When a Public Health Alert Goes Wrong

At 8:07 am on January 13, 2018, people in Hawaii received an emergency alert advising them to seek shelter from an incoming ballistic missile.

A very long 38 minutes later, the message was retracted via the same systems that had sent it—the Wireless Emergency Alert system, which sends location-based warnings to wireless carrier systems, and the Emergency Alert System, which sends television and radio alerts.

The Federal Communications Commission report that covered the debacle noted that, among other errors, the employee responsible for triggering the false alert believed the missile threat was real. Moreover, the exercise plans did not document a process for disseminating an all-clear message. And on top of that, the established ballistic missile alert checklist did not include a step to notify the Hawaii Emergency Management Agency’s public information officer responsible for communicating with the public, media, other agencies, and other stakeholders during an incident.

Researchers from the CDC and Hawaii Department of Health analyzed tweets sent during 2 periods: early (8:07-8:45 am), the 38 minutes during which the alert circulated; and the late period (8:46-9:24 am), the same amount of elapsed time after the correction had been issued.

They found 4 themes dominated the early period: information processing, information sharing, authentication, and emotional reaction (shock, fear, panic, terror). Information processing was defined as any indication of initial mental processing of the alert. Many of the tweets dealt with coming to terms with the threat.

During the late period, information sharing and emotional reaction persisted, but they were joined by new themes that, according to the researchers, were “fundamentally different” from the early-period themes and reflected reactions to misinformation: denunciation, insufficient knowledge to act, and mistrust of authority. “Insufficient knowledge to act” involved reacting to the lack of a response plan, particularly not knowing how to properly take shelter. Denunciations blamed the emergency warning and response, especially the time it took to correct the mistake. Mistrust of authority involved doubting the emergency alert system or governmental response.

How can a situation like this be better handled? The researchers say public health messaging during an emergency is complicated. For instance, it is influenced by how messages are perceived and interpreted by different people, and by the fact that messages need to be sent over multiple platforms to ensure that the information is disseminated accurately and quickly.

Which is why social media is both a handicap and a boon in public health emergencies. Tweets spread misinformation as fast as information (if not faster), so the first messages are critical. In addition to conveying timely messages, the researchers advise, public health authorities need to address the reactions during each phase of a crisis. They also need to establish credibility to prevent the public from mistrusting the public health message and its issuers.

Most important, perhaps: Alerts should carry clear instructions for persons in the affected area to carry out during an emergency.

At 8:07 am on January 13, 2018, people in Hawaii received an emergency alert advising them to seek shelter from an incoming ballistic missile.

A very long 38 minutes later, the message was retracted via the same systems that had sent it—the Wireless Emergency Alert system, which sends location-based warnings to wireless carrier systems, and the Emergency Alert System, which sends television and radio alerts.

The Federal Communications Commission report that covered the debacle noted that, among other errors, the employee responsible for triggering the false alert believed the missile threat was real. Moreover, the exercise plans did not document a process for disseminating an all-clear message. And on top of that, the established ballistic missile alert checklist did not include a step to notify the Hawaii Emergency Management Agency’s public information officer responsible for communicating with the public, media, other agencies, and other stakeholders during an incident.

Researchers from the CDC and Hawaii Department of Health analyzed tweets sent during 2 periods: early (8:07-8:45 am), the 38 minutes during which the alert circulated; and the late period (8:46-9:24 am), the same amount of elapsed time after the correction had been issued.

They found 4 themes dominated the early period: information processing, information sharing, authentication, and emotional reaction (shock, fear, panic, terror). Information processing was defined as any indication of initial mental processing of the alert. Many of the tweets dealt with coming to terms with the threat.

During the late period, information sharing and emotional reaction persisted, but they were joined by new themes that, according to the researchers, were “fundamentally different” from the early-period themes and reflected reactions to misinformation: denunciation, insufficient knowledge to act, and mistrust of authority. “Insufficient knowledge to act” involved reacting to the lack of a response plan, particularly not knowing how to properly take shelter. Denunciations blamed the emergency warning and response, especially the time it took to correct the mistake. Mistrust of authority involved doubting the emergency alert system or governmental response.

How can a situation like this be better handled? The researchers say public health messaging during an emergency is complicated. For instance, it is influenced by how messages are perceived and interpreted by different people, and by the fact that messages need to be sent over multiple platforms to ensure that the information is disseminated accurately and quickly.

Which is why social media is both a handicap and a boon in public health emergencies. Tweets spread misinformation as fast as information (if not faster), so the first messages are critical. In addition to conveying timely messages, the researchers advise, public health authorities need to address the reactions during each phase of a crisis. They also need to establish credibility to prevent the public from mistrusting the public health message and its issuers.

Most important, perhaps: Alerts should carry clear instructions for persons in the affected area to carry out during an emergency.

At 8:07 am on January 13, 2018, people in Hawaii received an emergency alert advising them to seek shelter from an incoming ballistic missile.

A very long 38 minutes later, the message was retracted via the same systems that had sent it—the Wireless Emergency Alert system, which sends location-based warnings to wireless carrier systems, and the Emergency Alert System, which sends television and radio alerts.

The Federal Communications Commission report that covered the debacle noted that, among other errors, the employee responsible for triggering the false alert believed the missile threat was real. Moreover, the exercise plans did not document a process for disseminating an all-clear message. And on top of that, the established ballistic missile alert checklist did not include a step to notify the Hawaii Emergency Management Agency’s public information officer responsible for communicating with the public, media, other agencies, and other stakeholders during an incident.

Researchers from the CDC and Hawaii Department of Health analyzed tweets sent during 2 periods: early (8:07-8:45 am), the 38 minutes during which the alert circulated; and the late period (8:46-9:24 am), the same amount of elapsed time after the correction had been issued.

They found 4 themes dominated the early period: information processing, information sharing, authentication, and emotional reaction (shock, fear, panic, terror). Information processing was defined as any indication of initial mental processing of the alert. Many of the tweets dealt with coming to terms with the threat.

During the late period, information sharing and emotional reaction persisted, but they were joined by new themes that, according to the researchers, were “fundamentally different” from the early-period themes and reflected reactions to misinformation: denunciation, insufficient knowledge to act, and mistrust of authority. “Insufficient knowledge to act” involved reacting to the lack of a response plan, particularly not knowing how to properly take shelter. Denunciations blamed the emergency warning and response, especially the time it took to correct the mistake. Mistrust of authority involved doubting the emergency alert system or governmental response.

How can a situation like this be better handled? The researchers say public health messaging during an emergency is complicated. For instance, it is influenced by how messages are perceived and interpreted by different people, and by the fact that messages need to be sent over multiple platforms to ensure that the information is disseminated accurately and quickly.

Which is why social media is both a handicap and a boon in public health emergencies. Tweets spread misinformation as fast as information (if not faster), so the first messages are critical. In addition to conveying timely messages, the researchers advise, public health authorities need to address the reactions during each phase of a crisis. They also need to establish credibility to prevent the public from mistrusting the public health message and its issuers.

Most important, perhaps: Alerts should carry clear instructions for persons in the affected area to carry out during an emergency.