In this cross-sectional study, which was conducted in Fars province, Iran, 2021, the study population was all nursing staff (nurses, midwives and operating room staff) of Shiraz University of Medical Sciences. Having at least 3 months of work experience, employment in medical departments for at least 12 hours per week, and consent to participate in the study were considered as inclusion criteria, and cases who did not respond the questionnaire completely were excluded from the study.
Based on the estimated prevalence of needle stick in Shiraz teaching hospitals in previous studies [6–8], and using NCSS PASS 15, a sample size of 280 was calculated for the study (α = 0.05, d = 0.05, non-response rate 10%). Participants were recruited by multistage random sampling method. For this purpose, first, five teaching and general hospitals of Shiraz University of Medical Sciences were randomly selected from 47 hospitals, then the participants were selected from each hospital according to the number of staff through simple random sampling method.
The study data collection tool consisted of three parts: The first part was a demographic information form (age, sex, work experience, level of education, field of study, and awareness about safe injection guideline). The second part contains information about the participants’ experiences about occupational exposure to needle sticks, patients' blood and fluids during their career time, including three items (with not at all, once, twice, three times, four and more times options) The third part includes a researcher-made questionnaire which was developed based on protection motivation theory, including perceived threat (6 items), internal and external rewards (6 items), response costs (3 items), perceived efficacy (6 items), fear (3 items) all with a 5-point Likert scale (strongly agree to strongly disagree), protection motivation (7 items) with a 5-point Likert scale (not at all to definitely), and finally behavior (7 items) with four points scale (not at all, rarely, sometimes, always).
the face validity of the questionnaire was confirmed by ten nurses who were interviewed face to face and examined the items in terms of level of difficulty, appropriateness and ambiguity. Content validity ratio (CVR) and content validity index (CVI) were calculated to evaluate the content validity of the questionnaire, using the opinions of a panel of experts consisting six health education and promotion professionals and four nurses. The CVR for all of the items were more than 0.75, which were acceptable based on Lawshe criteria [25], and CVI for all of the items were more than 0.81 which were acceptable based on Waltz and bussel’s criteria [26]. Internal consistency of questionnaire was calculated through Cronbach alpha (> 0.62) for each construct. The external consistency was assessed by intra class correlation [ICC] in a test- retest on a pilot study with two weeks’ interval (N = 30, ICC = 0.71, P = 0.01). Because of covid-19 pandemic, questionnaires were administered online using an Porsline, which is an online survey platform in Iran.
The study was approved by ethical committee of Shiraz University of Medical Sciences, Iran (code: IR.SUMS.REC.1397.770). An informed consent form was completed and signed by all participants in the study. The questionnaires were anonymous and participants were assured that their participation is voluntary and their information would remain confidential.
Data were analyzed using SPSS 22 statistical software at a significance level of < 0.05. The normality assumption of the variables was checked and confirmed through Kolmogorov-Simrnov test. Frequency descriptive statistics were used to report the frequency of participants' demographic information. The correlation between mean scores of constructs were analyzed and reported by Pearson’s correlation coefficient. Multiple linear regression analysis was used to identify factors predicting protection motivation and NSI preventive behaviors. Structural Equation Modeling (SEM) by AMOS 24 software was used to evaluate the fit indices of proposed model. Goodness-of-fit indexes such as Chi-Square/Degrees of Freedom Ratio (X2/Df), Root Mean Square Error of Approximation (RMSEA), Goodness-of-Fit Index (GFI), Comparative Fit Index (CFI) and Incremental Fit Index (IFI) were used to assess the final model fit[27, 28].