The study was conducted among nurses working in hospital or community settings using convenience sampling with open recruitment in Shanghai, China. At baseline, advertisements for the study were distributed through WeChat, one of the most popular social media in China. Interested nurses entered the study by scanning a QR code, and peer sharing and referral was used to increase the response rate. Eligibility was assessed using an inclusion and exclusion form on the first page of the survey. Inclusion criteria were age ≥18 years; registered nurses; aware of the content of the survey and willing to participate. Registered nurses are individuals who have completed relevant nursing professional education, obtained a nursing qualification certificate, and are registered and legally licensed to practice in medical institutions including hospitals and community settings in China. Those with any clinical diagnosis of mental health disorders were excluded. After the eligibility page, participants who clicked “I have read the informed consent form carefully and agree to participate” could respond to survey items. Implied consent to participate was indicated when participants provided responses to survey items. For assessing the one-month test-retest reliability, short massages containing a follow-up survey link were sent to participants, using telephone numbers collected at baseline; their names were not asked. The survey at baseline and one-month follow-up were programmed to allow only one completion per device to prevent duplicate submissions.

Sample size was calculated based on the requirement for conducting CFA. A total sample of 1000 participants is regarded as “excellent” in the statistical viewpoints and previously published literature [27]. For the sample of follow-up survey, a minimal sample size of 100 was recommended for a high reliability [28].

PETS-C brief was adapted from a 40-item instrument initially developed for assessing knowledge and perceptions of Telemedicine Technology in clinicians in Iran, each item was on a five-point Likert scale [20]. The original instrument included six factors as follows: knowledge (7 items), perception of the advantages (7 items), perception of the disadvantages (8 items), perception of necessity (6 items), perception of ease of usage (6 items), and perception of security (6 items). Kwan and colleges adapted the instrument in Chinese nurses through item modification and translation, and a broader concept of eHealth technology was used [14]. A four-factor structure with 19 items was yielded by conducting EFA and CFA. The four factors were knowledge (4 items), perception of the advantages (4 items), perception of the disadvantages (6 items), and perception of eHealth applications (5 items) [14]. Each item of PETS-C brief was scored on a five-point Likert scale from 1=“very low” to 5=“very high.” Total scores range 19‐95, with higher scores indicating better perceptions of eHealth technology. PETS-C Brief was set as a compulsory question, thereby no data were missing.

Sociodemographic characteristics included sex, age, educational attainment, marital status, monthly household income. Younger age [25,26] and higher educational attainment [24] were suggested as potential factors influencing nurses’ perception of eHealth technologies. Apart from the two variables, the selection of sex, marital status, and monthly household income was consistent with studies on Chinese nurses’ working experiences with Internet Plus Nursing Service for result comparison [29,30]. Working-related characteristics included year of employment, professional title, and setting.

CFA with diagonally weighted least squares estimation for ordinal data was conducted to verify the four-factor structure of PETS-C Brief previously reported [14], with full information maximum likelihood was used for handling missing data. The goodness of fit was determined by a combination of the following indicators: comparative fit index (CFI; ≥0.90), root mean square error of approximation (RMSEA; <0.08), and standardized root mean squared residual (SRMR; <0.08) [31]. We reported results of the χ² test for descriptive purpose but not for evaluating the model fit (cutoff for a good fit: P >.05), because the result is always statistically significant in large samples [32]. Standardized factor loadings of 19 items on the four factors were evaluated, with a value ≥0.70 indicating a strong measure of the factor [27]. Cronbach α was calculated for internal consistency reliability. One-month test–retest reliability was determined by intraclass correlation coefficients (ICC) calculated based on two-way mixed-effects model with absolute agreement, by the following criteria: poor (<0.4), fair (0.4‐0.6), good (0.6‐0.75), and excellent (>0.75) [33]. Associations of sociodemographic and working-related characteristics with PETS-C Brief total scores and subscores of four factors were analyzed using multivariable linear regression, with complete case analysis was used for handling missing data. Known-group validity was assessed by examining associations of age and educational attainment with PETS-C Brief total scores. All analyses were conducted using STATA (version MP 15.1; StataCorp). P<.05 was considered statistically significant.

The study procedures were carried out in accordance with the Declaration of Helsinki. Participants were informed of the right to refuse to participate or to withdraw consent to participate at any time without reprisal by contacting the research team; all data were deidentified and no compensation was provided to participants. The research protocol was approved by the Shanghai Ninth Peoples Hospital, Shanghai Jiao Tong University School of Medicine Ethics Committee (SH9H2024-T99-3).

