Enhancing Communications
Communicating with the Workforce During Emergencies: Developing an Employee Text Messaging Program in a Local Public Health Setting Hilary N. Karasz, PhD, MAa Sharon Bogan, MPHa Lindsay Bosslet, MPHa
ABSTRACT Short message service (SMS) text messaging can be useful for communicating information to public health employees and improving workforce situational awareness during emergencies. We sought to understand how the 1,500 employees at Public Health – Seattle & King County, Washington, perceived barriers to and benefits of participation in a voluntary, employer-based SMS program. Based on employee feedback, we developed the system, marketed it, and invited employees to opt in. The system was tested during an ice storm in January 2012. Employee concerns about opting into an SMS program included possible work encroachment during non-work time and receiving excessive irrelevant messages. Employees who received messages during the weather event reported high levels of satisfaction and perceived utility from the program. We conclude that text messaging is a feasible form of communication with employees during emergencies. Care should be taken to design and deploy a program that maximizes employee satisfaction.
Public Health – Seattle & King County, Seattle, WA
a
Address correspondence to: Hilary N. Karasz, PhD, MA, Public Health – Seattle & King County, 401 Fifth Ave., Ste. 1300, Seattle, WA 98104; tel. 206-263-8705; fax 206-296-0166; e-mail . ©2014 Association of Schools and Programs of Public Health
Public Health Reports / 2014 Supplement 4 / Volume 129
61
62 Enhancing Communications
Advances in technology mean that there are more ways than ever to reach employees during an emergency. For example, short message service (SMS) text messaging can be used to augment existing communication channels.1 While relying on a single technology is not recommended, and concerns exist about the reliability of SMS during large-scale emergencies,2 text messaging is particularly valuable when power is out, voice-to-voice lines are overloaded, or people are away from a computer.3 SMS may be more useful during some emergency situations because of cell phone ubiquity: In 2013, adult cell phone ownership in the United States reached 91%,4 and an estimated 80% of cell phone owners send or receive text messages.5 These rates are even higher among people of color (of all ages), who text more frequently than non-Hispanic white people do.6 A 2008 Institute of Medicine report outlined steps to strengthen public health systems by improving emergency preparedness and response in the U.S.7 Improving employee readiness and emergency communications were identified as key research objectives. One important way to strengthen the public health system and aid continuity of operations during a crisis is to assure that employees can be reached during an emergency.8 However, anecdotal evidence suggests that local health departments (LHDs) are not broadly adopting texting alert systems for their employees. We were unable to locate any previous research addressing how such systems may be implemented by LHDs, so this research may provide new information for LHDs that wish to improve employee emergency communication. A growing number of institutions use text messaging alert systems. For example, following the shootings at Virginia Tech in 2007, hundreds of universities purchased text alert systems, where cell phone and text message adoption among students is nearly universal.6,9,10 During a tornado advisory, a large university used a combination of text messages, e-mails, and online instant messaging to communicate with students and staff.11 In a health services institutional context, a study of emergency communications in a hospital’s anesthesiology department found its current phone tree system was outdated, inconsistent, time intensive, and ultimately ineffective. This department tested an SMS system and found that it was likely to improve employee mobilization efforts during an emergency.12 Carrier rules prohibit requiring people to subscribe to text messaging services, so these systems are optional for subscribers.13 Fifteen hundred employees of Public Health – Seattle & King County (PHSKC) provide population and personal health services across 40 locations to
two million residents of King County, Washington. We describe PHSKC’s experience planning and implementing an employee emergency SMS program to improve awareness, safety, and continuity of operations. The objective was to better understand whether an optional public health employer-based texting program would be acceptable and useful for staff. Informed by Fishbein and Ajzen’s technology acceptance model (TAM), which explains the process by which individuals accept new information technologies, we found that attitudes, subjective norms, and perceived intrusiveness of a text alert program predicted acceptance of such a program. While this study did not aim to evaluate the effectiveness of TAM to predict employee uptake of an emergency SMS alert system, TAM does point to the value of understanding employee attitudes and perceived barriers and benefits to opting in, so that a program can be successfully developed and marketed.3,14,15 Furthermore, because telecommunications industry rules prohibit sending text messages to recipients without permission and people must opt in, it was important to understand employee concerns prior to developing and marketing a system.13 METHODS Formative audience research We developed an 18-item electronic survey to gauge employee interest in receiving text message alerts about a variety of health topics, including emergency alerts and personal health tips. The survey was reviewed and approved by department administrators and union representatives. Employees were recruited by e-mail. About half of the 1,536 staff (n5828) responded. Results revealed several themes: (1) emergency and site closures were topics of interest to a majority of respondents, (2) interest in signing up for a text messaging service was inversely correlated with age, and (3) employees across the age spectrum had significant concerns about receiving messages from work. Topics. Survey respondents were given a list of topics and asked to choose the topics that interested them most. Worksite closure information and emergencies were of interest to about half of the respondents (54% and 56%, respectively), while public health news (10%) and health tips (4%) were the least interesting topics. About one-third of respondents (37%) indicated no interest in receiving messages on any topic. Age. Younger PHSKC workers were more likely than older workers to say they would be willing to sign up for a text program from PHSKC. While about one-third
