User login
Smartphone‐Enabled Communication System
Previous studies have advocated the importance of effective communication between clinicians as a critical component in the provision of high‐quality patient care.[1, 2, 3, 4] There is increasing interest in the use of information and communication technologies to improve how clinicians communicate in hospital settings. A number of hospitals have implemented different solutions to improve communication. These solutions include alphanumeric pagers,[5] smartphones,[6] e‐mail,[7] secure text messaging,[8] and a Web‐based interdisciplinary communication tool.[9]
These systems have different limitations that render them inefficient and likely inhibit collaborative care. Current systems, such as pagers, rely on the sender to ensure the message was received and are successful in delivering messages approximately 67% of the time.[5, 9, 10] Although alphanumeric pagers and secure text messaging can increase the likelihood of delivery, these messages are often isolated and not easily viewable by the whole care team.[11] Improved systems should also reduce unnecessary interruptions by providing support for both urgent and delayed messages. Finally, messages should be stored and retrievable to enable increased accountability and allow for review for quality improvement initiatives.
It is also important to consider the unintended consequences of technology implementations.[12] Moving communication to text messages and smartphones has the potential to reduce interprofessional relations and can increase confusion if used for complex issues.[10, 13] In this article, we present a system designed to improve interprofessional communication on general internal medicine wards by incorporating these desired features and describe the usage and attitudes toward the system, specifically assessing for effects on multiple domains including efficiency, interprofessional collaboration, and relationships.
METHODS
Research Question
Will nurses and physicians use a system designed to improve interprofessional communication and will they perceive it to be effective and improve workflow?
Setting
The study took place on the general internal medicine wards at Toronto General Hospital and Toronto Western Hospital, 2 large academic teaching hospitals. There are several general internal medicine wards at each site with approximately 80 beds at each site. At each site there are 4 clinical teaching units and 1 hospitalist team. The study was approved by the research ethics board at the University Health Network.
Intervention
To address issues with communication, we developed a systemClinical Message (CM)that included 2 main components: a physician handover tool and secure messaging module. The focus of CM was to improve communication and information flow among different healthcare providers (physicians, nurses, pharmacists, social workers and therapists) through a secure, shared platform.
Physician Handover
The physician handover tool was designed to facilitate the physician handover process at shift change and is used as a patient rounding tool for day‐to‐day management of patients. It is also accessed by nurses and other clinicians to view the physicians' notes and to stay informed on the overall care plan. The tool contains standard elements including a list of patients with the following information on each patient: demographics, diagnosis, code status, medical history, active issues, and discharge plans (Figure 1).
Secure Messaging
Secure messaging was designed around our dominant communication: nurses sending messages to physicians who would then respond. Nurses and other health professionals sent messages to the medical teams by accessing CM, selecting the appropriate patient, and filling out a message template. The system automatically populated the To field with the team assigned to the selected patient. Messaging for each team was centralized around a single team smartphone that was carried 24 hours a day, 7 days a week by a physician on that team. This removed the guesswork of trying to identify the individual physician responsible for that patient. For each message, a subject or issue and content were entered (Figure 2). Logic was also incorporated to reduce the amount of unnecessary interruptions. Senders would choose to send the message immediately as an interrupt message (urgent) for urgent/time sensitive issues or as an allow time to respond message (delayed). For the latter, the message was posted to the system where physicians could check and answer them. Interrupt messages were sent to the team smartphone using the Short Message Service (SMS) protocol. To try and ensure the communication loop on any issues was closed, when a message requested a response and did not receive it, the system sent another message. For urgent messages, a repeat message was initiated after 15 minutes. For delayed messages, the sender defined when they needed a response, typically within 2 to 8 hours. Senders were also able to select the mode of response that would best meet their needs from a workflow perspective: call back, text reply, or to specify that a reply was not required. Senders were also able to verify if the messages were received by the physician's smartphone. Physicians could view the messages within CM and reply. For messages that went to their team smartphone, physicians could respond from the smartphone through a secure Web link.
Because the messages were linked to the patients, they were visible to the entire care team, not just the message sender and recipient. If the care of the patient was transferred from 1 clinician to the next, the new clinician could easily review prior messages to understand recent patient events. The system was accessible through a browser on the intranet. The system regularly pulled patient demographic details such as name, age, medical record number, and location from our electronic medical record through a 1‐way interface. Information from this communication system was not considered part of the medical record but was retrievable.
The system was introduced as the new standard method of communication for nurses to reach physicians for all of the general internal medicine wards and for all medical teams at site 1 on May 2, 2011 and site 2 on June 6, 2011. The system replaced a text‐based Web‐paging system and supplemented the numeric pager carried by residents. Initial training of a half hour was provided to all nurses and residents.
Message Analysis for Usage Statistics
We analyzed messages created and sent via the CM system from May 2011 until August 2012. The extracted message information included date and time sent, issue, level of urgency, response type requested, roles of clinicians involved from the associated team, hospital site (senders and receivers), and message details. The following inclusion criteria were used for the analyses: (1) the senders and receivers of the messages could not be CM support staff, and (2) the messages sent were intended for the team smartphones used by the respective medical teams, not individual clinicians. Descriptive statistics and frequency analysis were performed using Microsoft Excel (Microsoft Corp., Redmond, WA) and IBM SPSS (IBM, Armonk, NY).
Survey
Development of the Survey
We used standard methods to develop a survey to assess staff perceptions on the impact of the new communication system. Relevant questionnaire items were compiled from a systematic review of the literature for communication surveys and communication issues that included the following domains: efficiency, accountability, accuracy, collaboration, timeliness, richness of the communication medium, and impact on interprofessional relationships and verbal communication.[10, 14, 15] We carried out pilot testing with 5 nurses and physicians, and modified the questionnaires based on their feedback.
Sampling and Data Collection of the Survey
Survey participants consisted of 2 groups of clinicians: (1) medical trainees that included medical residents, medical interns, and clinical fellows, and (2) nursing staff that included part‐time and full‐time nurses. To qualify for inclusion, participants had to have used the CM system for at least a month prior to administration of the questionnaire.
