Keywords COVID-19 - smartphones - in-hospital communication - secure messaging
Background and Significance
Background and Significance
In March 2020, in response to the coronavirus disease 2019 (COVID-19) pandemic, hospitals
and health care systems introduced drastic policy and practice changes that were necessary
for protecting the health and safety of patients and health care personnel. With more
than 3,000 Chinese health care workers inadvertently infected in the early months
of the pandemic, several recommendations were issued regarding the need for agile
crisis communication strategies in the hospital setting.[1 ]
[2 ]
[3 ] In line with these recommendations, leadership at the Hospital of the University
of Pennsylvania (HUP) recognized the need to ensure hundreds of hospital employees
across all departments and disciplines received information about infection prevention
strategies, proper personal protective equipment (PPE) use, testing protocols, and
symptom screening in a timely and consistent way, and with a mechanism for feedback.[1 ]
[4 ]
[5 ]
[6 ] Without this approach, inconsistent messages contributed to variations in practice
and deviations from organizational standards developed by internal experts. These
deviations placed employees and patients at risk for exposure. Further, given the
uncertainty surrounding COVID-19, inconsistent messaging unintentionally created confusion,
exacerbated fear, and eroded trust.[7 ]
To minimize inconsistencies, an innovative, multimodal communication strategy needed
to be rapidly implemented. Traditional communication methods, which were already being
employed, each had limitations. For instance, email allows leadership to deliver consistent
messages to large groups, but readership is often poor or delayed among employees.[8 ] Conversely, in-person methods, like huddles, engage employees directly but are limited
by the need for physical proximity and lack consistent delivery across multiple different
groups.
Objectives
Over the last decade, smartphone-based text messaging via Health Insurance Portability
and Accountability Act (HIPAA)-compliant applications has increased in popularity
for both two-way and team-based in-hospital communication as a way to improve clinical
communication.[8 ]
[9 ]
[10 ] Leveraging this type of technology, our team implemented an enterprise text-messaging
intervention that complemented existing communication methods to (1) deliver consistent
and timely COVID-related information directly to hospital employees during the COVID-19
surge, (2) promote uptake of this information to minimize COVID-19 exposure events,
and (3) provide a mechanism to facilitate front-line feedback or questions about pandemic
policies.
Methods
Implementation
At the start of the COVID-19 surge, information about evolving practices was predominantly
distributed by word of mouth among care teams. Expeditiously, several communication
strategies, including in-person huddles between leaders and employees, a dedicated
Web site with policies and guidelines, and mass emails from the hospital executive
team to all employees, were implemented to disseminate information more readily. Enterprise
text messaging was added as a component of the hospital-wide strategy on March 26,
2020 as the COVID-19 census surged; it ended on June 19, 2020 after the surge concluded
locally.
Technology Platform
We used an existing text-messaging tool, Cureatr (Cureatr Inc., New York, New York,
United States) to implement our intervention. Cureatr is a commercially available
HIPAA-compliant secure text-messaging application that has been used at HUP since
2013.[9 ] All hospital employees, including but not limited to physicians, nurses, advanced
practice providers, social workers, environmental service staff, therapists, and unit
secretaries, have access to Cureatr on hospital-issued devices (e.g., smartphones
or computers). Prior to our intervention, these groups used the platform only to communicate
with other care team members to coordinate direct patient care. The platform was never
previously utilized enterprise-wide for communication from hospital leadership to
employees.
Enterprise Texting Workflow
We created a sender account titled, “Penn Infection Control” to mirror a health system
department widely recognized as experts responsible for developing infection control
policies. When a message was sent from this account, it was immediately distributed
to all on-duty Cureatr users who were assigned to an active inpatient clinical service.
The message appeared on their devices as a push notification from “Penn Infection
Control” with a subject line “COVID19 Advisory.” Since Cureatr is a two-way communication
tool, recipients had the ability to reply to “Penn Infection Control” with feedback
or questions ([Fig. 1 ]). Cureatr also allowed us to track “read” receipts when each user opened each message.
Read receipts could not be disabled by end users.
