Methods
The objective of the surveys undertaken in this study was to explore the activities being undertaken by children’s centres to prevent thermal injuries, falls and poisonings in children aged < 5 years. Two cross-sectional national surveys of children’s centres in England were undertaken. A survey of fire-related injury prevention took place in 2010,217 whereas the focus of the 2012 survey was the prevention of falls, poisonings and scalds. The 2010 study population consisted of managers of children’s centres in 30 PCTs [three in each of 10 strategic health authority (SHA) areas] across England. Two of the 30 PCTs had merged by 2012, so the 2012 study population consisted of managers of all children’s centres in 29 PCTs. When managers managed more than one centre, they were asked to complete a questionnaire for each centre.
The questionnaires included questions about the management and organisation of children’s centres, child health priority areas and injury prevention activities. We used questions from previous surveys when possible.218–220 We assessed face validity by asking members of the Faculty of Medicine and Health Sciences at the University of Nottingham who were not injury prevention researchers to review the questionnaires. We checked content validity by asking injury prevention experts and the lay research adviser within the study group to review the questionnaire.221 We used a variety of methods to enhance the reliability of questions, including adhering to the ‘principles of questionnaire design’,222 expert advice and piloting.222,223 We piloted the questionnaires using managers from four children’s centres from PCTs across the country who were not part of the final sample. Minor changes were made to the questionnaires following piloting. Copies of the questionnaires are provided in Appendix 3.
Methods that have previously been shown in systematic reviews to increase response rates were used. These included using reminders, providing further questionnaires, keeping the questionnaires as short as possible, providing Freepost reply envelopes, providing an assurance of confidentiality, using the NHS logo on the envelope and questionnaire to try and influence saliency and using university logos on study documentation.69,70 Questionnaires were sent out in March 2010 (fire-related injury prevention) and January 2012 (fall, poisoning and scald prevention). Three reminders were used to improve the response rate.69,70 For the first survey, a random one in 10 sample of questionnaires was double entered and discrepancies identified and corrected. The data entry error rate was 0.19%. For the second survey, all data were double entered by an external data entry company and discrepancies identified and corrected. Data were analysed using Stata/SE 11.0 (StataCorp LP, College Station, TX, USA).
Sample size estimations, based on the prevalance of responses to questions on injury activity in primary care groups (PCGs),219 indicated that for an unclustered design, 196 responses would allow the prevalance of the following to be estimated with a maximum 95% CI of ±7%: unintentional injuries ranked as least important of a range of health topics (66%), taking action to prevent injuries (34%), having a written injury prevention strategy (29%) or believing that the organisation can be effective in preventing injuries (58%).
In 2010 there were 2918 children’s centres in England and 147 PCTs, giving an average of 20 children’s centres per PCT. Assuming a 65% response rate from children’s centres, based on health professionals’ response rates in previous similar surveys218–220 gave an average cluster size of 13 responses per PCT. The design effect to account for cluster sampling assuming an intraclass correlation coefficient (ICC) of 0.05 and an average cluster size of 13 is 1.6. Therefore, 314 responses were required from a total of 25 PCTs (n = 314/13). To ensure a national spread of children’s centres, PCTs were stratified by SHA (n = 10) and three PCTs were sampled at random within each SHA to give 30 PCTs in total. Questionnaires were sent to all children’s centres in those 30 PCTs. By the time of the 2012 survey, more children’s centres had been established and two PCTs had merged, so questionnaires were sent to all children’s centres in the 29 PCTs.
The majority of attitudinal questions required responses on a five-point Likert scale. For the purposes of analysis we combined the ‘strongly agree’ and ‘agree’ responses into an ‘agree’ category and the ‘disagree’ and ‘strongly disagree’ responses into a ‘disagree’ category. Responses to open questions were categorised by generating a coding list and assigning responses to categories. This was undertaken by two researchers working independently and any disagreements were handled by consensus-forming discussions. Categorical variables are described using frequencies and percentages. No adjustment for clustering has been made as these are purely descriptive statistics. Approval for the study was granted by North Nottinghamshire Research Ethics Committee (reference number 09/H0407/44).
Results
In the 2010 survey, 694 questionnaires were posted, five were returned undelivered and one recipient was no longer a children’s centre. The response rate was 56% (384/688). In the 2012 survey, 851 questionnaires were posted, eight were returned undelivered and, of the 526 returned, nine were blank, giving a response rate of 61% (517/843). The denominators vary for responses to individual questions presented in the following sections as not all respondents answered all questions.