Data Analysis
Responses were recorded into an Excel spreadsheet that was imported into SPSS for analysis. Categorical variables were described using proportions. Survey comments were grouped into common themes, and themes mentioned by more than 1 respondent were reported.
RESULTS
Usage Analysis
A total of 60,969 messages were sent using CM between May 2, 2011 and August 19, 2012. On average, a team would receive 14.8 messages per day. Of all messages, 76.5% requested a text reply, 7.7% requested a call‐back, and 15.7% did not request a response. More than two‐thirds of messages at both hospitals were sent as immediate. Of the nonurgent messages, 86% were not replied to within the desired time, requiring a repeat message to be sent. Examples of different types of messages are shown in Table 1.
| Sender | Issue | Details | Priority | Desired Response Type | Time Created | Time Sent | Reply | Time Replied |
|---|---|---|---|---|---|---|---|---|
| ||||||||
| Nurse | Vital sign | Pt's BP is 182/95, HR is 108 now. Previous at 0800 was 165/78; HR was 99. PT is not on antihypertensive meds. | Allow time to respond (23:00) | Text reply | 21:43 | 23:02 | OK. Will assess. | 23:03 |
| Nurse | NG tube | NG tube is in place. Can you please enter portable chest x‐ray to check placement ASAP? | Immediate | Text reply | 16:58 | 16:58 | Will do. | 17:00 |
| Nurse | Bloodwork | Pt creat=216. Pt has NS @ 75 cc/hr. Pt has noted crackles throughout lung fields and has productive cough; eating and drinking well. Would you like it continued as well? Pt O2Sat 93% RA; would you like 4 L of O2 continued? Pls call for telephone order. | Immediate | Call back | 12:53 | 13:04 | Dealt with it on phone. | 13:05 |
| Nurse | Pain control | Hello! Pt has been getting 1 mg hydromorphone IV q 1 hr and pain is still not controlled. Pt remains awake and alert. Thanks! | Immediate | Info only | 15:41 | 15:41 | Thank you. | 15:42 |
For messages requesting a text reply, 8.6% did not receive a reply. The median response time was 2.3 minutes (interquartile range of 5.8 minutes), but some messages did not receive a response even after a week, which skewed the distribution of response times. For those messages that did receive a reply, 68.9% of them were responded to within 5 minutes, and 84.5% were responded to within 15 minutes. Messages were predominantly received between 9 am and midnight (see Supporting Figure 1 in the online version of this article). Because the sending of some messages was delayed, there appeared to be fewer messages received during protected educational times (89 am and 121 pm) as well as between midnight and 7 am compared to other times.
Survey Results
Between April 2013 and June 2013, 82 of 86 medical trainees (95.3%) and 83 of 116 nurses (71.6%) completed the survey, for an overall response rate of 81.7%. Clinicians perceived that CM appeared to have a positive impact on efficiency. In particular, 82.8% of physicians and 78.3% of nurses agreed or strongly agreed that CM helped speed up daily work tasks (Table 2). The majority of physicians and nurses agreed that the system increased accountability, increased timeliness of communication, and improved interprofessional relationships. It was not seen to be effective for communicating complex patient issues.
| No. of Subitems in Survey | Physician (% Agree, Strongly Agree), n=82 | Nurse (% Agree, Strongly Agree), n=83 | |
|---|---|---|---|
| |||
| Positive impact on efficiency. | 7 | 58.9% | 66.6% |
| The CM system helps speed up my daily work tasks. | 82.8% | 78.3% | |
| Positive impact on physician‐nurse collaboration. | 6 | 55.3% | 58.5% |
| The CM system increases the amount of communication between nurses and physicians. | 50.6% | 67.1% | |
| Improved timeliness of communication. | 5 | 54.2% | 50.5% |
| Communication through the CM system helps me resolve patient issues within the appropriate time frames. | 66.7% | 55.6% | |
| Increased accountability. | 2 | 67.1% | 73.2% |
| Improved accuracy of communications. | 3 | 41.6% | 50.7% |
| Improved interprofessional relationships. | 2 | 62.2% | 53.6% |
| Increased verbal communications. | 2 | 35.1% | 25.3% |
| Richness of the communication medium. | 6 | 40.7% | 48.3% |
| I find the CM system useful for communicating complex patient issues. | 35.8% | 26.3% | |
| I would prefer CM over standard hospital communication methods such as numeric paging. | 1 | 68.3% | 76.5% |
| I enjoy using the CM system for clinical communication on the wards. | 1 | 63.0% | 79.0% |
| Communication through the CM system helps to reduce interruptions for physicians. | 1 | 45.7% | |
Survey comments revealed that nurses perceived a lack of desired response, whereas physicians noted being interrupted with low‐value information through the system (Table 3). Both commented that further functionality, such as an active message stream, would be of benefit. Difficulty in communicating complex issues was also noted.
| Issue | Occurrences | Example | ||
|---|---|---|---|---|
| MD | RN | Total | ||
| ||||
| Lack of response | 1 | 10 | 11 | It depends if they respond quickly or not. A few times I send the 2nd message to remind them of the issue. I also spend more time to check if they answer it or not. I even call their Blackberries at last to get a response. |
| Message stream | 3 | 4 | 7 | I wish that I could see follow‐up messages after my initial reply (ie, it would be nice to have an open message stream). |
| Difficult to communicate complex issues | 1 | 5 | 6 | Difficult to communicate complex issues. Takes a lot of time to respond, and it becomes inefficient when responding to nonurgent CM because it interrupts workflow. |
| Many messages are low‐value interrupts | 3 | 0 | 3 | CM is useful for handover between clinicians, but often it slows down the clinician when they are used for information‐related low‐value/noncritical messages between nurses and clinicians |
| Lack of detailed response | 0 | 3 | 3 | Specific messages regarding response to care is required most times. For example, acknowledged is not a favorable response. |
| Technical issues | 2 | 0 | 2 | I find CM very useful. We have had multiple issues with our Blackberry this month, and CM was not working. When it is up and running, however, it is a wonderful tool. |
| Discrepancy in perceived urgency | 2 | 0 | 2 | Discrepancy between what nurses find urgent and what we find urgent. |
DISCUSSION
We describe an implementation of a system to improve clinical communication in hospitals. The system was highly used and was perceived to improve communication by both nurses and physicians. Specifically, users found that the system increased efficiency, accountability, timeliness, and collaboration, but that there were issues with message clarity for complex medical issues.
