Public Perceptions of the “Remove,Remove,Remove” Information Campaign Before and During a Hazardous Materials Incident: A Survey

Abstract

In the event of an incident involving the release of hazardous material (eg, chemical, biological, radiological, or nuclear agents), key actions can be taken to reduce harm and protect the people involved. The quicker actions can be taken, the less harm will occur. Guidance is, therefore, needed to help nonspecialist members of the public to act rapidly before emergency responders arrive. The “Remove, Remove, Remove” campaign includes critical information for anyone who is at or near the immediate scene of a hazardous material release. Using a representative sample of 1,000 members of the UK population, this study examined the impact of the information campaign on knowledge and perceptions both before and during the immediate response to an incident. Findings showed that perceptions of the information were positive, with respondents stating that the information was useful and that the recommended actions would be effective. Respondents also stated they would be willing and able to perform the actions during a real incident. Additionally, the respondents' knowledge and confidence in taking protective actions increased significantly after receiving the campaign information, and they overwhelmingly agreed that they would want to receive this information if it were available before an incident. The findings of the survey support the use of the “Remove, Remove, Remove” information before and during hazardous materials incidents.

Introduction

The likelihood of deliberate and accidental releases of hazardous materials (eg, chemical, biological, radiological, or nuclear [CBRN] agents) is increasing due to technological advances, the willingness of terrorists to use unconventional weapons, and ease of trafficking such materials.^1,2 Recent examples of hazardous materials incidents involving CBRN agents include the attempted assassination of Sergei Skripal with the nerve agent Novichok in Salisbury, England, in March 2018,³ and numerous other incidents involving acid.⁴ In the event of an incident involving the suspected release of a hazardous material, key actions can be taken to protect the people involved. For example, upon arrival at the scene of a hazardous material release, emergency responders may decide to initiate emergency decontamination procedures. These can include asking those at or near the scene to evacuate to a safe distance, disrobe, and use either water or absorbent materials to remove any contaminant from their skin.⁵ Research increasingly indicates that speed is essential during decontamination and that actions should be taken as soon as possible to prevent the harmful effects of a contaminant.⁶

A recent Home Office initiative, the Initial Operational Response (IOR) program,⁷ sought to address the need for rapid action during a hazardous material incident. The program put in place guidance and procedures to enable nonspecialist first responders to initiate decontamination rapidly, before the arrival of specialist teams and equipment. IOR specifies that key initial actions in the management of a hazardous materials incident are evacuation and removal of contaminated clothing (disrobing). Specifically, IOR suggests that as soon as an incident involving chemical contamination is suspected, the people involved should be evacuated upwind (and ideally uphill) of the contamination source and asked to remove the top layers of their clothes. Following this, it may be beneficial to initiate improvised dry or wet decontamination. Improvised dry decontamination is the removal of a contaminant from the skin using any available absorbent material, such as blue roll, whereas improvised wet decontamination is the removal of a contaminant from the skin using any available water source. IOR guidance suggests that dry decontamination should be carried out as soon as possible, unless the suspected contaminant is a caustic chemical or particulate contaminant. Current protocols state that dry decontamination should then be followed by wet decontamination to remove any contaminant from casualties' hair. This IOR guidance was used to develop the “Remove, Remove, Remove” information campaign* to quickly educate people on the 3 key actions⁸: remove themselves from the current location, remove their outer clothing, and remove the substance from their bodies. This campaign is now in use among emergency services and professionals in the crowd safety and security sector. The IOR program is designed to allow decontamination procedures to be initiated as rapidly as possible, without the need to wait for specialist teams and equipment.

Despite the importance of rapidly initiating decontamination, exercises have shown that no systems are in place to advise the public during the first 4 hours of a hazardous materials incident. Although emergency responders will likely attend the scene within the first few minutes, it may take at least 4 hours for formal scientific advice to be developed and provided to the public. There is a need for basic advice that can be deployed rapidly to the public during the immediate response phase of a hazardous materials incident. Advice will necessarily vary depending on where people are in relation to the incident (eg, those further away from the incident may be asked to stay inside and shut windows). The “Remove, Remove, Remove” information campaign represents key pieces of advice for anyone at or near the scene of the incident. It is, therefore, important to understand public perceptions about the use of this information in the immediate aftermath of a hazardous materials incident.

