While levels of community distress and anxiety typically decline in the months following major terrorist attacks, population shifts towards ‘protective’ behaviours may persist for longer periods 1 2 . Such changes include altered use of public transport systems and air travel, avoidance of places of perceived high risk and increased substance abuse 2 3 4 5 . Collectively, such behaviours may have substantial economic and health impacts. Significantly reduced travel was observed on the London underground 12 months after the 2005 London transport bombing, while reductions in U.S. air travel continued up to two years after the 9/11 attacks 5 6 . Paradoxically, this latter shift towards presumed ‘safer’ travel resulted in an estimated 1500 additional road fatalities in the first year following the attacks 7 .

Previous research regarding coping responses to terrorism threat has examined ‘problem-focused’ coping, which includes preparedness activities to address specific issues (e.g. learning evacuation plans, establishing emergency contact plans) and ‘emotion-focused’ coping which aims to manage the associated stress 8 . More critical issues may relate to whether such responses are primarily cognitive or behavioural in nature and particularly whether they constitute avoidance coping 4 9 10 . There is evidence that terrorism preparedness and avoidance represent qualitatively distinct factors 9 and are associated with adaptive and maladaptive outcomes respectively. For example, preparedness responses after the 9/11 attacks were associated with lower psychological distress, while forms of mental and behavioural avoidance predicted higher distress and psychopathology 11 12 . Similar findings have been observed in Canada, despite its history of limited direct exposure 9 .

Behavioural responses to terrorism threat are also associated with specific cognitive, affective and demographic factors. Threat appraisal models show that such responses depend on our judgement of specific threat elements (i.e. perceived likelihood, seriousness) but are often more influenced by our perceived ability to cope with them 8 13 . Coping perceptions or ‘self efficacy’ has been shown to be one of the strongest cognitive predictors of terrorism preparedness behaviours 14 , while affective states such as worry and concern are associated with both preparedness and avoidance 14 15 . Despite such findings, few studies have simultaneously examined concern and perceived efficacy as predictors of terrorism preparedness in the general community 14 . This is notable in that efficacy perceptions are increasingly targeted as part of natural hazard preparedness 16 and are key predictors of health worker willingness to respond to terrorist incidents 17 18 . Further research is needed regarding the role of these factors in individual preparedness for terrorism, as this may critically inform preparedness initiatives for the general population.

Research examining community preparedness also needs to consider such responses across a range of terrorism event phases. While avoidance and preparedness may be high following an attack, information is also required about response factors in situations of perceived pending threat 9 . Australia’s recent history of limited direct exposure but high assumed threat 19 provides a suitable population for such research. The aim of this study was to determine whether high perceived coping in relation to terrorism and concurrent high concern are associated with terrorism preparedness in Australia.

A search of existing survey instruments was conducted to identify items that would address the study aim and be most relevant to the Australian threat context. The terrorism question module consisted of validated and reliable items adopted from two previous surveys 3 10 . These assessed: perceived likelihood of terrorism; concerns for self/family; perceived ability to cope in the event of a terrorist attack; emergency evacuation intentions; and current avoidance/preparedness behaviours. For the latter, a prologue to the question set framed a general response period and threat examples: “In recent years terrorist attacks, such as bombings and shootings, have occurred in a number of countries.” All study protocols and procedures were approved by the New South Wales Ministry of Health and the University of Western Sydney ethics committee (protocol no. H7143).

A total of 2,038 state residents completed the terrorism survey module. The survey response rate was determined in accordance with NSW Health Survey methodology and calculated as total completed interviews of eligible participants, divided by the combined total of completed interviews and refusals 22 . The survey process yielded 3548 eligible participants (2,038 completions and 1510 refusals) and a final response rate of 57.4%.

The prevalence estimates regarding avoidance and preparedness behaviours and evacuation intentions are presented in Table 1. With regard to terrorism-related avoidance, 13.7% reported (moderate to extreme) avoidance of some public places or events, 5.6% reported changes in their use of public transport and 20.4% had deferred or changed plans for overseas holiday travel. With regard to terrorism-related preparedness, 45.0% reported moderate to extreme increases in their vigilance for suspicious behaviours/activities, 39.0% had learned the evacuation plan of a building they occupy frequently and 19.7% developed an emergency family contact plan. In relation to evacuation intentions, 74.9% reported being very/extremely willing to evacuate from their home in the context of a possible terrorist threat and 88.0% reported being very/extremely willing to evacuate from a workplace/public facility in such circumstances.

1. Lower confidence interval (LCI), upper confidence interval (UCI).

2. N/A – not applicable to respondent circumstances.

The current multivariate analysis highlights that terrorism-related avoidance and preparedness behaviours are most consistently associated with cognitive and affective factors, specifically higher concern and perceived coping. The combined concern/coping indicator was positively associated with five of the six current behavioural responses examined in the study and accounted for a broader range of responses than the other cognitive, socio-demographic and health-related factors. Importantly, this combined indicator was associated with all of the current preparedness behaviours as well as evacuation intentions regarding a possible future incident. These findings support other recent data showing that coping self efficacy is strongly associated with terrorism-related preparedness behaviours 14 . Consistent with wider hazard preparedness however, this relationship is mediated by affective factors with higher concern generally being needed to motivate such responses 16 27 .

