiness, technology acceptance, and organisational readiness among endocrinology nurses in Shanxi Province, China. The findings suggest that this workforce is interested in nurse-led telehealth and remote glucose monitoring, but its readiness is uneven. Nurses reported strong openness to digital care, yet organisational support, infrastructure, and structured implementation were only moderate. Perceived usefulness was the strongest correlate of behavioural intention, but intention alone is unlikely to translate into sustained adoption when training, workflow clarity, and organisational support remain limited. The qualitative themes, especially the “invisible” nature of digital workload and the lack of structured training, help explain this gap between willingness and practice. Interpretation and comparison with previous literature The observation of high openness to digital tools alongside lower perceived support mirrors recent international and domestic findings. Consistent with Dermody et al. [9], who showed that nursing readiness differs across settings and depends heavily on organisational support, our participants reported strong personal motivation but lower scores for support and resources. A similar pattern was reported by Yu-Tong et al. [19] among Chinese clinical nurses, where willingness to engage in telehealth coexisted with concerns about workflow, role clarity, and practical support. In addition, the systematic review by Alotaibi et al. [20] suggests that digital capability improves most when training is repeated, practice-based, and supported by local facilitation. This helps explain why nurses in our study who reported recent training also showed broader digital exposure and more favourable readiness patterns. Our results extend this literature by focusing on a specialist endocrinology workforce and by showing that readiness is linked not only to attitudes, but also to exposure and organisational context. The specific profile of technology adoption, with widespread use of telephone and WeChat-based communication but limited use of structured remote patient monitoring (RPM) and CGM portals, also deserves attention. This pattern reflects the broader “Internet + Healthcare” landscape in China, in which low-threshold communication tools have spread faster than fully integrated clinical platforms [21]. Similar implementation gaps have been reported internationally. For example, Kirkland et al. [22] found that RPM programmes for diabetes often show clear clinical value but remain difficult to embed because of staffing, workflow, and reimbursement barriers. Our qualitative finding that digital tasks are experienced as “invisible workload” offers a plausible explanation for why simpler tools were more common in our sample: nurses may choose communication methods that fit existing practice even when more structured systems are available. In this sense, low-threshold tools may function as an entry point for digital care, but they are unlikely to substitute for formal protocols, time allocation, and implementation support when services become more complex. Theoretical constructs from the Technology Acceptance Model (TAM) and organisational readiness theory proved useful in this context. Perceived usefulness emerged as the strongest predictor of behavioural intention, which is in line with Rahimi et al. [11], who identified usefulness as a persistent determinant of technology acceptance in health informatics. However, the present findings also show that usefulness is only part of the picture. Perceived ease of use and digital readiness both remained independently associated with intention, suggesting that nurses are more likely to endorse digital care when systems appear practical, manageable, and compatible with their skill level. This is consistent with Strudwick’s review [16], which emphasised that ease of use and confidence remain central to nurses’ willingness to adopt new systems. Crucially, our study bridges individual- and system-level perspectives by demonstrating a clear link between organisational readiness (ORIC) and individual nurse outcomes. Nurses who perceived their hospitals as having stronger commitment and efficacy for change reported higher digital readiness and stronger intention to use telehealth and remote monitoring. This finding supports Weiner’s theory [12] and is consistent with studies linking organisational climate to digital implementation success in telehealth and electronic record implementation [23, 24]. In our data, organisational readiness did not replace the importance of individual beliefs; instead, it appeared to shape the conditions under which positive beliefs could be translated into action. This interpretation is also supported by the qualitative accounts: where local leadership, IT support, and workflows were clearer, nurses described more confidence in day-to-day digital care. The subgroup patterns also add nuance to interpretation. Training status differentiated nurses more clearly than age. Although older nurses reported somewhat lower ease-of-use scores, age was not independently associated with readiness in the adjusted model. This suggests that differences sometimes attributed to age may instead reflect unequal opportunities for training and routine digital practice. Likewise, qualitative accounts from pilot sites highlighted uncertainty about procedures and expectations, whereas nurses in more established services described clearer implementation pathways. Taken together, these subgroup findings indicate that readiness is context-sensitive and potentially modifiable, and they reinforce the practical value of investing in structured training and staged implementation rather than assuming fixed staff resistance. The qualitative emphasis on leadership and shared vision further reinforces Matthews and Burgess’s [25] argument that workforce capacity depends on visible, supportive leadership. In our sample, where leadership was described as proactive and telehealth was framed as part of routine care, nurses expressed greater confidence. Where leadership was fragmented, digital care was more likely to be seen as an unstable add-on to existing duties [10]. Implications for practice, education, and policy The findings have several implications for workforce development and service design in Chinese endocrinology. First, the strong relationship between recent training and readiness underscores the need for targeted educational interventions. Generic digital literacy is unlikely to be enough; nurses require role-specific preparation in interpreting remotely transmitted glucose data, managing virtual communication, documenting telehealth activity, and addressing