Our validation study supported the four-factor structure of PETS-C Brief with satisfactory validity and reliability in Chinese nurses. The results of CFA showed a four-factor structure. Though the χ² test was statistically significant, this has been common in large samples [32]. Other model fit indicators including comparative fit index (CFI), standardized root mean squared residual (SRMR), and root mean square error of approximation (RMSEA) were within the prespecified cutoff values, suggesting that the four-factor model was acceptable. The original scale had 40 items on six items for assessing perception of telemedicine in clinicians in Iran, including knowledge, perception of the advantages, perception of the disadvantages, perception of necessity, perception of ease of usage, and perception of security [20]. The scale was then adapted in English for assessing perception of telemedicine in health care practitioners in Saudi Arabia, and knowledge, advantages, disadvantages, necessity, issues affecting telemedicine, and effectiveness obtained from patient feedback were identified as six factors [34]. The English version had two different factors compared with the original one, which might be due to cultural differences in understanding telemedicine. The Spanish adapted version reduced the item number to 13 and identified four factors including knowledge, disadvantages, utility, and knowledge of security in Ecuadorian clinicians [35]. Compared with telemedicine focusing on remote health care services, eHealth technologies include a broader range including mobile health, electronic health record, wearable devices, and online health education resources. Nurse-led eHealth technology use has been increasingly promoted by the Chinese government [21,22]. In the Chinese context, Kwan and colleges modified the scale into PETS-C Brief for assessing perceptions of eHealth technologies in nurses, and knowledge, perception of the advantages, perception of the disadvantages, and perception of eHealth applications were identified as four factors of the scale [14]. Our study complemented the study by the use of a large sample and confirmation of the four-factor structure. Nevertheless, more studies in nurse populations are warranted as the scale was less examined in this group than in clinicians.

Internal consistency reliability of PETS-C brief and its four factors were good in our study. Specifically, Cronbach α of the scale was higher than those conducted in Chinese nurses previously [11,14]. Changes in the sociocultural environment over years and the increased recognition of eHealth may have affected the way nurses understood and responded to the scale items, which in turn may have affected the internal consistency of the scale. Our study complemented previous findings by supporting one-month test–retest reliability of PETS-C Brief and its four factors, showing that the scale could be reliable over a short time. However, test–retest reliability findings must be interpreted with caution because of the relatively small sample size (n=123) and short time interval. Longitudinal studies following more nurses are needed to confirm the test–retest reliability of PETS-C Brief.

Known-group validity was supported by higher PETS-C Brief total scores associated with younger age and higher educational attainment. These findings were consistent with studies on digital divide in nurses showing that age [25,26,36] and educational attainment [11,24] could influence nurses’ perception of eHealth technologies. The study using PETS-C Brief in Chinese nurses further confirmed that higher educational attainment was associated with higher PETS-C Brief total scores [11]. Specifically, age was inversely associated with subscores of knowledge and perception of the disadvantage of eHealth technologies. Younger nurses may have earlier exposure in daily life and work, hence becoming more tech-savvy [36]. This may increase their knowledge and critical opinions of eHealth technologies. Educational attainment was positively associated with subscores of perception of the advantages and perception of eHealth applications. These findings were consistent with previous results on higher levels of digital literacy and comfort with technology adoption reported by nurses with higher educational attainment [37]. Nurses with higher educational attainment may have more trainings in use of eHealth technologies for clinical, research, and administration purposes, hence increasing their perceptions of advantages and eHealth applications. These findings supported the TAM, indicating that individual experience and training could influence perception of technologies [17]. Intervention and training programs can be delivered to nurses with older age and lower educational attainment for promoting eHealth technology use.

This study had several limitations. First, the original version of PETS-C Brief was initially developed to assess perceptions of telemedicine in Iranian clinicians, containing 40 items on six factors [20]. The scale was then adapted to 19 items on four factors in Chinese nurses through translation, item reduction, and validation [14]. Though poorly performing items were reduced and the new scale structure has been validated, differences in professional background, culture, and technology exposure may affect the meaning of items and factors. Second, the PETS-C Brief measures a broader concept of eHealth, which might reduce its usefulness for guiding targeted interventions adopting specific technologies, such as telemedicine. Third, convergent validation was unclear. Future studies may analyze correlations of PETS-C Brief with other widely used instruments such as TAM for confirming the convergent validation. such as instruments based on TAM. Fourth, participants were recruited using convenience sampling, and peer sharing or referral was used to increase the response rate. This may lead to mutual introduction, referring to the inclusion of a group of people with similar views, resulting in selective bias. Duplicate submissions by the same participant using different devices cannot be avoided. However, telephone numbers were checked, and no duplicate number was found. Future studies with random sampling are needed. Fifth, educational attainment and monthly household income were collected as categorical data, which inform future studies using continuous variables for providing more details. Last, all data were self-reported, which could be subjected to recall bias.

Our validation study supported the four-factor structure of PETS-C Brief with satisfactory validity and reliability in Chinese nurses. With the pervasion of advanced eHealth technologies, PETS-C Brief can be a useful tool for continuously assessing Chinese nurses’ perception of eHealth technologies.