Implementing the national priorities for injury surveillance

Decreasing the global burden of injuries is one of the main challenges for public health in the 21st century.1 In Australia, injuries account each year for around 370 000 hospital admissions and 7900 deaths, and cost the health system around $4.1 billion.2,3 Injury is the leading cause of death in Australia for individuals aged 1 to 44 years.4 Unintentional injuries rank third highest across all health outcomes for years of life lost, fifth highest in terms of incidence and total mortality, and sixth highest for disability-adjusted life years.5 Most importantly, all these injuries are potentially preventable.

Information about public health expenditure on injury prevention is difficult to obtain, but available data indicate that this expenditure is disproportionate to the magnitude of the problem. For example, in 2004–05, national expenditure on health promotion activities for injury prevention, nutrition, physical activity and sun protection combined represented only 8.6% of government public health expenditure6 (and only 0.1% of total government health expenditure7). Furthermore, unintentional injury received only 1.9% of total National Health and Medical Research Council (NHMRC) funding during 2002–03.5 The gap between the size of the public health problem of injury and national efforts to reduce it needs to be addressed.

Successful injury prevention rests on a foundation of quality surveillance data. Australia has a relatively advanced system of injury surveillance compared with many other high-income countries; this includes the recently established National Coroners Information System (NCIS), state-based trauma registries and external-cause data collected nationally for hospital admissions. However, aspects of Australia’s injury surveillance could be improved. The current National Injury Prevention and Safety Promotion Plan8 includes recommendations on injury surveillance, yet most remain to be implemented.

Here, we describe some of the actions needed to implement this plan, as well as other opportunities to enhance injury surveillance in Australia. We also discuss barriers to implementation and how they might be overcome.

Injury surveillance has been defined as “the ongoing, systematic collection, analysis and interpretation of [injury] data essential to the planning, implementation, and evaluation of [intervention strategies], closely integrated with the timely dissemination of these data to those who need to know . . . and the application of these data to prevention and control”.9 Injury surveillance systems are needed to estimate the community burden of injury; to monitor injury trends over time; to detect clusters of injury events by time, person and place; to prioritise strategies for injury prevention; to evaluate injury-prevention strategies; to set directions for injury-prevention research; and to inform the development of injury prevention policies.10-13 Optimal injury surveillance systems have the capacity to identify what injuries occurred, where, to whom, when and why.

During the past 20 years, the health sector has played a major role in developing injury policy in Australia,14 culminating in the National Injury Prevention and Safety Promotion Plan, which covers the decade 2004–2014.8 Recommendations on injury surveillance have been integral to most of these policies and plans (Box 1). However, of the 22 recommendations relevant to injury surveillance included in Australian health policies since 1986, only three have been completely implemented, and the recommendations in the current plan are yet to be completed.

Strategies to improve the collection of information on injury are also contained in other national policy initiatives. These include the:

• National Sports Safety Framework;21

• National Occupational Health and Safety Strategy 2002–2012;22

• National Road Safety Strategy 2001–2010;23 and

• National Water Safety Plan 2004–2007.24

All recommend enhancing injury surveillance through improved data coverage, more timely identification of existing and emerging injury risks, the use of consistent definitions, improved data for identifying causality, and timely dissemination of surveillance data.

Other recent national initiatives have also laid a path for improved injury surveillance. These include:

• the adoption of the International classification of diseases, 10th revision, Australian modification (ICD-10-AM),25 which has extended the information collected on injury;

• the creation of a new national mortality database, the NCIS; and

• the development of national classification standards, such as the National Minimum Data Set for Injury Surveillance26 and the Type of Occurrence Classification System27 for work-related injuries, which allow comparisons across the whole of Australia.

While these initiatives are welcome, there is still much to be done to improve injury surveillance in Australia.

To further develop national capacity for injury surveillance and to promote further innovation in this surveillance, some key initiatives need to be implemented. These group into four themes (Box 2):

• improving the current injury mortality and morbidity data collection systems;

• filling the gaps in injury surveillance by extending surveillance beyond the most serious injury outcomes, ensuring that all geographic areas are covered, and maintaining vigilance over data quality;

• increasing the integration and accessibility of injury data through data warehousing and data linkage; and

• developing technical expertise in surveillance among researchers and data coders.