Characteristics of children’s centres
shows the characteristics of the children’s centres that participated in the 2010 and 2012 surveys. The percentage of respondents from phase 3 children’s centres increased between the 2010 and 2012 surveys, with a concomitant reduction in the percentages of respondents from phase 1 and phase 2 centres. Most centres were led (i.e. managed) by local authorities or education departments at both time points, with very few having a NHS lead. Responses were fairly evenly distributed across SHA areas in both the 2010 survey and the 2012 survey. In both surveys, a larger proportion of responses came from the South East Coast SHA (20% in 2010 and 19% in 2012).
Characteristics of children’s centres participating in the 2010 and 2012 surveys
Children’s centre priority areas and injury prevention strategies
shows the priority afforded to injury prevention by children’s centres in 2010 and 2012. A similar percentage of respondents considered injury prevention to be among their three main child health priorities in 2010 (58%) and 2012 (60%). In 2010, 16% (59/374, 10 missing) placed injury prevention first whereas in 2012, 16% (80/485, 32 missing) placed injury prevention first. Fewer than half the respondents in 2010 (47%) and 2012 (42%) stated that their children’s centre had an injury prevention strategy, and most did not know if their PCT/local authority had an injury prevention strategy at each time point (61% for PCTs/local authorities in 2010, 65% for local authorities in 2012 and 74% for PCTs in 2012), as shown in .
Injury prevention strategies
Knowledge and attitudes
In the 2010 survey, the potential for improving knowledge was demonstrated as few (11%, 38/348) knew that the most common cause of death was choking and suffocation or that falls are the most common non-fatal injury (33%, 115/350). Respondents’ attitudes towards injury prevention from the 2010 survey are shown in . Attitudes towards injury prevention were positive, with the majority believing that most child accidents were preventable (94%, 358/379) and that children’s centres could effectively prevent accidents (99%, 377/381) and most disagreeing that other agencies had greater responsibilities for preventing child accidents than children’s centres (64%, 244/379).
Attitudes towards injury prevention among respondents. From Watson et al. under the Creative Commons Attribution Non-Commercial 3.0 Unported License (see https://creativecommons.org/licenses/by-nc/3.0/).
Respondent attitudes towards injury prevention from the 2012 survey are also shown in and are very similar to the findings from 2010. In the 2012 survey, knowledge of the main cause of child injury deaths in the under-fives in the home remained poor, with only 12% (51/435) knowing that most child injury deaths resulted from choking and suffocation and 47% (211/445) knowing that falls were the most common non-fatal injuries.
In the 2010 survey, respondents were more likely to think that providing home safety equipment (89%, 330/372), providing one-to-one (88%, 329/372) or group home safety advice (86%, 316/368) or media campaigns on home safety (69%, 256/371) would be more effective than providing leaflets without additional advice (40%, 150/376). The questions on the effectiveness of prevention activities were not asked in the 2012 survey.
Injury prevention activities
Injury prevention activities in 2010 and 2012 are shown in . In the 2010 survey, 97% (364/376) of centres were involved in some form of injury prevention including displaying posters on child safety (97%, 371/382), participating in Child Safety Week (93%, 348/376), inviting outside speakers to talk to parents (78%, 293/378), collecting data on child accidents (56%, 205/365), lobbying or campaigning on child safety issues (34%, 122/364), working with local media (17%, 63/372) or issuing first aid kits (15%, 55/375). The involvement in injury prevention in 2012 was very similar to that in 2010.
Injury prevention activities undertaken by children’s centres.
Centres provided advice on a range of fire-related injury prevention topics (). Advice was most commonly provided on general fire safety, smoking cessation and bonfire and firework safety and least commonly on barbecue safety, candle safety, handling hot irons and making fire escape plans. Providing leaflets was the approach most commonly used to address most of these topics.
Advice provided by children’s centres on fire prevention (2010 survey). From Watson et al. under the Creative Commons Attribution Non-Commercial 3.0 Unported License (see https://creativecommons.org/licenses/by-nc/3.0/).
Centres provided advice on a range of falls, poisonings and scalds prevention topics (–). For falls prevention, advice was most commonly provided on stair safety, not leaving children on high surfaces, what to do if a child has a head injury and general falls prevention and was least commonly provided on non-slip bath mats, baby walker safety and high chair and pushchair safety. For poisoning prevention, advice was most commonly provided on safe storage of hazardous substances such as medicines and household chemicals and general poisoning prevention and least commonly provided on poisonous plants and disposal of unwanted medicines. For scalds prevention, advice was most commonly provided on handling hot drinks, general scald prevention and cooking safety and least commonly provided on thermostatic mixer valves (TMVs). It appears that advice in groups and one-to-one advice was being used more commonly for falls, poisonings and scalds prevention in 2012 than for fire-related injury prevention in 2010.
