Globally, 21% of road traffic deaths occur among children aged 0–19 year [1
]. Road traffic injuries (RTIs) are the leading cause of death amongst children aged 1–19 years with global rates being estimated at 10.7 per 100,000 population [2
] Of these, 93% occur in low- and middle-income countries (LMIC) such as South Africa (SA) where the overall (all age) road traffic fatality rate is estimated to be 33.2 per 100,000 population. Children are especially vulnerable road users in SA, with road traffic crashes being the second leading cause of death for children aged 5–14 years [4
]. There is a growing concern for those children being transported in vehicles as vehicle ownership increases and there is a greater reliance on public service vehicles [5
The World Health Organization (WHO) has declared that the main risk factor for young children being injured as a result of a road traffic collision is the lack of or improper use of an appropriate restraint and that this risk could be significantly reduced through the use of child restraint systems (CRS) [1
]. Child restraint systems, including rear- and front-facing five-point restraint car seats and booster seats, are highly effective mechanisms to prevent child passenger injuries [7
]. CRS, when correctly installed and used, can be very effective with evidence showing that they reduce the need for hospitalization by 69% in the under 5 age group and the risk of death by approximately 70% for infants and by 47% to 54% for toddlers (aged 1 to 4 years) [10
]. A child up to 4 years of age has a 50% lower risk of injury in a forward-facing child restraint and 80% lower in a rear-facing seat [12
]. This compares with injury reductions of only 32% when an adult seatbelt is worn. For children aged 5–9 years, child restraints reduce injury by 52%, whereas for seat- belts alone the reduction is only 19%. For older children aged 10–14 years seatbelts reduce injury by 46% [12
Despite this evidence CRS use rate is low in many countries; for example, the rate of use among children under 5 years of age ranged from 7.9% to 17.4% in two Mexican cities [13
]. This is similar to the observed child restraint use in Ghana with a proportion of 13.2% [14
]. Even in countries with high use rate, correct and appropriate restraint use can still be a problem [15
]. Observational data show that children and young adults are less likely to use seatbelts than adults [16
]. According to statistics from the Child Accident Prevention Foundation of South Africa, 84% of children in vehicles are not restrained and 80% of children who had been injured in collisions were not restrained [17
]. A previous observation study conducted at the main gate of the Red Cross War Memorial Hospital in Cape Town in 2008 showed that only 3% of observed children were adequately restrained [18
Comprehensive legislation which includes laws surrounding child restraint use is a critical component in a child road traffic injury prevention effort [1
]. The latest World Health Organization Global Status Report on Road Safety indicates that only 33 countries have a child restraint law in place which aligns with best practice [19
]. Consequently, improving child restraint usage is one of 12 performance targets recently adopted by the United Nations General Assembly. The South African law on child restraints does not satisfy best practice. The amendment to the road safety law that was introduced in 2014 mandates the use of a South African Bureau of Standards-approved child restraint for all child passengers under the age of 3 years. This does not adhere to the United Nations (UN) prescribed best practice law that requires the use of suitable child restraints for children up to the age of 12 based on age, height and weight [20
]. Experience from high-income countries shows that even when a comprehensive legislation exists, high rates of use are not possible without complimentary educational programmes, increased enforcement and supportive programmes to distribute CRS such as loan incentive schemes [10
]. Barriers to increased use, particularly in LMICs include the cost of the child restraint, lack of knowledge and awareness of parents and caregivers and relatively low safety standard for vehicles [15
]. However, there is little empirical data to support this. To guide the development of interventions and to persuade policy makers to focus on this under-served population, studies are needed that explore these barriers. This study aims to address the current gap in the scientific literature surrounding the barriers to CRS use in a middle-income setting by determining the availability and affordability of certified child restraints amongst different populations in Cape Town. This study was conducted as part of a larger mixed method study of child restraint use in Cape Town and therefore is not exhaustive in terms of data collected and analysis. The findings reported in this article will be combined with a systematic review of child restraint usage, a quantitative analysis of hospital data in Cape Town, qualitative data from focus groups and in-depth interviews of road safety experts in Cape Town and a review of legislation, which will serve to develop a brief intervention that will be evaluated through an appropriate study.
This study was done in collaboration with the George Institute for Global Health UK, University of Oxford, the ChildSafe team at the Red Cross Children’s Hospital, the University of Cape Town and the SA Medical Research Council. Ethical approvals were obtained from both the University of Oxford and the University of Cape Town.
2. Materials and Methods
To gain in-depth insight into the barriers surrounding CRS use we utilized both an unobtrusive observational study and a cross sectional survey between September 2019 to November 2019 in Cape Town, South Africa.