Other systems and approaches have been implemented to improve communication. These included the use of alphanumeric pagers, e‐mail, secure texting, and smartphones. There is evidence that more advanced systems can improve efficiency for senders.[16] A recent randomized trial of secure text messaging found that it was perceived to be more efficient than paging, but overall usage was low and inconsistent.[8] There is also evidence that smartphones may increase interruptions, worsen interprofessional relationships, and cause issues with professional behavior.[10] Unfortunately, there are a limited number of interventions that improve communication, with some improving efficiency but none demonstrating improved patient‐oriented outcomes.[16, 17] This study evaluated a novel system, with functionality to link communication to patients, and created a system that aligned with the workflow of the clinicians. Messages were linked to the patient, not the sender or receiver, so other clinicians in the patient's circle of care could easily view the communication. Moreover, the system was designed to improve message response rates and allow for nonurgent messages.
Our communication system uses standard, commercially available components (smartphones, SMS), and relatively basic functionality (handover, secure messaging). Important findings are that the current system of paging can be transformed to a more efficient system that users will readily adopt. We found positive effects with components of the system. It appeared to improve efficiency and increase accountability. Accountability is crucial and moves from undocumented conversation to fully documented details of interactions. This can be used for both incident review and to review for quality improvement.
Using the system, physicians perceived that they were bothered by low‐value information, whereas nurses perceived a lack of response, and both found that the system was not ideal for complex messages. The mismatch between what physicians and nurses perceive as important has been attributed to their different timeframes and context.[18] For nurses with an upcoming change of shift, they wanted resolution of issues before handover. A physician on a different ward may not appreciate the context of a nurse having to directly interact with an irate family member. These different perceptions likely contributed to the lack of response to 8.6% of text messages. This is still better than other systems, such as paging, which can be as high as 33%.[10] For nonurgent items, clinicians would ideally check and clear items regularly from the system using a desktop computer, responding within the allotted timeframe. Unfortunately, this never became part of routine physician workflow, likely due to their busy workload, so many physicians would only respond when items became overdue. However, having a method to deal with nonurgent messages may have prevented some interruptions during protected educational times of trainees. The system was also not ideal for urgent or complex items. Complex items can be difficult to convey using the rarified communication medium of text messages.[19, 20] Urgent or complex issues are likely best resolved with a face‐to‐face or telephone conversation.
There are several limitations in our study that should be considered when interpreting the results. It is a study of usage and perceptions after implementation. Although more rigorous study is required to evaluate the effects, we see this as a first step in process improvement. Future research should measure the impact on improving patient care of this system and on patient outcomes such as adverse events. The study and intervention was limited to general internal medicine wards in 2 academic hospital settings where there are frequent rotations of medical personnel. The findings may not be generalizable to other hospital settings.
Future directions should be to further improve on the communication system and to educate and train staff on how to effectively communicate. Survey results showed that although users perceived increased efficiency, there was still significant opportunity to improve. One way to improve would be to have a mobile application in which physicians can easily review nonurgent items. Improvements could also be realized by educating clinicians on the use of the system and providing immediate feedback. Providing feedback to physicians on how well they respond could address nurses' issues around lack of timely response. By creating consensus between nurses and physicians on what is of high and low value to communicate could increase satisfaction for all users.
In summary, we present the usage and perceptions of a system designed to improve hospital communication. We found that there was high uptake, and that users perceived it to improve efficiency, collaboration, and accountability, but it may not be useful for communicating complex issues.
ACKNOWLEDGEMENTS
The authors acknowledge the nurses, physicians, residents, and other health professions on the general internal medicine ward for their patience and support as we continue to try to innovate. The authors also acknowledge the members of the information systems department (Shared Information Management Systems, University Health Network) who helped to support the Communication System, and the software developer, QRS, that helped to codevelop the software system.
Disclosures: The hospital was in a codevelopment agreement that has since terminated. No researcher or hospital received any funds from private industry for any purpose including personal or research. The authors report no conflicts of interest.
- . When conversation is better than computation. J Am Med Inform Assoc. 2000;7(3):277–286.
- , , , et al. Incidence of adverse events and negligence in hospitalized patients. Results of the Harvard Medical Practice Study I. N Engl J Med. 1991;324(6):370–376.
- , , , et al. Gaps in pediatric clinician communication and opportunities for improvement. J Healthc Qual. 2008;30(5):43–54.
- , , , , . Quality in Australian health care study. Med J Aust. 1996;164(12):754.
- , , , . Implementation and evaluation of an alphanumeric paging system on a resident inpatient teaching service. J Hosp Med. 2009;4(8):E34–E40.
- , , , et al. Demonstrating the BlackBerry as a clinical communication tool: a pilot study conducted through the Centre for Innovation in Complex Care. Healthc Q. 2008;11(4):94–98.
- , , , . The use of wireless email to improve healthcare team communication. J Am Med Inform Assoc. 2009;16(5):705–713.
- , , , , , . Smarter hospital communication: secure smartphone text messaging improves provider satisfaction and perception of efficacy, workflow. J Hosp Med. 2014;9(9):573–578.
- , , , , . Beyond paging: building a web‐based communication tool for nurses and physicians. J Gen Intern Med. 2009;24(1):105–110.
- , , , 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. 2013;20(4):766–777.
- , , , et al. Improving hospital care and collaborative communications for the 21st century: key recommendations for general internal medicine. Interact J Med Res. 2012;1(2):e9.
- , , , , , . 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. 2011;18(1):82–90.
- , , , et al. An evaluation of the use of smartphones to communicate between clinicians: a mixed‐methods study. J Med Internet Res. 2011;13(3):e59.
- , , , , . Organizational assessment in intensive care units (ICUs): construct development, reliability, and validity of the ICU nurse‐physician questionnaire. Med Care. 1991;29(8):709–726.
