i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 278–284
journal homepage: www.ijmijournal.com
Short message service or disService: Issues with text messaging in a complex medical environment Robert Wu a,b,∗ , Lora Appel a,c , Dante Morra d , Vivian Lo a , Simon Kitto e,f,g , Sherman Quan d a
Centre of Innovation in Complex Care, University Health Network, Toronto, ON, Canada Department of General Internal Medicine, University Health Network, Toronto, ON, Canada c Department of Communication and Information, Rutgers University, New Brunswick, NJ, USA d Trillium Health Partners, Mississauga, ON, Canada e Office of Continuing Education and Professional Development, University of Toronto, ON, Canada f Li Ka-Shing Knowledge Institute of St. Michael’s Hospital, Toronto, ON, Canada g Department of Surgery, University of Toronto, Toronto, ON, Canada b
a r t i c l e
i n f o
a b s t r a c t
Article history:
Background: Hospitals today are experiencing major changes in their clinical communication
Received 10 June 2013
workflows as conventional numeric paging and face-to-face verbal conversations are being
Received in revised form
replaced by computer mediated communication systems. In this paper, we highlight the
20 December 2013
importance of understanding this transition and discuss some of the impacts that may
Accepted 10 January 2014
emerge when verbal clinical conversations are replaced by short text messages.
Keywords:
were conducted on the General Internal Medicine wards at five academic teaching hospitals
Hospital communication system
in Toronto, Canada.
Methods: In-depth interviews (n = 108) and non-participatory observation sessions (n = 260 h)
Email
Results: From our analysis of the qualitative data, we identified two major themes. De-
Interprofessional communication
contextualization of complex issues led to an increase in misinterpretation and an increase
Smartphones
in back and forth messaging for clarification. Depersonalization of communication was due to less verbal conversations and face-to-face interactions and led to a negative impact on work relationships. Conclusions: Text-based communication in hospital settings led to the oversimplification of messages and the depersonalization of communication. It is important to recognize and understand these unintended consequences of new technology to avoid the negative impacts to patient care and work relationships. © 2014 Elsevier Ireland Ltd. All rights reserved.
1.
Background
Smartphones are increasingly being adopted as a means of communication within hospitals [1–3]. Smartphones have the ability to support both voice and text messaging, and they
create a platform for asynchronous communication in addition to providing access to medical applications and the Internet. For these reasons, experts believe that smartphones may not only serve as efficient communication tools but may also improve the coordination of patient care in the clinical setting [4].
∗ Corresponding author at: 200 Elizabeth Street, 14EN222, Toronto General Hospital, Toronto, ON, Canada M5G 2C4. Tel.: +1 416 340 4567; fax: +1 416 595 5826. E-mail address: [email protected] (R. Wu). 1386-5056/$ – see front matter © 2014 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijmedinf.2014.01.003
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 278–284
However, there may be issues with their use within hospitals where the complexity of medical work is compounded with a tradition of professional fragmentation and a diffusion of accountability [5]. There is increasing recognition that the implementation of information technology in the complex health care environment can result in unintended consequences with increased errors. Increased errors have been seen with computerized physician order entry, clinical decision support systems, and barcode medication administration [6,7], and therefore, it is important to identify the unintended consequences of replacing pagers with smartphones in order to mitigate them. The use of smartphones is moving conversations that were traditionally face-to-face or by telephone to a more text-based communication mode [8]. Such computer-mediated communication (CMC) has known effects and has been described previously in literature external to healthcare. CMC is a term used to describe the broad concept of communication, the transfer of information, via any technological means. This includes telephone calls, Skype calls, short message service, and instant messaging among many others. Of the CMC mediums, text-based is the least media rich, after voice-only and voice and video communication. Although CMC is similar to conversation by being “interactive, relatively spontaneous and generally unplanned,” [9] it changes the way people connect [10]. Although CMC affords users the ability to alleviate barriers in space and time, it is sometimes described as a rarefied form of conversation because it does not always support meta-communicative features like facial expression, posture and tone of voice which add richness to communication [11]. As opposed to other forms of text-based communication such as email, text messaging dilutes social presence and media richness further as communication is reduced to a short message, typically less than 140 characters [12]. In a clinical environment, text messaging can be especially limiting when complex information is conveyed with only a few characters and can lead to miscommunication between senders and receivers. This can damage an already sensitive interprofessional relationship [13]. Recent research on CMC in hospitals, and specifically the use of smartphones and texting between clinicians, has shown improvements in efficiency, but it is important that we also understand the unintended consequences [13,14]. Textbased messaging can also lead to ineffective communication and deteriorate interprofessional relations. In order to understand the impact of text-based communication, we examined
279
this aspect through the lens of computer-mediated communication.
2.
Methods
2.1.
Study design
To provide a more comprehensive understanding of clinicians’ experiences with the communication technologies in the hospital settings, we adopted an exploratory case study approach [15,16] consisting of data triangulation from multiple sources that included (a) semi-structured interviews and (b) non-participatory ward observation methods. This dataset is part of a larger multi-site study that evaluated the impact of different types of communication interventions and initiatives in hospital settings [8]. The original study was conducted on the General Internal Medicine (GIM) wards at five academic teaching hospitals that are affiliated with the University of Toronto, Canada. In this paper, we explore the implications of moving towards text-based communications on interprofessional relationships among medical teams and clinicians.
2.2.
Setting
Each participating hospital has two to four GIM wards that are located on different floors within the respective medical facilities. Each site has clinical teaching units with typically four medical teams, each consisting of an attending physician, a senior resident, junior residents and medical students. Four of the five hospitals have adopted text-based communication channels as part of clinicians’ routine workflow (Table 1).
2.3.
Data collection
Qualitative data consisting of in-depth interviews and nonparticipatory ward observations were collected from June 2009 to September 2010 across the five hospital sites. Ethical approval was obtained from the respective institutions’ Research Ethics Board committees.
2.3.1.
Interviews
Participants were selected and solicited through purposeful and criterion-based sampling techniques to ensure a variety of appropriate data sources were obtained. The criteria for recruitment were that participants had to be practicing
Table 1 – Different communication devices and methods adopted at the five hospitals. Site
Communication method
1
Numeric pages
2
Alphanumeric pages, Smartphones
3.4
Smartphones, intranet-based messaging system Intranet-based task-management messaging system that queued non-urgent messages.
5
Typical communication process Nurses page physicians, and physicians respond by calling back to a telephone number. Nurses send text messages to physicians’ pagers. If necessary, physicians would call back using their smartphones and/or landline phones. Nurses use the intranet-based messaging system to send messages to physicians’ smartphones. Physicians respond by email or call back using the smartphones. Nurses enter a text-based message using the intranet-based messaging system, and physicians are notified on their pagers of a new message, which can be read on the same system. Physicians reply using the task-management messaging system.
280
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 278–284
clinicians who have experiences using the communication technologies and modalities that were implemented at their designated hospital site and wards. We also employed a maximum variation sampling approach by identifying participants from different clinical roles that included attending physicians, residents, nurses, medical students and allied health personnel [17]. We adopted a confirming/disconfirming strategy in which we asked the interviewees to recommend potential participants who might: (a) have a similar experience/viewpoint to their own; and/or (b) have a divergent experience/viewpoint from their own. A total of 108 interviews were conducted by a member of the research team and audio taped at all the five sites. Each interview lasted between 15 and 45 min and was carried out at a mutually convenient location within the respective sites. The interview protocol consisted of a series of open-ended questions with appropriate follow-up probes that explored the impact of different communication interventions (smartphone, numeric paging, and texting) on inpatient care delivery, workflow, and clinician relationships. The interviews were then professionally transcribed and data were entered into a qualitative analysis software tool (NVivo 8, QSR International) to facilitate coding and analysis.
