Health Promotion International, Vol. 19, No. 2, 215-226, June 2004
HEALTH PROMOTION INTERNATIONAL Vol. 19. No. 2 © Oxford University Press 2004. All rights reserved
A study of safety-promoting activities for children and adolescents in 25 Swedish municipalities
1Department of Public Health Sciences, Karolinska Institute, Stockholm, Sweden and 2National Institute of Public Health, Stockholm, Sweden
Address for correspondence: Karin Guldbrandsson, Department of Public Health, Karolinska Institutet, Social Medicine, Novrbacka 2nd floor, Karolinska Hospital, SE-171 76 Stockholm, Sweden, E-mail: karin.guldbrandsson{at}phs.ki.se
SUMMARY
The purpose of this study was to investigate two elements of the community approach to safety promotion for children and adolescents. The first concerns whether there is any correlation between safety-promoting activities in different municipal settings, and the second whether socio-economic and demographic characteristics affect the level of safety-promoting activities. A cross-sectional study was performed comprising 25 municipalities in Stockholm County. The extent of safety-promoting activities was studied in four settings: the municipal setting in general, the traffic setting, the pre-school setting, and the school and leisure activity setting. Data on nine demographic and socio-economic characteristics of the municipalities were collected from public statistics. There were three main findings. First, safety-promoting activities in the municipal setting in general, the pre-school setting, and the school and leisure activity setting, but not in the traffic setting, were apparently associated with total municipal safety-promoting activities. Secondly, socio-economic characteristics, measured as the proportion of adults with >12 years of education, did not predict the extent of safety-promoting activities. Finally, growing municipalities with a large proportion of children reported fewer safety-promoting activities than non-growing municipalities. This study demonstrated weak correlations between safety-promoting activities in different municipal settings. The finding that growing municipalities have fewer safety-promoting activities provides pertinent lessons for future action and research. The study offers a method by which it may be possible to better assess the complex components of the community approach to safety promotion.
Key words: cross-sectional; municipality; safety promotion
INTRODUCTION
In both safety promotion and health promotion [World Health Organisation (WHO), 1998
; Ytterstad et al., 1998; Bracht, 1998; United Nations, 2001], intervention at the local community level is often emphasized. In European high-income countries, the most important local community structure is the municipality. A municipality typically comprises a total population of 10 000 to 100 000 inhabitants. At this level, there is often an administrative structure with responsibility for public services, including roadwork, schools and various social services. The specific arrangements, however, may vary considerably between countries. For instance, in Nordic countries the municipalities have unique conditions concerning autonomy, democratic control, tax equalization and a comprehensive public sector. The municipal agencies are often responsible for most of the safety-promoting activities. These agencies are internally organized in separate departments, implying, for example, that the planning of roadwork is separated from the planning of school- or leisure-safety activities.
There are two major considerations that might explain the extent of safety activities in a community. The first concerns the way in which safety promotion is organized, and the second is the relative affluence of the municipality. During recent years, a multi-sectorial community approach to the organization of municipal safety has been advocated (Svanström et al., 1995; Andersson, 1996). This approach is used in contrast to an organization in which each sector (like roadwork and schools) carries out its activities independently of one another. The multi-sectorial community approach presumes that there is an interaction between the separate sectors and that this interaction might be enhanced. Thus, if safety promotion in one sector is advanced, the probability of safety promotion being advanced in other sectors in that specific municipality will increase as well. This interaction between different sectors within an agency is crucial for the effectiveness of the community approach, yet this interaction has scarcely been studied empirically.
In a paper by Klassen et al. (Klassen et al., 2000), 32 separate studies evaluating the impact of community-based safety promotion efforts on childhood injuries, safety behaviours and the adoption of safety devices were systematically reviewed. The results indicate that community-based approaches are effective in increasing some safety practices, such as bicycle helmet use and car seat use among children. Furthermore, Klassen et al. suggest that strong evidence supporting the effectiveness of community-based interventions is lacking, in part because few studies used randomized controlled designs and few examined injury rates among children and adolescents as outcome measures. Klassen and colleagues drew three conclusions from their findings. The first conclusion is that the use of multiple strategies grounded in a theory of behaviour change is critical. The second is the notion that to maximize success, interventions should be integrated into the community and approaches should be tailored to meet unique community needs. The third conclusion is the belief that community stakeholders should be included in the development of community-based strategies. In a systematic review of the international literature on health promotion interventions in reducing childhood injuries, Dowswell et al. (Dowswell et al., 1996) reason that there are three factors that lead to success for community-based campaigns: sustained use of surveillance systems, commitment to interagency cooperation, and time set aside for the development of professional networks and the implementation of interventions.
