A preliminary study of further attempt at the development, testing and application of an independent primary screening stool card

Development and introduction of a stool card applet

We collected fecal images in the relevant departments of the Second Affiliated Hospital of Harbin Medical University. We used most of the pictures to develop the applet and retained some for subsequent testing. Then, the classification logic was established according to the color and shape of feces. We divided the color of feces into yellow, brown, green, white, red and black. The shape of feces was divided into 7 types according to the Bristol Stool Form Scale (BSFS)²⁷.

We classified the collected feces according to the classification logic and analyzed each kind of feces to inform patients about their feces normality or abnormality, the diseases they may be indicative of, and the appropriate next steps. The information provided was reviewed by senior and experienced physicians who have been working in the Second Affiliated Hospital of Harbin Medical University for many years to ensure the accuracy of the information provided.

Then, WXML language and JavaScript was used to develop the WeChat applet, which is named the Doctor Friend Primary Screening Stool Card (DFPSSC). The applet supports both Chinese and English. The interface of the WeChat applet is presented in Fig. 1.

Interface display of Doctor Friend Primary Screening Stool Card. (A) QR code display of applet. (B) Mode selection interface. (C) Screening mode—color selection interface. (D) Screening mode—shape selection interface. (E) Fecal information display interface. (F) Learning mode interface.

Study 1: preliminary validation of effectiveness of DFPSSC

Participant recruitment and grouping method

We recruited freshmen from Harbin Medical University for testing. Freshman at this university have not systematically learned relevant professional knowledge, and their cognition of feces is therefore at the same level as that of nonmedical professionals. We randomly selected a dormitory building for freshmen and selected 6 bedrooms on different floors (a total of 3 floors, 18 bedrooms and 108 people).

People who agreed to participate were tested, excluding those who scored higher (score greater than 80%), because this project was aimed at those who did not have ample knowledge about feces. The remaining people were divided into three groups according to their respective floors. The first group was the group using the stool card applet.

The detail of the test

All subjects meeting the inclusion criteria will be tested again after the trial intervention. The tests focused on the identification of common stool types, diseases indicated by feces and countermeasures in different situations. The tests consist of 30 multiple-choice questions, each of which had 4 response options with single or multiple correct answers. Each question had a value of two points, for a total score of 60 points.To ensure that the difficulty of the pretest was consistent with the post test, questions and the question options were the same, but the question order was changed. The participants were not informed of the answer after each test. Finally, four clinicians from the Department of Gastroenterology of the Second Affiliated Hospital of HMU reviewed the test questions to ensure that they were suitable for this experiment. The questions list is provided as Multimedia Appendix 1.

Although the subjective factors of raters were excluded when testing in the form of multiple-choice questions, participants might have chosen the correct answer without mastering the related knowledge because of their skills and the hints provided in the options in multiple-choice questions. Therefore, we simulated simple situations in which abnormal stool occurred and then asked participants to give solutions to these situations. Finally, their answers were scored by a clinician. The stimulation test is provided as Multimedia Appendix 2.

The test was conducted in the dormitory of the subject to ensure fairness, each test was carried out under our supervision. Participants were required to keep a certain distance from each other, and they were not allowed to read any materials or use electronic equipment.

Implement method

We posted the QR code for the stool card applet on the door of the public toilet on the floor so that the participants in this group could easily use the applet when defecating. The second group was the traditional paper group. A paper stool card was distributed to each selected bedroom; the content of the paper version was the same as that of the applet. The third group was the control group without any intervention.

One month after the pretest, the participants were tested again, and the results of the post test were compared with those of the pretest. The comparison of the two tests and the simulation test results will jointly verify the effectiveness of DFPSSC.

Study 2: preliminary validation of usability of DFPSSC

Participants selection and implement method

After completing the effectiveness test, the subject group members using the applet will complete a questionnaire to assess the usability of the applet. Then, we also invited 20 clinicians with rich clinical experience. They were all engaged in the diagnosis and treatment of gastrointestinal diseases, and all had more than 5 years of work experience. After using the applet for 1 week, they will finish a questionnaire designed for them.

The detail of questionnaire

On the basis of previous practice, Andrea Fairman et al. designed MAUQ, which categorizes the types of app by two dimensions²⁸. The first dimension is interaction modes (interactive or standalone), and the second dimension is target users of app (patients or health services providers), these two dimensions divide the app into 4 kinds, then MAUQ provides 4 questionnaires suitable for different situations based on different classifications. According to the above classification, the stool card applet adopted the questionnaire designed for standalone mHealth apps (the Cronbach alpha = 0.914). There are 17 items in the questionnaire, covering three aspects (Ease of Use, Interface and Satisfaction, Usefulness). Patients and clinicians took the different version of questionnaire to carry out the research. Their differences are reflected in the items in Usefulness. The version for patients focused on whether users can get better health services, and the one for clinicians focused on whether users can practice better health services. And the items of the questionnaires were rated on a Likert scale (1–5 points: strongly disagree to strongly agree).

Statistical analysis

Continuous variables are expressed as the means with standard deviation (SD). Categorical variables are reported using frequencies and percentages. The normality assumption for test scores was verified with the Shapiro–Wilk test. When the data approximately conformed to a normal distribution, if they satisfied the assumption of homoscedasticity, one-way analysis of variance (ANOVA) was performed to determine the significance of differences, and Tukey’s test was used for post hoc testing; if the assumption of homoscedasticity was not met, Welch’s ANOVA was performed to determine the significance of the differences, and the Games-Howell test was used for post hoc testing. Non normally distributed data were tested with Kruskal–Wallis ANOVA to determine the significance of differences in distribution and Bonferroni’s test adjusted for P values. The t test for paired samples was performed to compare the efficiency of the intervention. P values < 0.05 were considered significant. All data were analyzed using Statistical Package for Social Sciences 26.0 (SPSS, Inc. Chicago, Illinois, USA).

Ethical approval

Our study has been approved by the Ethics Committee of the Second Affiliated Hospital of Harbin Medical University (KY2020-270). All procedures performed in studies involving human participants were in accordance with Helsinki declaration.

Informed consent

Informed consent was obtained from all individual participants included in the study.