Public Health Reports / 2014 Supplement 4 / Volume 129
An Employee Text Messaging Program for Emergencies 63
of respondents (n5253) said they would be likely or very likely to sign up, 51% (n527) of respondents aged 18–29 years said they were likely or very likely to sign up, and only 18.5% (n521) of respondents aged $60 years said they were likely or very likely to sign up for a text program. Texting experience also correlated with age: of 18- to 29-year-old respondents, 64% (n534) reported texting every day and only 6% (n53) reported rarely or never texting. In contrast, 64% (n573) of respondents aged .60 years reported never texting. No comparable data from similar employee surveys were identified to suggest how these responses would compare with responses from other public health departmental workforces (data not shown). Areas of concern. The survey revealed employee concerns about the cost of receiving texts, receiving irrelevant information, and losing the distinction between being “on” and “off” the job. While many cell phone subscribers have unlimited texting packages,16 not everyone does, at least at the time the survey was conducted in 2011. Fifty-six percent of employees (n5442) said that cost would be a limiting factor in their decision to sign up for a PHSKC texting program. Many respondents stressed that the system must be voluntary because of cost. One respondent said, “It’s a great technology. It’s essential that employees have an opt-out privilege because of expense.” Another respondent noted, “Signing up for this service would depend on how many messages I’m likely to receive … I’ll be paying for them after all.” Other respondents commented that they would be more likely to opt in if PHSKC paid for their texts. One respondent wrote, “I’d access this resource if [PHSKC] paid for my texting.” Another respondent wrote, “I do not use my personal cell phone for business. Even if the county were to pay for the text, would you pay for my time?” (data not shown). Concern about intrusion on personal time was a common theme. One respondent said, “I don’t want to be notified about work issues on my off time. PERIOD.” Another respondent said, “I do not need to sit at a ball game and receive work stuff. That is my downtime.” Survey respondents also voiced concern that work text messages would add to “information overload.” One respondent said, “There is far too much information being sent to me day and night. I want less technology, not more. It’s decreasing my quality of life.” Another respondent commented, “I think this is a great plan for site closures or other crucial information. I would be hesitant to sign up for emergency info because what I think is an emergency is different from what the media and [PHSKC] often identify as one. I worry that there
would be too many non-emergent or non-crucial text messages” (data not shown). Implementation of PHSKC’s employee text messaging program The texting program aimed to be responsive to employee concerns. It was voluntary and limited to emergency messaging only. We worked with our SMS vendor to build a system that allowed targeted messaging: Employees could opt to receive messages about their worksite only, about other PHSKC worksites, or about all PHSKC worksites. Resources to pay for the development and implementation of the system were pulled from the emergency preparedness and departmental communications units, but sustainability of the system remains an ongoing challenge. Marketing the program to employees consisted of two e-mails to introduce the program to staff in March and April 2011, and a reminder in winter preparedness e-mails in November 2011 and 2012. Our initial e-mail appeal aimed to respond to the concerns raised by employees who responded to the survey. We assured staff that (1) the program was voluntary and for their convenience only, (2) communications would be limited to emergencies and the occasional test message, (3) they could opt out at any time, (4) they would not be accountable for the information received by text message, and (5) their phone numbers would not be shared. By December 2011, 22.7% (n5349) of 1,536 employees had signed up for the program. The number of employees who signed up was low compared with what the formative research indicated, which may have been because the program was not aggressively marketed. In addition, post-marketing interviews with employees who did not opt in indicated that the reason they did not opt in was because they did not pay attention to the initial e-mails inviting them to do so, not because of concerns about irrelevant or too many texts. OUTCOMES In January 2012, King County, which has a geographical area of 2,307 square miles, experienced a significant winter storm in which 250,000 households and businesses lost electricity at some point during a five-day period. In parts of King County, one-third of the homes lost power.17 Many arterial roads were icy or not plowed. Buses and other transit were delayed or canceled. Five PHSKC health clinics lost power; four clinics lost power for four days. We texted our employee subscribers as soon as policy makers decided
Public Health Reports / 2014 Supplement 4 / Volume 129
64 Enhancing Communications
