Chronic non-communicable diseases (NCDs) are by far the leading cause of death in the world 1 . South Africa is no exception and is considered to be in the midst of a “profound health transition” that is characterised by a quadruple burden of communicable, non-communicable, perinatal and maternal- and injury-related disorders 2 . The burden of NCDs is increasing in urban and rural settings, but most profoundly in the urban poor 2 . The prevalence and burden of type 2 diabetes is of particular concern and has been referred to as an “epidemic” faced by African countries 3 .

Clearly, the health sector alone cannot bring about population-wide changes in moving toward healthier lifestyles. In this regard, the education sector has been shown to provide an effective setting for promoting healthy nutrition and physical activity behaviours in the prevention of NCDs 4 . In view of the high rates of childhood overweight and obesity worldwide, including South Africa, aiming prevention at a younger age group is essential in primary prevention 5 . Efforts to change the norms, habits, attitudes and preferences of children are still possible at a younger age, and are likely to track into adulthood 6 .

Intervening in schools in disadvantaged settings in South Africa has many challenges 7 . Even after 16 years of democracy, huge disparities still exist among schools in previously disadvantaged communities compared with those in more affluent areas. Schools in disadvantaged areas usually have little access to resources; have inadequate facilities 8 and poor parent involvement 9 . Even worse, educators and learners frequently have poor motivation levels with regard to trying new interventions to influence behaviour change 10 11 , and educators may feel inadequately prepared regarding curriculum changes which incorporate physical activity and nutrition 11 . There are, however, examples of school interventions which have been effective in disadvantaged settings. For example, Pathways, a randomised controlled trial to prevent obesity in American Indian school children, managed to produce significant positive changes in fat intake and food and health-related knowledge and behaviours 12 .

Both the World Health Organization (WHO) 1 and the American Centre for Disease Control (CDC) 13 have provided meaningful guidelines to promote health through the school system. One of the “core elements” of the guidelines proposed by the WHO is to conduct a situational analysis prior to intervention. Such an analysis provides a baseline of factors relating to the school environment and school policy which are likely to impact on the health of the educators, learners, parents, and the surrounding community. A situation analysis helps “to better understand the needs, resources and conditions that are relevant to planning interventions” 1 .

Informed by these guidelines, we undertook a situational analysis of the school health environment at 100 randomly selected primary schools in disadvantaged settings in two education districts of the Western Cape Province. One of the “major themes” of formative research is appropriateness 14 and as such one of the objectives of the situational analysis was to gain an understanding of the primary school environment of the selected schools. In this instance the health focus was on nutrition, physical activity and tobacco use. Our long-term aim was to gather information about the related environment and policies in order to develop a relevant school-based intervention programme 15 . This information was essential to construct the framework for the subsequent intervention mapping 16 used to develop the intervention. Another objective was to inform the purposive selection of 16 schools for the implementation of the HealthKick intervention. Other components included in our formative assessment were surveys of the health risk factors in educators from the 100 participating schools and another among parents of learners from those schools selected for the intervention. Only the results of the situational analysis are reported in this paper.

The aim of the situational analysis was to provide a baseline assessment of the environment in low-income primary schools relating to nutrition, physical activity and smoking behaviour. This would assist in designing an effective school-based intervention programme to prevent NCDs.

From the observational data, the general physical environment of the schools appeared to be in good condition, although concerns existed around the hygiene of toilet and bathroom facilities. This could have implications for hand hygiene, since the three primary measures associated with reduced incidence of infections include availability of a clean water supply, adequate disposal of waste (particularly for faeces), and hand hygiene 22 .

Another aspect of the general environment which needs addressing is the large number of signage boards on school premises sponsored by a soft-drink company. Displaying these boards would imply that the school supports and promotes the use of these particular soft drinks. Clearly this is not a health promotion message that should be encouraged, a view which is supported by findings from a qualitative study conducted in Australia with primary school learners 23 . This study found that children appeared to believe that school, and anything permitted at school, is inherently healthy. Further support came from parents, who postulated that the inconsistent messages about unhealthy energy-dense foods, including attractive marketing and advertising strategies, confused children 23 .