Fig. 1 Screenshot of enterprise text with reply from end user.
Message Content
Ensuring that the text messages sent to hospital employees were accurate and up to
date was critical to maintaining consistency.[2 ]
[4 ] Message quality control was maintained through multiple tiers of input and review.
To develop content, our team first selected and prioritized topic areas for each message
in collaboration with the hospital executive team. Message content focused primarily
on infection control strategies, such as appropriate PPE use, locations for isolating
COVID-19 patients, and the COVID inpatient census. Special consideration was given
to newly implemented or refined policies, questions frequently raised across departments,
observed deviations from existing guidance, and/or key learning from recent exposure
events. Once topics were selected, our team drafted each message with direct input
and approval from the Chief Medical Officer to ensure the language in the messages
aligned with content delivered through other modalities like mass emails. Other communication
channels concurrently used for COVID-19 communication were daily mass emails from
the hospital executive team, weekly department-specific emails, and daily in-person
huddles plus ad hoc huddles as needed.
To promote uptake of the information, content also needed to be succinct. Messages
were, therefore, typically limited to 140 characters to both optimize readability
and encourage a succinct writing style.[9 ]
[10 ] We used a consistent and predictable sender and subject line to alleviate perceptions
of overload and focus hospital employees' attention on the message content.[5 ] Additionally, messages were scheduled for Mondays, Wednesdays, and Fridays between
8 and 9 a.m., so recipients would become familiar with the messaging cadence. Occasionally,
urgent messages were sent as needed outside of this schedule. Lastly, messages were
only sent to on-service users to reduce the number of messages sent to off-duty employees.
If an on-service user logged on to the application at any point within 24 hours after
the initial message push, then that message would be automatically delivered.
Intervention Outcomes
In addition to message readership, we also evaluated trends in occupational exposures
that were reported to HUP Occupational Health over the course of the intervention
from March 3, 2020 to June 22, 2020. Users were also administered two cross-sectional
attitudinal surveys which were developed de novo by our team with the goal of being
short and elucidating user satisfaction with the intervention. The survey was distributed
via a link within the secure messaging platform. All statistical analyses of these
outcomes were completed using RStudio version 1.2.5033.
Results
Readership Rates
During the intervention period, 34 enterprise text messages were sent to an average
of 1,997 employees (range = 1,799–2,049). On average, 60% of the enterprise text messages
were read within 24 hours, translating to approximately 1,200 readers per message
([Fig. 2 ]). On average, 34% of employees (n = 680) read each text message within 2 hours, and 16% of employees (n = 325) read each text message within 10 minutes. Peak user readership occurred during
the second week of the intervention period, on April 1, 2020, with a message detailing
proper PPE use during aerosol-generating procedures; 79% of users (1,577/1,996) read this message within 24 hours. This delivery date coincided with increasing surge
activity. We reached our peak inpatient COVID-19 census on April 20. Readership fell
slightly following the first week of the intervention but remained consistently at
60% in subsequent weeks ([Fig. 2 ]). Notably, the readership rate for mass emails, which were sent daily to approximately
9,500 employees, regardless of clinical duty status, was 42%.
Fig. 2 Daily readership rates of enterprise text messages during the COVID surge at the
Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania, United States.
Blue dots indicate the percent of employees who read each message within 24 hours with smooth
local regression line . Red fill indicates the number of COVID patients admitted over the course of the surge.
Attitudinal Survey Responses and Direct Replies from Message Recipients
Twice during the intervention period, we administered Likert-scale surveys to assess
the utility of the enterprise text messages. Each survey was followed by a reminder
message sent 1 week later to increase participation. The first survey, administered
15 days after the intervention began, achieved a 10.1% (207/2,049) response rate and
showed that across specialties and roles, 79% (n = 163) of respondents found the messages “valuable.” Seventy-three percent (n = 151) of respondents “would recommend” the messages to their colleagues, and 55%
(n = 114) preferred the enterprise text messages to mass emails ([Table 1 ]). The second survey, administered 63 days after the intervention began, achieved
a 6.7% (137/2,049) response rate and showed that 80% (n = 109) of respondents found the messages “valuable.” When asked about how enterprise
text messaging could be used in the future, 31% of respondents suggested continuing
them for COVID-19 communications and 28% suggested using them for information about
changes in hospital operations or patient flow ([Table 1 ]).