2.3.2.
Observations
The communication processes in the hospitals were observed by documenting clinicians’ communication activities and interactions at the General Internal Medicine nursing stations – the hubs of many clinical communication exchanges. Using non-participatory observation, descriptions of communication events and issues were chronicled and time-stamped on field notes which were transcribed into Microsoft Word® documents [18]. No patient-related information was collected. Ward observations were conducted on both weekdays and weekends in 2-h time intervals across various timeslots. One member of the research team collected the data on sites 3 and 4 and trained another research assistant to collect data from sites 1, 2, and 5. A total of 131 observation sessions (n = 260 h) of the GIM nursing stations were recorded across all the five sites.
2.4.
Analysis
Data triangulation using the conventional thematic content analysis approach was undertaken to draw inferences about clinicians’ experiences around the use of text-based communication channels [19,20]. An initial sample of the interview transcripts and field notes from the ward observations were independently read and coded by two members of the research team to identify broad themes across all the sites. The coding process involved identifying descriptive categories in which the clinicians’ responses, communication events, and incidents would fit. The provisional thematic categories were discussed with the rest of the research team and then organized into a preliminary structure for coding the rest of the data with additional themes reported for each site as they emerged. Upon completion of the coding, the categories were again reviewed and refined.
3.
Results
In this study we identified two main themes regarding the unintended consequences of text-based communication: (1) the de-contextualization of complex issues leading to misinterpretation and increasing communication workload and (2) the depersonalization of communication as a barrier to interprofessional teamwork.
3.1. De-contextualization of complex issues leading to misinterpretation and increasing communication workload While the ability to communicate asynchronously can be very efficient [21], it can also lead to frustration in a hospital environment that can ultimately be detrimental to the optimal delivery of care. In the past, nurses would page physicians to alert them of a change in status with a patient. Physicians would then call back and have a conversation over the phone, or come to the nursing station in person and have a face-to-face conversation with the nurse. In these cases, both parties communicating have the ability to, in a short amount of time, exchange many back and forth utterances of speech. When dialogue is composed synchronously, people can more easily detect and correct for misinterpretations and misunderstandings in communication. This can be done by requesting additional detail, by repeating or stressing words to ensure the message is delivered, or by interpreting non-verbal cues, such as tone of voice, that can help convey the urgency of a situation. With text-based communication, messages are often reduced in length, sometimes to only a few characters, and this simplification and de-contextualization of content creates the possibility for misinterpretation. Replacing fluent synchronous conversation with asynchronous back and forth dialogue can result in a greater possibility of miscommunication. A common result is increased back and forth texting in order to resolve issues. Not only does the lack of detail lead to misunderstanding and additional questioning (back and forth texting), but it also compounds with the inability to quickly self-correct when communicating asynchronously. One nurse admitted that the curtness of her response could lead to miscommunication: “There is a possibility because our tendency is when you send messages, we kind of send the message in a text form as opposed to a paragraph or a story, right. So we might abbreviate it so much that it could be misinterpreted.” Another nurse complained that she receives short responses, albeit faster, but not with enough detail about the context of the problem to resolve the issue at hand: “But it might just be a situation where the patient’s condition has changed or I’m uncertain about a medication and why they want me to give it. Just something that would – it would be helpful to have a bit of back and forth conversation. And often I’ll get just a very brief response only to perhaps the most pertinent part of my message, but I feel that – I don’t really get as detailed a response as I would like.
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 278–284
So although I’m getting responses quicker, it’s not always as detailed as I’d like. It’s still – sometimes the fashion that people text message each other and I just get a yes or no answer and I need more information.” The following description of a scenario involving a prescription for an opioid medication was observed on the GIM ward, and it highlights how face-to-face communication can be vital to avoid miscommunication: • At 4:12 pm, patient talks to and gives Nurse A a prescription. • Nurse A tells Nurse E “I’m going to call down and see what she wants because I don’t understand what she wants”. Nurse A makes a call on phone #2 to outpatient pharmacy, and she asks about an incorrect prescription for narcotics/hydromorphone which was prescribed for 10 days. • Hangs up at 4:16, talks to patient: “I’m waiting for a response/clarification on the prescription”. • She sends a text at 4:19:04: -“Team 6, -Call back requested, -Message: Patient is back. Pharmacy would like amount for the narcotics, not days. He needs a new script” (-Sent at 4:20:22). • Nurse A tells patient “I paged them, let’s see what the answer is”. • At 4:48 (after obtaining new prescription) nurse A holds a paper in hand and says aloud “It is still not correct! where did [the doctor] go?” Looks around. • The resident approaches nurse A, and nurse A says “No! they want the amount, not the day, the amount!”. • [Resident] says “oh I see, I understand now, ok I’ll do that” • Problem is resolved at 4:56, nurse A sends patient away and tells [physician] “This is finally over!” (Field note, January 7, 2010, Toronto General Hospital, 14E). The observer noted that the nurse who was resolving the issue seemed to be getting irritated as she used multiple communication mediums to try to reach the person who had prescribed the drug to inform them that what needed to be specified was the amount of the drug required, not the number of days. Several communication attempts failed to convey this message and only when face-to-face conversation was employed could the issue be clarified. This problem, which was relatively common, took 45 min to resolve and involved numerous back and forth exchanges, yet could have been easily resolved through a short direct dialogue. The following quote highlights how disjointed, asynchronous conversation can actually be challenging for physicians. In addition to keeping track of the patients and their details, physicians now need to start remembering whether or not certain issues were resolved or “closed.” The interviewed physician then compares this to the traditional pager-callback protocol that often resolved issues on the spot. “So if I get a message saying, this is happening on the floor, I will follow up with questions that are not ordered. ‘So what are the latest vitals? Here’s a suggestion of what to do.’ I will not always hear back what was done. So then I have to remember in my head there was this issue pending. I started to create a list on myself of what happened, or I haven’t heard back, when there’s repeat. With phone, there in an instant,
281
close the door. With text, when it’s asynchronous communication, you have to start keeping track of what was closed and what isn’t.” Many clinicians complain that the back and forth texting could more easily and quickly be resolved by actually speaking to one another. Another clinician describes a situation she frequently faces when communicating with physicians: “See another thing I find that with pharmacy it’s kind of like a discussion back and forth when you go over the medication issues. It’s, like, well, why’s so-and-so’s medication on hold? Well, was the heart rate low? Like, what indication is it? With the [text messaging system] you can’t really use that because you can’t have a conversation. It has to be, like, 50 questions going back and forth just to do – versus like that could be done in a minute if you’re talking in person. So that’s another issue.” From our interviews we were also able to identify instances where clinicians became frustrated by the back and forth caused by messages that they themselves agreed were not urgent, but nonetheless required clarification. The following quote is from a registered nurse who expressed her frustration with a communication process that became lengthy due to texting: “Say we need to clarify a normal saline bolus order and sometimes that gets lost back and forth, back and forth. Because you wouldn’t want to call for that because it’s not an urgent issue, but it’s something where you need to be clear. So that’s when it gets stretched out longer than it needs to be.” Many of these problems can be ameliorated through either replacing or supplementing text messaging with verbal communications that can provide much needed context, even if it only adds a tonal aspect to the communication. The following quote from a nurse demonstrates how verbal communication helps to provide context to convey a message, which could not be done similarly through only text-messaging: “If you’re talking to a doctor, he hears the urgency in your voice as well. You can say to the doctor, could you please come now—even if you message “come now,” it doesn’t mean he’ll come now, but again, if you’re actually talking to him, he hears your reaction, he hears your voice and he will respond appropriately.” As evidenced by the above observations, text-based messaging can dilute the quality of communication by diminishing media cues (verbal and non-verbal), and by the inefficient content exchange (shortened message length), resulting in the de-contextualisation of messaging, the oversimplification of complex issues, an increase in interprofessional frustration and potentially can create patient care and safety issues.
