Unintentional injuries: prevention strategies for under 15s

General

1.1 This is one of three pieces of NICE guidance on how to prevent unintentional injuries among children and young people aged under 15. Several PDG members (including the chair) were co-opted as members of NICE's Public Health Interventions Advisory Committee (PHIAC) to advise on two pieces of guidance developed using NICE's public health intervention process. These covered unintentional injuries on the road and in the home and were published at the same time as this guidance.

1.2 The extent of participation in any activity (that is, someone's exposure to risk of injury) correlates with injury rates. However, multiple risk factors may also correlate with the number and type of injuries in any given situation. Therefore, the determinants of injury (such as exposure and context) need to be understood. Details such as the nature and duration of the activity – and number of people undertaking it – could be used to supplement injury data and develop this understanding. Care is required when interpreting children and young people's self-reported data, as they may be reluctant to report where they have been and what they have done. In addition, younger children do not have a well-developed sense of time, making their exposure difficult to estimate.

1.3 Many areas of the home, road and play and leisure environments have hazards which increase the risk of injury. Supervision, safety equipment and education are important to help keep children and young people safe. Equipment has to be maintained to be effective.

1.4 Some families may not be receptive to advice on how to prevent unintentional injury because of 'fatigue' from repeated contact about other health problems, such as cardiovascular disease (CVD) and cancer.

1.5 Injury prevention interventions can be passive or active. Passive interventions do not require an active change in behaviour (as an example, they could include the presence of fire resistant materials or air bags in cars).

1.6 Children are not just small adults. Their physical, psychological and behavioural characteristics make them more vulnerable to injuries than adults. For example, the small stature of young children increases their risk on the road, where they may be masked by parked cars. Similarly, a given amount of a poisonous substance is likely to be more toxic for a child who has a much smaller body mass than an adult (Peden et al. 2008).

1.7 Targeting specific groups may help reduce health inequalities. However, it will have a limited impact on overall injury rates. Targeted and universal approaches are required to reduce both the overall injury rate and health inequalities.

1.8 Preventing serious injury is important. For every death, there are many more serious injuries which result in hospitalisation and most of these are avoidable.

Legislation, regulation and enforcement

1.9 Caution should be exercised when considering evidence from other countries as different contexts often apply. For example, the drafting and introduction of UK legislation is often preceded by extensive consultation, which is not the case in all countries.

1.10 Legislation can cover everyone, not just children and young people. For example, home safety regulation that requires gas inspections generally benefits everyone in the home.

1.11 Numerous mechanisms are available to encourage compliance with safety procedures (for example, enforcement, insurance, health and safety legislation and the use of penalty points for drivers). However, enforcement activities may be more acceptable in public spaces such as on roads than in private spaces such as the home.

1.12 Levels of compliance with legislation and regulation are dependent upon having a structured and comprehensive inspection process. For example, Australian studies on swimming pools have found that compliance with safety regulations is more likely if: there is a register of households with swimming pools, there is an annual inspection programme, and penalties are enforced for any breach of the regulations.

Injury surveillance

1.13 In 2002, the Home Accident Surveillance System (HASS) and the Leisure Accident Surveillance System (LASS) both came to an end. Since then, there has been a lack of standardised data collection of unintentional injuries in the home and in leisure settings. 'An information revolution' (DH 2010) proposes that health data should be collected from multiple sources and disseminated by a single agency. It highlights the central role that high quality information can play in improving outcomes and narrowing inequalities.

1.14 The Programme Development Group (PDG) acknowledged a number of factors that may confound injury data. This includes the following:

1.15 Sharing injury data between organisations (for example, the ambulance service, hospitals and the police) is necessary to overcome gaps in knowledge and inconsistencies in recording such injuries. However, the PDG was aware that organisations can find it difficult to share data. Barriers can be institutional or relate to the confidentiality and security of personal information.

1.16 Injury rates may vary according to the time of year. For example, children and young people's activity patterns may be different during the school term compared with the school holidays.

