For children with disability, it is essential to provide current and evidence-based interventions to ensure best treatment. Professionals may find it difficult to keep up to date with the latest developments in cerebral palsy (CP) intervention because of the substantial increase of systematic reviews published on CP intervention in the last decade (Anttila et al. 2008; Donald et al. 2014; Novak et al. 2013; Reedman, Boyd & Sakzewski 2017; Thomason & Graham 2014; Wiart, Darrah & Kembhavi 2008). Novak and her Australian colleagues reported on the state of evidence for interventions to specifically support children with CP (Novak 2014; Novak et al. 2013). Even though the results of Novak’s comprehensive systematic review were questioned by some Australian experts as these experts were of the opinion that the majority of the studies included in the review seemed to be sponsored by industry (Thomason & Graham 2014), it is important to consider effective evidence-based CP interventions apart from surgery. In school settings, pain management could contribute to children with CP’s optimal learning opportunities and participation in classroom activities. This study, which is part of a larger project that aims to facilitate discussions about best practices (Adolfsson, Johnson & Nilsson 2018; Johnson, Nilsson & Adolfsson 2015; Nilsson, Johnson & Adolfsson 2016), explores how professionals working in South African schools for children with special educational needs perceive pain management intervention for children with CP, and how their ideas about interventions comply with knowledge about evidence-based interventions as reported by Novak et al. (2013).

Cerebral palsy includes a group of permanent disorders of the development of movement and posture (Rosenbaum, Paneth & Leviton 2007) that could limit persons with CP to participate in activities of daily living (such as eating, sleeping and walking), physical activities (such as gross and fine motor activities), communication, learning and social engagement (Breau 2011; Johnson et al. 2015; Lauruschkus et al. 2017). In addition to individual factors, environmental factors may further impact the independence of a child living with CP, such as physical accessibility, lack of assistive materials or support networks (Novak et al. 2013).

The focus of this article will be on the classification of chronic nociceptive musculoskeletal pain according to the International Classification of Diseases 11th Revision (ICD-11; World Health Organization 2019). Chronic nociceptive musculoskeletal pain is often under-recognised even though it is the most common comorbidity in children with CP (Novak 2014; Westbom, Rimstedt & Nordmark 2017). Causes of pain may be multi-factorial and could be attributed to spasticity (occurring in 75% of children with CP), contractures (in 80% of children with CP) and/or inability to walk (in 33% of children with CP) (Novak 2014). Children with spasticity are largely at risk of contractures. Contractures may lead to hip displacement and further progress into painful hip dislocation. Hip dislocation means that the femoral head is completely displaced laterally out of the acetabulum (Hägglund et al. 2016). Displacement and/or dislocation of the hip is a major hindrance for children with CP. It is present in one out of three children with CP, manifesting itself from 2 to 3 years of age (Huser, Mo & Hosseinzadeh 2018; Novak et al. 2013). Children with bilateral involvement who cannot walk have the highest risk of hip displacement or dislocation. In severe cases, a windswept syndrome may occur, sometimes combined with scoliosis (Hägglund et al. 2016). The windswept hip deformity usually starts from the lower extremities with abduction and external rotation of one hip, with the opposite hip in adduction and internal rotation (Hägglund et al. 2016). Windswept syndrome is difficult to treat and could affect a child’s ability to stand or make sitting and any lying down uncomfortable.

Research in recent years has proven that prevention approaches could manage a child’s pain and other comorbidities, such as hip displacement, epilepsy or sleep disorders, as well as lessen the development of contractures that could worsen the child’s health outcome (Donald et al. 2014; Hägglund et al. 2014). One means of prevention is to systematically follow-up the health status of children with CP, with a focus on the hip displacement and muscle contractures. Systematic hip surveillances have been created by orthopaedic surgeons in collaboration with therapists and used successfully in, for example, Nordic countries, Scotland and Australia (Hägglund et al. 2014; Wynter et al. 2014). However, no such systematic management process or preventive pharmacological treatment is available in South Africa (Donald et al. 2014; Nilsson et al. 2016).

