2.1 Development of item pool
The PHBS was developed using classical measures. The semi-structured interview method was used to conduct a qualitative interview on the theme of "perineal hygiene behavior" among gynecology and nursing experts. The sample size was determined by the "information saturation method" (N=8), and the interview was terminated when the interviewer could not obtain new valid information from additional respondents.
The interview outline on the theme of ' perineal hygiene behavior ' includes:
① What poor perineal hygiene and care behaviors do you think women may have in their daily life?
② What poor perineal hygiene and care behaviors do you think women may have in their sexual activity?
The items were extracted and analyzed by Colaizzi's 7-step method, coded, and pooled to construct the pool of items for PHBS.
Using the Delphi expert consultation method, 5 experts in the fields of gynecology, psychology, nursing, linguistics, etc. (experts in the expert consultation method are different from those in the qualitative interview method) were invited to score the content validity of the pool of items, which ranged from 0 to 10, with higher scores representing a higher degree of conformity between the content of the items and the theme of "perineal hygiene behaviors".
Using cognitive interviewing techniques, 20 women living in rural communities were selected for the pre-survey, and based on the feedback, the items were further revised to finalize the draft version of the PHBS.
2.2 Scale scores
The draft version of the PHBS consisted of 16 items, divided into two dimensions: “daily life” and “sexual life”. Item 2, 7, 8, 11, and 12 were scored as "Never = 0, Occasionally = 1, Often = 2, and Always = 3" ("No sexual behavior = 0" in item 12). Item 1, 3, 4, 5, 6, 9, 10, 13, 14, 15, and 16 were scored as "Never = 3, Occasionally = 2, Often = 1, Always = 0" ("No sexual behavior = 0" in item 13, 14, 15, and 16). The higher the score, the more obvious the respondents ' poor perineal hygiene-related behaviors.
2.3 Survey tools
The survey instrument was designed as a self-assessment electronic questionnaire. It consisted of five parts:
① A questionnaire for the collection of sociodemographic characteristics: including nationality, residence, marriage status, economic status, job category and partner-related conditions, etc.
② A questionnaire for the collection of clinical information: This section mainly includes menstrual history, pregnancy and childbirth history, sexual life history, self-reported RTI history, etc.
③ Draft version of Perineal Health Behavior Scale.
2.4 Data collection
This study was an exploratory cross-sectional study. A 2-week cross-sectional survey was conducted from January 17 to January 30, 2022, in rural communities of Huaning County, Yuxi City. Judgmental sampling was used to select the sample according to the study's inclusion and exclusion criteria.
The inclusion criteria for participants were set as:
① Female respondents;
② ≥18 years old;
③ Resident in rural communities in Huaning County, Yuxi City, Yunnan Province;
④ Voluntarily and agree to participate in this survey;
⑤ Have literacy, reading, and writing skills to complete the electronic questionnaire independently.
The exclusion criteria for participants were set as:
① Patients with severe mental illness;
② Patients with cognitive dysfunction;
③ Visual impairment and unable to complete the questionnaire;
④ Hearing impairment and unable to complete the questionnaire.
The WJX.cn platform was used to compile the electronic questionnaire, and specially trained investigators pushed the questionnaire link or QR code to the respondents who met the research standards. Respondents followed the prompts to click on the link or scan the QR code and loaded the questionnaire home page. The study protocol was approved by the Ethics Committee. The study was conducted under the principle of complete voluntariness, anonymity, confidentiality, and no commercial interests. Respondents could suspend and withdraw from the survey at any time. To avoid missing data, questionnaire items were set as mandatory field.
2.5 Data Analysis
The data were exported from the background of WJX.cn system. The database was established by SPSS 19.0. The data were processed and statistically analyzed by SPSS 19.0 and AMOS 22.0. Confirmatory factor analysis (CFA) was used to test the structural validity of PHBS. Spearman correlation analysis was used to evaluate the content validity. The average variance extracted (AVE) and composite reliability (CR) were used to evaluate the convergent validity. The discrimination validity of PHBS was evaluated by cross-loadings and heterotrait-monotrait ratio (HTMT). Cronbach's α and McDonald's ω, Spearman-Brown coefficients were used to evaluate the internal consistency reliability and split-half reliability, respectively. Multiple linear regression analysis was used to explore the influencing factors of the PHBS score. The test level was specified as α=0.05.