digital ethics and privacy [19, 26]. Pre-registration, continuing education, and specialist endocrinology programmes should therefore include applied digital-care competencies rather than brief introductory content alone. Second, successful digital transformation requires organisational change as well as individual skill development. Hospital leaders need to address the “invisible workload” identified in the qualitative analysis by recognising tele-nursing as legitimate clinical work that requires time, staffing, and workflow support [22]. Readiness assessments such as ORIC may help leaders identify whether teams have the confidence, support, and infrastructure needed before introducing new services [24]. Third, service design should be incremental and nurse-informed. In this study, basic communication tools were much more common than structured monitoring platforms. This suggests that services may benefit from staged implementation, beginning with clear protocols, defined escalation pathways, and realistic staffing plans before expecting widespread use of advanced CGM portals or RPM systems. Involving nurses in workflow design may also improve perceived ease of use, which was an important correlate of intention in our model. Strengths, limitations, and future research This study strengthens the evidence base by integrating TAM and organisational readiness theory within a specialised nursing context in a high-burden diabetes region. It also combines survey data with qualitative findings, which improved interpretation of the statistical results. However, several limitations should be considered. First, the cross-sectional design prevents causal inference. Training, exposure, and organisational readiness were associated with more favourable outcomes, but the direction of these relationships cannot be confirmed. Second, the study relied mainly on self-report measures, which may have introduced recall error or social desirability bias and may overestimate actual competence or routine technology use. Third, although the qualitative data helped contextualise subgroup differences, the study was not powered to test complex interaction effects across multiple staff or site characteristics. The subgroup findings should therefore be interpreted as exploratory, and future work should test whether training effects differ systematically by age, role, or implementation stage. The use of mean substitution for handling missing data, although reasonable given the very low rate of item-level missingness (< 5%), may slightly underestimate standard errors and attenuate correlations; future studies with larger amounts of missing data should consider multiple imputation or maximum likelihood methods [29]. In addition, the digital readiness scale was adapted for this context and showed good internal consistency, but broader psychometric validation remains warranted. Finally, the sample was restricted to general hospitals in one province, limiting transferability to rural settings, primary care services, or hospitals with markedly different digital infrastructure and staffing models. Future research should focus on three areas. First, psychometric validation of the context‑specific digital readiness scale used here is warranted, alongside comparisons with established tools [15]. Second, longitudinal and interventional studies are needed to evaluate how specific training programmes or leadership interventions influence the trajectory of readiness and actual technology use over time. Finally, research should broaden to include the patient perspective and equity considerations. As digital health expands, understanding how nurse readiness interacts with patient readiness—particularly among older adults and underserved populations—is essential to prevent digital exclusion [27]. Conclusion In summary, this multi-centre study suggests that endocrinology nurses in Chinese general hospitals have generally favourable attitudes towards nurse-led telehealth and remote glucose monitoring, but that readiness remains uneven and closely linked to organisational context. Perceived usefulness, ease of use, digital readiness, and organisational readiness were all associated with stronger behavioural intention, while training and broader exposure to digital tools appeared to be important modifiable enablers. At the same time, qualitative findings showed that limited formal training, under-recognised digital workload, and inconsistent leadership support may impede implementation. These findings indicate that digital transformation in endocrinology nursing requires more than technology availability. It also depends on practical training, workflow redesign, leadership support, and organisational recognition of digital nursing work. For hospital leaders and policymakers in China and similar settings, these areas may be important targets for narrowing the gap between policy ambitions and day-to-day practice. For researchers and educators, the integration of digital readiness, TAM constructs and organisational readiness in this study offers a useful framework for future work to test and refine, with the ultimate aim of ensuring that nurse-led telehealth and remote monitoring contribute effectively and equitably to the care of people living with diabetes and other endocrine conditions. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Acknowledgments The authors wish to thank all participants for generously sharing their experiences, as well as the nursing staff at Shanxi Provincial People’s Hospital for their assistance with recruitment and interview arrangements. Author contributions Jilin Ren: Conceptualization, Methodology, Software, Investigation, Data Collection, Writing – Original Draft. Xiaoyun Wang: Investigation, Writing – Review & Editing. Yaoji Liu: Validation, Writing – Review & Editing. Jinfeng Su: Formal Analysis, Supervision, Project Administration, Writing – Review & Editing. All authors have read and approved the final manuscript. Funding This research received no external funding. Data availability The datasets used and analyzed during the current study are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate This study was approved by the Institutional Review Board of Shanxi Provincial People’s Hospital (No. (2025)-129). Written informed consent was obtained from all participants before enrollment in the study. All methods were performed in accordance with the relevant guidelines and regulations. This study was conducted in accordance with the principles of the Declaration of Helsinki (as revised in 2013) regarding ethical standards for research involving human subjects. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.