In addition to being used during an immediate response to a hazardous materials incident, the “Remove, Remove, Remove” information campaign also has the potential to be used before an incident to inform members of the public about actions they should take. Similar preincident information campaigns have been developed to inform members of the public about protective actions they should take during other types of terrorist incidents. These include “Run. Hide. Tell.”, which was designed for use during marauding terrorist firearm attacks,⁹ and “See it. Say it. Sorted.”, which was designed to reduce the risk from terrorist incidents on rail networks.¹⁰ It has been suggested that these types of preincident information campaigns are a key way in which public preparedness for CBRN incidents can be improved, enabling members of the public to take protective actions more rapidly.^11-14 This is in line with the Sendai Framework for Disaster Risk Reduction, which highlights the importance of including all members of society in planning for disasters and ensuring that members of the public have sufficient knowledge and efficacy to take action to protect themselves and others¹⁵; one way to do this is to develop effective preincident information campaigns.

To examine public perceptions of using the “Remove, Remove, Remove” information as part of a preincident information campaign, we conducted focus groups with a small representative sample of the London population.¹⁶ Findings from this previous research showed the information was perceived positively, with the majority of respondents reporting that they would like to receive the information if it was available before an incident.¹⁶ The information campaign also resulted in improved perceptions of knowledge and confidence in taking protective actions among focus group respondents. However, the previous research used a small sample size (N = 44) that was representative of the population in London, rather than United Kingdom. In addition, outcomes measured perceived public knowledge in taking protective actions, rather than actual knowledge. We determined it was necessary to use a larger representative sample of the UK population to examine public perceptions of the information in a preincident context and to examine its impact on actual knowledge and public perceptions.

For the current study we conducted a survey with a representative sample of 1,000 members of the UK population to examine the following aspects of the “Remove, Remove, Remove” information campaign: (1) public perceptions of the information for use before an incident, (2) public perceptions of the information for use during an incident, and (3) the impact of the information on knowledge and perceptions.

Method

Design

We conducted an online survey to capture public perceptions of the “Remove, Remove, Remove” information campaign. The survey expanded on our previous focus group research to examine the perceptions of a representative sample of the UK population. To detect a small effect size of 0.1 in this survey, an a priori power analysis indicated that a sample size of 788 was required. We increased our initial sample size to 1,000 respondents to ensure sufficient power. The survey used a within-subjects design, with 1 factor (time) and 2 levels: Time 1 – immediately before viewing the information, and Time 2 – immediately after viewing the information.

Respondents

To enroll in the study, respondents had to be at least 18 years old and fluent in English. A representative sample of 1,000 members of the UK population were enrolled.

Respondents were evenly divided between male (50.5%) and female (49%), with a small percentage identifying as other (0.5%). All age groups were well-represented within the sample; 14% were ages 18 to 24 years, 22% were 25 to 34, 20% were 35 to 44, 23% were 45 to 54, and 21% were 55 to 65. Respondents were also evenly split in terms of their location within the United Kingdom, with 22.9% from southern England, 19.4% from the Midlands, 17.7% from northern England, 10.7% from London, 8.9% from Yorkshire and Humberside, 8.2% from the east of England, 6.9% from Scotland, and 5.3% from Wales.

Procedure

Respondents were recruited via the Kantar market research company. They received information about the study from Kantar and were given an opportunity to read the information before deciding whether to take part in the study. Those who decided to enroll received a link to the online survey and were asked to check a series of boxes to provide their consent. Respondents then completed a series of screening questions to ensure they were eligible to take part in the survey and that survey quotas were met. At the start of the survey, respondents read a short scenario describing a hypothetical hazardous materials incident. Following this, they answered questions relating to how they thought they would feel and act during such an incident, including their perceived knowledge and confidence in being able to take actions to protect themselves and others. They were then shown a copy of the “Remove, Remove, Remove” information, before answering a series of questions relating to their perceptions of the information. At the end of the survey, respondents received a debriefing statement informing them about the nature of the study and directing them to further resources should they be concerned about any aspect of the study.

Materials

Scenario

A scenario was developed describing an incident involving members of the public splashing each other with a caustic liquid contaminant. Respondents were asked to read the scenario and visualize that they had been involved in the incident described, before answering a series of questions relating to their expected thoughts and actions during such a scenario.

Information Sheet and Survey

Respondents were asked a series of open-ended questions and questions on a Likert scale after reading the scenario. The questions focused on (1) actions respondents would take in the event of a chemical incident (eg, “What actions do you think you would take if you were involved in the incident described?”), and (2) respondents' knowledge and confidence in taking protective actions (eg, “If a real incident of this type were to occur, I would know what actions to take to protect myself”) (4 items, alpha = .92).