if, when, or where they should report to work. During the five-day event, subscribers received up to 15 text messages alerting them to late starts, clinic closures, and directions for getting more information. Typical messages were “PHSKC: KENT EAST HILL is closed due to lack of power. Check with supervisor about work plan for day. More info from HR to follow. Check PH web for updates,” and “Ice storm. Roads dangerous. PHSKC on late start at 10. If not critical staff, check with supervisor before coming. Check Web and hotline [number given] for info.” Following the storm, we gathered feedback from staff via electronic survey. We received 185 responses (36% of subscribers), and the feedback was primarily positive. Sixty-two percent of respondents (n5115) said the text messages were very relevant and helpful, 20% (n537) found them fairly relevant and helpful, and only 5% (n510) found them not relevant or helpful. About 70% (n5129) of recipients learned at least one piece of information via text message before they heard about it another way (e.g., the website, e-mail message, or hotline). With respect to the number of messages we sent (averaging three per day), 83% (n5154) reported that the number of messages was about right, 15% (n528) thought they received too few messages, and 2% (n54) said they received too many messages. Only two people opted out during this period, while 165 additional employees subscribed during the five days of the storm (data not shown). The texting service was helpful for employees who had lost electricity. A typical comment was, “Since I had no power and was unable to access my e-mail or Internet, I really appreciated the text messages.” Another comment was, “I was trying to get into [the Internet] to get an update and my power kept going in and out, then I received the text message. I had forgotten I had signed up for it but was very glad I did” (data not shown). Employees made it clear that they expect the texting service to provide timely information. One respondent noted, “If it could just arrive a teensy bit sooner, I can prepare better. I know that sometimes the information isn’t known right away. I think it has to do more with the planners for inclement weather, making a decision sooner.” Others complained that they had already left for work by the time they received a message about a late start or closure. One respondent said, “The first text message on Thursday about a late start time at 10 a.m. was texted too late. I don’t know when the decision for a late start time was made, but if it could be made by 6 a.m. and texted at that time, it would really help with the commute” (data not shown).
Usage and enrollment since the snowstorm PHSKC continues to use the emergency texting system, and analyses of the program’s success are still being reviewed. Since the 2012 winter storm event, we have employed the emergency text messaging program three additional times. Subscribers at three different health clinic sites were sent messages advising them of police actions in the vicinity of their clinics. The number of employees subscribed to the system has reached 664, and in three years, only 15 people have opted out of it. Marketing has been limited, primarily consisting of a half-dozen all-staff e-mails. All PHSKC-owned work cell phone numbers (n5248) have also been added as part of an enhanced enrollment effort, making a total of 912 phones that we have enrolled in the employee text messaging system. Employees may sign up with both personal cell phones and work phones, so it is unclear how many of the 912 total subscribers are individual employees, or if some of them have enrolled both their personal and work phones (Figure). LESSONS LEARNED Opt-ins dramatically increased during a snow emergency when the system was deployed, so we found that stated intentions did not necessarily match actual interest in the absence of an emergency event that may have increased salience. More research should be conducted to explore the difference between stated intentions to enroll vs. actual enrollment, although circumstantial evidence suggests that real-world events increase the salience for such a system for employees, possibly because the value of such a program may be more clearly demonstrated. Employers may consider reminding employees about the system immediately prior to and during an emergency to increase enrollment. Because not all employees will sign up to receive emergency alerts via SMS, text messaging will not replace, but can augment, other channels for communicating emergency information with employees. Text messaging can be particularly useful when other forms of communication (e-mail, website, or hotlines for employees to call in for emergency information) are not available or are inconvenient. Audience research with employees is critical to an emergency texting program’s success. Designing and marketing a program that employees will find attractive is critical, particularly when departmental policy does not permit managers to require that employees sign up for a texting program on their personal cell phones. Employees who choose not to enroll in a text messaging service on their personal phones may do so primarily because of
Public Health Reports / 2014 Supplement 4 / Volume 129
An Employee Text Messaging Program for Emergencies 65
Figure. Employee enrollment in an SMS program at Public Health – Seattle & King County, Washington, before, during, and after an ice storm, 2011–2012
SMS 5 short message service
concerns that they will receive messages that are not about emergencies and that work will encroach on their private lives. CONCLUSION It is critical to get procedures in place to send text messages during an emergency in the timeliest way possible. We speculate that there may be an end-user expectation that information delivered by text message is particularly timely, as one of the key features of text messaging is instantaneousness. It may be important to make clear to subscribers that messages are sent as soon as policy decisions or other critical information is known, but not before. In our health department, efforts will be made to further describe how texting will be used during emergencies to more clearly manage expectations regarding message timeliness and content. The authors thank collaborators at the Northwest Preparedness and Emergency Response Research Center and at Public Health – Seattle & King County, including Norah Herzog-Meyerson, Logan Kendall, Meredith Li-Vollmer, and Whitney Offenbecher. This work was supported by the Centers for Disease Control and Prevention (CDC) grant #5P01TP000297. The contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of CDC.