A recently published paper by Towner and Dowswell was based on an examination of 10 community-based injury prevention programmes (Towner and Dowswell, 2002). The authors reached the conclusion that increasing evidence is emerging regarding the effectiveness of such programmes. Important elements of community-based programmes include a long-term strategy, effective and focused leadership, multi-agency collaboration, use of local surveillance, and time to develop and coordinate networks. Towner and Dowswell also stressed an urgent need to develop and monitor indicators in order to assess and maintain a culture of safety, programme sustainability and long-term community involvement. Thus, there are theoretical arguments for the community approach to safety promotion. However, even if empirical evidence for its effectiveness is increasing, such evidence is not yet conclusive.
The possibilities for a municipality to create a safe environment may also depend on available resources. Generally, low socio-economic groups are more afflicted by injuries than affluent groups (Laflamme, 1998; Hjern and Bremberg, 2002). This differentiation has been ascribed to both variation in individual behaviour and the environment (Cubbin et al., 2000). Thus, in order to understand the basis for municipal safety promotion, it is appropriate to investigate the relation between the socio-economic status of individuals in a municipality and the extent of the safety-promoting activities that are carried out.
Municipal safety promotion affects all age groups, yet it will probably affect children, adolescents and elderly people more than adults of working age in that individuals in these age groups are often confined to the municipal geographical area. Accordingly, it is relevant to focus municipal safety-promoting activities on children and adolescents.
Thus, to investigate the usefulness of the community approach to safety promotion, the relationship between safety-promoting activities in different municipal settings was investigated. Moreover, the relative importance of the affluence in the municipality for the level of safety-promoting activities was studied.
METHODS
Study population and design
The study comprised 25 municipalities in Stockholm County with an average of 40 000 inhabitants. A cross-sectional design was used.
Outcome variable: evidence-based safety activities
A set of questionnaires was developed to describe evidence-based safety-promoting activities in the 25 municipalities from May 1999 to April 2001. Four settings were investigated with different questionnaires for each setting: the municipal setting in general, the traffic setting, the pre-school setting, and the school and leisure activity setting. In total, 31 questions were posed, of which 30 are presented in Table 2. Responsible civil servants for each of the settings were selected to serve as respondents to the questionnaires.
|
A prerequisite for the selection of safety-promoting activities to be included in the questionnaires was that they had to be shown by previous research to be effective (Table 1). In addition, they had to bear upon the environment, be present in Swedish municipalities and be accessible by the method used in this study. The safety-promoting activities were mainly identified from seven reviews (Towner et al., 1993
; Munro et al., 1995; Dowswell et al., 1996; Towner et al., 1996; Mølsted et al., 1999; Rivara et al., 2001; Towner et al., 2001).
|
The evidence concerning some of the safety-promoting activities that were chosen to mirror the safety level in the municipalities was found to be contradictory. For instance, Sacks et al. (Sacks et al., 1992
) noted that safety rounds were ineffective, whereas Bjärås et al. (Bjärås et al., 1990) found that safety rounds eliminated risk factors in a public environment. Lindqvist et al. (Lindqvist et al., 2002) included safety rounds in their Safe Community concept. In worksite safety promotion, safety rounds are perceived as effective (Menckel et al., 1996). With respect to training school children in how to swim, Rowe et al. showed that inability to swim is a risk factor for drowning, and Asher et al. suggested that swimming ability could be improved through qualified training (Rowe et al., 1977; Asher et al., 1995). However, the generalizability of these findings to drowning accidents among school children is unknown. Despite these non-conclusive findings, questions about safety rounds and training in swimming were included in the questionnaires.