- . Impact of communication medium on task performance and satisfaction: an examination of media‐richness theory. Inform Manag. 1999;35:295–312.
- , , , et al. Effects of clinical communication interventions in hospitals: a systematic review of information and communication technology adoptions for improved communication between clinicians. Int J Med Inform. 2012;81(11):723–732.
- , , , et al. Provider‐to‐provider electronic communication in the era of meaningful use: a review of the evidence. J Hosp Med. 2013;8(10):589–597.
- , , , et al. Perceptions of urgency: defining the gap between what physicians and nurses perceive to be an urgent issue. Int J Med Inform. 2013;82(5):378–386.
- , , , , , . Short message service or disService: issues with text messaging in a complex medical environment. Int J Med Inform. 2014;83(4):278–284.
- , , . Instant messaging at the hospital: supporting articulation work? Int J Med Inform. 2013;82(9):753–761.
Previous studies have advocated the importance of effective communication between clinicians as a critical component in the provision of high‐quality patient care.[1, 2, 3, 4] There is increasing interest in the use of information and communication technologies to improve how clinicians communicate in hospital settings. A number of hospitals have implemented different solutions to improve communication. These solutions include alphanumeric pagers,[5] smartphones,[6] e‐mail,[7] secure text messaging,[8] and a Web‐based interdisciplinary communication tool.[9]
These systems have different limitations that render them inefficient and likely inhibit collaborative care. Current systems, such as pagers, rely on the sender to ensure the message was received and are successful in delivering messages approximately 67% of the time.[5, 9, 10] Although alphanumeric pagers and secure text messaging can increase the likelihood of delivery, these messages are often isolated and not easily viewable by the whole care team.[11] Improved systems should also reduce unnecessary interruptions by providing support for both urgent and delayed messages. Finally, messages should be stored and retrievable to enable increased accountability and allow for review for quality improvement initiatives.
It is also important to consider the unintended consequences of technology implementations.[12] Moving communication to text messages and smartphones has the potential to reduce interprofessional relations and can increase confusion if used for complex issues.[10, 13] In this article, we present a system designed to improve interprofessional communication on general internal medicine wards by incorporating these desired features and describe the usage and attitudes toward the system, specifically assessing for effects on multiple domains including efficiency, interprofessional collaboration, and relationships.
METHODS
Research Question
Will nurses and physicians use a system designed to improve interprofessional communication and will they perceive it to be effective and improve workflow?
Setting
The study took place on the general internal medicine wards at Toronto General Hospital and Toronto Western Hospital, 2 large academic teaching hospitals. There are several general internal medicine wards at each site with approximately 80 beds at each site. At each site there are 4 clinical teaching units and 1 hospitalist team. The study was approved by the research ethics board at the University Health Network.
Intervention
To address issues with communication, we developed a systemClinical Message (CM)that included 2 main components: a physician handover tool and secure messaging module. The focus of CM was to improve communication and information flow among different healthcare providers (physicians, nurses, pharmacists, social workers and therapists) through a secure, shared platform.
Physician Handover
The physician handover tool was designed to facilitate the physician handover process at shift change and is used as a patient rounding tool for day‐to‐day management of patients. It is also accessed by nurses and other clinicians to view the physicians' notes and to stay informed on the overall care plan. The tool contains standard elements including a list of patients with the following information on each patient: demographics, diagnosis, code status, medical history, active issues, and discharge plans (Figure 1).
Secure Messaging
Secure messaging was designed around our dominant communication: nurses sending messages to physicians who would then respond. Nurses and other health professionals sent messages to the medical teams by accessing CM, selecting the appropriate patient, and filling out a message template. The system automatically populated the To field with the team assigned to the selected patient. Messaging for each team was centralized around a single team smartphone that was carried 24 hours a day, 7 days a week by a physician on that team. This removed the guesswork of trying to identify the individual physician responsible for that patient. For each message, a subject or issue and content were entered (Figure 2). Logic was also incorporated to reduce the amount of unnecessary interruptions. Senders would choose to send the message immediately as an interrupt message (urgent) for urgent/time sensitive issues or as an allow time to respond message (delayed). For the latter, the message was posted to the system where physicians could check and answer them. Interrupt messages were sent to the team smartphone using the Short Message Service (SMS) protocol. To try and ensure the communication loop on any issues was closed, when a message requested a response and did not receive it, the system sent another message. For urgent messages, a repeat message was initiated after 15 minutes. For delayed messages, the sender defined when they needed a response, typically within 2 to 8 hours. Senders were also able to select the mode of response that would best meet their needs from a workflow perspective: call back, text reply, or to specify that a reply was not required. Senders were also able to verify if the messages were received by the physician's smartphone. Physicians could view the messages within CM and reply. For messages that went to their team smartphone, physicians could respond from the smartphone through a secure Web link.
Because the messages were linked to the patients, they were visible to the entire care team, not just the message sender and recipient. If the care of the patient was transferred from 1 clinician to the next, the new clinician could easily review prior messages to understand recent patient events. The system was accessible through a browser on the intranet. The system regularly pulled patient demographic details such as name, age, medical record number, and location from our electronic medical record through a 1‐way interface. Information from this communication system was not considered part of the medical record but was retrievable.
The system was introduced as the new standard method of communication for nurses to reach physicians for all of the general internal medicine wards and for all medical teams at site 1 on May 2, 2011 and site 2 on June 6, 2011. The system replaced a text‐based Web‐paging system and supplemented the numeric pager carried by residents. Initial training of a half hour was provided to all nurses and residents.
Message Analysis for Usage Statistics
We analyzed messages created and sent via the CM system from May 2011 until August 2012. The extracted message information included date and time sent, issue, level of urgency, response type requested, roles of clinicians involved from the associated team, hospital site (senders and receivers), and message details. The following inclusion criteria were used for the analyses: (1) the senders and receivers of the messages could not be CM support staff, and (2) the messages sent were intended for the team smartphones used by the respective medical teams, not individual clinicians. Descriptive statistics and frequency analysis were performed using Microsoft Excel (Microsoft Corp., Redmond, WA) and IBM SPSS (IBM, Armonk, NY).