REFERENCES 1. Tucker C. Social media, texting play new role in response to disasters: preparedness, communication targeted. Nations Health 2011;41: 1-18. 2. Traynor P. Characterizing the security implications of third-party emergency alert systems over cellular text messaging services. IEEE Transactions on Mobile Computing 2012;11:984-94. 3. Karasz HN, Li-Vollmer M, Bogan S, Offenbecher W. Targeting young adult texters for public health emergency messages: a Q-study of uses and gratifications. In: Ahmed R, Bates BR, editors. Health communciation and mass media: an integrated approach to policy and practice. Surrey (UK): Gower Publishing; 2013. p. 115-31. 4. Rainie L. Cell phone ownership hits 91% of adults. Washington: Pew Internet and American Life Project, Pew Research Center; 2013 Jun 6. Also available from: URL: http://www.pewresearch. org/fact-tank/2013/06/06/cell-phone-ownership-hits-91-of-adults [cited 2013 Jul 10]. 5. Duggan M, Rainie L. Cell phone activities 2012. Washington: Pew Internet and American Life Project, Pew Research Center; 2012 Nov 25. Also available from: URL: http://pewinternet.org /Reports/2012/Cell-Activities/Main-Findings.aspx [cited 2013 Jul 10]. 6. Pew Research Center. Texting is nearly universal among young adult cell phone owners. Washington: Pew Research Center; 2012 Dec 14. Also available from: URL: http://www.pewresearch.org/ daily-number/texting-is-nearly-universal-among-young-adult-cellphone-owners [cited 2013 Jun 27]. 7. Institute of Medicine (US). Research priorities in emergency preparedness and response for public health systems: a letter report. Washington: National Academies Press; 2008. 8. McClain MA. Employee crisis communication and disaster assistance planning: providing disaster assistance to employees and their families. J Business Continuity Emerg Plann 2007;1:213-20. 9. Foster A. After Virginia Tech, campuses rush to add alert systems. Chronicle of Higher Education 2007 Sep 28 [cited 2013 Jul
Public Health Reports / 2014 Supplement 4 / Volume 129
66 Enhancing Communications
10. 11. 12. 13.
11]. Available from: URL: http://www.csun.edu/pubrels/clips /Sept07/09-28-07H.pdf Mastrodicasa J. Technology use in campus crisis. New Directions for Student Services 2008;2008:37-53. Sherman-Morris K. Tornado warning dissemination and response at a university campus. Nat Hazards 2010;52:623-38. Epstein RH, Ekbatani A, Kaplan J, Shechter R, Grunwald Z. Development of a staff recall system for mass casualty incidents using cell phone text messaging. Anesth Analg 2010;110:871-8. CTIA-The Wireless Association. CTIA compliance assurance solution: mobile compliance handbook [cited 2014 Jan 1]. Available from: URL: http://www.wmcglobal.com/assets/ctia-mobile-commercecompliance-handbook-v1-2.pdf
14. Welch M. Appropriateness and acceptability: employee perspectives of internal communication. Public Relations Rev 2012;38:246-54. 15. Lee D, Chung JY, Kim H. Text me when it becomes dangerous: exploring the determinants of college students’ adoption of mobilebased text alerts short message service. Comput Human Behav 2013;29:563-9. 16. Lenhart A, Ling R, Campbell S, Purcell K. Teens and mobile phones. Washington: Internet and American Life Project, Pew Research Center; 2010 Apr 20. Also available from: URL: http:// www.pewinternet.org/2010/04/20/teens-and-mobile-phones [cited 2013 Dec 12]. 17. Public Health – Seattle & King County. Winter storm response January 17, 2012–January 21, 2012: after action report/improvement plan [cited 2013 Sep 21]. Available from: URL: http://www.apctoolkits .com/documents/vp/AAR-IP_Winter_Storm_Jan_2012_vpat.pdf
Public Health Reports / 2014 Supplement 4 / Volume 129
Communicating with the Workforce During Emergencies: Developing an Employee Text Messaging Program in a Local Public Health Setting Hilary N. Karasz, PhD, MAa Sharon Bogan, MPHa Lindsay Bosslet, MPHa
ABSTRACT Short message service (SMS) text messaging can be useful for communicating information to public health employees and improving workforce situational awareness during emergencies. We sought to understand how the 1,500 employees at Public Health – Seattle & King County, Washington, perceived barriers to and benefits of participation in a voluntary, employer-based SMS program. Based on employee feedback, we developed the system, marketed it, and invited employees to opt in. The system was tested during an ice storm in January 2012. Employee concerns about opting into an SMS program included possible work encroachment during non-work time and receiving excessive irrelevant messages. Employees who received messages during the weather event reported high levels of satisfaction and perceived utility from the program. We conclude that text messaging is a feasible form of communication with employees during emergencies. Care should be taken to design and deploy a program that maximizes employee satisfaction.