Evidence-based medicine, with clear levels of evidence and grades of recommendation (Centre for Evidence-Based Medicine, 2002), has been developed in clinical medicine from an individual perspective. Community level activities, however, are not isolated phenomena. Thus, evidence is formulated differently in medicine and in public health. The complexity of health promotion requires measures that are more sophisticated and evaluation research design has to reflect this complexity (Nutbeam, 1998). According to Nutbeam, the use of a diverse range of data and information sources will generally provide more relevant and sensitive evidence of effects than a single ‘definitive’ study. The community-based approach suggests that a multi-agency strategy that combines a variety of interventions may be a promising approach to prevention (Munro et al., 1995). This approach implies that evaluation in terms of a traditional epidemiological study design is not straightforward because of multifaceted interventions and undefined populations. Thus, the different studies chosen to prove the outcome measures of this study reflect this complexity.
For the municipal setting in general, nine questions were posed based on the community-based approach (Table 2) (Svanström, 1993; Dowswell et al., 1996). For the pre-school, and school and leisure activity settings, 10 and nine questions were administered, respectively (Table 2). For the traffic setting there were two questions, both of which concerned speed restrictions (Table 2) (Sabey, 1995; Webster and Mackie, 1996; Tingvall, 1997; Towner et al., 2001).
For the traffic setting, an additional question concerning individual behaviour was included. This question pertained to local activities that encouraged the use of bicycle helmets by children and adolescents (Table 1) (Thompson et al., 1996; Attewell et al., 2001). Because this question did not pertain to the environment, it was analysed separately and hence is not included in Table 2.
The answer to each question was first given a preliminary score that ranged from 0 to 1. These scores were either based on percentage (e.g. ‘Percentages of pre-schools in the municipality with local injury registration: 76–100% = 1.00, 51–75% = 0.75, 26–50% = 0.50, 1–25% = 0.25’) or on rating scales (e.g. ‘How often are pre-school safety rounds performed outdoors? Once a month = 1.00, once every 6 months = 0.50, once a year = 0.20, no safety rounds = 0.00'). Dichotomous variables were also used (e.g. ‘Does the municipality get information about child injuries from the health care sector? Yes = 1.00, no = 0.00'). High level of activities corresponded to high scale values.
Standardized scores were computed for each separate question, for each setting and for each of the municipalities. The standardized score for each question was computed by dividing the preliminary score by the average score for the question. Standardized setting scores (the total score for each setting in each municipality) and standardized municipal sum scores (the total score for each municipality) were computed in an analogous manner.
Within each setting, several groups of questions pertaining to the same issue could be distinguished. These groups of questions were called ‘aspects’. Aspect scores were computed by adding the standardized scores of each question from each aspect. For example, within the municipal setting in general, there were three questions about injury registration (presence, frequency and measures taken). These question scores were added to one aspect score, which was presumed to mirror the injury registration aspect within the municipal setting in general.
The different activities were not weighed regarding the level or quality of evidence or the level of reach into the communities. Thus, all activities were given equal weight. This was done not with the aim of finding single activities, but complexes of activities that could mirror the organizational perspective. The rationale for computing sum scores was the assumption that setting safety and total municipal safety were the result of a large number of safety activities.
Independent variables
To investigate the affluence of the municipality, nine demographic, political, economic and social variables for the 25 municipalities were collected (Public Statistics, 2001).
Procedures
The questionnaires were pre-tested in five municipalities. The WHO Collaborating Centre on Community Safety Promotion in Stockholm had previously appointed four of these municipalities as ‘safe communities’. Three of the five municipalities were located outside Stockholm County, while two were inside. The two pilot municipalities inside Stockholm County were later included in the main study. The final questionnaires were sent to the respondents in the 25 Stockholm County municipalities in May 2001. All of the municipalities had responded by August 2001.
Statistical analysis
Statistical analysis was performed using Pearson's correlation coefficients to assess possible associations between different activities. Statistical significance was set at p < 0.05. For the number of observation units (25 municipalities), p-values <0.05 were obtained if r > 0.3882.
Validation
Answers to three of the questions in this study were compared with answers to similar questions that had been collected from eight municipalities in Stockholm County by the Swedish National Institute of Public Health as a part of a national survey in 2001 (Eriksson, 2001).
RESULTS
Complete or near complete questionnaires from all 25 municipalities were obtained. The percentages of questions that were answered regarding the four municipal settings were 100% (municipal setting in general), 86% (traffic setting), 88% (pre-school setting) and 96% (school and leisure activity setting).
Validity of the municipal reports
It was possible to make 24 comparisons with the national study (three similar questions and eight municipalities). Eighteen answers were identical and six differed. Of the six answers that differed, four could be explained by different coding procedures in the present and the national study, while two could not be explained.