3.2. Depersonalization of communication as a barrier to interprofessional teamwork Despite the advantages, such as being place and time independent, text-based communication is often criticized for
282
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 278–284
depersonalizing relations; senders and receivers lack the ability to express tone of voice, facial expression and body language. Social presence theory states that the fewer the number of cue systems (visual, verbal, aural, etc.) per communication medium used, the less interpersonal connection that communicators enact and experience with one another [22]. Our interviews revealed numerous descriptions of such cases where clinicians felt they wanted a more personalized work relationship, and that more personable dialogue would result in improved communication and lead to better quality of care. When asked about the change from meeting in person to using text messaging, one nurse said: “Face-to-face communication, I like to look into people’s face, see their eyes and you see their expression. It’s different. Being on the computer is kind of cold, ‘cause you don’t know whom you’re talking to, for one thing. And you don’t hear their voice. And a lot of meaningful interaction is lost because it’s just the cold, hard thing you’re sending. I’m a people person, so, I mean, I like talking to people.” Other nurses admitted to not knowing as many of the physicians since the adoption of text-based messaging. When using pagers one nurse stated, “we’d actually know who the patients’ doctors were and, you’d actually get to talk to them and build a rapport with them,” in contrast to another nurse’s current experience, “I do not recall even knowing who the resident is.” Depersonalization of the workforce is a phenomenon well described in much of the mediated communication literature that looks at the health of relationships in trans-national work teams [23,24]. We observed similar issues in the clinical environment, where clinicians complained that they did not know their peers and that this affected their day-to-day interactions. The likely effect of depersonalization is a decrease in interprofessional teamwork. Key dimensions of interprofessional teamwork are good communication, efforts to breakdown stereotypes and barriers, and mutual role understanding [25]. Increasing depersonalization and decreasing socialization likely negatively impact these dimensions.
4.
Discussion
Our study identified two main negative effects of textbased communication: the de-contextualization of complex issues and the depersonalization of communication. Decontextualization of complex issues led to an increase in misinterpretation and an increase in further back and forth messaging for clarification. Depersonalization of communication was due to less verbal or face-to-face interactions and led to a negative impact on work relationships. There are a small number of studies looking at the impact of text messages on the clinical environment. Iversen et al. analyzed one month’s worth of instant messages sent through a perioperative coordination and communication system at a Danish hospital and found that instant messaging was used extensively for the coordination and logistics of patient care; messages were kept short, and it was not used for complex issues [26]. Text messaging has also been found to create gaps in what nurses and physicians perceive to be urgent clinical
issues [27]. These gaps appear to be due to a lack of conveying the context of the message. Urgency due to nursing change of shifts and anxious family members are examples of some conditions associated with discordant prioritizations between inter-professional staff. Previous communication literature describes the misunderstandings that can arise with the loss of non-verbal components associated with solely text based communication [28,29]. The two most basic non-verbal features of face-to-face conversation are (1) the collaborative commitment of participants and the co-formulation of the message and (2) the feedback which allows the social meaning of the message to be processed immediately [11]. These aspects are missing in CMC, and studies have shown that the lack of these features reveal significant differences in the degrees of social presence and media richness. Social presence is the user’s perception of the ability of the means of communication to marshal and focus the presence of communicating subjects, while media richness is the ability of the means of communication to interlink a variety of topics, render them less ambiguous, and enable users to learn about them within a given time-span [21]. Both these features are desirable for effective communication [11], however our study identified them as lacking among clinicians communicating with text-based messaging systems. The inability to have synchronous feedback increases the amount of written messages exchanged (back and forth messages) and lengthens the time for resolving issues. The reduction of social presence and media richness that accompanies text-based messaging depersonalizes communication, and therefore adds tension to already strained interprofessional relationships. In addition to lack of non-verbal cues, text-based communication differs fundamentally in how users package and send information. The examples highlighted in the section discussing the reduction of complex issues are well described by relevance theory, which posits that in contrast to the traditional conduit model of communication, the sender says just enough to communicate what they intend, relying on the audience to fill in the details that they did not explicitly communicate [30]. This alternate way of conceiving how thoughts are communicated explains why text messages are often decoded ineffectively when they are simplified and contain too much implicit information. It also helps explain the differing sender and receiver expectations of text message length. The importance of these findings is that with the shifting of communication from verbal to text messages, there appears to be predictable negative effects such as increased frustration and decreased interprofessional relationships. In order to reduce these effects, we recommend efforts from clinicians and hospital administration to address the issues. Clinicians need to collectively come to an agreement on norms and protocols for efficient communication. It is important to determine which messages are relevant and appropriate to send via text versus other means of communication. Until norms for message importance, urgency, and relevance are co-created and agreed upon both by senders and receivers, receivers will continue to feel as though they are being unnecessarily interrupted, and senders will continue to think their messages are given little importance and
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 278–284
are going unread, ultimately deteriorating interprofessional relations [27]. Furthermore, clinicians should be familiar with rules of netiquette, the set of social conventions and norms that facilitate effective and inoffensive interaction over mediated networks. Physicians should promote mutual respect, and should not undermine the importance of verbal communication. CMC may not always be ideal for every situation; the clinical environment may dictate that text-based communication is suitable for simple issues, whereas talking over the phone or face-to-face is required when more detailed or complex responses are desired. The ability and benefits of switching media when required to conduct more complex clinical discussions has also previously been described [26]. Clinicians and hospital administration should demand better health information technology systems that support clinical communication. With the rapid advancement of personal communication systems such as smartphones and social media, we hope to see similar clinical communication systems that can help clinicians communicate with each other effectively and efficiently while fostering positive interprofessional relationships. These systems could aid in providing additional context and non-text based information to aid in communication. Similar to social media platforms, we could imagine that an improved clinical communication system would have multiple communication methods. This could include the use of voice messaging, push-to-talk calls, video messaging, and video calls for more complex issues all of which, if functioning properly, have increased media richness and social presence over text messages.
5.
Limitations
There were limitations to this study. Most notably, observations were drawn from a larger study of addressing general communication on the General Internal Medicine wards, and were not focused on text-based messaging specifically. The study was also conducted at five academic hospital sites; thus, generalizing to other institutions with different hospital cultures may yield different results. However, the intervention we analyzed used standard protocol for pagers, smartphones, and intranet-based messaging systems, and other academic hospitals may be able to learn from this experience. Future studies could look at the impact of such text-based communication when proper training is provided to all clinical staff.
6.
Conclusions
This paper describes the unintended consequences of moving from verbal communication to text-based communication in a hospital setting. While routine adoption of smartphones by physicians and their use of text-based messaging appear to improve efficiency, it came with some disadvantages; decontextualization of complex issues and depersonalization of communication. In order to mitigate some of these issues we suggest that clinicians need to collaboratively establish norms of communication and that better communication systems are required.