Table 1
Attitudinal surveys of enterprise text recipients
Survey 1
Question
Strongly disagree
Disagree
Neither agree nor disagree
Agree
Strongly agree
Enterprise text messages are valuable for clarifying real-time best practices
4.37% (9)
5.83% (12)
10.68% (22)
40.29% (83)
38.83% (80)
Would recommend these text messages to my colleagues
3.88% (8)
6.8% (14)
15.53% (32)
37.38% (77)
36.41% (75)
Prefer to receive future messages via text opposed to email
9.27% (19)
18.05% (37)
17.07% (35)
30.24% (62)
25.37% (52)
Survey 2
Question
Strongly disagree
Disagree
Neither agree nor disagree
Agree
Strongly agree
Enterprise text messages are valuable for clarifying real-time best practices
7.35% (10)
6.62% (9)
5.88% (8)
38.97% (53)
41.18% (56)
Question
Operations/hospital flow
COVID-related
Informatics
Patient safety
Other (specify)
Preferred areas for future text messages (select all that apply)
28.03% (104)
31.54% (117)
16.71% (62)
22.1% (82)
1.62% (6)
Additionally, 41 message recipients (2%) utilized Cureatr's direct reply feature to
reply to an enterprise text message with a question or comment. Although the number
of replies was small, we categorized them by theme to better understand what content
resonated most with hospital employees. The four themes included PPE use (21/41),
personal safety (9/41), feedback about the text-messaging strategy (8/41), and specific
questions pertaining to patient care (3/41).
Employee Exposure Events
We compared the number of COVID-19 exposure events among hospital employees before
and after implementation of enterprise messaging ([Fig. 3 ]). Employee exposure events were defined as an employee having at least a direct
contact within 6 feet with a patient or staff member who was or has since been confirmed
as having COVID-19 without wearing a proper PPE. Prior to implementing our intervention,
the median number of employee exposure events per day totaled 9.5 (interquartile range
[IQR] = 20.5), and the maximum number of exposure events in 1 day reached 113 (March
13, 2020). Following the implementation of our intervention on March 26, 2020, the
median number of exposure events per day dropped to 1 (IQR = 4) with a maximum of
31 events in 1 day (May 8, 2020).
Fig. 3 Daily occupational exposure trend before and during enterprise text-message intervention
at the Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania, United
States. Red fill indicates daily occupational exposure events. Blue fill indicates a smoothed line, estimated using local regression with 0.25 smoothing span,
for employee readership.
Discussion
We provide outcomes from a novel smartphone-based communication strategy for keeping
hospital employees abreast of updated clinical policies and safety practices during
a time of turbulent change. In our case study, smartphone-based guidance had consistently
high readership, which was associated with reduced occupational exposures. Our experience
with this intervention demonstrates several key takeaways. First, during crises like
COVID-19, hospital employees across all departments and disciplines must have access
to vital information in a consistent, timely, and efficient way to minimize safety
risks. Second, leveraging pre-existing clinical informatics infrastructure, enterprise
text messaging can be rapidly implemented to deliver targeted information to hospital
employees and meet this need. Third, to mitigate notification burden, confusion, and
fear, we believe enterprise text messages should be succinct, delivered only to on-duty
employees, and deliberately aligned with messages delivered via other channels.
Given the volume of information that hospital employees were receiving during the
surge, we realized that additional text messages may feel burdensome rather than helpful.
Perceptions of high digital alert volume are associated with symptoms of physician
burnout.[11 ] Minimizing notification burden was, therefore, essential. We therefore deliberately
designed the enterprise text messages to only highlight crucial updates/changes and
reminders with a succinct style that limited content and maximized clarity.