Survey
Development of the Survey
We used standard methods to develop a survey to assess staff perceptions on the impact of the new communication system. Relevant questionnaire items were compiled from a systematic review of the literature for communication surveys and communication issues that included the following domains: efficiency, accountability, accuracy, collaboration, timeliness, richness of the communication medium, and impact on interprofessional relationships and verbal communication.[10, 14, 15] We carried out pilot testing with 5 nurses and physicians, and modified the questionnaires based on their feedback.
Sampling and Data Collection of the Survey
Survey participants consisted of 2 groups of clinicians: (1) medical trainees that included medical residents, medical interns, and clinical fellows, and (2) nursing staff that included part‐time and full‐time nurses. To qualify for inclusion, participants had to have used the CM system for at least a month prior to administration of the questionnaire.
Data Analysis
Responses were recorded into an Excel spreadsheet that was imported into SPSS for analysis. Categorical variables were described using proportions. Survey comments were grouped into common themes, and themes mentioned by more than 1 respondent were reported.
RESULTS
Usage Analysis
A total of 60,969 messages were sent using CM between May 2, 2011 and August 19, 2012. On average, a team would receive 14.8 messages per day. Of all messages, 76.5% requested a text reply, 7.7% requested a call‐back, and 15.7% did not request a response. More than two‐thirds of messages at both hospitals were sent as immediate. Of the nonurgent messages, 86% were not replied to within the desired time, requiring a repeat message to be sent. Examples of different types of messages are shown in Table 1.
| Sender | Issue | Details | Priority | Desired Response Type | Time Created | Time Sent | Reply | Time Replied |
|---|---|---|---|---|---|---|---|---|
| ||||||||
| Nurse | Vital sign | Pt's BP is 182/95, HR is 108 now. Previous at 0800 was 165/78; HR was 99. PT is not on antihypertensive meds. | Allow time to respond (23:00) | Text reply | 21:43 | 23:02 | OK. Will assess. | 23:03 |
| Nurse | NG tube | NG tube is in place. Can you please enter portable chest x‐ray to check placement ASAP? | Immediate | Text reply | 16:58 | 16:58 | Will do. | 17:00 |
| Nurse | Bloodwork | Pt creat=216. Pt has NS @ 75 cc/hr. Pt has noted crackles throughout lung fields and has productive cough; eating and drinking well. Would you like it continued as well? Pt O2Sat 93% RA; would you like 4 L of O2 continued? Pls call for telephone order. | Immediate | Call back | 12:53 | 13:04 | Dealt with it on phone. | 13:05 |
| Nurse | Pain control | Hello! Pt has been getting 1 mg hydromorphone IV q 1 hr and pain is still not controlled. Pt remains awake and alert. Thanks! | Immediate | Info only | 15:41 | 15:41 | Thank you. | 15:42 |
For messages requesting a text reply, 8.6% did not receive a reply. The median response time was 2.3 minutes (interquartile range of 5.8 minutes), but some messages did not receive a response even after a week, which skewed the distribution of response times. For those messages that did receive a reply, 68.9% of them were responded to within 5 minutes, and 84.5% were responded to within 15 minutes. Messages were predominantly received between 9 am and midnight (see Supporting Figure 1 in the online version of this article). Because the sending of some messages was delayed, there appeared to be fewer messages received during protected educational times (89 am and 121 pm) as well as between midnight and 7 am compared to other times.
Survey Results
Between April 2013 and June 2013, 82 of 86 medical trainees (95.3%) and 83 of 116 nurses (71.6%) completed the survey, for an overall response rate of 81.7%. Clinicians perceived that CM appeared to have a positive impact on efficiency. In particular, 82.8% of physicians and 78.3% of nurses agreed or strongly agreed that CM helped speed up daily work tasks (Table 2). The majority of physicians and nurses agreed that the system increased accountability, increased timeliness of communication, and improved interprofessional relationships. It was not seen to be effective for communicating complex patient issues.
| No. of Subitems in Survey | Physician (% Agree, Strongly Agree), n=82 | Nurse (% Agree, Strongly Agree), n=83 | |
|---|---|---|---|
| |||
| Positive impact on efficiency. | 7 | 58.9% | 66.6% |
| The CM system helps speed up my daily work tasks. | 82.8% | 78.3% | |
| Positive impact on physician‐nurse collaboration. | 6 | 55.3% | 58.5% |
| The CM system increases the amount of communication between nurses and physicians. | 50.6% | 67.1% | |
| Improved timeliness of communication. | 5 | 54.2% | 50.5% |
| Communication through the CM system helps me resolve patient issues within the appropriate time frames. | 66.7% | 55.6% | |
| Increased accountability. | 2 | 67.1% | 73.2% |
| Improved accuracy of communications. | 3 | 41.6% | 50.7% |
| Improved interprofessional relationships. | 2 | 62.2% | 53.6% |
| Increased verbal communications. | 2 | 35.1% | 25.3% |
| Richness of the communication medium. | 6 | 40.7% | 48.3% |
| I find the CM system useful for communicating complex patient issues. | 35.8% | 26.3% | |
| I would prefer CM over standard hospital communication methods such as numeric paging. | 1 | 68.3% | 76.5% |
| I enjoy using the CM system for clinical communication on the wards. | 1 | 63.0% | 79.0% |
| Communication through the CM system helps to reduce interruptions for physicians. | 1 | 45.7% | |
Survey comments revealed that nurses perceived a lack of desired response, whereas physicians noted being interrupted with low‐value information through the system (Table 3). Both commented that further functionality, such as an active message stream, would be of benefit. Difficulty in communicating complex issues was also noted.