283
Summary points What was already known on this topic? • Interprofessional clinical communication affects the quality of patient care. • There is a shift in hospitals to using text-based messaging systems of communication What this study added to our knowledge? • There are unintended consequences to using textbased messaging systems of communication: decontextualization may increase workload and frustration and depersonalization may reduce interprofessional collaboration. • Adoption of health information technology such as text-based messaging creates unintended consequences. • By removing context and important details, text-based messaging systems may increase workload and frustration. • By reducing socialization, text-based messaging may increase depersonalization and may reduce interprofessional collaboration.
Authors’ contributions RW and SQ conceived the idea of the paper. VL performed the data collection activities. RW, LA and VL conducted the analysis. LA created the first draft. RW, DM, LA, VL, SQ, and SK contributed to critically revising the manuscript and approved the final version.
Conflicts of interest The authors declare that they have no competing interests.
Funding The authors would like to thank the following for their financial support towards this project: Alternate Funding Plans for Academic Health Science Centres (AFP-AHSC) created by Ontario Medical Association and the Ministry of Health and Long-Term Care; and the Department of Medicine at the University of Toronto. These sponsors provide unrestricted funds and have no role in the study design, data collection and analysis or manuscript write up.
Acknowledgement The authors would like to thank all the clinicians who participated in the study.
references
[1] R.C. Wu, K. Tran, V. Lo, K.J. O’Leary, D. Morra, S.D. Quan, L. Perrier, Effects of clinical communication interventions in
284
[2] [3] [4] [5]
[6]
[7]
[8]
[9]
[10] [11]
[12]
[13]
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 278–284
hospitals: a systematic review of information and communication technology adoptions for improved communication between clinicians, Int. J. Med. Inform. 81 (Nov (11)) (2012) 723–732. Industry Survey Results: Critical Healthcare Messaging on Smartphones & Tablets Amcom Software, 2011. B. Runyon, J. Lovelock, Plot a proactive path from the pager to the smartphone, Gartner Inc., Stamford, CT, USA, 2012. Poor communication is common cause of errors, Healthcare Benchmarks Qual. Improv. 1 (August (2)) (2002) 18–19. L. Varpio, P. Hall, L. Lingard, C.F. Schryer, Interprofessional communication and medical error: a reframing of research questions and approaches, Acad. Med. 83 (October (10 Suppl.)) (2008) S76–S81. M. Bloomrosen, J. Starren, N.M. Lorenzi, J.S. Ash, V.L. Patel, E.H. Shortliffe, Anticipating and addressing the unintended consequences of health IT and policy: a report from the AMIA 2009 Health Policy Meeting, J. Am. Med. Inform. Assoc. 18 (January (1)) (2011) 82–90. J.S. Adelman, G.E. Kalkut, C.B. Schechter, J.M. Weiss, M.A. Berger, S.H. Reissman, H.W. Cohen, S.J. Lorenzen, D.A. Burack, W.N. Southern, Understanding and preventing wrong-patient electronic orders: a randomized controlled trial, J. Am. Med. Inform. Assoc. 20 (March (2)) (2013) 305–310. R.C. Wu, V. Lo, D. Morra, B.M. Wong, R. Sargeant, K. Locke, R. Cavalcanti, S.D. Quan, P. Rossos, K. Tran, et al., The intended and unintended consequences of communication systems on general internal medicine inpatient care delivery: a prospective observational case study of five teaching hospitals, J. Am. Med. Inform. Assoc. 20 (Jul-Aug (4)) (2013) 766–777. N.K. Baym, The performance of humor in computer-mediated communication, J. Comput. Mediat. Commun. 1 (2) (1995). Sproull l, S. Kiesler, Don’t forget Dr. Frankenstein, Computerworld 26 (5) (1992) 25. G. Riva, Say not to Say, in: L. Anolli, R. Ciceri, G. Riva (Eds.), Say not to Say: New persectives on Miscommunication, IOS Press, Milan, 2002. C. Sanger, A. Taylor, J. Vincent, An SMS History, in: L. Hamill, A. Lasen, D. Diaper (Eds.), Mobile World, Springer, London, 2005, pp. 75–91. R. Wu, P. Rossos, S. Quan, S. Reeves, V. Lo, B. Wong, M. Cheung, D. Morra, An evaluation of the use of smartphones to communicate between clinicians: a mixed-methods study, J. Med. Internet Res. 13 (3) (2011) e59.
[14] R.C. Wu, D. Morra, S. Quan, S. Lai, S. Zanjani, H. Abrams, P.G. Rossos, The use of smartphones for clinical communication on internal medicine wards, J. Hosp. Med. 5 (Nov-Dec (9)) (2010) 553–559. [15] R.K. Yin, Case Study Research, Design and Methods, Sage Publications, Newbury Park, 2002. [16] B. Flyvbjerg, Five misunderstandings about case-study research, Qual. Inquiry 12 (2) (2006) 219–245. [17] S.C. Kitto, J. Chesters, C. Grbich, Quality in qualitative research, Med. J. Aust. 188 (Febuary (4)) (2008) 243–246. [18] P.L. Rice, D. Ezzy, Qualitative Research Methods: A Health Focus, Oxford University Press, Melbourne, 1999. [19] H.F. Hsieh, S.E. Shannon, Three approaches to qualitative content analysis, Qual. Health Res. 15 (November (9)) (2005) 1277–1288. [20] U. Flick, Triangulation revisited: strategy of validation or alternative? J. Theor. Soc. Behav. 22 (2) (1992) 175–197. [21] E. Coiera, When conversation is better than computation, J. Am. Med. Inform. Assoc. 7 (May (3)) (2000) 277–286. [22] J.A. Short, J.A. Williams, B. Christie, The Social Psychology of Telecommunications, John Wiley & Sons, New York, 1976. [23] G. Hertel, S. Geister, U. Konradt, Managing virtual teams: a review of current empirical research, Hum. Resour. Manage. Rev. 15 (1) (2005) 69–95. [24] D.L. Shapiro, S.A. Furst, G.M. Spreitzer, M.A. Von Glinow, Transnational teams in the electronic age: are team identity and high performance at risk? J. Org. Behav. 23 (4) (2002) 455–467. [25] S. Reeves, S. Lewin, S. Espin, M. Zwarenstein, Interprofessional Teamword for Health and Social Care, Wiley-Blackwell, West Sussex, 2010. [26] T.B. Iversen, L. Melby, P. Toussaint, Instant messaging at the hospital: supporting articulation work? Int. J. Med. Inform. 82 (September (9)) (2013) 753–761. [27] S.D. Quan, D. Morra, F.Y. Lau, W. Coke, B.M. Wong, R.C. Wu, P.G. Rossos, Perceptions of urgency: defining the gap between what physicians and nurses perceive to be an urgent issue, Int. J. Med. Inform. 82 (May (5)) (2013) 378–386. [28] E. Williams, Experimental comparisons of face-to-face and mediated communication: a review, Psychol. Bull. 84 (5) (1977) 963–976. [29] S. Kiesler, J. Siegel, T.W. McGuire, Social psychological aspects of computer-mediated communication, Am. Psychol. 39 (10) (1984) 1123–1134. [30] D. Wilson, D. Sperber, Relevance theory, in: L.R. Horn, G. Ward (Eds.), The Handbook of Pragmatics, Blackwell Science, Oxford, 2004, pp. 607–632.