One limitation of this intervention was the limited ability to further evaluate the
value of this messaging strategy for hospital employees. Although the advent of this
intervention coincided with an overall decrease in occupational exposure events, there
were several confounders that may have also reduced exposures. These included increased
employee comfort with PPE policies, an overall decrease in regional COVID-19 cases,
and increased awareness through other communication channels. Additionally, the 60%
24-hour readership rate may have been lowered by messages sent to on-service, but
night-shift, or logged-off employees who did not check their hospital-issued device
within 24 hours of message delivery.
End-user feedback was an additional limitation. We expected that one source of feedback
would be directly from employees leveraging Cureatr's two-way messaging capability.
Indeed, some hospital employees sent questions or feedback, but most did not. We therefore
solicited feedback using an attitudinal survey distributed via a hyperlink embedded
within an enterprise text message. However, this feedback may have been biased by
the low response rate and the mode of delivery being to engage text-messaging users.
Recognizing that most recipients were unable to prioritize taking a survey during
this critical time period and hoping to avoid additional notification burden, we did
not further solicit attitudinal feedback via other mechanisms, such as direct interviews.
Given our promising initial results, we are exploring additional uses for this technology
beyond the pandemic response, ranging from hospital-wide operational changes to department-level
initiatives. Additionally, given the Centers for Disease Control's predictions for
a resurgence of COVID-19 in our region during the upcoming months,[12 ] we will be well equipped to rapidly reimplement enterprise text messaging to deliver
updates on safety policies and patient census if needed. As we discover more use cases,
we recommend adherence to several guiding principles, including the creation of succinct
and consistent messaging, monitoring for notification fatigue, and gathering readership
data to evaluate this communication method.[4 ]
[13 ]
Conclusion
Enterprise text messaging can be effectively utilized as a crisis communication strategy
to supplement traditional communication methods and deliver vital information directly
to hospital employees. We found that rapidly implementing this platform was feasible
using existing infrastructure for secured text messaging, and hospital employees value
receiving information in this targeted way.
Clinical Relevance Statement
Clinical Relevance Statement
Leveraging secure messaging technology to make information readily accessible to hospital
employees during the COVID-19 surge likely helps them adapt to required, frequent
workflow changes and adopt safe practices.
Multiple Choice Questions
Multiple Choice Questions
Which of the following may lead to higher occupational exposure to COVID-19 among
hospital employees?
Inconsistent messaging about proper PPE use
Staffing shortages
Use of email to deliver PPE guidance
Use of in-person meetings to deliver PPE guidance
Correct Answer: The correct answer is option a. Inconsistent messaging about proper PPE use may lead
to confusion and mistrust among employees leading to improper PPE use and consequent
exposures. Email and in-person meetings both have strengths for delivering guidance
about the pandemic response including email having a large number of recipients, and
in-person meetings ensuring delivery of the information. However, it is critical to
maintain consistent messaging across these varied platforms.
When implementing an enterprise-text crisis communication strategy, which of the following
strategies can minimize notification fatigue among hospital employees?
Highly detailed message content
Limiting the number of message recipients
Succinct message length
Varied timing of message delivery
Correct Answer: The correct answer is option c, succinct message length. Given that higher digital
alert volume is associated with more physician burnout, added messaging modalities
such as text alerts should provide content that is direct and streamlines the recipient's
attention. Consistent messages should be delivered to all hospital employees, and
predictable timing and subject lines of messages sent allows recipients to focus attention
on message content.
Which of the following topics may be the most crucial messaging content area for improving
employee safety during the COVID-19 pandemic?
Testing procedures for patients under investigation
Symptom screening among hospital employees
Proper practices for employee PPE use
Inpatient COVID-19 patient census
Correct Answer: The correct answer is option c, proper practices for employee PPE use. Given that
COVID-19 is primarily transmitted through respiratory droplets, airway and physical
barriers are critical for safety among front-line employees who must have direct patient
contact. Surgical masks, respirators, and eye protection may only serve this purpose
when worn properly and in the appropriate settings. As policies on PPE use rapidly
changed it was critical to deliver up-to-date and consistent instruction.