In recent studies, professionals (teachers and therapists) working with children with CP in school settings in South Africa and Sweden were asked about the intervention strategies they used to manage the children’s recurrent pain in school settings (Adolfsson et al. 2018; Johnson et al. 2015; Nilsson et al. 2016). The professionals in the said studies reported on a variety of intervention strategies. Subsequently, action-and-reaction approaches followed by healthcare professionals during pain management were identified (Adolfsson et al. 2018). Action refers to strategies professionals use to prevent pain, while reaction refers to professionals’ strategies to intervene when a child is already experiencing pain. A difference was noted between strategies employed by South African and Swedish professionals, as a reaction approach was more commonly used in South Africa (Adolfsson et al. 2018). Reasons for this finding could be that the Swedish health professionals (who are typically consultants in school settings) had access to interventions to prevent children’s pain. Such interventions are a systematic follow-up surveillance programme for children with CP – also referred to as the Cerebral Palsy follow-Up Programme (CPUP) (Degerstedt, Wiklund & Enberg 2017), as well as pharmacological pain management strategies, for example, botulinum toxin (BoNT) and baclofen intended for the reduction of spasticity and recurrent pain (Hägglund et al. 2014). In contrast, in South Africa, pain management strategies are based on traditional analgesics, that is, paracetamol and non-steroidal anti-inflammatory agents. These drugs are mostly used for acute pain (Wren et al. 2019). In the South African schools where the study was conducted, teachers or nurses sometimes only offered children a glass of water as an alternative to pain medication (Adolfsson et al. 2018; Nilsson et al. 2016). Possible explanations for strategies used by South Africa professionals working in school settings are (1) legislation stipulating that only doctors can prescribe pharmacological treatment, (2) caregivers (e.g. participants in the said study) avoiding analgesics because of potential side effects, (3) medical doctors and orthopaedic surgeons who are only available outside school settings and (4) patients (e.g. children with CP) who have to wait very long periods to get appointments with specialists in public hospitals (Nilsson et al. 2016).

In comparison with Swedish healthcare professionals, the school-based South African health professionals were able to provide ‘hands-on’ or immediate pain management when a child experienced pain. These action–reaction approach differences may also be the reason why there is a higher incidence of children with severely affected CP (Gross Motor Function Classification System [GMFCS] levels IV and V) reported in Africa compared to those in European countries (such as Sweden) and North America (Donald et al. 2014, 2015). Nevertheless, one previous study in this project (Adolfsson et al. 2018) found that the South African participants valued other action interventions implemented globally to support the well-being of children with CP. They suggested additional resources needed to secure a satisfactory pain management intervention for these children.

A comprehensive systematic meta-review was conducted by Novak et al. (2013) to obtain current knowledge about the best available published intervention evidence for children with CP. Data in the current study have been sorted and analysed based on the results and recommendations by Novak (2014) and Novak et al. (2013).

This study focused on pain management for children with CP in South African schools for children with special educational needs. The aim was to investigate how interventions that the professionals reported as beneficial for the child’s pain management complied with published international evidence-based interventions for children with CP, as reported by Novak et al. (2013) at the time of data collection. Another aim was to reflect on the participants’ perceptions about interventions or resources perceived missing to obtain and secure a satisfactory pain management strategy for children with CP. The intention was to provide support for an update of knowledge on both individual level (i.e. professionals) and system level (i.e. decision makers).

The study followed a descriptive with a directed, qualitative approach (Hsieh & Shannon 2005). Using a combination of manifest and latent content analyses (Graneheim & Lundman 2004), professional statements from focus groups were analysed and reflected against published interventions for CP reported as effective by Novak et al. (2013, 2014). The study is limited to pain management in children with CP even though the larger project, of which this study is a part, addressed pain assessment and pain management (Adolfsson et al. 2018; Johnson et al. 2015; Nilsson et al. 2016). Ethics approval was obtained from the Research Ethics Committee of the University (GW20140201HS) and permission was granted from the Gauteng Department of Education (D2014/226), South Africa.

Principals from five public schools for children with special educational needs which accommodate children with CP in the Gauteng area of South Africa gave permission for focus groups to be conducted with staff at their schools in February 2014. The school principals (or a designated person appointed by the principal) completed a short questionnaire to provide the investigators with detailed information on the school, for example, the ages of the children accommodated in the school, the number of children in the school, the number of children with CP in the school as well as whether the school has a hostel facility (see Table 2). The participating schools accommodated children with various types of disabilities, including CP. Three of the five schools were boarding schools where children from rural areas were accommodated in hostels. As the children from rural areas had to attend a boarding school, they typically start schooling at the age of seven. It was reported by some of the participants that these children received limited to no early intervention (EI) services, resulting in hip displacement or dislocations and chronic nociceptive musculoskeletal pain already present.