Public Health – Seattle & King County, Seattle, WA
a
Address correspondence to: Hilary N. Karasz, PhD, MA, Public Health – Seattle & King County, 401 Fifth Ave., Ste. 1300, Seattle, WA 98104; tel. 206-263-8705; fax 206-296-0166; e-mail . ©2014 Association of Schools and Programs of Public Health
Public Health Reports / 2014 Supplement 4 / Volume 129
61
62 Enhancing Communications
Advances in technology mean that there are more ways than ever to reach employees during an emergency. For example, short message service (SMS) text messaging can be used to augment existing communication channels.1 While relying on a single technology is not recommended, and concerns exist about the reliability of SMS during large-scale emergencies,2 text messaging is particularly valuable when power is out, voice-to-voice lines are overloaded, or people are away from a computer.3 SMS may be more useful during some emergency situations because of cell phone ubiquity: In 2013, adult cell phone ownership in the United States reached 91%,4 and an estimated 80% of cell phone owners send or receive text messages.5 These rates are even higher among people of color (of all ages), who text more frequently than non-Hispanic white people do.6 A 2008 Institute of Medicine report outlined steps to strengthen public health systems by improving emergency preparedness and response in the U.S.7 Improving employee readiness and emergency communications were identified as key research objectives. One important way to strengthen the public health system and aid continuity of operations during a crisis is to assure that employees can be reached during an emergency.8 However, anecdotal evidence suggests that local health departments (LHDs) are not broadly adopting texting alert systems for their employees. We were unable to locate any previous research addressing how such systems may be implemented by LHDs, so this research may provide new information for LHDs that wish to improve employee emergency communication. A growing number of institutions use text messaging alert systems. For example, following the shootings at Virginia Tech in 2007, hundreds of universities purchased text alert systems, where cell phone and text message adoption among students is nearly universal.6,9,10 During a tornado advisory, a large university used a combination of text messages, e-mails, and online instant messaging to communicate with students and staff.11 In a health services institutional context, a study of emergency communications in a hospital’s anesthesiology department found its current phone tree system was outdated, inconsistent, time intensive, and ultimately ineffective. This department tested an SMS system and found that it was likely to improve employee mobilization efforts during an emergency.12 Carrier rules prohibit requiring people to subscribe to text messaging services, so these systems are optional for subscribers.13 Fifteen hundred employees of Public Health – Seattle & King County (PHSKC) provide population and personal health services across 40 locations to
two million residents of King County, Washington. We describe PHSKC’s experience planning and implementing an employee emergency SMS program to improve awareness, safety, and continuity of operations. The objective was to better understand whether an optional public health employer-based texting program would be acceptable and useful for staff. Informed by Fishbein and Ajzen’s technology acceptance model (TAM), which explains the process by which individuals accept new information technologies, we found that attitudes, subjective norms, and perceived intrusiveness of a text alert program predicted acceptance of such a program. While this study did not aim to evaluate the effectiveness of TAM to predict employee uptake of an emergency SMS alert system, TAM does point to the value of understanding employee attitudes and perceived barriers and benefits to opting in, so that a program can be successfully developed and marketed.3,14,15 Furthermore, because telecommunications industry rules prohibit sending text messages to recipients without permission and people must opt in, it was important to understand employee concerns prior to developing and marketing a system.13 METHODS Formative audience research We developed an 18-item electronic survey to gauge employee interest in receiving text message alerts about a variety of health topics, including emergency alerts and personal health tips. The survey was reviewed and approved by department administrators and union representatives. Employees were recruited by e-mail. About half of the 1,536 staff (n5828) responded. Results revealed several themes: (1) emergency and site closures were topics of interest to a majority of respondents, (2) interest in signing up for a text messaging service was inversely correlated with age, and (3) employees across the age spectrum had significant concerns about receiving messages from work. Topics. Survey respondents were given a list of topics and asked to choose the topics that interested them most. Worksite closure information and emergencies were of interest to about half of the respondents (54% and 56%, respectively), while public health news (10%) and health tips (4%) were the least interesting topics. About one-third of respondents (37%) indicated no interest in receiving messages on any topic. Age. Younger PHSKC workers were more likely than older workers to say they would be willing to sign up for a text program from PHSKC. While about one-third
Public Health Reports / 2014 Supplement 4 / Volume 129
An Employee Text Messaging Program for Emergencies 63