Safety promotion in the municipal setting in general
Inter-sectorial safety promotion groups were the most common general safety activity (Table 2). In two municipalities, all three general safety-promoting activities (injury registration, inter-sectorial groups and general policy) were found. In addition, these two municipalities reported the presence of health-service sector representatives in their inter-sectorial safety promotion groups and safety promotion policies in common with the health sector. At the other end of the range, none of the three general safety-promoting activities were recorded in 11 of the 25 municipalities in the study. Thus, there was a wide range of activity levels. Within the municipal setting in general, there was only one significant correlation, that between the presence of injury registration and the presence of safety promotion policies (r = 0.42).
The traffic setting
Speed restriction initiatives were implemented in only a few of the road systems of the 25 municipalities (Table 2). Only four municipalities had >10% of their road systems covered by a speed limit of 30 km/h and only five municipalities had >10% of their pedestrian road crossings provided with speed restrictions (e.g. speed humps). One municipality implemented both these measures in >10% of their road system. These two traffic measures (speed humps and a speed limit of 30 km/h) did not correlate significantly with one another.
The answers given to the additional question of bicycle helmets demonstrated that municipal promotion campaigns were widespread. Eighty per cent of the municipalities in Stockholm County reported between one and three activities in this field. There was a significant negative correlation between the presence of a 30 km/h speed limit and bicycle helmet campaigns (r = –0.47).
Pre-school setting
Safety-promoting activities were widespread in the pre-school setting, although most developed in four of the 25 municipalities (Table 2). In these four municipalities, injury registration was performed in more than half of their pre-schools, and safety measures were regularly carried out when an injury had been reported. In three of the four municipalities, safety education of pre-school staff was executed regularly. At the other end of the range, neither injury registration nor safety education of pre-school staff was exercised in two of the 25 municipalities.
The school and leisure activity setting
Plans for safety education were the least developed safety-promoting activity within the school and leisure activity setting (Table 2). In three municipalities safety-promoting measures were most developed, with injury registration performed in >50% of the schools and safety measures regularly carried out when an injury had been reported. Furthermore, plans for safety education had been made for schoolchildren in these three municipalities. In two of the municipalities there were plans for safety education for staff. In contrast, none of these activities had been implemented in the three least developed municipalities. In the school and leisure activity setting, a significant correlation was found between injury registration and plans for the safety education of staff (r = 0.57).
Correlations between activities in the four settings
Correlations between total setting scores in general, the pre-school setting, and the school and leisure activity setting were all positive, although none reached statistical significance (Table 3). However, these three setting scores correlated positively and significantly with the total municipal sum scores.
|
Negative correlations were found between the additional question on bicycle helmet campaigns and municipal activities in all four settings. However, only one of these correlations reached statistical significance: the correlation between bicycle helmet campaigns and the total municipal sum scores (r = –0.49).
The use of Pearson's correlation coefficients implied that the variables were distributed near normally, an assumption that was confirmed visually by histograms and because the mean and the median scores differed only slightly.
Demographic and socio-economic municipal determinants
Information on nine demographic and socio-economic characteristics, presumed to mirror the 25 municipalities’ relative wealth, is presented in Table 4. Based on correlations between these characteristics, three dimensions were selected to represent the level of affluence in the municipalities. Socio-economic variables formed the first dimension. The proportion of adults with >12 years of education was selected to represent this dimension because of its high correlation with the other characteristics within this dimension: average annual income (r = 0.89), proportion of unemployed (r = –0.42), proportion of recipients of social allowance (r = –0.47) and proportion of socialists in municipal government (r = –0.80). Population growth was selected to represent the second dimension. It was highly correlated with the proportion of 0- to 19-year-olds (r = 0.52) and with the proportion of inhabitants in sparsely populated areas (r = 0.46). Population size formed the third dimension. Thus, the variable ‘proportion of adults with >12 years of education’ was used as an indicator of socio-economic conditions, and the variable ‘annual population growth’ was used to represent both population growth and the proportion of children and adolescents aged 0–18 years.
