This was a cross-sectional study conducted in special schools accommodating LSB in two provinces of South Africa, that is, the Limpopo and Gauteng provinces. The schools were classified as public schools for learners with special needs (LSN). The 2005 policy statement of the Department of Basic Education defines a school for LSN as a resource to deliver education to learners requiring high-intensity education and other support on either a full-time or a part-time basis (Department of Education 2005). The Limpopo province is one of the poorest provinces in South Africa (Young 2020) and comprises a predominantly rural population of approximately 6 million people (Statistics South Africa 2020). Gauteng province is the most populous and has a population of approximately 16 million people (Statistics South Africa 2019). Gauteng province is considered the financial hub of the African continent; it is highly urbanised and considered the wealthiest province in South Africa (Statistics South Africa 2006). The choice of Gauteng province and Limpopo province was because of convenience for the principal investigator, and also to provide some insight into the state of infection control in special schools that represent a rural and an urban province.

Limpopo province has 35 special schools for LSN, of which three accommodate learners with physical disabilities. Gauteng province has 120 special schools for LSN. For this study, 28 special schools accommodating learners with physical disabilities were purposively selected, including 25 from Gauteng province and three from Limpopo province.

The study population comprised principals and staff members working directly with LSB in selected special schools. Staff categories included educators, nurses, health promoters, physiotherapists, psychologists, speech therapists, occupational therapists, social workers and house mothers. In this context, a house mother referred to a staff member responsible for the care of LSB residing in school boarding facilities during the academic term. The inclusion criteria were school principals from the schools accommodating LSB, and staff in any of the aforementioned categories who were involved in managing LSB.

An email was sent to the principals of each of the 28 identified special schools informing them of the aim of the study and inviting their participation. Principals who consented to participate were asked to inform staff working with LSB about the study. Staff who expressed interest in participating agreed to have their contact details shared with the principal investigator (PI).

Of the 28 schools initially selected, six were excluded as a result of the absence of enrolled LSB. Of the remaining 22 schools, 13 did not provide consent to participate. In line with ethical requirements, no further data were collected from these schools, and their identifying information was not disclosed.

The sample was secured from the nine schools, representing the accessible portion of the population at the time of data collection. The principals of these nine schools and a total of 121 staff members consented to participate. The final sample size was therefore determined by institutional access and the willingness of eligible participants. The use of consent-driven sampling is widely accepted in educational and organisational research, particularly in studies involving defined professional groups and institution-based populations (Lohr 2019). In such contexts, the emphasis is placed on obtaining information-rich data from participants with direct relevance to the study objectives. Although the achieved sample size was limited to 121 participants, this number is consistent with widely cited methodological guidelines for exploratory and descriptive survey research, which indicate that samples of 100–200 participants are adequate for stable estimates and meaningful analysis (Israel 1992).

To address the potential gatekeeping role of school principals, it was emphasised that their role was limited to granting institutional access and circulating study information. Principals were not informed about the identities of the staff members who elected to participate, and no individual responses were shared with principals or school management.

Recruitment was slower than anticipated because of delayed responses from some schools. In certain cases, publicly listed email addresses were outdated or non-functional. Follow-up reminder emails were sent, and alternative contact details were sought to facilitate communication.

The questionnaires were developed by the PI, guided by existing literature and established infection control standards. To enhance content validity, the questionnaire was reviewed by a subject matter expert in SB. A pre-test was conducted with a small group of school staff working at a school accommodating LSB to assess clarity, relevance, and comprehension, and to refine the wording prior to full-scale data collection. The school selected for the pre-test was not part of the nine schools included in the main study.

The situational analysis was conducted using a structured questionnaire of closed and open-ended questions in English that was emailed to the school principals. The questionnaire elicited basic information relating to the number and age range of learners enrolled at the school, the number of LSB, the type of accommodation available for LSB, the number and category of staff employed and staff training requirements. In addition, the age and number of years of experience of the principals were obtained.

A closed-ended questionnaire in English was used to assess knowledge, attitudes and practices of infection control among members of staff. Section one elicited demographic variables and years of experience in special education settings. Section two explored participants’ knowledge of infection control. The questionnaires included information that clean intermittent catheterisation is a procedure to empty a bladder for LSB, heels, ankles, hips and elbows are prone to pressure sores on LSB, absent skin sensation in the trunk and buttocks can cause wound formation, headache, nausea or vomiting are the symptoms of shunt infection, coronavirus is transmitted through direct contact with respiratory droplets of an infected person and fever, cough, sore throat, headaches and difficulty in breathing are the symptoms of coronavirus.

Section three explored the participants’ attitudes toward infection control for LSB. Questionnaires included information about the staff’s level of attitude on the risk of infection in LSB, infection control is an important aspect in the management of LSB, the importance of infection control was emphasised in the schools, being comfortable assisting LSB affected with infections, staff were well trained to work with LSB and to prevent the spread of coronavirus on LSB.