1.17 Shortcomings in injury data collection may result from a lack of awareness of the benefits of monitoring and surveillance. For example, emergency department staff may consider data collection an unnecessary burden. Greater awareness of the use and benefits of this information may lead to a greater commitment to data collection among these staff.

Home safety

1.18 The recommendations on home safety assessments and the supply and installation of home safety equipment are aimed at preventing unintentional injuries among all children and young people aged under 15. However, they prioritise households where children and young people are at greater than average risk of unintentional injuries due to one or more factors. For example, those aged under 5 and those living in social, rented or temporary accommodation with families on a low income are particularly vulnerable.

1.19 Extensive evidence suggests that socioeconomic disadvantage increases the risk of childhood injury. Forty-four per cent of lone parents with dependent children are social tenants (Communities and Local Government 2009). Social tenants and often, tenants of private landlords have less income than owner-occupiers.

Given the extent of unintentional injuries among children under 5 in the home – and the increased risk of injuries among disadvantaged families, the PDG has made specific recommendations for these groups. 

1.20 The physical environment may have an influence on the rate and type of injuries that occur. For example, high-rise flats often have potential hazards such as balconies, communal stairs and unsecured windows (Child Accident Prevention Trust 2010). In such situations, tenants may not have permission or the resources to make alterations.

1.21 The evidence available focused on items that need to be fitted to use at home, such as smoke alarms, window restrictors and thermostatic mixing valves (although there was no evidence about some equipment, including carbon monoxide alarms). It does not cover safety devices that do not need installing (for example, those already fitted onto lighters).

1.22 When interpreting the evidence it should be noted that:

1.23 It became compulsory to fit thermostatic mixing valves to bath taps in all new homes in England and Wales from 6 April 2010. Thermostatic mixing valves are usually fitted near to the tap, so that most stored hot water remains at a high enough temperature to kill the bacterium that causes Legionnaires' disease.

1.24 With the exception of window restrictors, all age groups would benefit from home safety equipment (smoke and carbon monoxide alarms and thermostatic mixing valves). Window restrictors should benefit children aged over 2 as they are capable of climbing and falling from an unguarded window. The age at which window restrictors become ineffective is not clear. However, it is likely that most children can overcome child-resistant mechanisms by the time they reach the age of 5. Key-operated locks (where the key is inaccessible to a child) tend to be effective for longer. It is important to note the need to open windows in a fire emergency.

1.25 As more smoke alarms are installed than any other type of safety equipment, there is less potential to use them to reduce health inequalities.

1.26 Gaining access to people's homes needs sensitive consideration. The PDG acknowledge that the home is a private space and access will involve discussion and negotiation with residents.

Outdoor play and leisure

1.27 The PDG agreed with the Royal Society for the Prevention of Accidents (RoSPA) that children should be "as safe as necessary, not as safe as possible". Children and young people learn, develop and mature when playing and taking part in activities that challenge them and that sometimes involves taking risks. Play and leisure activities help children and young people to learn about the complex relationship between themselves and the world in which they live. Exposure to a degree of challenge may be beneficial during these activities. However, a distinction should be made between manageable and unmanageable situations:

1.28 Parents' and carers' and their child's perception of safety can influence the amount of time children and young people spend on outdoor play and leisure activities. These perceptions can be influenced by the media. In addition, fear of litigation can influence the nature and extent of activities provided by educational and play organisations.

1.29 It is difficult to regulate activities such as canyoning and wild swimming and the settings in which they take place. It is also difficult to regulate inland waterways not currently used for supervised recreation.

1.30 The classification of a leisure activity is not always clear. For example, when a child is cycling it's not always clear whether cycling is a leisure activity or is being used as a form of transport. Similarly, it's not always clear whether a child or young person is playing in water or swimming, playing with a ball or participating in sport.

1.31 Media campaigns to promote injury prevention activities may increase health inequalities, as uptake is likely to vary among different groups. For example, disadvantaged families are less likely to respond to health information than families who are more advantaged.