journal homepage: www.ijmijournal.com
Short message service or disService: Issues with text messaging in a complex medical environment Robert Wu a,b,∗ , Lora Appel a,c , Dante Morra d , Vivian Lo a , Simon Kitto e,f,g , Sherman Quan d a
Centre of Innovation in Complex Care, University Health Network, Toronto, ON, Canada Department of General Internal Medicine, University Health Network, Toronto, ON, Canada c Department of Communication and Information, Rutgers University, New Brunswick, NJ, USA d Trillium Health Partners, Mississauga, ON, Canada e Office of Continuing Education and Professional Development, University of Toronto, ON, Canada f Li Ka-Shing Knowledge Institute of St. Michael’s Hospital, Toronto, ON, Canada g Department of Surgery, University of Toronto, Toronto, ON, Canada b
a r t i c l e
i n f o
a b s t r a c t
Article history:
Background: Hospitals today are experiencing major changes in their clinical communication
Received 10 June 2013
workflows as conventional numeric paging and face-to-face verbal conversations are being
Received in revised form
replaced by computer mediated communication systems. In this paper, we highlight the
20 December 2013
importance of understanding this transition and discuss some of the impacts that may
Accepted 10 January 2014
emerge when verbal clinical conversations are replaced by short text messages.
Keywords:
were conducted on the General Internal Medicine wards at five academic teaching hospitals
Hospital communication system
in Toronto, Canada.
Methods: In-depth interviews (n = 108) and non-participatory observation sessions (n = 260 h)
Results: From our analysis of the qualitative data, we identified two major themes. De-
Interprofessional communication
contextualization of complex issues led to an increase in misinterpretation and an increase
Smartphones
in back and forth messaging for clarification. Depersonalization of communication was due to less verbal conversations and face-to-face interactions and led to a negative impact on work relationships. Conclusions: Text-based communication in hospital settings led to the oversimplification of messages and the depersonalization of communication. It is important to recognize and understand these unintended consequences of new technology to avoid the negative impacts to patient care and work relationships. © 2014 Elsevier Ireland Ltd. All rights reserved.
1.
Background
Smartphones are increasingly being adopted as a means of communication within hospitals [1–3]. Smartphones have the ability to support both voice and text messaging, and they
create a platform for asynchronous communication in addition to providing access to medical applications and the Internet. For these reasons, experts believe that smartphones may not only serve as efficient communication tools but may also improve the coordination of patient care in the clinical setting [4].
∗ Corresponding author at: 200 Elizabeth Street, 14EN222, Toronto General Hospital, Toronto, ON, Canada M5G 2C4. Tel.: +1 416 340 4567; fax: +1 416 595 5826. E-mail address: [email protected] (R. Wu). 1386-5056/$ – see front matter © 2014 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijmedinf.2014.01.003
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 278–284
However, there may be issues with their use within hospitals where the complexity of medical work is compounded with a tradition of professional fragmentation and a diffusion of accountability [5]. There is increasing recognition that the implementation of information technology in the complex health care environment can result in unintended consequences with increased errors. Increased errors have been seen with computerized physician order entry, clinical decision support systems, and barcode medication administration [6,7], and therefore, it is important to identify the unintended consequences of replacing pagers with smartphones in order to mitigate them. The use of smartphones is moving conversations that were traditionally face-to-face or by telephone to a more text-based communication mode [8]. Such computer-mediated communication (CMC) has known effects and has been described previously in literature external to healthcare. CMC is a term used to describe the broad concept of communication, the transfer of information, via any technological means. This includes telephone calls, Skype calls, short message service, and instant messaging among many others. Of the CMC mediums, text-based is the least media rich, after voice-only and voice and video communication. Although CMC is similar to conversation by being “interactive, relatively spontaneous and generally unplanned,” [9] it changes the way people connect [10]. Although CMC affords users the ability to alleviate barriers in space and time, it is sometimes described as a rarefied form of conversation because it does not always support meta-communicative features like facial expression, posture and tone of voice which add richness to communication [11]. As opposed to other forms of text-based communication such as email, text messaging dilutes social presence and media richness further as communication is reduced to a short message, typically less than 140 characters [12]. In a clinical environment, text messaging can be especially limiting when complex information is conveyed with only a few characters and can lead to miscommunication between senders and receivers. This can damage an already sensitive interprofessional relationship [13]. Recent research on CMC in hospitals, and specifically the use of smartphones and texting between clinicians, has shown improvements in efficiency, but it is important that we also understand the unintended consequences [13,14]. Textbased messaging can also lead to ineffective communication and deteriorate interprofessional relations. In order to understand the impact of text-based communication, we examined
279
this aspect through the lens of computer-mediated communication.
2.
Methods
2.1.
Study design
To provide a more comprehensive understanding of clinicians’ experiences with the communication technologies in the hospital settings, we adopted an exploratory case study approach [15,16] consisting of data triangulation from multiple sources that included (a) semi-structured interviews and (b) non-participatory ward observation methods. This dataset is part of a larger multi-site study that evaluated the impact of different types of communication interventions and initiatives in hospital settings [8]. The original study was conducted on the General Internal Medicine (GIM) wards at five academic teaching hospitals that are affiliated with the University of Toronto, Canada. In this paper, we explore the implications of moving towards text-based communications on interprofessional relationships among medical teams and clinicians.
2.2.
Setting
Each participating hospital has two to four GIM wards that are located on different floors within the respective medical facilities. Each site has clinical teaching units with typically four medical teams, each consisting of an attending physician, a senior resident, junior residents and medical students. Four of the five hospitals have adopted text-based communication channels as part of clinicians’ routine workflow (Table 1).
2.3.
Data collection
Qualitative data consisting of in-depth interviews and nonparticipatory ward observations were collected from June 2009 to September 2010 across the five hospital sites. Ethical approval was obtained from the respective institutions’ Research Ethics Board committees.
2.3.1.
Interviews
Participants were selected and solicited through purposeful and criterion-based sampling techniques to ensure a variety of appropriate data sources were obtained. The criteria for recruitment were that participants had to be practicing
Table 1 – Different communication devices and methods adopted at the five hospitals. Site
Communication method
1
Numeric pages
2
Alphanumeric pages, Smartphones
3.4
Smartphones, intranet-based messaging system Intranet-based task-management messaging system that queued non-urgent messages.
5
Typical communication process Nurses page physicians, and physicians respond by calling back to a telephone number. Nurses send text messages to physicians’ pagers. If necessary, physicians would call back using their smartphones and/or landline phones. Nurses use the intranet-based messaging system to send messages to physicians’ smartphones. Physicians respond by email or call back using the smartphones. Nurses enter a text-based message using the intranet-based messaging system, and physicians are notified on their pagers of a new message, which can be read on the same system. Physicians reply using the task-management messaging system.
280
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 278–284
clinicians who have experiences using the communication technologies and modalities that were implemented at their designated hospital site and wards. We also employed a maximum variation sampling approach by identifying participants from different clinical roles that included attending physicians, residents, nurses, medical students and allied health personnel [17]. We adopted a confirming/disconfirming strategy in which we asked the interviewees to recommend potential participants who might: (a) have a similar experience/viewpoint to their own; and/or (b) have a divergent experience/viewpoint from their own. A total of 108 interviews were conducted by a member of the research team and audio taped at all the five sites. Each interview lasted between 15 and 45 min and was carried out at a mutually convenient location within the respective sites. The interview protocol consisted of a series of open-ended questions with appropriate follow-up probes that explored the impact of different communication interventions (smartphone, numeric paging, and texting) on inpatient care delivery, workflow, and clinician relationships. The interviews were then professionally transcribed and data were entered into a qualitative analysis software tool (NVivo 8, QSR International) to facilitate coding and analysis.
