Injury, Int. J. Care Injured 41S (2010) S16–S19
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Special issues in injury prevention research: Developing the science of program implementation Roderick John McClure a,*, Elizabeth Davis b, Emily Yorkston c, Per Nilsen d, Philip Schluter e, Lyndal Bugeja a a
Accident Research Centre, Monash University, Melbourne, Australia1 Injury Prevention and Control (Australia) Ltd, Brisbane, Australia c ARTD Consultants, Sydney, Australia d Department of Medicine and Health Sciences, Division of Social Medicine and Public Health Science, Linko¨ping University, SE-58183 Linko¨ping, Sweden e Faculty of Health and Environmental Sciences, Auckland University of Technology, Auckland, New Zealand
b
A R T I C L E I N F O
A B S T R A C T
Keywords: Models Policy Public health Implementation research
Introduction: Despite widespread application of the public health approach to injury prevention, there is an acknowledged limitation in the extent to which it facilitates translation of research evidence to injury prevention practice. Aim: In this paper we clarify the ecoepidemiological principles that underpin the public health approach to injury control in order to support explicit efforts to address the multilevel component causes and improve the evidence base on which effective program implementation depends. Keypoints: If injury causation is understood in terms of the ecoepidemiological model rather than the proximal risk factor model, then quantification of the context in which the event occurs is evidently as important as quantification of the energy exchange event itself. The ecoepidemiological model of injury causation recognises barriers and facilitators to injury prevention as component causes of population burden of injury and thus delineates these factors as legitimate targets for intervention. Injury prevention programs that are designed to specifically address the factors causally related to the program’s implementation, as well as the contextual factors that determine the characteristics of the energy exchange event, are more likely to be implemented and more likely to result in effective uptake of recommended energy exchange countermeasures. Conclusion: Interventions to reduce the burden of injury in the population should address the individual level factors that increase the risk of injury, but also the upstream factors that influence the extent to which there is widespread adoption by individuals of the recommended countermeasures. ß 2010 Elsevier Ltd. All rights reserved.
Introduction Three seminal papers have contributed to explaining the paradigm within which injury prevention is currently practiced; On the escape of tigers: An ecological note,6 The epidemiologic basis for injury prevention,24 and From discovery to delivery: Injury prevention at CDC.22 As a result of the work of authors such as these, the public health approach to injury prevention recognises energy as the causal agent of injury, and the contextual importance of the social and physical environment in which energy exchange
* Corresponding author at: Accident Research Centre, Room 126, Building 70, Monash University, Clayton, Victoria 3800, Australia. Tel.: +61 3 9905 4372; fax: +61 3 9905 4363. E-mail address:
[email protected] (R.J. McClure). 1 MUARC Webpage: www.monash.edu.au/muarc. 0020–1383/$ – see front matter ß 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.injury.2010.05.001
occurs.2,4,25,9,21 The scientific rationale for basing efforts to prevent injury on a thorough understanding of its causation is now established.22,8 Whilst the rational basis for the public health approach is well accepted, the practice of program implementation has been hampered by the lack of a clear epidemiological explanation of what has been referred to as ‘the web of injury causation’.19 The ‘proximal risk factor approach’ of Modern Epidemiology20 remains the predominant epidemiologic model underpinning etiologic and evaluation research and prevention practice.18 Consequently, injury prevention practice has tended to concentrate on specific countermeasures related to energy control rather than on interventions that address the contexts in which the energy exchange occurs. The risk factor approach to program implementation22 provides limited insight to support changes in leadership, infrastructure, and capacity that are prerequisites of successful uptake of evidence-based safety interventions. Despite widespread
R.J. McClure et al. / Injury, Int. J. Care Injured 41S (2010) S16–S19
application of the public health approach to injury prevention, there is an acknowledged limitation in the extent to which it facilitates translation of research evidence to injury prevention practice.17,3,10 In this paper we clarify the epidemiological principles that underpin the public health approach to injury control in order to support explicit efforts to address the multilevel component causes and improve the evidence base on which effective program implementation depends. We argue that a full elucidation and quantification of the ecoepidemiological23 cause of injury includes an identification of the social systems that are responsible for the circumstances leading to the energy exchange event. By recasting injury causation in terms of the ecoepidemiological model, the barriers and facilitators to program implementation are incorporated as causal factors in the explanation of the population injury burden. We conclude that interventions to reduce the burden of injury in the population should address the individual level factors that increase the risk of injury, but also the upstream factors that influence the extent to which there is widespread adoption by individuals of the recommended countermeasures. This