2.1. Study Setting
Cape Town is the largest city of the Western Cape Province in South Africa and is home to 64% of the Western Cape’s population. It is the second most populous city in South Africa after Johannesburg. Seven suburbs across Cape Town were selected to give a good representation of ethnic groups and socio-economic status. To try and maximize response rates and increase denominators, study sites within these suburbs were purposively selected to include hospital settings and childcare facilities. Hospitals and daycare centers were approached by the research team to request permission to carry out data collection. Table 1
shows the site selection that was used for both observational studies and the survey.
2.2. Observational Study
For the observational study, a standardized observation form was developed with the local ChildSafe team. Any vehicle with at least one child occupant age 14 years or under (estimated based on observations) seen to be entering the car park of the study site was eligible for the study. A team of local fieldworkers were recruited based on previous connections and knowledge of ChildSafe work. They were trained using a two-step process that included theoretical and practical knowledge. The theoretical aspect involved “in-classroom” teaching sessions to explain the nature of the study, to familiarize the fieldworkers with the survey forms and study protocols and to help them identify the different types of CRS that are currently available in South Africa. Fieldworkers were also trained on how to estimate the age of children. This was followed by the practical session in the car park of the hospital to give the fieldworkers an opportunity to use the forms and practice estimating ages of children in the front/backs of vehicles. This was used as a pilot test of the form and subsequent modifications were made to make data collection easier and more accurate. Teams of two fieldworkers were allocated specific sites and time slots. Due to the high rates of crime and violence in South Africa, the protection of fieldworkers was paramount. Homicide accounts for 56% of fatal injuries amongst individuals aged 15–34 years in South Africa [23
]. The safety of fieldworkers was therefore ensured by the local study team with hi-visibility jackets being provided and advice given on safe zones for data collection. Data were collected in time slots of two hours throughout the day from 8am to 5pm ensuring that drop-off and pick-up times at the childcare facilities was captured. No night-time data collection occurred, again to safeguard the fieldworkers from potentially dangerous situations. Fieldworkers stationed themselves where vehicles were forced to slow down near the entrance/exit barriers to centres. Both fieldworkers were instructed to observe the cars and collect data on each vehicle, verbally verifying with each other the details of CRS usage. Forms were collated at the end of each day and taken to the ChildSafe offices for data cleaning, quality checks and data capturing.
For the survey, a questionnaire comprising of 30 questions was developed based on previous studies and modified to take local culture and context into consideration by translating into two additional official languages spoken in the city (Afrikaans and Xhosa) together with English. Any driver who was seen to have children was approached to respond to the survey. Asking parents/carers who had lost a child because of a road traffic injury to answer the survey was felt to be too sensitive therefore any respondent who made the fieldworkers aware of such an event was excluded from the study. Similar to the observational study, local fieldworkers were trained in both “in-classroom” sessions and field work piloting. The in-classroom teaching went through each question of the survey and interview techniques to try and maximise a positive response rate. The practice session served as a pilot and based on feedback from both interviewers and interviewees, modifications were made to the questionnaire to shorten it and to reword certain questions in South African vernacular and to restructure questions which may have been construed as being too sensitive in nature (e.g., on income level). Teams of two fieldworkers were allocated to the study sites on specific days. These fieldworkers were chosen based on whether their linguistic proficiencies in speaking the local dialect and English. Data forms were collated at the end of each day and taken to ChildSafe for data cleaning, quality checks and capturing.
Quality was assured by the ChildSafe study team for both studies with random spot checks being undertaken.
Descriptive statistics were used to summarize the main findings of the observational data with Fishers Exact test and Chi-squared statistics being used to test the relationship between variables.
The observational study showed that, despite the passing of legislation in 2014, CRS use among children remains very low in Cape Town (7.8%). This is disappointingly lower than previous studies from 2008 and 2019 in South Africa which indicated rates of 11% and 18% respectively [18
]. This proportion is significantly lower than some high-income countries where rates of 86–99% have been reported [26
]. However, the prevalence is comparable to other developing countries such as Mexico, Ghana and China [13
] where restraint use for children under 5 years of age is less than 4%. This is further supported by the findings from the survey and other studies which suggest that higher education levels and income levels are linked with increased CRS use and ownership [30
A large proportion of children (84%) were observed to be sitting in the front seat of the vehicles, unrestrained. It is known that children who are unrestrained or sitting in the front seat face the greatest risk for death in road traffic crashes [31
]. The relationship between driver seatbelt use and CRS use is well documented in the literature [33
]. This potentially has implications in terms of opportunities to influence parental/carer seatbelt usage through increased enforcement and targeted education programs surrounding seatbelt legislation in South Africa [35
In this study there was a significant association between CRS use and driver seatbelt use and multiple children. These findings are found elsewhere in the literature particularly in LMIC such as Nigeria and Brazil [36
]. This supports the premise that if efforts to increase seatbelt use amongst adults were strengthened, it may lead to increased child restraint use if parents are more sensitized to the importance of seatbelts and restraints and if children, who are known to learn by example and replicate the actions of their parents also engage in the process.