Respondents were then shown a “Remove, Remove, Remove” information sheet produced by the National CBRN Centre (Figure 1), with advice about what they should do during a hazardous materials incident, before being asked a series of questions relating to their perception of information. These questions focused on:

Figure 1.

“Remove, Remove, Remove” information sheet. Reproduced with permission from the National Chemical, Biological, Radiological and Nuclear Centre.

Respondents' knowledge and confidence in taking protective actions (eg, “If a real incident of this type were to occur, I would know what actions to take to protect myself”) (4 items, alpha = .94)

Respondents' perceptions of the information (eg, “If this were a real incident, I would find it useful to receive the ‘Remove, Remove, Remove’ information”)

Whether respondents felt that taking the recommended actions would be effective (eg, “I think that taking the actions recommended in the ‘Remove, Remove, Remove’ information would be an effective way to protect myself”)

Whether respondents would be willing to take the recommended actions (eg, “I would be willing to take the actions recommended in the ‘Remove, Remove, Remove’ information sheet”)

Whether respondents would feel able to take the recommended actions (eg, “I think I would be able to take the actions recommended in the ‘Remove, Remove, Remove’ information sheet”)

How respondents would feel about receiving this type of information before an incident (eg, “If the ‘Remove, Remove, Remove’ information was available before an incident occurred, I would find it useful”)

Which organization(s) respondents would want to receive the information from

Analysis

Quantitative data were analyzed using IBM SPSS Statistics for Windows version 25.0 (IBM Corp, Armonk, NY, USA). One-sample t tests were carried out to establish any differences between variable mean scores and the scale midpoint. A paired sample t test was carried out to examine differences between variable scores at Time 1 and Time 2. An analysis of variance (ANOVA) was used to examine differences between variable scores across demographic categories.

Qualitative survey responses were analyzed using the “framework approach.” This is a type of thematic analysis commonly used within research that has implications for policy and practice.¹⁷ For each of the 7 open-text response survey items, a thematic framework was identified, based on the aims of the study. Additional themes also emerged from the data.

Results

Demographic Data

ANOVA was used to determine whether there were any significant differences in demographic characteristics for each of the outcome variables. Because there were 8 dependent variables, a Bonferroni adjustment was applied, with the significant alpha value subsequently set at .006. A 1-way ANOVA revealed significant differences between male and female respondents in terms of whether they would want to receive “Remove, Remove, Remove” information before an incident (F = 5.24, P = .005), with female respondents (mean [M] = 6.06) reporting significantly greater desire for the information before an incident than male respondents (M = 5.81, P = .004). However, the mean for both male and female respondents was significantly above the midpoint value of 4, indicating positive perceptions of the information.

The only significant difference between age groups was in terms of knowledge and confidence in taking protective actions after receiving “Remove, Remove, Remove” information (F = 4.05, P < .001), with those ages 35 to 44 years (M = 5.77) reporting significantly greater knowledge and confidence in taking action after receiving the information than those ages 18 to 24 years (M = 5.31, P = .001). Again, the mean for both groups was significantly above the midpoint value of 4, indicating a good level of confidence in taking protective actions after receiving the information.

Significant differences existed by employment status in terms of perceived knowledge and confidence in taking protective actions before receiving “Remove, Remove, Remove” information (F = 3.60, P = .001). Higher level professional workers (M = 4.88) reported significantly greater knowledge and confidence before receiving the information compared with junior professionals (M = 4.16, P = .001) and semiskilled or unskilled manual workers (M = 4.19, P = .006). However, there were no significant differences between these groups in terms of their perceived knowledge and confidence after receiving the information. In addition, there were no significant differences between those living in different areas.

Quantitative Data

Perceptions of Information

A series of 1-sample t tests were carried out to compare scores for various items with the midpoint value (4) of the scale. The aim of these tests was to demonstrate whether mean responses were significantly different from the scale midpoint, which respondents could select if they neither agreed nor disagreed with a statement. A score that was significantly greater than 4 demonstrated positive perceptions of the “Remove, Remove, Remove” information. The results showed that both perceived usefulness of the information and perceived effectiveness of the actions were significantly greater than the scale midpoint (perceived usefulness of the information during an incident: M = 5.83, t₉₉₉ = 45.79, P < .001; perceived effectiveness of the recommended actions: M = 5.81, t₉₉₉ = 52.18, P < .001) (Table 1).