conclusion creates a substantial challenge for injury prevention research, because whilst the concept of ecological causation of injury is straightforward, empirical quantification of the relationships between components of the web of causation is extremely difficult. To date there have been no reported studies in the literature that have delineated the relevant multilevel causal factors for a given injury type and then developed, implemented and evaluated an injury prevention program based on this ecoepidemiological framework. Injury causation Physical injury results when the magnitude of the biologically relevant dose1 of energy delivered to a tissue of the body is beyond the limits compatible with the tissue’s physiological function. Psychological injury occurs when a person’s subjective experience of an event is beyond the limits that can be accommodated by their functional psychological response, irrespective of whether physical injury was sustained. The cause of a given injury is the set of conditions or attributes that, had the set not existed, the given injury would not have occurred. Any condition or attribute that could affect the range within which a person can continue their uninterrupted physiological or psychological function, or that could extend the energy exposure or psychological experience beyond that range, is potentially an element of the causal set. In isolation, these conditions and attributes rarely lead to injury, but in combination they may contribute to an aggregation that is sufficient for an injury to result. Whilst all the elements in the causal set have their ultimate effect at a single point in time, many elements are products of other elements that preceded them. Although the effect of the causal set is experienced by the injured individual at the personal level, elements of the set may be properties of groups of people, a geographical area, or an ambient environment.23 In addition to the known and observable elements, the causal set for a particular injury can also include a number of unobservable and/or unknown elements. This is particularly true of elements relating to the contexts in which injuries occur that have a part to play in an injury’s causation, but that are not evident to an observer at the time of injury occurrence.11 The large number of unknown elements in most injury causal sets gives the incorrect impression that accidents are random events rather than the deterministic outcomes of a causal process. A causal set may also contain elements which are observed in circumstances where injuries do not eventuate. This is because in those circumstances the total aggregation of elements does not constitute a sufficient cause for injury.
S17
The above account of injury causation can be exemplified by a hypothetical case of a fall-related hip fracture in an older woman. In the case where an older person makes an uncontrolled turn, trips on an uneven surface and falls to the ground landing on her hip, the potential energy when erect is converted to kinetic energy during the falling process and transmitted through the greater trochanter at the point of impact. If the energy is greater than can be absorbed by the structures and tissues it passes through, a fracture of the neck of femur results. In this example the causal set includes elements relating to the uneven surface, the uncontrolled turn, the woman’s muscle strength, balance and bone density, and the underlying nutritional and behavioural risk factor history that has contributed to her frailty and osteoporosis. The set includes the woman’s economic resources, access to health care, and healthrelated knowledge, attitudes and behaviour that have contributed to her health status, together with the social factors that determined the quality of the physical environment in which her fall occurs. Importantly, the set also includes those upstream and systems factors, related to industry, government and community priorities that explained why an effective injury prevention program was not operational at the time of the fall. The elements that relate to the cause of the fall are a subset of the total cause of the injury sustained. In this example, the combination of these elements constitutes a causal set that is sufficient to lead to the fractured neck of femur. Injury prevention It follows from the above explanation of injury causation that if one could prevent the occurrence of a causal set by affecting at least one of the elements, then one would have prevented the occurrence of this particular injury. Injury prevention is the process of managing individual and environmental attributes in such a way as to minimise the prevalence of injury causal sets and thus decrease the probability of an injury outcome. Population-level social policies determine the contexts in which the potential injury events may arise. Effective policies result in a reduced prevalence of injury risk factors. They encourage widespread adoptions of processes and practices that operate at each of the multiple levels of the ecoepidemiological determination of injury to encourage the minimisation of individual injury risk exposures for people within the population. In the previous example of the fall-related fractured neck of femur, the ecoepidemiological approach to falls injury prevention advocates for the specific addition of health-ageing policies to support the adoption of individual risk assessments and targeted interventions.15,16 The healthy-ageing policies are designed to: (i) enable the individual and societal behaviours (e.g. improved social inclusion, financial independence, appropriate housing and transport access) that in turn; (ii) support a reduction in the intermediate determinants of falls injury (e.g. physical inactivity, poor nutrition, smoking, excess alcohol consumption, hazardous environments), that leads to; (iii) decreases in the probability of aggregations of proximal risk factors (e.g. osteoporosis, lack of muscle strength and balance, uneven surface) being sufficient for an injury to occur. Challenges for injury prevention research The orthodox account of the public health approach to injury prevention based on the risk factor epidemiological model describes four stages: Surveillance; Risk and protective factor definition; Development and evaluation of countermeasures, and finally; Prevention program implementation.22 In accordance with this model, the usual approach to prevention is to carefully accumulate evidence relating to problem definition, causation, and