Section four explored their practices of infection control towards LSB. The section included questionnaires on LSB were checked daily for pressure sores, burns and other injuries, LSB washed their hands frequently with soap and water, LSB were assisted to do wheelchair push-ups every 20–30 min, schools had bladder and bowel management programme, bladder and bowel management programme were practised daily at a scheduled time, schools were having catheterisation programme, schools were providing diapers for LSB with urine or bowel leakage, and learners with shunts were prevented from participating in contact sports.

With references to the section on knowledge and attitudes, they were asked to indicate their degree of agreement on each statement by using a 5-point Likert scale where 1 = strongly agree, 2 = agree, 3 = strongly disagree, 4 = disagree and 5 = neutral. Participants were asked to respond either by using ‘Yes’, ‘No’, or ‘I don’t know’ on statements relating to practices. The PI contacted each staff member either telephonically or via email, and questionnaires were emailed to those who had chosen to complete the questionnaire via email. All the completed questionnaires were returned to the PI. Data was collected from August 2020 to May 2021.

Data from the questionnaires were entered into a Microsoft Excel spreadsheet. For data analysis, study participants were categorised into four groups based on their roles: educators, allied health (physiotherapists, speech therapists, occupational therapists and social workers), health staff (professional nurses, and auxiliary nurses), and house mothers. These groupings allowed for comparison of knowledge, attitudes and practices across different types of staff working with LSB.

For the 11 items assessing knowledge, 1 point was given for every correct answer. The six items assessing attitude were measured using a five-point Likert scale and scored as follows: strongly agree = 5, agree = 4, neutral = 3, disagree = 2 and strongly disagree = 1. Higher scores indicated more positive attitudes toward infection control. The responses were collapsed into three main categories: agree (combining strongly agree and agree) and disagree (combining disagree and strongly disagree) and neutral. The 13 practice items were measured using binary response options: Yes = 1 and No or I don’t know = 0. Percentages were then calculated and presented in tables to summarise participants’ responses.

Descriptive statistics, including frequencies, percentages, means and standard deviations, were calculated to summarise knowledge, attitudes and practices scores, demographic characteristics and school-level situational data. Spearman’s rank correlation was used to assess the association between knowledge, attitude and practice scores. A p-value < 0.05 was considered statistically significant.

Ethical approval for the study was obtained from the University of KwaZulu-Natal Biomedical Research Ethics Committee (BREC/00000854/2019). Permission was obtained from the Gauteng and Limpopo Provincial Departments of Education and the principals of each of the schools to conduct the study. As a result of the COVID-19 pandemic and associated restrictions, in-person contact between the researcher and participants was minimised. As a result, informed consent was obtained and recorded telephonically prior to participation.

The University of KwaZulu-Natal Biomedical Research Ethics Committee approved a COVID-19 related amendment to the study protocol, permitting electronic data collection methods. Participants were informed of the study’s purpose, procedures, and their right to withdraw at any time. Participant anonymity was maintained by the removal of identifying information. Data were stored securely in digital files protected by a password and in locked physical storage, with access restricted to members of the research team only.

Nine schools accommodating LSB were included: Six primary schools (Grades R–7), two combined schools (Grades R–12), and one junior primary school (Grades R–5). The total number of learners per school ranged from 184 to 389, with the number of LSB ranging from 1 to 25. Learners’ ages ranged from 5 years to 25 years (Table 1).

All nine schools (100%) reported having running water, flushable toilets and wheelchair-accessible bathrooms equipped with basins, soap, and paper towels. Four schools (44%) had boarding facilities. None of the schools had separate classrooms for LSB.

The average years of experience as a school principal were seven years, with a range of 10 years to 29 years. Four (44%) schools did not have a house mother, and five (56%) schools reported having a house mother. The schools employed various professional staff (refer to Table 1). Four schools (44%) did not employ a house mother, while five schools (56%) did. The minimum qualifications varied: one school (11%) required Grade 10, two schools (22%) required Grade 12, two schools (22%) required a home-based care qualification and four schools (44%) did not require any specific qualification.

Training related to infection control and the management of LSB varied across schools. Only one (11%) principal reported that house mothers received school-based infection control training. In contrast, six (67%) principals indicated that educators had received training, although the format and focus differed between institutions. Reported training topics included augmentative and alternative communication (AAC), workshops to increase awareness of SB, inclusive education workshops, infection prevention and support for LSN and formal qualifications such as an Advanced Certificate in Education for LSN offered by universities. Training for educators on infection control and management of LSB was provided by the Provincial Department of Education.

Five (56%) principals reported that they attended training addressing the needs of LSN and the management of infection among LSB. This training primarily focused on managing educators, support staff and broader institutional responsibilities. Delivery methods included workshop-based sessions and formal academic programmes, such as higher degrees or Advanced Certificates in Education. Training for principals was provided by the Provincial Department of Education and universities.