|
Only trivial correlations were found between the scores on the safety-promoting activities and adults with >12 years of education (Table 5). The municipality sum score indicated a significant negative correlation with population growth (r = –0.43). Thus, in growing municipalities in which children made up a large proportion of the population, fewer safety activities were developed and applied than in non-growing municipalities. Safety promotion in the traffic setting, in comparison with safety promotion in other settings, appears to be related differently to demographic and socio-economic characteristics.
|
DISCUSSION
There were three main findings. The first finding is that safety-promoting activities in the municipal setting in general, the pre-school setting and the school leisure activity setting are apparently connected with the total safety-promoting activities in the municipalities. Safety-promoting activities in the traffic setting, however, were not related to activities in the other settings. The second finding is that socio-economic characteristics of the population, measured as the proportion of adults with >12 years of education, did not predict the amount of safety-promoting activities. The third finding is that growing municipalities with a large proportion of children and adolescents reported fewer safety-promoting activities than non-growing municipalities.
Limitations and strengths
This study is built on reports of a limited number of safety-promoting activities. The validation comparisons indicate that the respondents' answers regarding the municipal setting in general were reliable, however many safety-promoting features were not appraised. A multi-level study, based partly on the same questionnaires as the present study (except the traffic setting questionnaire), was performed to analyse whether the number of municipal safety-promoting activities corresponds with children's consumption of hospital care as a result of injury (Sellström et al., 2003). From their findings, Sellström et al. concluded that the implementation of multiple safety measures in a municipality had a significant effect on the risk for injury of pre-school children. In municipalities that implemented few safety activities, the risk of injury was 33% higher than in municipalities that implemented a large number of safety activities. Furthermore Sellström et al. observed a similar effect, albeit non-significant, in school-aged children. These results make it reasonable to draw the conclusion that the questionnaires were reliable.
The assessments for each setting might have had too low precision to allow for the detection of relations. The number of questions varied between the four settings (the municipal setting in general had nine questions, the traffic setting two questions, the pre-school setting 10 questions, and the school and leisure activity setting had nine questions). However, the correlations were not stronger between settings in which there were many questions. Thus, having too few questions probably cannot account for the pattern of the present results.
The comprehensiveness of the investigation, regarding both the safety-promoting activities covered and the inclusion of all municipalities in the area, is a strength of the study.
Interpretation of the findings
The findings lend some support to the community approach to safety promotion. However, there is one restriction, which is that the extent of traffic-setting safety initiatives did not vary with the other activities investigated. One reason could be that within the traffic sector there are several other responsible actors (e.g. the Swedish National Road Administration and private road keepers) in addition to the municipalities. This finding is an important exception in that of all injuries, traffic injury is the leading cause of injury death during childhood in most affluent countries. Perhaps municipal road safety is promoted more effectively by other organizations in Sweden (e.g. the Swedish National Road Administration). This finding might have a general applicability, at least in high-income countries. Accordingly, the community approach might only affect certain risks.
The finding of negative associations between bicycle helmet campaigns and other safety activities is intriguing. Campaigns are easier to implement than most other safety initiatives, therefore it cannot be excluded that helmet campaigns might serve as compensation in the absence of implementation of more comprehensive safety initiatives.
The lack of associations between municipal socio-economic characteristics and safety activities is noteworthy, given that a socio-economic gradient in child injuries has been reported (Scholer et al., 1997; Laflamme, 1998; Cubbin et al., 2000). In many countries, municipal resources are dependent on the affluence of their inhabitants. In Sweden, however, this is not the case because of an extensive national system for equalizing municipal resources. This equalization of resources might contribute to the low social gradient in early childhood injuries that is found in Sweden (Engström et al., 2002; Laflamme et al., 2002). This system takes into account the average income of the inhabitants and the proportion of children and elderly people, however it does not compensate for municipality growth. This fact may help to explain the lower level of safety promotion in expanding municipalities. Accordingly, in countries with accommodating systems for equalizing municipal resources, the socio-economic characteristics of the population might not relate to the level of safety-promoting activities.
CONCLUSIONS
This study examining community approaches to child and adolescent safety in Sweden showed weak correlations between selected environmental safety-promoting activities and overall municipal safety measures. In addition, no positive correlations for traffic safety were observed. The finding that growing municipalities had less safety-promoting activities than non-growing municipalities provides some pertinent lessons for future action and research. The study offers a method by which it may be possible to better assess the complex components of the community approach to safety promotion.