S18
R.J. McClure et al. / Injury, Int. J. Care Injured 41S (2010) S16–S19
intervention efficacy within the research domain and then move into the public domain in an attempt to translate research findings into practice. There is a growing literature however, demonstrating the limited extent to which quality evidence has lead to policy action when this evidence has simply been placed on the public agenda.14 As is illustrated by the above falls injury example, the ecoepidemiological specification of injury causation solves the implementation block by explicitly including in the causal model the contextual factors that facilitate program implementation. It is not sufficient to identify lack of strength and balance as a proximal risk factor for falls in older people and advocate strength and balance training as a prevention measure,5 without also understanding the personal and societal drivers of physical activity, and the social and political barriers to effective implementation of the strength and balance training as a population-level intervention. The ecoepidemiologicial model enables implementation programs to focus on upstream solutions that facilitate individual and societal behaviours and entrench physical activity, healthy nutrition, and active social involvement from an early age in addition to the clinical interventions that address medications, gait and balance.12 In addition, the model enables implementation programs to achieve local level engagement critical to the success of population-based interventions, whilst at the same time working with the social institutions responsible for the environments, conditions and attributes to ensure falls safe environments, policies, and practices in relation to physical activity are created and sustained.12 For the ecological approach to move beyond rhetoric, injury prevention researchers need to identify and measure the relevant contextual factors responsible for the circumstances and events that can lead to injury (including the factors that are barriers and facilitators to prevention program activity), as well as the risk factors relating to the potential energy exchange process. They need to quantify the prevalence of these factors and the relationships between these factors and risk of injury outcome. Researchers then need to work with practitioners to develop programs that specifically address the factors identified to be most important in the causation of the target injury. Evaluation can then be achieved by monitoring changes in prevalence of the identified factors, and consequent reductions in the injury of interest. As noted by Robertson,19 this is a complex expectation. In particular it requires researchers to be highly selective about what they identify to be relevant if there are going to be able to achieve the goal. The difficulty of this challenge is evidenced by the fact that there has not yet been an injury prevention program reported in the literature that has been developed, implemented and evaluated on the basis of an articulated ecoepidemiological characterisation of a selected target injury. Conclusions The increasing burden of injuries is an international challenge of global importance.7 The challenge has been not so much ‘‘what works’’ in injury prevention but ‘‘how to make it work’’ at the population level.13 The ecoepidemiological approach, explained here in detail for the first time in the context of injury prevention, offers a solution by providing a methodology for quantifying the components of an energy change event as well as the context within which the event occurs. The ecoepidemiological model is thus the unifying theory which links causation and implementation. It thus enables an examination of the effectiveness of any specific countermeasure to be understood in terms of the circumstances surrounding the operationalisation of the countermeasure.
The empirical design and evaluation of injury prevention programs should be consistent with the logic that underpins our understanding of injury causation. The science of implementation can be defined as the methodologies used to understand, develop and quantify the effectiveness of complex ecoepidemiological interventions. Implementation science based upon the ecoepidemiological understanding of injury causation provides the framework to enable better empirical application of Haddon’s original ecological concepts. What this topic is about: 1. The importance of basing efforts to prevent injury on a thorough understanding of its causation is well established. 2. The ecoepidemiological model is the unifying theory that links causation and prevention practice.
Common problems and challenges: 1. A common problem is to consider injury causal factors in isolation from their contexts. 2. The challenge is to recognise and measure the critical contextual factors relevant to effective implementation of prevention programs and be able to quantify the role of these factors in the causation of injury-related harm.