The data gathered from a large percentage of the principals provided insight into the surrounding communities. Principals expressed their concerns regarding the existence of poverty, unemployment, crime, violence and especially child abuse in these communities. Principals from the urban district identified their learners were facing these problems to a greater extent than the principals from rural schools. Clear differences were observed in the perceptions of urban and rural district principals that were not linked to the poverty grade (quintile). These findings are interesting because poverty is, as elsewhere in South Africa, a rural phenomenon, with the rural poverty rate in the Western Cape estimated at 26.1% compared to 20.1% in urban areas 24 . The reasons for this difference in perceptions about poverty and employment could possibly be explained by the fact that poverty and unemployment were referred to in the same question, and schools in the urban district are mostly located in the Cape Town municipality, which has the fourth highest unemployment rate in the Western Cape, despite having the lowest poverty rate 24 . The learners in these schools were, therefore, not only from communities facing high unemployment, but also from the poorest households in the urban district. On the other hand, learners in the rural district were often from agricultural households that face lower unemployment levels, although they have a higher poverty rate than their non-agricultural counterparts. The reason for the different perceptions about child abuse and neglect is less clear as it appears that living in deep poverty increases the vulnerability of children for abuse and neglect 25 .

Perceptions about higher crime and violence in the urban district are supported by findings that the homicide victimisation rates for men aged 15–29 years in the Cape Town townships, where many urban schools are located, are more than twice the average for the country 26 . Furthermore, crime and violence are closely related to alcohol and substance abuse. A review of studies on substance abuse trends found that the Western Cape had the second highest (7.1%) 12-month prevalence of substance use disorders and the highest (18.5%) lifetime prevalence of substance use disorders compared to other provinces 27 28 . In South Africa, the Western Cape had the highest alcohol consumption among males and females. Therefore, it is not surprising that principals reported substance abuse to be one of the top three health priorities for parents.

Regarding health and health-related priorities and programmes, it is clear from the results that school principals considered lifestyle-related health issues to be priorities for learners, educators and parents. They also indicated a need for programmes to address these health priorities. However, various barriers to implement these were identified, with lack of time and financial resources being the most important ones. Lack of time has been raised elsewhere as being a major barrier to health promotion programmes 29 30 . A study of Intermediate Phase (grades 4–6) educators also raised time limitation as a major problem for the implementation and presentation of Life Orientation, a compulsory learning area for primary school children. This learning area was introduced as part of the outcomes-based education system implemented in 1997 to eradicate the inequalities of the apartheid education system 31 . Life Orientation comprised four learning outcomes for the Intermediate Phase, i.e. health promotion, social development, personal development, and physical development and movement 32 , which made this the ideal vehicle for health promotion programmes. Changes have been made to the National Curriculum since the implementation of the HealthKick intervention. Life Orientation is now called Life Skills and is divided into three study areas: Personal and Social Well-being, Physical Education and Creative Arts 33 . At this stage, there is no clarity how external health promotion programmes, such as HealthKick, may be integrated into this subject. The WHO Health Promoting Schools (HPS) initiative however provides an existing framework for school health programmes 34 . Strengthening this initiative is furthermore listed as a strategy to address inequity and social determinants of health by the South African National Department of Health 35 .

Overall, parental and community involvement appeared to be poor. Only a third of the schools had parents who were involved in the school health and safety committee. This was also reflected by a lack of involvement of the parents in the SGB and the small number of schools where parent or SGB involvement in the school tuck shop was reported. A study conducted in 2004 on parental participation in SGBs, showed that lack of participation could be attributed to the low education level of parents in disadvantaged communities, language barriers and difficulty in attending meetings 9 .

The school food environment has been identified as one of the most important components in effective school-based interventions to promote healthy eating 1 36 . Apart from the finding that most of the schools participated in the NSNP, a large number had tuck shops and vendors selling food items. The findings clearly point to a need for intervention, since sweets (candy), chocolates, cold drinks and crisps were the main items sold by these providers.