2.3.2.
Observations
The communication processes in the hospitals were observed by documenting clinicians’ communication activities and interactions at the General Internal Medicine nursing stations – the hubs of many clinical communication exchanges. Using non-participatory observation, descriptions of communication events and issues were chronicled and time-stamped on field notes which were transcribed into Microsoft Word® documents [18]. No patient-related information was collected. Ward observations were conducted on both weekdays and weekends in 2-h time intervals across various timeslots. One member of the research team collected the data on sites 3 and 4 and trained another research assistant to collect data from sites 1, 2, and 5. A total of 131 observation sessions (n = 260 h) of the GIM nursing stations were recorded across all the five sites.
2.4.
Analysis
Data triangulation using the conventional thematic content analysis approach was undertaken to draw inferences about clinicians’ experiences around the use of text-based communication channels [19,20]. An initial sample of the interview transcripts and field notes from the ward observations were independently read and coded by two members of the research team to identify broad themes across all the sites. The coding process involved identifying descriptive categories in which the clinicians’ responses, communication events, and incidents would fit. The provisional thematic categories were discussed with the rest of the research team and then organized into a preliminary structure for coding the rest of the data with additional themes reported for each site as they emerged. Upon completion of the coding, the categories were again reviewed and refined.
3.
Results
In this study we identified two main themes regarding the unintended consequences of text-based communication: (1) the de-contextualization of complex issues leading to misinterpretation and increasing communication workload and (2) the depersonalization of communication as a barrier to interprofessional teamwork.
3.1. De-contextualization of complex issues leading to misinterpretation and increasing communication workload While the ability to communicate asynchronously can be very efficient [21], it can also lead to frustration in a hospital environment that can ultimately be detrimental to the optimal delivery of care. In the past, nurses would page physicians to alert them of a change in status with a patient. Physicians would then call back and have a conversation over the phone, or come to the nursing station in person and have a face-to-face conversation with the nurse. In these cases, both parties communicating have the ability to, in a short amount of time, exchange many back and forth utterances of speech. When dialogue is composed synchronously, people can more easily detect and correct for misinterpretations and misunderstandings in communication. This can be done by requesting additional detail, by repeating or stressing words to ensure the message is delivered, or by interpreting non-verbal cues, such as tone of voice, that can help convey the urgency of a situation. With text-based communication, messages are often reduced in length, sometimes to only a few characters, and this simplification and de-contextualization of content creates the possibility for misinterpretation. Replacing fluent synchronous conversation with asynchronous back and forth dialogue can result in a greater possibility of miscommunication. A common result is increased back and forth texting in order to resolve issues. Not only does the lack of detail lead to misunderstanding and additional questioning (back and forth texting), but it also compounds with the inability to quickly self-correct when communicating asynchronously. One nurse admitted that the curtness of her response could lead to miscommunication: “There is a possibility because our tendency is when you send messages, we kind of send the message in a text form as opposed to a paragraph or a story, right. So we might abbreviate it so much that it could be misinterpreted.” Another nurse complained that she receives short responses, albeit faster, but not with enough detail about the context of the problem to resolve the issue at hand: “But it might just be a situation where the patient’s condition has changed or I’m uncertain about a medication and why they want me to give it. Just something that would – it would be helpful to have a bit of back and forth conversation. And often I’ll get just a very brief response only to perhaps the most pertinent part of my message, but I feel that – I don’t really get as detailed a response as I would like.
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 278–284
So although I’m getting responses quicker, it’s not always as detailed as I’d like. It’s still – sometimes the fashion that people text message each other and I just get a yes or no answer and I need more information.” The following description of a scenario involving a prescription for an opioid medication was observed on the GIM ward, and it highlights how face-to-face communication can be vital to avoid miscommunication: • At 4:12 pm, patient talks to and gives Nurse A a prescription. • Nurse A tells Nurse E “I’m going to call down and see what she wants because I don’t understand what she wants”. Nurse A makes a call on phone #2 to outpatient pharmacy, and she asks about an incorrect prescription for narcotics/hydromorphone which was prescribed for 10 days. • Hangs up at 4:16, talks to patient: “I’m waiting for a response/clarification on the prescription”. • She sends a text at 4:19:04: -“Team 6, -Call back requested, -Message: Patient is back. Pharmacy would like amount for the narcotics, not days. He needs a new script” (-Sent at 4:20:22). • Nurse A tells patient “I paged them, let’s see what the answer is”. • At 4:48 (after obtaining new prescription) nurse A holds a paper in hand and says aloud “It is still not correct! where did [the doctor] go?” Looks around. • The resident approaches nurse A, and nurse A says “No! they want the amount, not the day, the amount!”. • [Resident] says “oh I see, I understand now, ok I’ll do that” • Problem is resolved at 4:56, nurse A sends patient away and tells [physician] “This is finally over!” (Field note, January 7, 2010, Toronto General Hospital, 14E). The observer noted that the nurse who was resolving the issue seemed to be getting irritated as she used multiple communication mediums to try to reach the person who had prescribed the drug to inform them that what needed to be specified was the amount of the drug required, not the number of days. Several communication attempts failed to convey this message and only when face-to-face conversation was employed could the issue be clarified. This problem, which was relatively common, took 45 min to resolve and involved numerous back and forth exchanges, yet could have been easily resolved through a short direct dialogue. The following quote highlights how disjointed, asynchronous conversation can actually be challenging for physicians. In addition to keeping track of the patients and their details, physicians now need to start remembering whether or not certain issues were resolved or “closed.” The interviewed physician then compares this to the traditional pager-callback protocol that often resolved issues on the spot. “So if I get a message saying, this is happening on the floor, I will follow up with questions that are not ordered. ‘So what are the latest vitals? Here’s a suggestion of what to do.’ I will not always hear back what was done. So then I have to remember in my head there was this issue pending. I started to create a list on myself of what happened, or I haven’t heard back, when there’s repeat. With phone, there in an instant,
281
close the door. With text, when it’s asynchronous communication, you have to start keeping track of what was closed and what isn’t.” Many clinicians complain that the back and forth texting could more easily and quickly be resolved by actually speaking to one another. Another clinician describes a situation she frequently faces when communicating with physicians: “See another thing I find that with pharmacy it’s kind of like a discussion back and forth when you go over the medication issues. It’s, like, well, why’s so-and-so’s medication on hold? Well, was the heart rate low? Like, what indication is it? With the [text messaging system] you can’t really use that because you can’t have a conversation. It has to be, like, 50 questions going back and forth just to do – versus like that could be done in a minute if you’re talking in person. So that’s another issue.” From our interviews we were also able to identify instances where clinicians became frustrated by the back and forth caused by messages that they themselves agreed were not urgent, but nonetheless required clarification. The following quote is from a registered nurse who expressed her frustration with a communication process that became lengthy due to texting: “Say we need to clarify a normal saline bolus order and sometimes that gets lost back and forth, back and forth. Because you wouldn’t want to call for that because it’s not an urgent issue, but it’s something where you need to be clear. So that’s when it gets stretched out longer than it needs to be.” Many of these problems can be ameliorated through either replacing or supplementing text messaging with verbal communications that can provide much needed context, even if it only adds a tonal aspect to the communication. The following quote from a nurse demonstrates how verbal communication helps to provide context to convey a message, which could not be done similarly through only text-messaging: “If you’re talking to a doctor, he hears the urgency in your voice as well. You can say to the doctor, could you please come now—even if you message “come now,” it doesn’t mean he’ll come now, but again, if you’re actually talking to him, he hears your reaction, he hears your voice and he will respond appropriately.” As evidenced by the above observations, text-based messaging can dilute the quality of communication by diminishing media cues (verbal and non-verbal), and by the inefficient content exchange (shortened message length), resulting in the de-contextualisation of messaging, the oversimplification of complex issues, an increase in interprofessional frustration and potentially can create patient care and safety issues.