of respondents (n5253) said they would be likely or very likely to sign up, 51% (n527) of respondents aged 18–29 years said they were likely or very likely to sign up, and only 18.5% (n521) of respondents aged $60 years said they were likely or very likely to sign up for a text program. Texting experience also correlated with age: of 18- to 29-year-old respondents, 64% (n534) reported texting every day and only 6% (n53) reported rarely or never texting. In contrast, 64% (n573) of respondents aged .60 years reported never texting. No comparable data from similar employee surveys were identified to suggest how these responses would compare with responses from other public health departmental workforces (data not shown). Areas of concern. The survey revealed employee concerns about the cost of receiving texts, receiving irrelevant information, and losing the distinction between being “on” and “off” the job. While many cell phone subscribers have unlimited texting packages,16 not everyone does, at least at the time the survey was conducted in 2011. Fifty-six percent of employees (n5442) said that cost would be a limiting factor in their decision to sign up for a PHSKC texting program. Many respondents stressed that the system must be voluntary because of cost. One respondent said, “It’s a great technology. It’s essential that employees have an opt-out privilege because of expense.” Another respondent noted, “Signing up for this service would depend on how many messages I’m likely to receive … I’ll be paying for them after all.” Other respondents commented that they would be more likely to opt in if PHSKC paid for their texts. One respondent wrote, “I’d access this resource if [PHSKC] paid for my texting.” Another respondent wrote, “I do not use my personal cell phone for business. Even if the county were to pay for the text, would you pay for my time?” (data not shown). Concern about intrusion on personal time was a common theme. One respondent said, “I don’t want to be notified about work issues on my off time. PERIOD.” Another respondent said, “I do not need to sit at a ball game and receive work stuff. That is my downtime.” Survey respondents also voiced concern that work text messages would add to “information overload.” One respondent said, “There is far too much information being sent to me day and night. I want less technology, not more. It’s decreasing my quality of life.” Another respondent commented, “I think this is a great plan for site closures or other crucial information. I would be hesitant to sign up for emergency info because what I think is an emergency is different from what the media and [PHSKC] often identify as one. I worry that there
would be too many non-emergent or non-crucial text messages” (data not shown). Implementation of PHSKC’s employee text messaging program The texting program aimed to be responsive to employee concerns. It was voluntary and limited to emergency messaging only. We worked with our SMS vendor to build a system that allowed targeted messaging: Employees could opt to receive messages about their worksite only, about other PHSKC worksites, or about all PHSKC worksites. Resources to pay for the development and implementation of the system were pulled from the emergency preparedness and departmental communications units, but sustainability of the system remains an ongoing challenge. Marketing the program to employees consisted of two e-mails to introduce the program to staff in March and April 2011, and a reminder in winter preparedness e-mails in November 2011 and 2012. Our initial e-mail appeal aimed to respond to the concerns raised by employees who responded to the survey. We assured staff that (1) the program was voluntary and for their convenience only, (2) communications would be limited to emergencies and the occasional test message, (3) they could opt out at any time, (4) they would not be accountable for the information received by text message, and (5) their phone numbers would not be shared. By December 2011, 22.7% (n5349) of 1,536 employees had signed up for the program. The number of employees who signed up was low compared with what the formative research indicated, which may have been because the program was not aggressively marketed. In addition, post-marketing interviews with employees who did not opt in indicated that the reason they did not opt in was because they did not pay attention to the initial e-mails inviting them to do so, not because of concerns about irrelevant or too many texts. OUTCOMES In January 2012, King County, which has a geographical area of 2,307 square miles, experienced a significant winter storm in which 250,000 households and businesses lost electricity at some point during a five-day period. In parts of King County, one-third of the homes lost power.17 Many arterial roads were icy or not plowed. Buses and other transit were delayed or canceled. Five PHSKC health clinics lost power; four clinics lost power for four days. We texted our employee subscribers as soon as policy makers decided
Public Health Reports / 2014 Supplement 4 / Volume 129
64 Enhancing Communications
if, when, or where they should report to work. During the five-day event, subscribers received up to 15 text messages alerting them to late starts, clinic closures, and directions for getting more information. Typical messages were “PHSKC: KENT EAST HILL is closed due to lack of power. Check with supervisor about work plan for day. More info from HR to follow. Check PH web for updates,” and “Ice storm. Roads dangerous. PHSKC on late start at 10. If not critical staff, check with supervisor before coming. Check Web and hotline [number given] for info.” Following the storm, we gathered feedback from staff via electronic survey. We received 185 responses (36% of subscribers), and the feedback was primarily positive. Sixty-two percent of respondents (n5115) said the text messages were very relevant and helpful, 20% (n537) found them fairly relevant and helpful, and only 