A Health Promoting School is defined by the World Health Organisation (WHO, 1998, p. 2) as a school that constantly strengthens its capacity as a healthy setting for living, learning and working. It focuses on fostering learning in health, engaging all school stakeholders, from pupils to education officials; it strives to create a healthy, sustainable environment; and it implements relevant policies and practices and takes a lead within its setting on striving to improve the health of its community members (ibid, p. 3). For over a decade, the Healthy Schools programme has been evaluated internationally. In Hong Kong, a study by Lee, Cheng, Fung and St. Ledge (2006) demonstrated that the programme improved the health of young people. In their study, data were collected from nine schools, representing 962 primary and 1,221 secondary school children who received The Health Promoting School programme for two years. The study found that children’s school performance increased post intervention and this increase was significant among secondary school children (p<.001). Significant improvement was also reported on children’s dietary patterns among both primary and secondary school children post intervention. Antisocial behaviours, mental health and smoking were among other outcomes that improved for both primary and secondary school children. However, alcohol consumption increased for both primary and secondary school children post-intervention. In Canada, the concept of a health promoting school, entitled Healthy Schools BC3, was first introduced in 1933. Since then, the association has evolved to become Physical and Health Education Canada (PHE Canada) and aims to support schools in becoming ‘Health Promoting Schools’, and to foster healthy school communities.

Within Ireland, the WHO Healthy Schools programme was originally implemented and  evaluated in an urban disadvantaged region of Dublin from 2008 to 2012 (Comiskey et al., 2012). Comiskey et al. (2012) examined how the programme was implemented in schools and assessed the baseline results from the primary school children’s physical and mental health status. The key findings from the baseline data reported no significant changes in children’s overall health and well-being for both the intervention and comparison schools. However, a later report (Comiskey et al., 2015) highlighted the challenges of school-based, local implementation and the impact of this design on evaluation results. For example, schools felt they were not equipped to identify the health needs of the children. They also reported concerns of being held responsible if the programme did not work and expressed concerns with the expectation that they would lead the programme implementation. These challenges negatively impacted on the overall programme and as a result no positive impacts were found on the children’s overall health outcomes by the end of evaluation. Hyland, Hyland, Banka and Comiskey (under peer review) also found that children experienced high levels of victimisation in schools; and Banka, Hyland Hyland and Comiskey (under peer review) found relationships between depression and children’s body mass index. 

The aim of the present study was to explore these challenges in more depth from an international perspective and to share the findings with both practitioners and academics. The objectives were to compare victimisation, depression and body mass index among school children internationally and to share the findings from an Irish context with a view to sustaining and extending this learning internationally. This research was supported by the Children’s Research Network Prevention and Early Intervention Research Initiative Senior Scholars Grant Scheme in 2018. 

Within the first objective, seven work packages (WP) were designed, as described below

  • WP1 Measure and describe health risks and victimisation of pre-teen children in Irish school settings – Trinity College Dublin (TCD) team to lead.
  • WP2 Measure and describe health risks and victimisation of similar aged children in another EU setting (The Netherlands) - Toegepast Natuurwetenschappelijk Onderzoek (TNO) team to lead on data extraction and analysis, with support from the TCD team. WP3 Measure and describe health risks and victimisation of similar aged children in a wider, international, high income country setting (data from English children) - University of Technology Sydney (UTS) team to lead on data extraction and analysis to be supported by the TCD team.
  • WP4 Measure (if possible) and describe health risks and victimisation of similar aged children in a low to middle income setting (South Africa) - University of Western Cape (UWC) team to lead on data extraction, and analysis to be supported by the TCD team if required.
  • WP5 Provide a comparative analysis across the range of international settings - TCD team to lead on analysis with support from all other teams.
  • WP6 Provide a meta-analysis of outcomes across the settings if it is possible to obtain common outcome data across the country settings - TCD team to lead supported by all other teams.
  • WP7 Share the learning across the settings in terms of implications for school and national educational and health policies - each team to lead within country domain.


Building on existing and new collaborations, the TNO, UTS and UWC4, teams were invited to participate in a shared analysis of children’s health and victimisation  data. Regular Skype meetings were held; details and descriptions of the four international data sources were discussed; and details of the variables within their measurement instruments and choice of year(s) of data were decided. Permission to access the data was obtained and protocols for data extraction were drawn  up. The data from the urban disadvantaged region of Dublin were gathered among 449 children at baseline (when children were aged between eight to twelve years), representing fifty-five per cent of the estimated sample frames. Following this, data were collected from 428 children in Year 1, and 315 children in Year 2. Children completed the age appropriate self-report psychometric surveys to measure depressive symptoms and Health Related Quality of Life (HRQoL). Depressive symptoms were measured using the Children’s Depression Inventory-Short (CDI-S; Kovacs, 1985). HRQoL was measured using KIDSCREEN-27 (KIDSCREEN Group Europe, 2006)  which consists of twenty-seven items. Body Mass Index (BMI) [weight / (height) 2] was measured by a qualified children’s nurse. Demographic information was also  gathered, such as gender and age. Scores were converted to age and gender specific z scores using AnthroPlus software to categorise weight using the WHO (2007)  thresholds. 