Table 1.

Perceptions of “Remove, Remove, Remove” Information

Variable	M	SD	t	P Value
Perceived usefulness of information	5.83	1.26	45.79	< .001^a
Perceived effectiveness of information	5.81	1.10	52.18	< .001^a
Perceived willingness to perform actions	5.95	1.08	56.98	< .001^a
Perceived ability to perform actions	5.82	1.11	51.71	< .001^a
Would want to receive information	5.93	1.22	50.03	< .001^a
Information is useful	5.98	1.18	52.91	< .001^a

Indicates the mean value is significantly different from the scale midpoint value of 4 (neither agree nor disagree).

Abbreviations: M, mean; SD, standard deviation.

Willingness and Ability to Take Actions

A further series of 1-sample t tests was carried out to examine respondents' perceived willingness and ability to take recommended “Remove, Remove, Remove” actions. Results showed that both perceived ability to take the recommended actions and willingness to take the recommended actions were significantly greater than the scale midpoint (ability to take actions: M = 5.82, t₉₉₉ = 51.71, P < .001; willingness to take actions: M = 5.95, t₉₉₉ = 56.98, P < .001).

Impact of Information on Public Understanding and Preparedness

A paired samples t test showed a significant increase in knowledge and confidence in taking protective actions before receiving information (M = 4.38) and after receiving information (M = 5.59, t₉₉₉ = 28.75, P < .001).

When respondents were asked whether they would want to receive “Remove, Remove, Remove” information before an incident and whether they thought it would be useful to receive the information before an incident, results for both items were significantly higher than the midpoint value of 4 (wanting to receive information before an incident: M = 5.93, t₉₉₉ = 50.03, P < .001; useful to receive information before an incident: M = 5.98, t₉₉₉ = 52.91, P < .001). Only 4% of people said they would not want to receive the information before an incident, and only 3.2% felt that receiving the information before an incident would not be useful.

Preferred Organizations

In terms of the organizations that people wanted to receive “Remove, Remove, Remove” information from, the most commonly chosen was the National Health Service (n = 632), followed by ambulance service (n = 537) and police (n = 488). The least commonly selected were local council (n = 180), Home Office (n = 191), and employer (n = 203). The majority of people selected between 1 and 4 organizations they would like to receive information from (n = 698).

Qualitative Data

Before receiving the “Remove, Remove, Remove” information, the most common of the 3 actions that people reported they would carry out was removing the substance from the skin using water or available absorbent materials (n = 349). This increased to 410 respondents after receiving the information. The next most common action that people reported before receiving the information was removing themselves from the immediate hazard (n = 58). This number increased substantially after receiving the information, with 325 people reporting they would move away from the immediate hazard. The least commonly reported action before receiving the information was removing outer layers of clothing, reported by just 12 people. After receiving the information, this number increased to 377. Before receiving the information, only 1 respondent specifically stated they would carry out the 3 “Remove, Remove, Remove” actions in the recommended order (remove from immediate hazard, remove outer layers of clothing, remove substance). After receiving the information, this number increased to 212. Another commonly reported action was reporting the incident to the emergency services (before receiving information, n = 643; after receiving information, n = 402).

After rating how useful they found the “Remove, Remove, Remove” information, respondents were asked to state why they did or did not find the information useful. For those who stated they found the information useful, the most commonly cited reasons were: it is informative, it is easy to remember, and it is clear and simple advice. The number of people reporting that the information was not useful was very small (n = 46), and just over half (n = 24) of those (n = 24) provided valid responses to this question. Reasons provided regarding why the information is not useful included: the information is not clear, the information would not help in the situation described, and more information is needed.

When asked if they did not understand any of the information, the vast majority (n = 799) of respondents indicated that there was nothing they did not understand and that the information was clear. When respondents stated there were things they did not understand, these fell broadly into 3 categories: existing information is not clear (eg, difficult to remember), practical difficulties with taking actions (eg, difficulty removing clothing without lifting over the head), and further information is necessary (eg, when to seek medical help, further contact information, whether information applies to all substances).

Just under a third (n = 332) of respondents described further information they would like to receive immediately following a hazardous materials incident. The majority of responses fell into 3 broad themes: what the substance was or how dangerous the substance was (n = 65); whether to call emergency services or go to hospital (n = 64); and sources of ongoing support, such as advice about seeking medical support, sources of aftercare, and symptoms to look out for (n = 51).