The present findings are consistent with Social-Cognitive models of emergency preparedness 16 but also specific threat- and efficacy-based models such as Witte’s Extended Parallel Process Model. The EPPM addresses the interplay of ‘threat’ and ‘efficacy’ relating to preparedness behaviours and predicts that individuals facing uncertain risks are more likely to engage in proactive behaviours when they regard the threat as legitimate and believe they can perform the required behaviours efficaciously 13 . As such, efficacy perceptions alter the way in which risks are perceived and have been shown to be a key predictor of health worker readiness to respond to major disasters, including terrorism 17 18 . The current data support a similar case regarding general population preparedness for terrorism. High concern/high coping was associated with all of the preparedness behaviours and future intentions, while high concern/low coping predicted increased vigilance alone. As the EPPM and Social-Cognitive models also predict, the combined variables which included ‘low concern’ (e.g. low concern/high coping) were generally not associated with behavioural response. This suggests that high concern may be a necessary (but not sufficient) condition for general population preparedness in this context.

Increased vigilance was the only preparedness behaviour which showed some inconsistency with EPPM, being associated with both high and low coping perceptions. This may be due to the broader range of situations this could entail when compared to the other preparedness responses. Since the first Bali bombing in 2002, the behavioural focus of Australian Government information campaigns has primarily related to increased vigilance (e.g. “Be alert, not alarmed”, “Let’s look out for Australia”) and the reporting of suspicious behaviours to authorities 28 . The current survey item was intended to examine such vigilance behaviours, although the findings may reflect a spectrum of response; from reactive/avoidant coping (characterised by low control/coping) to proactive monitoring and reporting 29 . Recent international campaigns adopted in Australia reflect attempts to direct awareness towards suspicious behaviours (not racial or cultural identifiers) and to operationalise appropriate responses (e.g. “If you see something, say something”) 30 . The greater effect size for vigilance noted with high coping respondents may indicate more proactive monitoring by this group. This prospect could be tested with more detailed research.

Socio-demographic factors such as low education, older age and minority ethnicity may be associated with more reactive/avoidant responses to terrorism threat 31 32 . In this study, low education and low income were among the strongest predictors of avoidance. This supports data from a study in Canada 31 . A possible explanation for these avoidance responses is lack of resources (e.g. financial/physical). Such groups are recognised as key planning partners, 33 and public health emergency planners should work through relevant issues with them to support their preparedness in this context.

Once cognitive and affective factors were accounted for in the present analysis, only three demographic variables independently predicted preparedness: female gender (learned evacuation plan); having dependent children (contact plan); and ethnic minority status (contact plan, vigilance). The importance of establishing family emergency contacts plans was highlighted after the 2005 London bombing 2 . Its current association with parental status is consistent with reported responses of Los Angeles residents one year after the September 11 attacks 34 . Terrorism-related learning of evacuation plans has not previously been found to be associated with gender 31 . This more proactive response in the present findings may relate to care-provider responsibilities or possibly females’ greater general concern with the welfare of others 35 . It is also consistent with the current findings that both female gender and parental status were positive predictors of evacuation willingness. These findings suggest such groups may engage more readily in community risk mitigation efforts for terrorism and possibly act as ‘influence leaders’ for other demographic groups.

It is notable that rural respondents reported being less willing to comply with evacuation requests of non- residential buildings. Although urban areas are presumed to represent higher profile terrorist targets many rural areas house critical infrastructure including potentially vulnerable transport systems and agricultural resources. Rural regions may be the subject of direct terrorist threats requiring infection control, decontamination and evacuation orders 17 . These findings may indicate the need for increased awareness within such communities which contextualises potential threats and details basic preparedness and response requirements.

When considering the implications of the current findings for community preparedness, it is important to note that coping self-efficacy is not defined as a static personal trait, but as competency beliefs that can be altered through experience or education 29 . Recent studies of health workers show that efficacy appraisals are a key factor mediating willingness to respond to major threats, including chemical, biological and radiological (CBR) terrorism 17 18 . Efficacy-focused education for these workers may constitute cognitive ‘interventions’ that promote lower concern and greater response willingness; outcomes that may be crucial in maintaining emergency ‘surge’ capacity 17 . Among Australian paramedics for example, receipt of terrorism-related training focused on key competencies was found to be a stronger predictor of response readiness than career incident experience 36 .

Research examining efficacy effects on community preparedness for terrorism remains limited. However, recent evidence indicates that people actively seek efficacy information when facing such threats. Focus groups responding to CBR terrorism scenarios sought: (1) appropriate identification of the risk, and (2) information about concrete actions they could take to increase safety 37 . In practice, personal safety information in pre-event phases is often quite limited (e.g. online pamphlets) 38 . More comprehensive initiatives can be seen in places such as Singapore where citizens can undertake competency based education to enhance their terrorism preparedness. This extends to ‘hands on’ exercises such as the use of ‘In-Place-Protection’ simulators to practice the proper sealing of rooms 39 . While the current data support a relationship between existing efficacy perceptions and preparedness, further research could enhance its real world application by determining: (1) the extent to which personal efficacy appraisals can be raised in this threat context, (2) optimal modalities to achieve this, and (3) and the conditions under which enhanced efficacy appraisals translate to preparedness outcomes.

Practical preparedness activities related to terrorism threats are consistently associated with personal coping appraisals, while also being mediated by levels of personal concern. In contrast, socio-demographic and other cognitive/affective factors were found to be less predictive of such preparedness. Perceived individual coping with terrorism, at both general and incident-specific levels, represent viable intervention targets as part of population preparedness initiatives and may support broader community adaptation to this threat.

The authors declare that they have no competing interests.

GS and MT conceived the idea and designed the study. GS carried out the statistical analysis and wrote the manuscript. All authors made contributions to the interpretation of results and revised the manuscript for important intellectual content. All authors read and approved the final version of the manuscript.