This prospective validation trial was approved by the Ethics Committee of Fujian Provincial Hospital, and written consent was obtained from all participants. The study protocol was registered at the Chinese Clinical Trials Registry (http://www.chictr.org.cn, ChiCTR2100049555, 2/8/2021). We conducted this trial at Fujian Provincial Hospital, Fuzhou, China, in accordance with Consolidated Standards of Reporting Trials (CONSORT) guidelines.

The inclusion criteria were patients aged equal to or over 18 years, scheduled for elective surgery under general anesthesia, and familiar with smartphone use. Patients with cognitive disorders, inability to understand and read Mandarin, known alcohol or substance abuse, completion of a similar questionnaire before, or any conditions that impeded the completion of the questionnaire within 24 h after surgery were excluded.

The EVAN-G scale is intended to measure satisfactory conditions for perioperative patients. We developed the EVAN-GC scale according to the methods adopted by the International Quality of Life Assessment (IQOLA) project [15]. The translation process from French to Mandarin was forward and backward translation. Above all, the original French EVAN-G scale was translated into Chinese by a Chinese medical student proficient in French and a native Chinese speaker without a medical background. Second, the two translators and a recording observer discussed, modified, and synthesized the first translation graft. Then, another two bilingual French speakers back translated the Chinese version into French. After that, an expert committee discussed discrepancies and produced the prefinal version. Finally, pilot testing on 36 patients was performed to ensure understanding of all questions and ease of administration. No specific cultural adaptation was made after the pilot testing.

An electronic medical record system was used to screen for potential participants. The study procedure and information about the planned surgery were informed orally on the day before surgery. In addition, basic information about the enrolled patients was recorded, such as age, sex, the extent of surgery, and educational background. We classified the extent of surgery as grade I − IV based on a deterministic patient classification approach, which groups patients according to the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM3) codes [16]. The educational background was classified as follows: junior high school or below, high school, associate’s or bachelor’s degree, and master’s degree or above [17].

On the day of surgery, patients were invited to complete the EVAN-GC scale, the Spielberger State-Trait Anxiety Inventory (S-STAI) [18], and the McGill pain questionnaire (MGPQ) [19] within 4 to 24 h after surgery. The S-STAI is a reliable measurement for clinical screening and behavioral research. The MGPQ contributes to evaluating pain intensity. These scales were performed under the support of the WeChat mini program. The enrolled patients were provided with a link to the questionnaires, and the assessor collected data from the background program. The duration of completing each dimension was recorded. The 26-item EVAN-GC scale was evaluated using a five-point Likert scale, in which 1 point equaled much less than expected, 2 points meant less than anticipated, 3 points stood for as expected, 4 points equaled more than expected, and 5 points equaled much more than expected [20]. In other words, a higher score indicates excellent satisfaction, while a lower score reveals poor satisfaction.

The sample size was calculated based on our 65-patient pilot test. According to the formula N = Z2pq/δ2 (α = 0.05, δ = 0.07, p = 56%, q = 1-p), [21, 22] 193 patients were required for a statistical power of 90% power at the 0.05 significance level. Assuming a 15% dropout rate, 227 patients were deemed for this study.

IBM SPSS for Windows version 25.0 software (SPSS Inc., Chicago, IL, USA) was used for statistical analyses. Data were reported as mean (standard deviation, SD), median [interquartile range, IQR], or number (percentage, %) as appropriate. For Gaussian data, correlations were calculated with the Pearson correlation coefficient, and for non-Gaussian data, correlations were calculated with the Spearman correlation coefficient. Inferential analysis was performed using the Mann–Whitney or Kruskal–Wallis tests.

First, we measured construct validity, convergent validity, and discriminate validity. To validate the construct validity, we identified interitem, item-dimension, interdimensional correlations, and principal component factors. We analyzed them by several methodologies, including Spearman’s correlation coefficient (ρ), principal component analysis, and factor analysis, as appropriate. To calculate the convergent validity, we measured the correlation between the EVAN-G scale and S-STAI and the correlation between the EVAN-GC scale and MGPQ using ρ. To assess the discriminant validity, we calculated the differences between different patient groups (such as age, sex, and educational background) using the Mann–Whitney or Kruskal–Wallis tests. Second, we assessed internal consistency reliability and test-retest reliability. Cronbach’s alpha coefficients were computed to analyze the internal consistency reliability. The test-retest reliability was evaluated by Spearman’s correlation coefficient on a 36-patient-scale subsample estimated twice between a 15-day interval. Third, acceptability and feasibility were valued with successful completion rate and duration for patients to complete the questionnaire.