Discussion

“Remove, Remove, Remove” information⁸ is currently used by emergency responders and crowd safety and security professionals in the management of incidents involving hazardous materials. This study used a survey of public perceptions to examine the potential of providing the information to members of the public, both before an incident and during the initial response to an incident. Overall, findings suggest that the information may be effective if provided before or during the initial response to an incident.

In terms of the potential use of the information for increasing public preparedness before an incident, findings were positive. Respondents overwhelmingly reported that they would find the “Remove, Remove, Remove” information useful if they received it before an incident and that they would want to receive it before to an incident if it were available. In addition, respondents' perceived knowledge and confidence in taking protective actions during a hazardous chemical incident increased significantly after receiving the information. Along with increased perceptions of knowledge and confidence, findings from the qualitative survey showed that respondents' actual knowledge of taking appropriate actions during this type of incident increased. The only reported behavior that decreased following provision of the information was reporting the incident to emergency services. Highlighting the need to report an incident to emergency services more prominently within the information would, therefore, be beneficial.

The “Remove, Remove, Remove” information was also perceived as being useful during an incident involving a hazardous substance, and respondents felt that the recommended actions would be effective during this type of incident. Further, respondents reported they would feel both willing and able to perform the recommended actions during a real incident of this type. Public perceptions of the information for use during an immediate response to an incident were therefore very positive. Overall, findings strongly support the use of “Remove, Remove, Remove” information both before an incident and during the immediate response to an incident. Findings for this study are in line with our previous focus group study of public responses to the information.¹⁶

Although respondents reported they would find it valuable to receive the “Remove, Remove, Remove” information before an incident, and their knowledge and confidence in taking recommended actions was improved immediately after receiving the information, it is unclear to what extent this information would be retained over time. A recent review of the efficacy of preincident information campaigns for improving public preparedness for CBRN incidents shows that studies rarely assess public ability to retain preincident information over the medium to long term (more than 3 months after receiving information).¹³ Further research is therefore needed to examine the best method of providing “Remove, Remove, Remove” information to members of the public to ensure retention of information over the medium to longer term. The review also recommends that, where possible, preincident information for CBRN incidents should be provided using multiple methods (eg, traditional and social media) to increase the breadth of coverage and enhance information retention.

Limitations

This study provides valuable insight into public perceptions of the “Remove, Remove, Remove” information, using a large-scale survey with a representative sample of the UK population. However, some limitations should be noted. First, the study examined public perceptions of the information using a hypothetical scenario, and the ecological validity of the study may, therefore, be questioned. However, previous research has shown good engagement with similar hypothetical hazardous materials scenarios,^16,18 including scenarios presented online.¹⁹ Second, respondents' perceptions of knowledge and confidence in taking protective actions, and actual knowledge of protective actions, were measured immediately after viewing the information. Further research is, therefore, needed to examine the extent to which the improvement in respondents' knowledge would be maintained over the medium to long term. Third, while there were very few differences between those in different demographic groups and perceptions of the information across all demographic groups were positive, we included only a limited measure of socioeconomic status (eg, employment status) and we did not measure ethnic background. It is possible that perceptions of the “Remove, Remove, Remove” information may differ among those of different socioeconomic status or ethnic background. Relatedly, it should be noted that respondents were recruited using a market research company and the sample was not random. It is possible that, although the sample is broadly representative in terms of demographics, those who sign up to take part in market research may differ in some way from those who do not, and this may have affected the outcomes of this research.

Conclusion

It is increasingly recognized that all members of the public should be involved in planning for disasters¹⁵ and that effective preincident information campaigns are a crucial way to achieve this. Such information campaigns ensure that people know what actions to take during disasters and emergencies, including incidents involving hazardous materials, and that they can act quickly to protect themselves and others.^11-14 The evidence presented here suggests that the information has the potential to be used before and during an immediate response to a hazardous materials incident. Given how important it is that people act quickly during these incidents,⁶ such preincident information could potentially save lives.

Footnotes

Acknowledgments

This research was funded by the National Institute for Health Research Health Protection Research Unit in Emergency Preparedness and Response at King's College London in partnership with Public Health England, in collaboration with the University of East Anglia and Newcastle University. The views expressed are those of the authors and not necessarily those of the National Health Service, National Institute for Health Research, Department of Health and Social Care, or Public Health England.