Regarding the frequency of training for the principals, nine schools offered training very often. A similar pattern was observed for educators, often reported in seven schools. Notably, six of the nine schools did not offer infection control training specifically for LSB. Training was extended to parents of LSB in three schools and to non-teaching staff in one school. All nine schools reported receiving COVID-19-related training provided by the respective Provincial Departments of Education.

A total of 121 staff members participated. The majority were female (n = 91, 75.2%), with males constituting 24.8% (n = 30). Most participants worked in primary schools (67.8%), followed by combined schools (21.5%) and junior secondary schools (10.7%). Participants’ ages were distributed as follows: 28–39 years (11.6%), 40–49 years (46.3%) and 50+ years (42.1%). The largest occupational group was educators (57.9%), followed by house mothers (13.2%), nurses (12.4%), occupational therapists (7.4%), physiotherapists (4.1%), social workers (3.3%) and speech therapists (1.7%). Years of experience were distributed as 0–9 years (32.2%), 10–19 years (44.6%) and 20+ years (23.1%). Geographically, 60.3% of participants were from Gauteng province and 39.7% from Limpopo province.

Table 2, Table 3, and Table 4 present responses about knowledge, attitudes and practices of infection control among staff working with LSB in Gauteng province and Limpopo province.

The mean knowledge score among staff was 3.80 (standard deviation [s.d.] = 2.56), indicating generally low to moderate knowledge. According to the knowledge statements, only a minority of health staff recognised that clean intermittent catheterisation is a procedure to empty a bladder. Similarly, a comparable proportion correctly identified the heels as being prone to pressure sores (Table 2).

The staff who agreed that ankles are prone to pressure sores were eleven (15.7%) educators and five (25.0%) allied health. Similarly, low levels of knowledge were reported across all staff categories about hips and elbows being prone to pressure sores. Most health staff agreed that the absence of skin sensation in the trunk and buttocks can cause wound formation, and most staff reported knowing how the coronavirus was transmitted and the symptoms thereof.

The mean attitude score was 17.21 (s.d. = 3.90), ranging from 11 to 30, reflecting moderately positive attitudes. A larger proportion of house mothers agreed that they were concerned about the risk of infection in LSB compared to other staff categories. The proportion of staff who agreed that infection control was an important aspect in the management of LSB varied across the staffing categories (Table 3). Ten health staff (66.7%) agreed that they were comfortable assisting LSB affected with infections. Very few staff agreed that they were well-trained to work with LSB.

The mean practice score was 1.64 (s.d.= 2.27), with a range of 0–12, indicating very low adherence to recommended practices. Most health staff agreed that LSB were checked daily for pressure sores, burns, and injuries. The proportion of staff who agreed that LSB were assisted with washing their hands frequently with soap and water varied from 25 educators (35.7%), 12 health staff (80.0%), seven allied health (35.0%) and six house mothers (37.5). Eight educators (11.4%), two allied health professionals (10.0%), six health staff (40.0%) and one house mother (6.3%) indicated that their school had a bladder management programme. Only a small proportion of staff across the different categories of occupation reported that the programme was practised daily at a scheduled time. A small proportion of participants across staff categories reported that their schools had a bowel management programme, and even fewer indicated that the programme was practised daily at a scheduled time. Similarly, fewer categories of staff reported that catheterisation programmes exist at their schools. Very few participants reported that diapers were provided for learners experiencing urine or bowel leakage.

Spearman’s rank order correlation analysis demonstrated statistically significant positive correlations among all three domains (p < 0.001). Knowledge was moderately correlated with practice (ρ = 0.423, p < 0.001) and attitude (ρ = 0.521, p < 0.001). Attitude was also moderately correlated with practice (ρ = 0.434, p < 0.001).

This study has several limitations that should be considered when interpreting the findings. Firstly, it was conducted in only two provinces in South Africa (Gauteng province and Limpopo province), which limits the generalisability of the findings. However, these provinces reflect disparities in urban and rural settings, and schools in other provinces likely face similar challenges regarding infrastructure, staffing and infection control.

Secondly, the study was conducted as a cross-sectional survey and did not include multiple sources of information typically used in a full situational analysis. Therefore, the findings reflect staff-reported knowledge, attitudes and practices regarding infection control and may not capture all aspects of the school context. Future studies incorporating multiple data sources could provide a more comprehensive situational analysis and inform policy and practice for inclusive and safe school environments.

Thirdly, practices were self-reported rather than directly observed. This raises the possibility of information bias, where participants may overestimate adherence or report socially desirable behaviours. Fourthly, the response rate from schools was low, with less than 50% of eligible schools consenting to participate, which may limit the generalisability of the findings. Schools that declined participation may differ systematically from those that participated, introducing potential selection bias.

A further limitation relates to the use of school principals as gatekeepers. While necessary to gain access, this approach may have influenced participation. Principals could have selectively facilitated access to certain staff members, introducing selection bias.