ACKNOWLEDGEMENTS
The authors would like to acknowledge the municipal officials who carefully answered the questionnaires. This study has been funded by the Swedish National Institute of Public Health.
REFERENCES
Andersson, R. (1996) Regional strategies for injury prevention: experiences from six county-wide programs in Sweden. International Journal for Consumer Safety, 3, 57–65.
Asher, K., Rivara, F., Felix, D., Vance, L. and Dunne, R. (1995) Water safety training as a potential means of reducing risk of young children's drowning. Injury Prevention, 1, 228–233.[Abstract]
Attewell, R., Glase, K. and McFadden, M. (2001) Bicycle helmet efficacy: a meta-analysis. Accident Analysis and Prevention, 33, 345–352.
Bjärås, G., Danielsson, K., Schelp, L., Sjöberg, D. and Skjönberg, G. (1990) Safety rounds in public environments: experience of a new tool for prevention of accidental injuries. Accident Analysis and Prevention, 22, 223–228.
Bracht, N. (ed.) (1998) Health Promotion at the Community Level: New Advances. Sage Publications, London.
Centre for Evidence-Based Medicine (2002) Levels of Evidence and Grades of Recommendations. Research and Development, Oxford (http://minerva.minervation.com/cebm/NHS).
Cubbin, C., LeClere, F. B. and Smith, G. S. (2000) Socioeconomic status and injury mortality: individual and neighbourhood determinants. Journal of Epidemiology and Community Health, 54, 517–524.
Dowswell, T., Towner, E. M., Simpson, G. and Jarvis, S. N. (1996) Preventing childhood unintentional injuries—what works? A literature review. Injury Prevention, 2, 140–149.[Abstract]
Engström, K., Diderichsen, F. and Laflamme, L. (2002) Socio-economic differences in injury risks in childhood and adolescence: a nation-wide study of intentional and unintentional injuries in Sweden. Injury Prevention, 8, 137–142.
Eriksson, K. (2001) A Questionnaire about Injury Prevention (Unintentional Injuries) (in Swedish). National Injury Prevention Programme, National Institute of Public Health, Stockholm.
Hjern, A. and Bremberg, S. (2002) Social aetiology of violent deaths in Swedish children and youth. Journal of Epidemiology and Community Health, 56, 688–692.
Klassen, T. P., MacKay, J. M., Moher, D., Walker, A. and Jones, A. L. (2000) Community-based injury prevention interventions. The Future of Children, 10, 83–110.
Laflamme, L. (1998). Social Inequality in Injury Risks. Knowledge Accumulated and Plans for the Future. The Swedish National Institute of Public Health, Stockholm.
Laflamme, L., Engström, K. and Hasselberg, M. (2002) Social Differences in Injury Risks. A Report about the Distribution of Injury Risks among Children and Adolescents in Sweden Based on their Socioeconomic Status (summary in English). Swedish Child Safety Commision, Stockholm.
Laforest, S., Robitaille, Y., Lesage, D. and Dorval, D. (2001) Surface characteristics, equipment height, and the occurence and severity of playground injuries. Injury Prevention, 7, 35–40.
Lindqvist, K., Timpka, T., Schelp, L. and Risto, O. (2002) Evaluation of a child safety program based on WHO Safe Community model. Injury Prevention, 8, 23–26.
Menckel, E., Hallgren, L.-E. and Kullinger, B. (ed.) (1996) Fifteen Years of Occupational-Accident Research in Sweden. Swedish Council for Work Life Research, Stockholm.
Mølsted, H., Andersson, R., Hjern, A. and Bremberg, S. (1999) Child Safety in Practice. A Knowledge Summary of Evidence-based Injury Prevention (in Swedish). Injury Prevention Unit, Child and Adolescent Public Health Unit, Huddinge.
Mowat, D., Wang, F., Pickett, W. and Brison, R. (1998) A case-control study of risk factors for playground injuries among children in Kingston and area. Injury Prevention, 4, 39–43.
Munro, J., Coleman, P., Nicholl, J., Harper, R., Kent, G. and Wild, D. (1995) Can we prevent accidental injury to adolescents? A systematic review of the evidence. Injury Prevention, 1, 249–255.[Abstract]
Nutbeam, D. (1998) Evaluating health promotion—progress, problems and solutions. Health Promotion International, 13, 27–44.