The parent/carer survey indicated that there was a high level of reported knowledge surrounding child restraint legislation in South Africa (92%), however, when asked to accurately recall the age limits of the law only a small proportion could do so (32%). Whilst 74% of respondents correctly identified a fine as being the penalty received for breaking the law, no respondents could recall the precise fine amount with all responses estimating it to be higher than the current legal fine of 250 R (17 USD). The findings from the survey add strength to the argument that the current legislation in South Africa remains ambiguous as to the correct age that a child should be restrained and the appropriate positioning of the child [20
]. It also supports the notion that the current fine remains too low to act as a deterrent to drivers in South Africa driving with unrestrained child passengers [20
Results from the survey indicate that whilst most parents/carers (95%) believed that child restraints were important, only 51% actually owned a car seat and 36% of those who had one admitted to not always using it. The reported use was higher than in other countries such as Turkey (20%) and Kuwait (26%) [39
]. The relatively low proportion of parents/carers who admitted to not always using a car seat despite owning one (36%) demonstrates that although parents/carers are aware of the need to use a restraint, there is a perception that there are certain circumstances where it is acceptable to not use one (e.g., on short trips, if the child does not like it, or if an adult was able to hold a child on their lap). These findings of inconsistent use are highlighted in other studies and suggest the need for targeted education interventions for such populations [41
]. Utilising platforms such as well baby clinics, maternity wards, early learning centers to increase the general awareness of risks and debunking of certain myths surrounding CRS use would be beneficial. ChildSafe, a campaign run by the Child Accident Prevention Foundation of Southern Africa (CAPFSA) and Safe Kids Worldwide, promotes the use of CRS through organised events, workshops, targeted leaflets/posters and home safety education programs empowering parents/carers of previously injured children to prevent future such incidents. [43
]. Studies from other low-income countries have shown that road safety programs which combine policy, enforcement, education and advocacy can be effective in increasing the rate of both seatbelt and child restraint usage. In Lipetskaya, child restraint usage increased from 20.9% to 51.4% from baseline to four years after program implementation [44
Reasons given for not owning a CRS were diverse however the majority of respondents made the argument that the use of a seatbelt was an adequate alternative and reported that the cost of car seat was too high. Of those who did own a car seat, the majority of CRS had been bought new from a store (54%) or online (17%) suggesting that parents/carers were more willing to pay money for new car seats. The costs of car restraints in South Africa range from R 1000 (67USD) to R 4600 (310USD) which is more than what 55% of the respondents were willing to pay. As highlighted, there was a relationship between income level and purchasing of a new car seat with those in the lower income brackets being less likely to have purchased a new seat from a store or online. This finding is mirrored in other studies where those in lower socio-economic groups were less willing to pay for car seats [45
]. The provision of low-cost/subsidised CRS or borrowing schemes would help to mitigate against lower ownership rates. Organizations such as Wheel Well have been increasing the awareness of parents about the importance of CRS and increasing use through car seat exchange programs which give low-income families the opportunity to receive a second-hand car seat in exchange for an affordable donation [46
]. Given the large proportion of respondents stating that they would look online for information regarding their car seats, increasing ownership through targeted social media marketing may prove effective. Other financial motivations may include car insurance programs which provide incentives to parents who buckle up their children in appropriate CRS.
This study has several limitations. Available human, time and financial resources only permitted for a short window for data collection and therefore the sample sizes involved for both the survey and the observations is relatively low. Estimation of the age of children was also a limitation with the observational study. The fieldworkers were given some training in how to estimate such ages however, particularly with seated children being viewed through car windows, these estimates may not have been accurate. Given that the current legislation in South Africa pertained to children under 3 years of age, a decision was made to categorize children into under 3 year old and above 3 years in order to try and minimize such errors. This has implications for data analysis and its generalizability to the wider geographical area. However, this is the first such study in Cape Town which expands the site selection from a tertiary children’s hospital, which may integrate an element of bias if parents/carers felt the need to carry sick children in their arms, to include several suburbs with a mix of ethnic groups and rural/urban settings.