5% (n510) found them not relevant or helpful. About 70% (n5129) of recipients learned at least one piece of information via text message before they heard about it another way (e.g., the website, e-mail message, or hotline). With respect to the number of messages we sent (averaging three per day), 83% (n5154) reported that the number of messages was about right, 15% (n528) thought they received too few messages, and 2% (n54) said they received too many messages. Only two people opted out during this period, while 165 additional employees subscribed during the five days of the storm (data not shown). The texting service was helpful for employees who had lost electricity. A typical comment was, “Since I had no power and was unable to access my e-mail or Internet, I really appreciated the text messages.” Another comment was, “I was trying to get into [the Internet] to get an update and my power kept going in and out, then I received the text message. I had forgotten I had signed up for it but was very glad I did” (data not shown). Employees made it clear that they expect the texting service to provide timely information. One respondent noted, “If it could just arrive a teensy bit sooner, I can prepare better. I know that sometimes the information isn’t known right away. I think it has to do more with the planners for inclement weather, making a decision sooner.” Others complained that they had already left for work by the time they received a message about a late start or closure. One respondent said, “The first text message on Thursday about a late start time at 10 a.m. was texted too late. I don’t know when the decision for a late start time was made, but if it could be made by 6 a.m. and texted at that time, it would really help with the commute” (data not shown).
Usage and enrollment since the snowstorm PHSKC continues to use the emergency texting system, and analyses of the program’s success are still being reviewed. Since the 2012 winter storm event, we have employed the emergency text messaging program three additional times. Subscribers at three different health clinic sites were sent messages advising them of police actions in the vicinity of their clinics. The number of employees subscribed to the system has reached 664, and in three years, only 15 people have opted out of it. Marketing has been limited, primarily consisting of a half-dozen all-staff e-mails. All PHSKC-owned work cell phone numbers (n5248) have also been added as part of an enhanced enrollment effort, making a total of 912 phones that we have enrolled in the employee text messaging system. Employees may sign up with both personal cell phones and work phones, so it is unclear how many of the 912 total subscribers are individual employees, or if some of them have enrolled both their personal and work phones (Figure). LESSONS LEARNED Opt-ins dramatically increased during a snow emergency when the system was deployed, so we found that stated intentions did not necessarily match actual interest in the absence of an emergency event that may have increased salience. More research should be conducted to explore the difference between stated intentions to enroll vs. actual enrollment, although circumstantial evidence suggests that real-world events increase the salience for such a system for employees, possibly because the value of such a program may be more clearly demonstrated. Employers may consider reminding employees about the system immediately prior to and during an emergency to increase enrollment. Because not all employees will sign up to receive emergency alerts via SMS, text messaging will not replace, but can augment, other channels for communicating emergency information with employees. Text messaging can be particularly useful when other forms of communication (e-mail, website, or hotlines for employees to call in for emergency information) are not available or are inconvenient. Audience research with employees is critical to an emergency texting program’s success. Designing and marketing a program that employees will find attractive is critical, particularly when departmental policy does not permit managers to require that employees sign up for a texting program on their personal cell phones. Employees who choose not to enroll in a text messaging service on their personal phones may do so primarily because of
Public Health Reports / 2014 Supplement 4 / Volume 129
An Employee Text Messaging Program for Emergencies 65
Figure. Employee enrollment in an SMS program at Public Health – Seattle & King County, Washington, before, during, and after an ice storm, 2011–2012
SMS 5 short message service
concerns that they will receive messages that are not about emergencies and that work will encroach on their private lives. CONCLUSION It is critical to get procedures in place to send text messages during an emergency in the timeliest way possible. We speculate that there may be an end-user expectation that information delivered by text message is particularly timely, as one of the key features of text messaging is instantaneousness. It may be important to make clear to subscribers that messages are sent as soon as policy decisions or other critical information is known, but not before. In our health department, efforts will be made to further describe how texting will be used during emergencies to more clearly manage expectations regarding message timeliness and content. The authors thank collaborators at the Northwest Preparedness and Emergency Response Research Center and at Public Health – Seattle & King County, including Norah Herzog-Meyerson, Logan Kendall, Meredith Li-Vollmer, and Whitney Offenbecher. This work was supported by the Centers for Disease Control and Prevention (CDC) grant #5P01TP000297. The contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of CDC.