References

Schneidmiller

Nuclear smuggling shows terrorist WMD threat persists: State Department. Global Security Newswire. August 1, 2012. Accessed October 1, 2020. http://www.nti.org/gsn/article/state-report/

Shuster

Inside the uranium underworld: dark secrets, dirty bombs. Time. April 6, 2017. Updated April 10, 2017. Accessed October 1, 2020. https://time.com/4728293/uranium-underworld-dark-secrets-dirty-bombs/

Russian spy poisoning: what we know so far. BBC News. October 8, 2018. Accessed October 1, 2020. http://www.bbc.co.uk/new/uk-43315636

Marsh

Surge in acid attacks in England leads to calls to restrict sales. Guardian. July 7, 2017. Accessed October 1, 2020. https://www.theguardian.com/uk-news/2017/jul/07/surge-in-acid-attacks-in-england-leads-to-calls-to-restrict-sales

National Ambulance Resilience Unit (NARU). The Way the NHS Responds to HAZMAT/ CBRN Events Is Changing. Coventry, UK: NARU; 2013. Accessed October 1, 2020. https://naru.org.uk/wp-content/uploads/2014/05/NARUIOR-A4-X-3-v8a-2.pdf

Chilcott

RP.

Managing mass casualties and decontamination. Environ Int. 2014; 72:37-45.

Home Office. Initial Operational Response to a CBRN Incident. London, UK: Home Office; 2015. Accessed November 23, 2020. https://www.jesip.org.uk/uploads/media/pdf/CBRN%20JOPs/IOR_Guidance_V2_July_2015.pdf

National Ambulance Resilience Unit. ‘Remove, Remove, Remove’ – refreshed IOR messaging released by NARU. Published March 13, 2018. Accessed October 1, 2020. https://naru.org.uk/remove-remove-remove-refreshed-ior-messaging-is-released-by-naru/

City of London Police. Staying safe from terrorism. Accessed November 16, 2020. https://www.cityoflondon.police.uk/advice/advice-and-information/t/terrorism-in-the-uk/staying-safe-from-terrorism/

10.

British Transport Police. New National Rail security campaign starts today: “See it. Say it. Sorted.” Published November 1, 2016. Accessed October 1, 2020. https://www.btp.police.uk/latest_news/see_it_say_it_sorted_new_natio.aspx

11.

Carter

, Amlôt

Mass casualty decontamination guidance and psychosocial aspects of CBRN incident management: a review and synthesis. PLoS Curr. 2016;8:ecurrents.dis.c2d3d652d9d07a2a620ed5429e017ef5.

12.

Carter

, Gauntlett

, Rubin

, et al. Psychosocial and behavioural aspects of early incident response: outcomes from an international workshop. Global Secur Health Sci Policy. 2018; 3(1):28-36.

13.

Carter

, Drury

, Amlôt

. Recommendations for improving public engagement with pre-incident information materials for initial response to a chemical, biological, radiological or nuclear (CBRN) incident: a systematic review. Intl J Disaster Risk Reduction. 2020; 51:101796.

14.

Spencer

, Kindt

, Stans

. Public Resilience in CBRN Events: Lessons Learned from Seven Cases. US Air Force Counterproliferation Center Future Warfare Series no. 52. Maxwell Air Force Base, AL; USAF Counterproliferation Center; 2019. Accessed October 1, 2020. https://media.defense.gov/2019/Apr/11/2002115516/-1/-1/0/52PUBLICRES.PDF

15.

United Nations Office for Disaster Risk Reduction (UNISDR). Sendai Framework for Disaster Risk Reduction 2015–2030. Geneva: UNISDR; 2015. Accessed October 1, 2020. http://www.preventionweb.net/files/43291_sendaiframeworkfordrren.pdf

16.

Carter

, Weston

, Symons

, Amlôt

. Public perceptions of ‘Remove, Remove, Remove’ pre-incident information materials for initial response to a chemical incident. Disaster Prev Manag. 2019; 28(5):565-584.

17.

Pope

, Ziebland

, Mays

. Analysing qualitative data. BMJ. 2000; 320:114.

18.

Carter

, Weston

, Betts

, Wilkinson

, Amlôt

. Public perceptions of emergency decontamination: effects of intervention type and responder management strategy during a focus group study. PLoS One. 2018b;13(4):e0195922.

19.

Carter

, Drury

, Amlôt

, Rubin

, Williams

. Effective responder communication, perceived responder legitimacy, and group identification predict public cooperation and compliance in a mass decontamination visualisation experiment. J Appl Soc Psychol. 2015; 45(3):173-189.