1.32 The PDG acknowledged that dividing on- and off-road cycling into two separate activities was an artificial division, particularly in relation to older children. The scope of the guidance did not include equipment used to prevent against unintentional injuries on the road. However, it did cover outdoor play and leisure, so the use of helmets in parks, on bridleways and in other environments was reviewed. (Children often fall off their bikes, especially when they are learning to ride a bicycle and when they are learning BMX and mountain bike skills, so there is a need to protect them from unnecessary injury.)

1.33 Recommendations have been made about promoting cycle helmets but not about making them compulsory. The PDG was aware of the debate on cycle helmets.

1.34 The PDG considered a number of issues in relation to the use of helmets including the:

1.35 Current playground standards aiming to reduce the incidence of traumatic brain injury are important, as it is a potentially serious injury. Protection against broken arms and legs is also needed, as these are common and can result in disability and deformity.

1.36 Interventions that have been shown to reduce firework injuries in other countries may not, necessarily, have the same effect in England. For example, in countries with drier weather conditions, the danger from unexploded fireworks is greater and so measures to clear them up are likely to have a greater impact. Enforcing firework regulations in England is also different because they are only on sale here for short periods of time. For example, retailers and display organisers are granted temporary licences to sell them in advance of Bonfire Night and other festivals.

Road safety

1.37 The PDG noted several demographic differences in child pedestrian injuries. For example, more boys than girls are injured. In addition, children aged 10 and under are more likely to be injured on minor urban roads, while those aged 11 and over are more likely to be harmed on main roads. It also noted that children living in deprived areas (and those from some minority ethnic groups) are more likely than the general population to make journeys alone or only supervised by an older sibling.

1.38 Most studies on traffic speed are conducted on the main road network. Fewer are conducted on minor residential roads where children and young people are more likely to be present.

1.39 The PDG acknowledged that injury prevention activities should take into account the importance of public transport and sustainable travel modes, such as walking and cycling, which have known health benefits. Reducing traffic speed should help to encourage physically active modes of travel.

1.40 Most studies focus on the evaluation of legislation which is enforced by imposing sanctions on those who break the rules. This is because data on the effect of such interventions are more readily available than for less punitive measures. Although the latter may be equally effective, they have not been recommended due to a lack of evidence.

1.41 Transport studies tend to use a 'before-and-after' design. They estimate the relationship between two or more factors using data collected at a number of specified intervals over a period of time. They require an adequate control to demonstrate causality.

1.42 Children and young people cannot influence the speed or general manner in which vehicles are driven or whether seatbelts are available. In addition, they often have little or no choice about their mode of travel.

1.43 The evidence review on the effectiveness of safety cameras which informed evidence statement 3.1 only included systematic reviews. One of these has since been updated (Wilson et al. 2010) and evidence statement 3.1 has been amended in appendix C of this guidance to include its findings. The systematic reviews in the original report to NICE did not capture evidence from relevant primary sources that report differential effectiveness. However, the PDG noted that the National Safety Camera Programme (Gains et al. 2005) reports differential effects on children, for urban and rural environments and for fixed and mobile cameras. The cost effectiveness review which informed evidence statement 6.5 used primary sources, including the National Safety Camera Programme.

Limitations of the evidence

1.44 The recommendations reflect the evidence identified and the PDG's discussions. The absence of recommendations on any particular measures to prevent unintentional injuries is a result of a lack of evidence that met the inclusion criteria for the evidence reviews. It should not be taken as a judgement on whether or not any such measures are effective and cost effective.

1.45 Repeated testing of outcome measures can affect the validity of an evaluation. For example, a variable that is extreme when first measured will tend to be closer to the mean when measured later. If this statistical effect is not taken into account, caution will need to be exercised when interpreting any conclusions about an intervention's effectiveness.

1.46 Many injury prevention programmes do not lend themselves to the use of 'blinding' (whereby participants are not aware which research study group they have been allocated to). However, it is often possible to have evaluators who are 'blind' to group allocation.

1.47 Although interventions often include adults, children and young people, the outcomes for children and young people are not reported separately.

Studies of the effectiveness of strategic approaches to injury prevention (such as legislation and enforcement) did not provide a strong evidence base for economic modelling.  As a result, most of the assumptions or variables used in the modelling are based on very limited or estimated data and the conclusions should be treated with caution.