The dataset from UTS included data from 5335 young people aged eleven, thirteen and fifteen years (fifty-two per cent boys) who participated in the Health  Behaviour in School-aged Children (HSBC; www.hbsc.org) study for England. Data from fifteen-year-olds were omitted, as the Irish data included children aged less  than thirteen years. Only data from an urban disadvantaged population were extracted for comparison with the current study. The UTS England study used the Health Behaviour in School-aged Children (HBSC) questionnaire which collects data every four years on eleven, thirteen and fifteen-yearold children’s health and well-being, health behaviours and social environments (Roberts et al., 2009). Data from TNO were sourced from the Local Health Services in Leiden and from the bullying intervention project at TNO. Currently, overall health status and demographic data are in the process of being extracted for urban disadvantaged children aged between eight and twelve years of age. Descriptive and correlational analysis, with mediation and moderation modelling will be conducted on this data.

Preliminary results

Findings on children: As of September 2018, data extraction was ongoing. To compare with the Irish participants from the Healthy Schools evaluation, the UK/UTS5 and TNO teams will extract data by matching the profiles of the Irish children who were recruited through schools designated as urban and disadvantaged. Preliminary analysis of the Irish data on victimisation by Hyland et al. (under peer review) across three waves (baseline, Year 1 and Year 2) found that, relative to findings published recently on the baseline data of the Healthy Schools programme (Hyland, Hyland and Comiskey, 2017), the frequency of victimisation has decreased in Year 2 compared to earlier waves. This was observed for both single-item and multiple item measures of victimisation behaviour. The baseline data showed significantly impaired levels in all aspects of health-related quality of life, as well as increased levels of childhood depression in frequent or occasional victims compared to non-victims. Frequent or occasional victims are those who reported being bullied ‘sometimes’ and ‘always’, while nonvictims are those who reported that they have never been bullied. However, only lower levels of autonomy and parental relationships, and social support and peer relations, were different across victim  categories in Year 2, with no differences observed in Year 1. In terms of childhood depression, Banka et al. (under peer review) found, using mediation models incorporating age, and controlling for gender, that health status  did not change for these urban disadvantaged children. Children with initially poor or good health maintained their health status over the three-year period. In  addition, increases in BMI and decreases in depressive symptoms were also observed.

Findings for practice: Key findings in terms of leading and sustaining a healthy school include: the importance of good leadership within a school; the importance of identifying a local school-based need that was endorsed and recognised by the whole school community, and the readiness of a school for local implementation (Comiskey et al., 2015). The importance of leadership, addressing a need and organisational readiness has been addressed by Fixsen, Naoom, Blase, Friedman and Wallace (2005) in their synthesis of the implementation science literature. Burke, Morris and McGarrigle (2012) have presented a clear framework to guide implementation across time and have provided a list of enablers and barriers organisations need to be aware of. Examples of enablers, from the framework designed by Burke et al. (2012) include leadership, resources, implementation teams, buy-in from stakeholders, and communication. Examples of barriers include vested interest, resistance to change, and external environment. Evidence of a desire to participate but a lack of current leadership, and hence readiness, was found within the UWC. Staff originally involved in their Healthy Schools project had retired and other staff were not directly involved in the original work with the teachers. This also demonstrated a need for not only leadership but distributable leadership within schools. ‘Distributable leadership’ recognises that the “collective interactions among leaders, followers, and their situation are paramount. The situation of leadership isn’t just the context within which leadership practice unfolds; it is a defining element of leadership practice” (Spillane, 2006, p. 4).

Conclusions and implications for future research

While the UWC were not able to participate on this occasion, the learning from this international initiative has potential for African, EU and other countries. Banka, Comiskey and colleagues have initiated a pilot Healthy Schools project in Mauritius. Currently, the team from TCD and the University of Mauritius are applying for funding from the Mauritius Tertiary Education Commission to assess the physical and mental health outcomes among disadvantaged children in Mauritius. The data  will be compared to the data from the Healthy Schools programme in Ireland. Furthermore, the work with local healthy schools in Ireland was instrumental in providing background experience for a further project by Galligan and Comiskey (2017) on young people and alcohol use and distribution in the region, which was supported by the Tallaght Local Drugs and Alcohol Task Force. This in turn has led to an international collaboration with the University of Valencia on the study of binge drinking by young people and associated hospital attendances and health consequences in the region. The sharing of learning from the original Healthy Schools programme has extended beyond the original project and practice and has contributed not only to the region, but also to international practice.