| Issue | Occurrences | Example | ||
|---|---|---|---|---|
| MD | RN | Total | ||
| ||||
| Lack of response | 1 | 10 | 11 | It depends if they respond quickly or not. A few times I send the 2nd message to remind them of the issue. I also spend more time to check if they answer it or not. I even call their Blackberries at last to get a response. |
| Message stream | 3 | 4 | 7 | I wish that I could see follow‐up messages after my initial reply (ie, it would be nice to have an open message stream). |
| Difficult to communicate complex issues | 1 | 5 | 6 | Difficult to communicate complex issues. Takes a lot of time to respond, and it becomes inefficient when responding to nonurgent CM because it interrupts workflow. |
| Many messages are low‐value interrupts | 3 | 0 | 3 | CM is useful for handover between clinicians, but often it slows down the clinician when they are used for information‐related low‐value/noncritical messages between nurses and clinicians |
| Lack of detailed response | 0 | 3 | 3 | Specific messages regarding response to care is required most times. For example, acknowledged is not a favorable response. |
| Technical issues | 2 | 0 | 2 | I find CM very useful. We have had multiple issues with our Blackberry this month, and CM was not working. When it is up and running, however, it is a wonderful tool. |
| Discrepancy in perceived urgency | 2 | 0 | 2 | Discrepancy between what nurses find urgent and what we find urgent. |
DISCUSSION
We describe an implementation of a system to improve clinical communication in hospitals. The system was highly used and was perceived to improve communication by both nurses and physicians. Specifically, users found that the system increased efficiency, accountability, timeliness, and collaboration, but that there were issues with message clarity for complex medical issues.
Other systems and approaches have been implemented to improve communication. These included the use of alphanumeric pagers, e‐mail, secure texting, and smartphones. There is evidence that more advanced systems can improve efficiency for senders.[16] A recent randomized trial of secure text messaging found that it was perceived to be more efficient than paging, but overall usage was low and inconsistent.[8] There is also evidence that smartphones may increase interruptions, worsen interprofessional relationships, and cause issues with professional behavior.[10] Unfortunately, there are a limited number of interventions that improve communication, with some improving efficiency but none demonstrating improved patient‐oriented outcomes.[16, 17] This study evaluated a novel system, with functionality to link communication to patients, and created a system that aligned with the workflow of the clinicians. Messages were linked to the patient, not the sender or receiver, so other clinicians in the patient's circle of care could easily view the communication. Moreover, the system was designed to improve message response rates and allow for nonurgent messages.
Our communication system uses standard, commercially available components (smartphones, SMS), and relatively basic functionality (handover, secure messaging). Important findings are that the current system of paging can be transformed to a more efficient system that users will readily adopt. We found positive effects with components of the system. It appeared to improve efficiency and increase accountability. Accountability is crucial and moves from undocumented conversation to fully documented details of interactions. This can be used for both incident review and to review for quality improvement.
Using the system, physicians perceived that they were bothered by low‐value information, whereas nurses perceived a lack of response, and both found that the system was not ideal for complex messages. The mismatch between what physicians and nurses perceive as important has been attributed to their different timeframes and context.[18] For nurses with an upcoming change of shift, they wanted resolution of issues before handover. A physician on a different ward may not appreciate the context of a nurse having to directly interact with an irate family member. These different perceptions likely contributed to the lack of response to 8.6% of text messages. This is still better than other systems, such as paging, which can be as high as 33%.[10] For nonurgent items, clinicians would ideally check and clear items regularly from the system using a desktop computer, responding within the allotted timeframe. Unfortunately, this never became part of routine physician workflow, likely due to their busy workload, so many physicians would only respond when items became overdue. However, having a method to deal with nonurgent messages may have prevented some interruptions during protected educational times of trainees. The system was also not ideal for urgent or complex items. Complex items can be difficult to convey using the rarified communication medium of text messages.[19, 20] Urgent or complex issues are likely best resolved with a face‐to‐face or telephone conversation.
There are several limitations in our study that should be considered when interpreting the results. It is a study of usage and perceptions after implementation. Although more rigorous study is required to evaluate the effects, we see this as a first step in process improvement. Future research should measure the impact on improving patient care of this system and on patient outcomes such as adverse events. The study and intervention was limited to general internal medicine wards in 2 academic hospital settings where there are frequent rotations of medical personnel. The findings may not be generalizable to other hospital settings.
Future directions should be to further improve on the communication system and to educate and train staff on how to effectively communicate. Survey results showed that although users perceived increased efficiency, there was still significant opportunity to improve. One way to improve would be to have a mobile application in which physicians can easily review nonurgent items. Improvements could also be realized by educating clinicians on the use of the system and providing immediate feedback. Providing feedback to physicians on how well they respond could address nurses' issues around lack of timely response. By creating consensus between nurses and physicians on what is of high and low value to communicate could increase satisfaction for all users.
In summary, we present the usage and perceptions of a system designed to improve hospital communication. We found that there was high uptake, and that users perceived it to improve efficiency, collaboration, and accountability, but it may not be useful for communicating complex issues.
ACKNOWLEDGEMENTS
The authors acknowledge the nurses, physicians, residents, and other health professions on the general internal medicine ward for their patience and support as we continue to try to innovate. The authors also acknowledge the members of the information systems department (Shared Information Management Systems, University Health Network) who helped to support the Communication System, and the software developer, QRS, that helped to codevelop the software system.
Disclosures: The hospital was in a codevelopment agreement that has since terminated. No researcher or hospital received any funds from private industry for any purpose including personal or research. The authors report no conflicts of interest.
Previous studies have advocated the importance of effective communication between clinicians as a critical component in the provision of high‐quality patient care.[1, 2, 3, 4] There is increasing interest in the use of information and communication technologies to improve how clinicians communicate in hospital settings. A number of hospitals have implemented different solutions to improve communication. These solutions include alphanumeric pagers,[5] smartphones,[6] e‐mail,[7] secure text messaging,[8] and a Web‐based interdisciplinary communication tool.[9]
These systems have different limitations that render them inefficient and likely inhibit collaborative care. Current systems, such as pagers, rely on the sender to ensure the message was received and are successful in delivering messages approximately 67% of the time.[5, 9, 10] Although alphanumeric pagers and secure text messaging can increase the likelihood of delivery, these messages are often isolated and not easily viewable by the whole care team.[11] Improved systems should also reduce unnecessary interruptions by providing support for both urgent and delayed messages. Finally, messages should be stored and retrievable to enable increased accountability and allow for review for quality improvement initiatives.
