The study was conducted at the DCCs in Alexandria, Egypt using a cross-sectional design. The sample size was calculated using Epi Info 7.1.3.3 program (CDC, Atlanta). Based on the assumption of a prevalence of good knowledge among caregivers in DCCs of 50% and 5% confidence limit, the minimum required sample size at 95% confidence level was 384 caregivers. The sample was rounded to 400 and 402 caregivers were included. Each caregiver was observed 3 times consecutively with a total of 1206 observations.
According to the Directorate of Social Solidarity records (2013), Alexandria is divided into 6 districts with a total of 700 DCCs, and an average of 5 workers in each one of them (personal communication). Three districts were selected randomly, and caregivers were proportionately allocated according to the total number of caregivers in each selected district. A total of 59 DCCs licensed and supervised by the Ministry of Social Solidarity in Alexandria, Egypt were included and all caregivers in the chosen DCCs (teachers, nannies, kitchen workers, and cleaning workers) were also included (27 DCCs from district X with 208 caregivers, 12 DCCs from district Y with 75 caregivers and 20 DCCs from district Z with 119 caregivers).
Data was collected using a data collection sheet about the DCC features, a predesigned interviewing questionnaire with caregivers and an infection control checklist. The data collection sheet about the DCC features was filled by interviewing each of the DCCs directors, and by observing the DCC features, infrastructure and the infection control logistics. It included data about the nursery (design, location, number of children, number of classrooms, presence of playground, and source of drinking water), classrooms (aeration and using disinfectants in cleaning), bathroom (cleaning schedule and level of cleanliness), kitchen (level of cleanliness, food preservation method, and refrigerator cleanliness), and IC logistics (availability of soap and water, alcohol-based substances, personal protective equipment (PPE), diaper changing room, and its features if present).
A predesigned structured interviewing questionnaire was used to collect data from caregivers regarding their personal characteristics (age, sex, job, level of education, years of experience, marital state, infection control training, childcare training, number of children they are responsible for, and whether those children were located in one class or distributed in more than one class), knowledge regarding infectious diseases, modes of transmission, and the possible ways of infection prevention and control and knowledge regarding infection control practices. The total knowledge score was calculated by summing scores of all questions yielding a total score ranging from 0 to 50 and was classified into poor (< 50% or < 25 points), fair (50–< 75% or 25–< 38 points) and good or satisfactory level of knowledge (≥ 75% or ≥ 38 points).
Finally, an infection control checklist was used for observation of the following infection control practices: hand washing, use of PPE, waste disposal, diaper changing, environmental cleaning, and meals and feeds preparation. A special scoring system was constructed according to the job responsibilities of each caregiver. Accordingly, the practice score percent was calculated, and the percentage range was classified into poor (< 50%), fair (50–< 75%), and good or satisfactory levels of practice (≥ 75%).
Statistical analysis
The collected data were coded, entered, and cleaned using SPSS for Windows version 21.0 (SPSS Inc., Chicago, IL, USA) and SAS for Windows version 9.4 (SAS Institute Inc., Cary, NC, USA). For quantitative variables, mean and standard deviation were calculated. Multiple analysis of variance (MANOVA) was used to test the difference in two or more vectors of means (mean knowledge scores about infectious diseases and about infection control). Post hoc test using Tukey HSD (Honest Significant Difference) was used after MANOVA to determine which groups in the sample differ. Multiple regression analysis was used to identify factors affecting knowledge score of caregivers, while stepwise logistic regression analysis was carried out to estimate the magnitude of the association between different variables and the level of practice of caregivers (poor/fair and good). A p value < 0.05 was considered to be statistically significant.