Advice provided by children’s centres on falls prevention (2012 survey). From Watson MC, Mulvaney C, Timblin C, Stewart J, Coupland CA, Deave T, Hayes M, Kendrick D. Missed opportunities to keep children safe? National Survey of injury prevention (more...)
Advice provided by children’s centres on poisoning prevention (2012 survey). From Watson MC, Mulvaney C, Timblin C, Stewart J, Coupland CA, Deave T, Hayes M, Kendrick D. Missed opportunities to keep children safe? National Survey of injury prevention (more...)
Advice provided by children’s centres on scald prevention (2012 survey). From Watson MC, Mulvaney C, Timblin C, Stewart J, Coupland CA, Deave T, Hayes M, Kendrick D. Missed opportunities to keep children safe? National Survey of injury prevention (more...)
In the 2010 survey, two-thirds (64%, 245/384) of centres were aware of a home safety equipment scheme in their locality, whereas one in five (21%, 79/384) did not know if their area had a scheme. One-quarter of those with schemes (26%, 60/233) had schemes provided through the Royal Society for the Prevention of Accidents (RoSPA) national Safe At Home scheme and just over half the schemes (58%, 135/234) were based within children’s centres. Many schemes were fairly new, with 50% (122/245) being established within the preceding 18 months. Schemes provided, and in most cases fitted (78%, 186/238), a varying range of items of safety equipment, most commonly free (68%, 165/241) or at low cost (18%, 43/241). Equipment provided included stair gates, fireguards, cupboard locks, window catches and furniture corner covers. Stair gates were the most commonly provided (91%, 220/242) and furniture corner covers the least commonly provided (42%, 102/242).
In the 2012 survey, fewer centres (42%, 217/517) reported a home safety equipment scheme in their area and one-fifth (22%, 112/517) did not know if there was such a scheme. Of those that had a scheme, 7% (17/248) were part of the Safe At Home national scheme organised by RoSPA. Similar to the 2010 findings, almost half (47%, 101/217) the schemes had been in operation for < 18 months and 9% (19/217) had been in operation for > 4.5 years. Over half the schemes (53%, 111/211) operated from children’s centres. Schemes provided corner covers (68%, 104/154), devices to measure bathwater temperature (51%, 76/149), first aid kits (29%, 42/143), fridge locks (52%, 78/149), lockable medicine cupboards (10%, 15/143), safety catches for cupboards and drawers (74%, 111/150), safety gates (63%, 95/150), TMVs (4%, 5/140) and window locks (47%, 69/148). Most provided free (60%, 128/214) or low-cost equipment (34%, 73/214) and a smaller number loaned the equipment (5%, 10/214). Most (69%, 140/202) delivered equipment to homes and fitted it (55% 114/206).
Joint working
In the 2010 survey, few respondents (15%, 56/375) were aware of a local child accident prevention group. There was evidence of joint injury prevention work being undertaken with a range of organisations, most commonly community nursing services (86%, 331/384), fire and rescue services (69%, 266/384) and road safety organisations (61%, 233/384). There was also evidence of referral to other services such as NHS smoking cessation services (95%, 360/377) and fire and rescue services for smoke alarms (86%, 321/375) and fire safety risk assessments (85%, 309/362).
In the 2012 survey, 14% (68/503) of respondents knew of a child accident prevention group in their area and 59% (296/503) stated that they did not know whether or not there was such a group. In 2012, centres were not asked if they worked with other organisations on injury prevention. Some centres referred families to safety equipment schemes (47%, 221/466), to pharmacists for the safe disposal of unwanted medicines (49%, 230/472) and to organisations for home safety checks (53%, 251/473), most of which referred to fire and rescue services (62%, 122/198). Very few referred families to an organisation for TMVs (3%, 13/461).
Barriers to, and enabling factors for, injury prevention work
The main barriers to, and enabling factors for, injury prevention activities in the 2010 and 2012 surveys are shown in and , respectively. Lack of capacity in terms of staff time (39%, 131/339 in 2010; 39%, 162/417 in 2012) and lack of funding (33%, 111/339 in 2010; 52%, 216/417 in 2012) were the most frequently mentioned barriers. The most frequently mentioned enabling factors were access to families (45%, 113/249 in 2010; 39%, 121/312 in 2012) and working with other agencies (44%, 110/249 in 2010; 35%, 109/312 in 2012).
Barriers to injury prevention work.,
Facilitators for injury prevention work.,
Support for injury prevention activities
In the 2010 survey, most respondents stated that training (97%, 362/373), provision of educational materials (95%, 351/369), examples of good practice (94%, 341/363), help with planning injury prevention (94%, 343/366), support for working with partners (89%, 320/358) and communities (88%, 311/354) and help with evaluating activities (85%, 303/356) would be useful for their centre.