3.2. Depersonalization of communication as a barrier to interprofessional teamwork Despite the advantages, such as being place and time independent, text-based communication is often criticized for
282
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 278–284
depersonalizing relations; senders and receivers lack the ability to express tone of voice, facial expression and body language. Social presence theory states that the fewer the number of cue systems (visual, verbal, aural, etc.) per communication medium used, the less interpersonal connection that communicators enact and experience with one another [22]. Our interviews revealed numerous descriptions of such cases where clinicians felt they wanted a more personalized work relationship, and that more personable dialogue would result in improved communication and lead to better quality of care. When asked about the change from meeting in person to using text messaging, one nurse said: “Face-to-face communication, I like to look into people’s face, see their eyes and you see their expression. It’s different. Being on the computer is kind of cold, ‘cause you don’t know whom you’re talking to, for one thing. And you don’t hear their voice. And a lot of meaningful interaction is lost because it’s just the cold, hard thing you’re sending. I’m a people person, so, I mean, I like talking to people.” Other nurses admitted to not knowing as many of the physicians since the adoption of text-based messaging. When using pagers one nurse stated, “we’d actually know who the patients’ doctors were and, you’d actually get to talk to them and build a rapport with them,” in contrast to another nurse’s current experience, “I do not recall even knowing who the resident is.” Depersonalization of the workforce is a phenomenon well described in much of the mediated communication literature that looks at the health of relationships in trans-national work teams [23,24]. We observed similar issues in the clinical environment, where clinicians complained that they did not know their peers and that this affected their day-to-day interactions. The likely effect of depersonalization is a decrease in interprofessional teamwork. Key dimensions of interprofessional teamwork are good communication, efforts to breakdown stereotypes and barriers, and mutual role understanding [25]. Increasing depersonalization and decreasing socialization likely negatively impact these dimensions.
4.
Discussion
Our study identified two main negative effects of textbased communication: the de-contextualization of complex issues and the depersonalization of communication. Decontextualization of complex issues led to an increase in misinterpretation and an increase in further back and forth messaging for clarification. Depersonalization of communication was due to less verbal or face-to-face interactions and led to a negative impact on work relationships. There are a small number of studies looking at the impact of text messages on the clinical environment. Iversen et al. analyzed one month’s worth of instant messages sent through a perioperative coordination and communication system at a Danish hospital and found that instant messaging was used extensively for the coordination and logistics of patient care; messages were kept short, and it was not used for complex issues [26]. Text messaging has also been found to create gaps in what nurses and physicians perceive to be urgent clinical
issues [27]. These gaps appear to be due to a lack of conveying the context of the message. Urgency due to nursing change of shifts and anxious family members are examples of some conditions associated with discordant prioritizations between inter-professional staff. Previous communication literature describes the misunderstandings that can arise with the loss of non-verbal components associated with solely text based communication [28,29]. The two most basic non-verbal features of face-to-face conversation are (1) the collaborative commitment of participants and the co-formulation of the message and (2) the feedback which allows the social meaning of the message to be processed immediately [11]. These aspects are missing in CMC, and studies have shown that the lack of these features reveal significant differences in the degrees of social presence and media richness. Social presence is the user’s perception of the ability of the means of communication to marshal and focus the presence of communicating subjects, while media richness is the ability of the means of communication to interlink a variety of topics, render them less ambiguous, and enable users to learn about them within a given time-span [21]. Both these features are desirable for effective communication [11], however our study identified them as lacking among clinicians communicating with text-based messaging systems. The inability to have synchronous feedback increases the amount of written messages exchanged (back and forth messages) and lengthens the time for resolving issues. The reduction of social presence and media richness that accompanies text-based messaging depersonalizes communication, and therefore adds tension to already strained interprofessional relationships. In addition to lack of non-verbal cues, text-based communication differs fundamentally in how users package and send information. The examples highlighted in the section discussing the reduction of complex issues are well described by relevance theory, which posits that in contrast to the traditional conduit model of communication, the sender says just enough to communicate what they intend, relying on the audience to fill in the details that they did not explicitly communicate [30]. This alternate way of conceiving how thoughts are communicated explains why text messages are often decoded ineffectively when they are simplified and contain too much implicit information. It also helps explain the differing sender and receiver expectations of text message length. The importance of these findings is that with the shifting of communication from verbal to text messages, there appears to be predictable negative effects such as increased frustration and decreased interprofessional relationships. In order to reduce these effects, we recommend efforts from clinicians and hospital administration to address the issues. Clinicians need to collectively come to an agreement on norms and protocols for efficient communication. It is important to determine which messages are relevant and appropriate to send via text versus other means of communication. Until norms for message importance, urgency, and relevance are co-created and agreed upon both by senders and receivers, receivers will continue to feel as though they are being unnecessarily interrupted, and senders will continue to think their messages are given little importance and
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 278–284
are going unread, ultimately deteriorating interprofessional relations [27]. Furthermore, clinicians should be familiar with rules of netiquette, the set of social conventions and norms that facilitate effective and inoffensive interaction over mediated networks. Physicians should promote mutual respect, and should not undermine the importance of verbal communication. CMC may not always be ideal for every situation; the clinical environment may dictate that text-based communication is suitable for simple issues, whereas talking over the phone or face-to-face is required when more detailed or complex responses are desired. The ability and benefits of switching media when required to conduct more complex clinical discussions has also previously been described [26]. Clinicians and hospital administration should demand better health information technology systems that support clinical communication. With the rapid advancement of personal communication systems such as smartphones and social media, we hope to see similar clinical communication systems that can help clinicians communicate with each other effectively and efficiently while fostering positive interprofessional relationships. These systems could aid in providing additional context and non-text based information to aid in communication. Similar to social media platforms, we could imagine that an improved clinical communication system would have multiple communication methods. This could include the use of voice messaging, push-to-talk calls, video messaging, and video calls for more complex issues all of which, if functioning properly, have increased media richness and social presence over text messages.
5.
Limitations
There were limitations to this study. Most notably, observations were drawn from a larger study of addressing general communication on the General Internal Medicine wards, and were not focused on text-based messaging specifically. The study was also conducted at five academic hospital sites; thus, generalizing to other institutions with different hospital cultures may yield different results. However, the intervention we analyzed used standard protocol for pagers, smartphones, and intranet-based messaging systems, and other academic hospitals may be able to learn from this experience. Future studies could look at the impact of such text-based communication when proper training is provided to all clinical staff.
6.
Conclusions
This paper describes the unintended consequences of moving from verbal communication to text-based communication in a hospital setting. While routine adoption of smartphones by physicians and their use of text-based messaging appear to improve efficiency, it came with some disadvantages; decontextualization of complex issues and depersonalization of communication. In order to mitigate some of these issues we suggest that clinicians need to collaboratively establish norms of communication and that better communication systems are required.