One of the most immediate areas for action is to enhance the value of existing surveillance systems by improving the data collected. Easily implemented strategies include the addition of unique identifiers for patients or for the injury (eg, date or time that it occurred), better characterisation of the person injured (eg, Aboriginal or Torres Strait Islander status), reduced use of “dump” codes such as “other” and “unspecified”, and improved collection of details of the circumstances of the injury through the inclusion of narrative text. National implementation of a police notification of death form would also improve comparability and consistency of information on injury mortality around the country. This form has been adopted in all Australian jurisdictions except New South Wales, Victoria, and Western Australia. The police and attorney-general’s departments in these states should be encouraged to allocate resources for its implementation.

Currently, injury surveillance in Australia is limited almost entirely to tracking fatalities and serious acute injuries. While this produces valuable information, it means policy and practice decisions are restricted to these relatively rare events with almost immediate injury outcomes. We need to expand the scope of injury surveillance to encompass data from emergency departments, thereby including less serious and less acute injuries. The argument is similar for ensuring that injury surveillance is truly national, reflecting urban, regional, and even the most remote areas. In fact, a high and continuing priority for injury surveillance is ensuring that the collected data accurately describe the profile of injury, with no gaps in surveillance. Otherwise, prevention initiatives and resources may be misdirected. The validity of national injury data collections should be assessed to determine their accuracy for identifying national priorities. The agencies responsible for these data collections should receive resources for this assessment, and Australian injury research centres should assist in the evaluation process.

Key medium-term initiatives should involve activities to ensure the availability and most effective use of injury surveillance data. For example, the United States has developed the National Violent Death Reporting System to improve the accessibility of information on violent deaths by incorporating data from a variety of injury collections.29 There is no reason that a similar system of surveillance, in the form of a data warehouse, could not be implemented in Australia for injury-related mortality and morbidity data collections. This would require a national body, such as the Australian Bureau of Statistics, or a consortium to coordinate the data warehouse, and the development of partnerships and cooperative agreements between agencies on data access.

Lastly, ongoing improvements in injury surveillance rely on the work of clinical coders. Currently, there are question marks over the accuracy of coding of particular variables (eg, activity and location of incident). Comprehensive training for clinical coders regarding external causes would help improve the accuracy of injury information. An assessment of data quality in health data collections would help identify areas where further coding training is required. Organisations such as the National Centre for Classification in Health should be resourced to develop a training module for clinical coders, and national and state health departments should be encouraged to support this action. In addition, further investment in the continued development of a trained workforce that can undertake injury surveillance work should be fostered.

The actions related to injury surveillance in the National Injury Prevention and Safety Promotion Plan are unlikely to be achieved without an implementation plan, a performance management structure, sufficient resources and an appropriate national governance structure for the plan.14 These do not exist, and their absence is the only substantial barrier to advancing injury surveillance. The usually cited barriers to implementation (lack of financial resources and personnel time, and public concerns over privacy) are relative rather than absolute barriers and could be easily overcome if the government showed its commitment by addressing the above issues.

Information from routine data collections is widely used by national and state governments in Australia to formulate and evaluate public health policies and to allocate resources to areas deemed a high priority. High-quality injury surveillance data are essential to ensure the appropriateness of these policies. Thus, it is important to ensure that injury policy in Australia supports the development and progress of innovative injury surveillance capabilities. There is a substantial gap between the importance of the problem and the commitment of funds to support the implementation of recommended enhancements to injury surveillance systems in this country.

Australia has had a world first with the development of the NCIS, and some Australian states have been world leaders in surveillance initiatives (eg, WA and data linkage,30 and NSW and data text mining practices31). Implementation of the national priorities for injury surveillance that were identified in the most recent injury policy is paramount. For Australia to remain at the cutting edge of injury surveillance, we need to ensure that innovative surveillance practices are able to continue. What are we waiting for?