Upon consent, 38 staff members from the five schools participated in five separate focus groups. Criteria for inclusion were follows: any staff member (e.g. teachers, therapists, psychologists, social workers and personal assistants) who worked at the five schools with children with CP (see Table 3). All participants were women except for one who was the sibling and personal assistant of a child with severe CP. The age range of the participants was 22–64 years, with an average age of 44.4 years. In total, 76% of the 11 teacher participants had at least 4 years of university education with an average of 11.8 years experience in working with children with CP, attesting to their knowledge of this condition. The 26 participating clinicians represented six professions: nursing (n = 5), occupational therapy (OT; n = 8), physiotherapy (PT; n = 6), psychology (n = 1), social work (n = 1) and speech-language therapy (n = 5). Therapists worked in therapy rooms and school classrooms with the children.

The five focus group interview sessions included two identically applied parts, focusing on pain assessment and pain management, respectively. In the beginning of each part, one main question was presented, followed by three supporting sub-questions (Table 4). Piloted interview guides in English directed the introduction of the topic and the performance of the focus groups. Three researchers, of whom two were Swedish and one South African, conducted the focus group interviews that lasted between 70 and 110 min per session. The third investigator (a physiotherapist [PT] having experience with children with CP as well as conducting focus groups) acted as the moderator (Wibeck et al. 2007). The first investigator (a special needs teacher who specialises in pain communication) typed all statements on a laptop and assisted the participants in Afrikaans where necessary. The second investigator (a paediatric nurse with specialisation in pain management) reflected on the statements and asked for more information when needed. The statements were projected onto a wall as they were typed, and the discussions were audio-recorded to be used as a reference for the researchers’ data analysis. At the end of the sessions, a member check was performed. All statements were jointly reviewed and revised or extended where necessary. To validate the data and enable participants to determine that the statements truly represented their experiences, two final questions were asked (White & Verhoef 2005): (1) do these findings accurately represent your experiences? and (2) is there anything we have missed that you feel should be included? For the purpose of this study, statements considered included professionals’ information about their use of methods, knowledge about methods that would be beneficial in their opinion and information about interventions or resources they perceived missing to address pain management of children with CP.

The statements (mostly equal to one meaning unit) were entered on separate spreadsheets for the five focus groups. A few comprehensive statements included two meaning units (Graneheim & Lundman 2004). For analyses in this study, all statements related to pain management were merged onto one spreadsheet, that is, both statements related to current strategies and requested additional resources. Reductions were made because of duplications and statements not related to CP or pain experiences.

All statements were initially reviewed and interpreted together by all investigators. Because the strategies about pain management often included several underlying meanings, the statements had to be interpreted according to the context, which is why a manifest approach was not sufficient. Statements about interventions that focused directly on pain management were linked to one of the categories of evidence-based intervention options for children with CP, as identified by Novak et al. (2013), and for the purpose of this study determined as most relevant, that is, spasticity management, contracture management or improved motor activities and function.

In the next step, the statements were condensed, labelled with a code and sorted into two categories that had previously been identified in the larger project by Nilsson et al. (2016): treatment strategies (i.e. ‘hands-on’ or immediate treatment strategies and medication to relieve pain) and environmental strategies (i.e. interventions influencing a child’s environment). To be clear about the underlying latent content in statements, all three investigators reviewed the linkages and jointly discussed professional interpretations. Thereby, the interpretation could lean on a multidisciplinary background knowledge. Finally, identified interventions were analysed and related to the updated knowledge reported in the contemporary meta-review by Novak et al. (2013). Before the discussion of results, the four levels of evidence were dichotomised into higher and lower levels to indicate intervention options that should be used (S+ and W+) and that should not be used (W- and S-) for pain management.

The five focus groups generated 164 statements that included information about available intervention options and 47 statements on the perceived need for additional interventions or resources. For the analysis, 21 of the statements about additional resources were excluded as they did not focus on pain management but rather on the need for extended resources. Examples of such resources were emotional support and speech training, help with self-care and schoolwork, or support at home that were not the focus of this article.