Discussion
Main findings
In both surveys we found that around 60% of children’s centres identified unintentional injuries as one of their three main priorities but fewer than half had a written injury prevention strategy. Providing leaflets to parents was the most common approach for delivering injury prevention information in 2010 and it remained a common method in 2012, despite this being perceived as less effective than other methods. Although managers held positive attitudes towards injury prevention, they had gaps in their knowledge about injury prevention and about important local initiatives. Two-thirds of centres had access to a local home safety equipment scheme in 2010 but this had fallen to 42% in 2012, with fewer schemes in 2012 providing and fitting free equipment. Our findings suggest that most centres do not have an evidence-based strategic approach, that child injury prevention appears to be a neglected area within children’s centres given the scale of the problem and that most centres would welcome help and support in planning, delivering and evaluating child injury prevention. The findings suggest the considerable scope for improving the provision of child injury prevention activities in children’s centres, which is the focus of the RCT (study M) undertaken in work stream 6 of this programme (see Chapter 7).
Strengths and limitations of the study
Our response rates are similar to those of surveys of other professional groups225 but, as in many surveys, non-response bias may have occurred whereby respondents may have been more interested and active in injury prevention than non-respondents. If this is the case the findings may overestimate injury prevention activity within children’s centres. Similarly, as we collected self-reported activity data, social desirability bias may have led to overestimation of ‘true’ activity levels. If either type of bias has occurred, given the scope for increasing injury prevention that we have demonstrated, this would strengthen our conclusions.
Our surveys provided a broad overview of the injury prevention activity taking places in children’s centres. The survey was not able to explore injury prevention activity in detail or the motivations for choosing particular ways of working or undertaking particular activities. Such information is important for designing interventions for delivery in children’s centres and for understanding and developing the role that children’s centres play in child injury prevention.
Data from the national database of children’s centres (Leila Allsopp, Department for Children, Schools and Families, 1 July 2009, personal communication), which we used as our sampling frame in 2010, indicated that 37% of centres were established in phase 1 (2004–6), 59% in phase 2 (2006–8) and 4% in phase 3 (2008–10). The database that we used as our sampling frame in 2012 (Shirley Best, Department for Education, 29 November 2011, personal communication) indicated that 31% of centres were set up in phase 1 (2004–6), 49% in phase 2 (2006–8) and 21% in phase 3 (2008–10). The phase of the centres responding to our 2010 and 2012 surveys was similar to that in the national sample, suggesting that our findings should be broadly generalisable to children’s centres across England. One-fifth of responses came from children’s centres located within the South East Coast SHA in the 2010 and 2012 surveys. This reflected the existence of a larger number of children’s centres in this area compared with other SHAs in the national database of children’s centres.
Comparisons with existing research
We were not able to find any other published studies exploring children’s centre injury prevention activities for comparison with our study. Previous surveys of injury prevention activities by health authorities218 and PCGs219 report findings similar to ours in terms of lack of capacity, lack of useful data and lack of prioritisation of injury prevention work.218,219
Despite national publications highlighting the importance of child injury prevention and the priority it should be afforded,1,19,22,25 child injury prevention was not among the top three child health priorities for two-fifths of children’s centres that responded to our survey. We also found lack of a strategic approach to injury prevention, with many centres not having a written injury prevention strategy and reliance on less effective methods of behaviour change such as providing leaflets,226 suggesting that better use could be made of the current evidence base, consistent with the conclusions in the Better Safe Than Sorry report.19 As most managers held positive attitudes to injury prevention, believed that children’s centres could be effective in preventing injuries and were keen to receive support, there is scope for further developing the injury prevention activities being delivered by children’s centres.
Although we found evidence of joint working with individual organisations, most respondents did not know of the existence of local injury prevention groups or strategies, suggesting suboptimal partnership working227–230 despite recent recommendations.1,19,22,25 Working effectively in partnership across agencies and organisations can be a complex process231–233 and it is likely that children’s centres will need support to do this. There have been numerous recommendations for the creation of local injury prevention co-ordinator posts19,22,234,235 and, if such posts are established, these could support children’s centres in their injury prevention work.
How these findings inform other research within the Keeping Children Safe programme
Our surveys confirm the scope for improving the provision of child injury prevention activities in children’s centres. The findings from study D have been used to inform the development of guides for interviews with children’s centre managers and staff to explore barriers to, and facilitators of, injury prevention (study G in work stream 4; see and Chapter 4). The findings have also been used to develop an injury prevention intervention (an IPB) for delivery by children’s centres, which was evaluated using a RCT (study M in work stream 6; see Chapter 7). Finally, the findings were used to develop a second IPB incorporating the findings from all of the studies in the KCS programme.