Tips for researchers: 1. Understand the context of the problem under consideration. 2. Formulate the injury prevention research question in terms of factors that are related to implementation of the anticipated findings.
Conflict of interests The authors have no competing interests in relation to this manuscript or its contents. Contributions The ideas for this paper arose from collective discussion over the course of several years during which the authors have collaborated on systematic reviews, intervention studies and program evaluations relating to the manuscript topic. All the authors have made an important contribution to the development of the ideas presented in the paper. The first author wrote the first draft and is the guarantor. All authors have been integrally involved in reviewing and extensively editing the text as it has progressed through several iterations. References 1. Armstrong BK, White E, Saracci R. Principles of exposure assessment in epidemiology. Oxford: Oxford University Press; 1994. 2. Blank D. Injury control from the perspective contextual pediatrics. J Pediatr (Rio J) 2005;81(5 Suppl.):S123–136. 3. Brussoni M, Towner E, Hayes M. Evidence into practice: combining the art and science on injury prevention. Inj Prev 2006;12:373–7.
R.J. McClure et al. / Injury, Int. J. Care Injured 41S (2010) S16–S19 4. Gielen A, Sleet D, DiClemente R, editors. Injury and violence prevention: behavioral science theories, methods and applications. San Francisco, CA: Jossey-Bass; 2006. 5. Gillespie LD, Robertson MC, Gillespie WJ, et al. Interventions for preventing falls in older people living in the community. Cochrane Database Syst Rev 2009 (2). 6. Haddon W. On the escape of tigers: an ecologic note. Am J Public Health 1970;60:2229–34. 7. Krug E. Injury surveillance is the key to preventing injuries. Lancet 2004;364 :1563–6. 8. Langley J, Brenner R. What is an injury? Inj Prev 2004;10:69–71. 9. Lu TH. Unalterable host factors? A social epidemiologist’s view of the Haddon Matrix. Inj Prev 2006;12:285–6. 10. Mallonee S, Fowler C, Istre GR. Bridging the gap between research and practice: a continuing challenge. Inj Prev 2006;12:357–9. 11. Martuzzi M. Science, policy, and the protection of human health: a European perspective. Bioelectromagnetics 2005;17:S151–156. 12. McClure RJ, Hughes K, Ren C, et al. The population approach to falls injury prevention in older people: findings of a two community trial. BMC Public Health 2010;10:79. 13. Moller J. Reconsidering community based interventions. Inj Prev 2004;10:2–3. 14. Peek-Asa C, Casteel CH. Documenting the need for translational research: an example from workplace violence prevention. Inj Prev 2010;16:50–2.
S19
15. Peel NM, McClure RJ, Hendrikz JK. Psychosocial factors associated with fallrelated hip fractures. Age Ageing 2007;36:145–51. 16. Peel NM, McClure RJ, Hendrikz JK. Health-protective behaviours and risk of fallrelated hip fractures: a population-based case–control study. Age Ageing 2006;35:491–7. 17. Pless I. Bridging from research to practice. Inj Prev 2004;10:1–2. 18. Rivara FP. Prevention of injuries to children and adolescents. Inj Prev 2002;8(Suppl. 4):iv5–8. 19. Robertson LS. Causal webs, preventive brooms, and housekeepers. Soc Sci Med 1998;46:53–8. 20. Rothman KJ. Modern epidemiology. Boston: Little Brown; 1986. 21. Runyan CW. Introduction: back to the future—revisiting Haddon’s conceptualization of injury epidemiology and prevention. Epidemiol Rev 2003;25: 60–4. 22. Sleet DA, Hopkins KN, Olson SJ. From discovery to delivery: injury prevention at CDC. Health Promot Pract 2003;4:98–102. 23. Susser M, Susser E. Choosing a future for epidemiology. II. From black box to Chinese boxes and eco-epidemiology. Am J Public Health 1996;86:674–7. 24. Waller JA. The epidemiologic basis for injury prevention. Pub Health Rep 1985;100:575–6. 25. Zwi A. Injury control in developing countries: context more than content is crucial. Inj Prev 1996;2:91–2.