It is also important to consider the unintended consequences of technology implementations.[12] Moving communication to text messages and smartphones has the potential to reduce interprofessional relations and can increase confusion if used for complex issues.[10, 13] In this article, we present a system designed to improve interprofessional communication on general internal medicine wards by incorporating these desired features and describe the usage and attitudes toward the system, specifically assessing for effects on multiple domains including efficiency, interprofessional collaboration, and relationships.
METHODS
Research Question
Will nurses and physicians use a system designed to improve interprofessional communication and will they perceive it to be effective and improve workflow?
Setting
The study took place on the general internal medicine wards at Toronto General Hospital and Toronto Western Hospital, 2 large academic teaching hospitals. There are several general internal medicine wards at each site with approximately 80 beds at each site. At each site there are 4 clinical teaching units and 1 hospitalist team. The study was approved by the research ethics board at the University Health Network.
Intervention
To address issues with communication, we developed a systemClinical Message (CM)that included 2 main components: a physician handover tool and secure messaging module. The focus of CM was to improve communication and information flow among different healthcare providers (physicians, nurses, pharmacists, social workers and therapists) through a secure, shared platform.
Physician Handover
The physician handover tool was designed to facilitate the physician handover process at shift change and is used as a patient rounding tool for day‐to‐day management of patients. It is also accessed by nurses and other clinicians to view the physicians' notes and to stay informed on the overall care plan. The tool contains standard elements including a list of patients with the following information on each patient: demographics, diagnosis, code status, medical history, active issues, and discharge plans (Figure 1).
Secure Messaging
Secure messaging was designed around our dominant communication: nurses sending messages to physicians who would then respond. Nurses and other health professionals sent messages to the medical teams by accessing CM, selecting the appropriate patient, and filling out a message template. The system automatically populated the To field with the team assigned to the selected patient. Messaging for each team was centralized around a single team smartphone that was carried 24 hours a day, 7 days a week by a physician on that team. This removed the guesswork of trying to identify the individual physician responsible for that patient. For each message, a subject or issue and content were entered (Figure 2). Logic was also incorporated to reduce the amount of unnecessary interruptions. Senders would choose to send the message immediately as an interrupt message (urgent) for urgent/time sensitive issues or as an allow time to respond message (delayed). For the latter, the message was posted to the system where physicians could check and answer them. Interrupt messages were sent to the team smartphone using the Short Message Service (SMS) protocol. To try and ensure the communication loop on any issues was closed, when a message requested a response and did not receive it, the system sent another message. For urgent messages, a repeat message was initiated after 15 minutes. For delayed messages, the sender defined when they needed a response, typically within 2 to 8 hours. Senders were also able to select the mode of response that would best meet their needs from a workflow perspective: call back, text reply, or to specify that a reply was not required. Senders were also able to verify if the messages were received by the physician's smartphone. Physicians could view the messages within CM and reply. For messages that went to their team smartphone, physicians could respond from the smartphone through a secure Web link.
Because the messages were linked to the patients, they were visible to the entire care team, not just the message sender and recipient. If the care of the patient was transferred from 1 clinician to the next, the new clinician could easily review prior messages to understand recent patient events. The system was accessible through a browser on the intranet. The system regularly pulled patient demographic details such as name, age, medical record number, and location from our electronic medical record through a 1‐way interface. Information from this communication system was not considered part of the medical record but was retrievable.
The system was introduced as the new standard method of communication for nurses to reach physicians for all of the general internal medicine wards and for all medical teams at site 1 on May 2, 2011 and site 2 on June 6, 2011. The system replaced a text‐based Web‐paging system and supplemented the numeric pager carried by residents. Initial training of a half hour was provided to all nurses and residents.
Message Analysis for Usage Statistics
We analyzed messages created and sent via the CM system from May 2011 until August 2012. The extracted message information included date and time sent, issue, level of urgency, response type requested, roles of clinicians involved from the associated team, hospital site (senders and receivers), and message details. The following inclusion criteria were used for the analyses: (1) the senders and receivers of the messages could not be CM support staff, and (2) the messages sent were intended for the team smartphones used by the respective medical teams, not individual clinicians. Descriptive statistics and frequency analysis were performed using Microsoft Excel (Microsoft Corp., Redmond, WA) and IBM SPSS (IBM, Armonk, NY).
Survey
Development of the Survey
We used standard methods to develop a survey to assess staff perceptions on the impact of the new communication system. Relevant questionnaire items were compiled from a systematic review of the literature for communication surveys and communication issues that included the following domains: efficiency, accountability, accuracy, collaboration, timeliness, richness of the communication medium, and impact on interprofessional relationships and verbal communication.[10, 14, 15] We carried out pilot testing with 5 nurses and physicians, and modified the questionnaires based on their feedback.
Sampling and Data Collection of the Survey
Survey participants consisted of 2 groups of clinicians: (1) medical trainees that included medical residents, medical interns, and clinical fellows, and (2) nursing staff that included part‐time and full‐time nurses. To qualify for inclusion, participants had to have used the CM system for at least a month prior to administration of the questionnaire.
Data Analysis
Responses were recorded into an Excel spreadsheet that was imported into SPSS for analysis. Categorical variables were described using proportions. Survey comments were grouped into common themes, and themes mentioned by more than 1 respondent were reported.