REFERENCES 1. Tucker C. Social media, texting play new role in response to disasters: preparedness, communication targeted. Nations Health 2011;41: 1-18. 2. Traynor P. Characterizing the security implications of third-party emergency alert systems over cellular text messaging services. IEEE Transactions on Mobile Computing 2012;11:984-94. 3. Karasz HN, Li-Vollmer M, Bogan S, Offenbecher W. Targeting young adult texters for public health emergency messages: a Q-study of uses and gratifications. In: Ahmed R, Bates BR, editors. Health communciation and mass media: an integrated approach to policy and practice. Surrey (UK): Gower Publishing; 2013. p. 115-31. 4. Rainie L. Cell phone ownership hits 91% of adults. Washington: Pew Internet and American Life Project, Pew Research Center; 2013 Jun 6. Also available from: URL: http://www.pewresearch. org/fact-tank/2013/06/06/cell-phone-ownership-hits-91-of-adults [cited 2013 Jul 10]. 5. Duggan M, Rainie L. Cell phone activities 2012. Washington: Pew Internet and American Life Project, Pew Research Center; 2012 Nov 25. Also available from: URL: http://pewinternet.org /Reports/2012/Cell-Activities/Main-Findings.aspx [cited 2013 Jul 10]. 6. Pew Research Center. Texting is nearly universal among young adult cell phone owners. Washington: Pew Research Center; 2012 Dec 14. Also available from: URL: http://www.pewresearch.org/ daily-number/texting-is-nearly-universal-among-young-adult-cellphone-owners [cited 2013 Jun 27]. 7. Institute of Medicine (US). Research priorities in emergency preparedness and response for public health systems: a letter report. Washington: National Academies Press; 2008. 8. McClain MA. Employee crisis communication and disaster assistance planning: providing disaster assistance to employees and their families. J Business Continuity Emerg Plann 2007;1:213-20. 9. Foster A. After Virginia Tech, campuses rush to add alert systems. Chronicle of Higher Education 2007 Sep 28 [cited 2013 Jul
Public Health Reports / 2014 Supplement 4 / Volume 129
66 Enhancing Communications
10. 11. 12. 13.
11]. Available from: URL: http://www.csun.edu/pubrels/clips /Sept07/09-28-07H.pdf Mastrodicasa J. Technology use in campus crisis. New Directions for Student Services 2008;2008:37-53. Sherman-Morris K. Tornado warning dissemination and response at a university campus. Nat Hazards 2010;52:623-38. Epstein RH, Ekbatani A, Kaplan J, Shechter R, Grunwald Z. Development of a staff recall system for mass casualty incidents using cell phone text messaging. Anesth Analg 2010;110:871-8. CTIA-The Wireless Association. CTIA compliance assurance solution: mobile compliance handbook [cited 2014 Jan 1]. Available from: URL: http://www.wmcglobal.com/assets/ctia-mobile-commercecompliance-handbook-v1-2.pdf
14. Welch M. Appropriateness and acceptability: employee perspectives of internal communication. Public Relations Rev 2012;38:246-54. 15. Lee D, Chung JY, Kim H. Text me when it becomes dangerous: exploring the determinants of college students’ adoption of mobilebased text alerts short message service. Comput Human Behav 2013;29:563-9. 16. Lenhart A, Ling R, Campbell S, Purcell K. Teens and mobile phones. Washington: Internet and American Life Project, Pew Research Center; 2010 Apr 20. Also available from: URL: http:// www.pewinternet.org/2010/04/20/teens-and-mobile-phones [cited 2013 Dec 12]. 17. Public Health – Seattle & King County. Winter storm response January 17, 2012–January 21, 2012: after action report/improvement plan [cited 2013 Sep 21]. Available from: URL: http://www.apctoolkits .com/documents/vp/AAR-IP_Winter_Storm_Jan_2012_vpat.pdf
Public Health Reports / 2014 Supplement 4 / Volume 129