Public Statistics (2001) Statistics Sweden (in Swedish). Statistics Sweden, Stockholm (http://www.scb.se).
Rivara, F., Beahler, C., Patterson, M., Thompson, D. and Zavitkovsky, A. (2001) Systematic Review of Childhood Injury Prevention Interventions. Harborview Injury Prevention Research Center, Washington, DC (http://depts.washington.edu/hiprc/childinjury/).
Rowe, M., Arango, A. and Allington, G. (1977) Profile of pediatric drowning victims in a water-oriented society. The Journal of Trauma, 17, 587–591.[ISI][Medline]
Sabey, B. (1995) Engineering safety on the road. Injury Prevention, 1, 182–186.[Medline]
Sacks, J., Brantley, M., Holmgreen, P. and Rochat, R. (1992) Evaluation of an intervention to reduce playground hazards in Atlanta child-care centers. American Journal of Public Health, 82, 429–431.
Scholer, S. J., Mitchel, E. F., Jr and Ray, W. A. (1997) Predictors of injury mortality in early childhood. Pediatrics, 100, 342–347.
Sellström, E. and Bremberg, S. (2000) Education of staff—a key factor for a safe environment in day care. Acta Paediatrica, 89, 601–607.[CrossRef][ISI][Medline]
Sellström, E., Guldbrandsson, K., Bremberg, S., Hjern, A. and Arnoldsson, G. (2003) Association between childhood community safety interventions and hospital injury records—a multilevel study.
Sibert, J., Mott, A., Rolfe, K., James, R., Evans, R., Kemp, A. et al. (1999) Preventing injuries in public playgrounds through partnership between health services and local authority: community intervention study. British Medical Journal, 318, 1595.
Sosin, D., Keller, P., Sacks, J., Kresnow, M. and van Dyck, P. (1993) Surface-specific fall injury rates on Utah school playgrounds. American Journal of Public Health, 83, 733–735.
Svanström, L. (1993) What is a Safe Community and How Can We Plan a Community Safety Programme? Department of Social Medicine, Sundbyberg.
Svanström, L., Ekman, R., Schelp, L. and Lindström, Å. (1995) The Lidköping accident prevention programme—a community approach to injury to preventing childhood injuries in Sweden. Injury Prevention, 1, 169–172.[Abstract]
Thompson, D. C., Rivara, F. P. and Thompson, R. S. (1996) Effectiveness of bicycle safety helmets in preventing head injuries. A case-control study. Journal of the American Medical Association, 276, 1968–1973.[Abstract]
Tingvall, C. (1997) Death Rates in Swedish Traffic are Among the Lowest in the World (in Swedish). Swedish Council on Technology Assessment in Health Care (SBU), Stockholm.
Towner, E. and Dowswell, T. (2002) Community-based childhood injury prevention interventions: what works? Health Promotion International, 17, 273–284.
Towner, E., Dowswell, T. and Jarvis, S. (1993) Reducing childhood accidents—the effectiveness of health promotion interventions: a literature review. Health Education Authority, London.
Towner, E., Simpson, G., Jarvis, S., Dowswell, T., Sowden, A., Sheldon, T. et al. (1996) Unintentional injuries in young people. Effective Health Care, 2, 1–16.
Towner, E., Dowswell, T. and Jarvis, S. (2001) Updating the evidence. A systematic review of what works in preventing childhood unintentional injuries: part 1. Injury Prevention, 7, 161–164.
United Nations (2001) Safety as a Human Right. 19/04/2001. E/C.12/2001/WP.3. (Other Treaty-Related Document). Twenty-fifth session. Item 5 of the provisional agenda. Economic and Social Council, Committee on Economic, Social and Cultural Rights, United Nations, Geneva.
Webster, D. and Mackie, A. (1996) Review of Traffic Calming Schemes in 20 mph Zones. TRL Report 215. Transport Research Laboratory, Crowthorn, England.
WHO (1998) Health 21—Health for All in the 21st century. WHO Europe, Copenhagen.
Ytterstad, B., Smith, G. and Coggan, C. (1998) Harstad injury prevention study: prevention of burns in young children by community based intervention. Injury Prevention, 4, 176–180.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Guldbrandsson, H. Back, and S. Bremberg The significance of policy documents in municipal child health promotion Scand J Public Health, May 1, 2008; 36(3): 228 - 234. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||