283
Summary points What was already known on this topic? • Interprofessional clinical communication affects the quality of patient care. • There is a shift in hospitals to using text-based messaging systems of communication What this study added to our knowledge? • There are unintended consequences to using textbased messaging systems of communication: decontextualization may increase workload and frustration and depersonalization may reduce interprofessional collaboration. • Adoption of health information technology such as text-based messaging creates unintended consequences. • By removing context and important details, text-based messaging systems may increase workload and frustration. • By reducing socialization, text-based messaging may increase depersonalization and may reduce interprofessional collaboration.
Authors’ contributions RW and SQ conceived the idea of the paper. VL performed the data collection activities. RW, LA and VL conducted the analysis. LA created the first draft. RW, DM, LA, VL, SQ, and SK contributed to critically revising the manuscript and approved the final version.
Conflicts of interest The authors declare that they have no competing interests.
Funding The authors would like to thank the following for their financial support towards this project: Alternate Funding Plans for Academic Health Science Centres (AFP-AHSC) created by Ontario Medical Association and the Ministry of Health and Long-Term Care; and the Department of Medicine at the University of Toronto. These sponsors provide unrestricted funds and have no role in the study design, data collection and analysis or manuscript write up.
Acknowledgement The authors would like to thank all the clinicians who participated in the study.
references
[1] R.C. Wu, K. Tran, V. Lo, K.J. O’Leary, D. Morra, S.D. Quan, L. Perrier, Effects of clinical communication interventions in
284
[2] [3] [4] [5]
[6]
[7]
[8]
[9]
[10] [11]
[12]
[13]
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 278–284
hospitals: a systematic review of information and communication technology adoptions for improved communication between clinicians, Int. J. Med. Inform. 81 (Nov (11)) (2012) 723–732. Industry Survey Results: Critical Healthcare Messaging on Smartphones & Tablets Amcom Software, 2011. B. Runyon, J. Lovelock, Plot a proactive path from the pager to the smartphone, Gartner Inc., Stamford, CT, USA, 2012. Poor communication is common cause of errors, Healthcare Benchmarks Qual. Improv. 1 (August (2)) (2002) 18–19. L. Varpio, P. Hall, L. Lingard, C.F. Schryer, Interprofessional communication and medical error: a reframing of research questions and approaches, Acad. Med. 83 (October (10 Suppl.)) (2008) S76–S81. M. Bloomrosen, J. Starren, N.M. Lorenzi, J.S. Ash, V.L. Patel, E.H. Shortliffe, Anticipating and addressing the unintended consequences of health IT and policy: a report from the AMIA 2009 Health Policy Meeting, J. Am. Med. Inform. Assoc. 18 (January (1)) (2011) 82–90. J.S. Adelman, G.E. Kalkut, C.B. Schechter, J.M. Weiss, M.A. Berger, S.H. Reissman, H.W. Cohen, S.J. Lorenzen, D.A. Burack, W.N. Southern, Understanding and preventing wrong-patient electronic orders: a randomized controlled trial, J. Am. Med. Inform. Assoc. 20 (March (2)) (2013) 305–310. R.C. Wu, V. Lo, D. Morra, B.M. Wong, R. Sargeant, K. Locke, R. Cavalcanti, S.D. Quan, P. Rossos, K. Tran, et al., The intended and unintended consequences of communication systems on general internal medicine inpatient care delivery: a prospective observational case study of five teaching hospitals, J. Am. Med. Inform. Assoc. 20 (Jul-Aug (4)) (2013) 766–777. N.K. Baym, The performance of humor in computer-mediated communication, J. Comput. Mediat. Commun. 1 (2) (1995). Sproull l, S. Kiesler, Don’t forget Dr. Frankenstein, Computerworld 26 (5) (1992) 25. G. Riva, Say not to Say, in: L. Anolli, R. Ciceri, G. Riva (Eds.), Say not to Say: New persectives on Miscommunication, IOS Press, Milan, 2002. C. Sanger, A. Taylor, J. Vincent, An SMS History, in: L. Hamill, A. Lasen, D. Diaper (Eds.), Mobile World, Springer, London, 2005, pp. 75–91. R. Wu, P. Rossos, S. Quan, S. Reeves, V. Lo, B. Wong, M. Cheung, D. Morra, An evaluation of the use of smartphones to communicate between clinicians: a mixed-methods study, J. Med. Internet Res. 13 (3) (2011) e59.
[14] R.C. Wu, D. Morra, S. Quan, S. Lai, S. Zanjani, H. Abrams, P.G. Rossos, The use of smartphones for clinical communication on internal medicine wards, J. Hosp. Med. 5 (Nov-Dec (9)) (2010) 553–559. [15] R.K. Yin, Case Study Research, Design and Methods, Sage Publications, Newbury Park, 2002. [16] B. Flyvbjerg, Five misunderstandings about case-study research, Qual. Inquiry 12 (2) (2006) 219–245. [17] S.C. Kitto, J. Chesters, C. Grbich, Quality in qualitative research, Med. J. Aust. 188 (Febuary (4)) (2008) 243–246. [18] P.L. Rice, D. Ezzy, Qualitative Research Methods: A Health Focus, Oxford University Press, Melbourne, 1999. [19] H.F. Hsieh, S.E. Shannon, Three approaches to qualitative content analysis, Qual. Health Res. 15 (November (9)) (2005) 1277–1288. [20] U. Flick, Triangulation revisited: strategy of validation or alternative? J. Theor. Soc. Behav. 22 (2) (1992) 175–197. [21] E. Coiera, When conversation is better than computation, J. Am. Med. Inform. Assoc. 7 (May (3)) (2000) 277–286. [22] J.A. Short, J.A. Williams, B. Christie, The Social Psychology of Telecommunications, John Wiley & Sons, New York, 1976. [23] G. Hertel, S. Geister, U. Konradt, Managing virtual teams: a review of current empirical research, Hum. Resour. Manage. Rev. 15 (1) (2005) 69–95. [24] D.L. Shapiro, S.A. Furst, G.M. Spreitzer, M.A. Von Glinow, Transnational teams in the electronic age: are team identity and high performance at risk? J. Org. Behav. 23 (4) (2002) 455–467. [25] S. Reeves, S. Lewin, S. Espin, M. Zwarenstein, Interprofessional Teamword for Health and Social Care, Wiley-Blackwell, West Sussex, 2010. [26] T.B. Iversen, L. Melby, P. Toussaint, Instant messaging at the hospital: supporting articulation work? Int. J. Med. Inform. 82 (September (9)) (2013) 753–761. [27] S.D. Quan, D. Morra, F.Y. Lau, W. Coke, B.M. Wong, R.C. Wu, P.G. Rossos, Perceptions of urgency: defining the gap between what physicians and nurses perceive to be an urgent issue, Int. J. Med. Inform. 82 (May (5)) (2013) 378–386. [28] E. Williams, Experimental comparisons of face-to-face and mediated communication: a review, Psychol. Bull. 84 (5) (1977) 963–976. [29] S. Kiesler, J. Siegel, T.W. McGuire, Social psychological aspects of computer-mediated communication, Am. Psychol. 39 (10) (1984) 1123–1134. [30] D. Wilson, D. Sperber, Relevance theory, in: L.R. Horn, G. Ward (Eds.), The Handbook of Pragmatics, Blackwell Science, Oxford, 2004, pp. 607–632.