Observations of learner spending during the survey showed that they spent little money at the tuck shop or food vendors. The small profit made is either used by the school to provide extra services to learners, or for vendors to make a living. This finding is supported by an evaluation report of the NSNP, which found that learners brought less than $0.3 to school to spend on food of “very poor nutritional value” 37 . The fear of losing the income (although small) generated by the sale of these items may however provide a barrier to the willingness of the school and the food vendors to sell healthier food items. An aspect that warrants further investigation, specifically in context of a developing country, has been recommended by Von Holy and Makhoane 38 . They suggest that baseline research is required to determine the safety and socio-economic importance of foods sold by vendors.

The feasibility of intervening in the nutrition-related environment of schools in disadvantaged settings where children’s buying power is limited, has received very little attention in the literature. A review on the impact of improving nutrition standards on school revenue concluded that in the North American situation very little evidence exists that revenues drop when healthier policies are adopted; however, this may not be the case in disadvantaged African settings 39 .

Very few schools in this study had clear policies guiding the food, nutrition and physical activity environment. This could possibly be ascribed to such policies not being required by the DOE. The WHO 1 , in their global strategy on the prevention of NCDs, urges government to draft policies that stimulate schools to promote healthy eating and encourage physical activity. Intervening in schools at the policy level has been successful in many studies 40 . In this regard South African schools are slow to follow.

Although principals from all the schools indicated that structured physical activity lessons were scheduled in the weekly timetable, this did not always translate to learners actually participating in physical activity outside the classroom. Possible explanations for this finding comes from a study in a similar sample by Van Deventer 11 , where educators indicated that the reasons for not presenting the Physical Development and Movement outcome had to do with a lack of time in the curriculum, educators who were not qualified to teach physical education and a lack of facilities and equipment. The latter is also relevant for extramural sport. Generally, the results showed that extramural sport was being offered at schools but that facilities, although available, were often not sufficient and/or in good condition. These findings reflect the legacy of apartheid when sport development and participation was a privilege set aside for only a small segment of the population 9 . Inadequate facilities were found to be the most important factor for non-participation in sport by black secondary school learners 8 .

Although children seem to be active and educators allocated for supervision during break times, no organised activities were observed. Leviton 41 referred to studies which showed that learners were most active when equipment and facilities, such as basketball hoops and better playgrounds, were available along with organised active games under supervision.

Lastly, tobacco use is an important preventable factor in the war against NCDs 42 . The finding that 13% of school principals reported their learners were frequently smoking on school grounds is of great concern considering their age. Findings from the 2008 South African Youth Risk behaviour survey support our results 19 . In this survey, 21% of the youth aged 13 to 18 years indicated smoking daily. Clearly schools and parents need to take additional means of preventing such behaviour. However, schools also need to address the smoking practices of educators as a substantial number of schools (n = 29) reported having smoking rooms for their staff. Although the tobacco control act 43 allowed this at the time, the negative modelling effect 44 this practice could have on learner behaviour should be considered.

The formative findings provide important information about the context in which primary school learners’ health behaviour takes place in disadvantaged schools in the Western Cape. Evidently, these show that school environments are not always conducive to healthy eating, sufficient physical activity or the prevention of tobacco use. A limitation of the study is that some of the findings, such as those on health priorities within the school community, rely only on the perspective of the school principals and the views of learners, parents and educators were not obtained. The findings do however contribute to the knowledge of key environmental and policy determinants that play a role in the health behaviour of learners, their parents and educators. This knowledge is essential to create an intervention programme in the school setting when trying to optimise the physical environment and develop relevant school policies.

The authors declare that they have no competing interests.

ADV: All aspects of the study: planning, implementation, writing up. NPS: All aspects of the study: planning, implementation, writing up. JMF: Questionnaire design and development; writing and editing final draft. CED: Questionnaire and observational schedule design and development, data collection, editing final draft. GB: Consultant: Western Cape Education Department; planning and implementation. CJL: Data analyses and interpretation. LD: Co-ordinator. ZA: Questionnaire design and development; writing up. EVL: All aspects of the study: planning, implementation, writing up.