From one of the statements, one could understand how a child in pain might affect everyone around:

Another statement indicated the breadth of the problems that could exist for the children with CP and that could be taken into consideration in a school setting:

These comments showed that not only musculoskeletal problems but also gastrointestinal dysfunction may cause pain in children with CP (Engel & Kartin 2006). Novak et al. (2012) also reported that constipation is a problem in more than 25% of children with CP.

Findings showed that the participants’ ideas about strategies beneficial for pain management of children with CP could fit into the three categories of outcomes as identified by Novak et al. (2013) and relevant for pain management (see Table 1). The subsequent discussion of the results will thus reflect participants’ concerns of their compliance with up-to-date knowledge as reported by Novak et al. (2013) (Table 5).

Most of the strategies in South African schools had a reaction approach, that is, strategies to intervene when a child is already experiencing pain (Adolfsson et al. 2018). In all focus groups, hip pain was mentioned as a main cause of pain, which might explain the participants’ focus on positioning. A transition to an action approach would most likely help by reducing the children’s spasticity and prevent severe contractures leading to hip displacement progressing into dislocation. Hip dislocation is preventable through early identification and intervention. As hip displacement is directly related to the level of GMFCS and most children remain at the same level from 2 years of age, this system could be used to identify hips at risk and indicate the need for interventions and systematic follow-ups in terms of hip surveillance programmes (Hägglund, Lauge-Pedersen & Wagner 2007).

Positive outcomes of hip surveillance programmes have been reported by researchers (Hägglund et al. 2007, 2014; Wynter et al. 2011). Such programmes include a standardised individual follow-up of gross motor function, clinical assessment and radiological review. It is an ongoing process that is jointly followed by PT, OT and orthopaedic surgeons and should continue until skeletal maturity so that the right interventions can be provided in a timely manner. Intervention plans include first non-surgical strategies, such as positioning, use of orthotics and assistive devices and spasticity-reducing pharmaceuticals. In addition, reconstructive hip surgery and/or SEMLS might be necessary. Preventive surgery can include adductor-psoas tenotomy, various osteotomies of the proximal femur or pelvic reconstruction (Hägglund et al. 2014). Thomason and Graham (2014) pinpoint surgery as essential, providing very good outcomes and an improved quality of life for many children with CP.

Novak (2014) refers to alternatives for effective rehabilitation intervention programmes that could also include pain management intervention. These should ‘include child-active learning-based interventions for motor and functional skill performance gains’ (Novak 2014:1151). Examples were given as bimanual therapy, CIMT, goal-directed training, home programmes and occupational therapy after BoNT. Other alternatives were orthopaedic and therapy interventions such as bisphosphonates, BoNT, casting, diazepam, fitness training and active hip surveillance, as well as compensatory and environmental interventions, such as context-focused therapy. In South African school settings, all the elements of recommended interventions are not relevant in addition to learning tasks. However, healthcare professionals who are working in the schools could keep the alternatives in mind and work for an enhanced focus on preventive pain interventions, implementation of a hip surveillance programme and improved collaboration with external doctors, such as orthopaedic surgeons.

The fact that the data collection was conducted 5 years ago is one of the limitations of the study. Evidence of pain management in children with CP may have changed over these past 5 years but up-to-date universal pain management strategies have been described and discussed in the article. Nonetheless, this study clearly showed a discrepancy between published guidelines and clinical practice within school settings. As a more recent meta-review has not been found, the scientific knowledge is deemed valuable even today. Another possible change during the years concerns the intervention options in the schools. Therefore, a follow-up of this study with new data collection evaluating whether the conditions may have changed since data collection is proposed. As the study was limited to the Gauteng province of South Africa, the results cannot be generalised for the whole country.

This study showed a discrepancy between published evidence and clinical practice for the management of chronic nociceptive musculoskeletal pain in children with CP as reported by professionals working in South African school settings. The results showed that even if evidence of best practice in pain management exists, it may not guarantee that children with CP receive this management in their daily care within school settings. If chronic nociceptive musculoskeletal pain is not acknowledged and treated, it might affect the children’s learning and development.