RESULTS
Usage Analysis
A total of 60,969 messages were sent using CM between May 2, 2011 and August 19, 2012. On average, a team would receive 14.8 messages per day. Of all messages, 76.5% requested a text reply, 7.7% requested a call‐back, and 15.7% did not request a response. More than two‐thirds of messages at both hospitals were sent as immediate. Of the nonurgent messages, 86% were not replied to within the desired time, requiring a repeat message to be sent. Examples of different types of messages are shown in Table 1.
| Sender | Issue | Details | Priority | Desired Response Type | Time Created | Time Sent | Reply | Time Replied |
|---|---|---|---|---|---|---|---|---|
| ||||||||
| Nurse | Vital sign | Pt's BP is 182/95, HR is 108 now. Previous at 0800 was 165/78; HR was 99. PT is not on antihypertensive meds. | Allow time to respond (23:00) | Text reply | 21:43 | 23:02 | OK. Will assess. | 23:03 |
| Nurse | NG tube | NG tube is in place. Can you please enter portable chest x‐ray to check placement ASAP? | Immediate | Text reply | 16:58 | 16:58 | Will do. | 17:00 |
| Nurse | Bloodwork | Pt creat=216. Pt has NS @ 75 cc/hr. Pt has noted crackles throughout lung fields and has productive cough; eating and drinking well. Would you like it continued as well? Pt O2Sat 93% RA; would you like 4 L of O2 continued? Pls call for telephone order. | Immediate | Call back | 12:53 | 13:04 | Dealt with it on phone. | 13:05 |
| Nurse | Pain control | Hello! Pt has been getting 1 mg hydromorphone IV q 1 hr and pain is still not controlled. Pt remains awake and alert. Thanks! | Immediate | Info only | 15:41 | 15:41 | Thank you. | 15:42 |
For messages requesting a text reply, 8.6% did not receive a reply. The median response time was 2.3 minutes (interquartile range of 5.8 minutes), but some messages did not receive a response even after a week, which skewed the distribution of response times. For those messages that did receive a reply, 68.9% of them were responded to within 5 minutes, and 84.5% were responded to within 15 minutes. Messages were predominantly received between 9 am and midnight (see Supporting Figure 1 in the online version of this article). Because the sending of some messages was delayed, there appeared to be fewer messages received during protected educational times (89 am and 121 pm) as well as between midnight and 7 am compared to other times.
Survey Results
Between April 2013 and June 2013, 82 of 86 medical trainees (95.3%) and 83 of 116 nurses (71.6%) completed the survey, for an overall response rate of 81.7%. Clinicians perceived that CM appeared to have a positive impact on efficiency. In particular, 82.8% of physicians and 78.3% of nurses agreed or strongly agreed that CM helped speed up daily work tasks (Table 2). The majority of physicians and nurses agreed that the system increased accountability, increased timeliness of communication, and improved interprofessional relationships. It was not seen to be effective for communicating complex patient issues.
| No. of Subitems in Survey | Physician (% Agree, Strongly Agree), n=82 | Nurse (% Agree, Strongly Agree), n=83 | |
|---|---|---|---|
| |||
| Positive impact on efficiency. | 7 | 58.9% | 66.6% |
| The CM system helps speed up my daily work tasks. | 82.8% | 78.3% | |
| Positive impact on physician‐nurse collaboration. | 6 | 55.3% | 58.5% |
| The CM system increases the amount of communication between nurses and physicians. | 50.6% | 67.1% | |
| Improved timeliness of communication. | 5 | 54.2% | 50.5% |
| Communication through the CM system helps me resolve patient issues within the appropriate time frames. | 66.7% | 55.6% | |
| Increased accountability. | 2 | 67.1% | 73.2% |
| Improved accuracy of communications. | 3 | 41.6% | 50.7% |
| Improved interprofessional relationships. | 2 | 62.2% | 53.6% |
| Increased verbal communications. | 2 | 35.1% | 25.3% |
| Richness of the communication medium. | 6 | 40.7% | 48.3% |
| I find the CM system useful for communicating complex patient issues. | 35.8% | 26.3% | |
| I would prefer CM over standard hospital communication methods such as numeric paging. | 1 | 68.3% | 76.5% |
| I enjoy using the CM system for clinical communication on the wards. | 1 | 63.0% | 79.0% |
| Communication through the CM system helps to reduce interruptions for physicians. | 1 | 45.7% | |
Survey comments revealed that nurses perceived a lack of desired response, whereas physicians noted being interrupted with low‐value information through the system (Table 3). Both commented that further functionality, such as an active message stream, would be of benefit. Difficulty in communicating complex issues was also noted.
| Issue | Occurrences | Example | ||
|---|---|---|---|---|
| MD | RN | Total | ||
| ||||
| Lack of response | 1 | 10 | 11 | It depends if they respond quickly or not. A few times I send the 2nd message to remind them of the issue. I also spend more time to check if they answer it or not. I even call their Blackberries at last to get a response. |
| Message stream | 3 | 4 | 7 | I wish that I could see follow‐up messages after my initial reply (ie, it would be nice to have an open message stream). |
| Difficult to communicate complex issues | 1 | 5 | 6 | Difficult to communicate complex issues. Takes a lot of time to respond, and it becomes inefficient when responding to nonurgent CM because it interrupts workflow. |
| Many messages are low‐value interrupts | 3 | 0 | 3 | CM is useful for handover between clinicians, but often it slows down the clinician when they are used for information‐related low‐value/noncritical messages between nurses and clinicians |
| Lack of detailed response | 0 | 3 | 3 | Specific messages regarding response to care is required most times. For example, acknowledged is not a favorable response. |
| Technical issues | 2 | 0 | 2 | I find CM very useful. We have had multiple issues with our Blackberry this month, and CM was not working. When it is up and running, however, it is a wonderful tool. |
| Discrepancy in perceived urgency | 2 | 0 | 2 | Discrepancy between what nurses find urgent and what we find urgent. |
DISCUSSION
We describe an implementation of a system to improve clinical communication in hospitals. The system was highly used and was perceived to improve communication by both nurses and physicians. Specifically, users found that the system increased efficiency, accountability, timeliness, and collaboration, but that there were issues with message clarity for complex medical issues.
Other systems and approaches have been implemented to improve communication. These included the use of alphanumeric pagers, e‐mail, secure texting, and smartphones. There is evidence that more advanced systems can improve efficiency for senders.[16] A recent randomized trial of secure text messaging found that it was perceived to be more efficient than paging, but overall usage was low and inconsistent.[8] There is also evidence that smartphones may increase interruptions, worsen interprofessional relationships, and cause issues with professional behavior.[10] Unfortunately, there are a limited number of interventions that improve communication, with some improving efficiency but none demonstrating improved patient‐oriented outcomes.[16, 17] This study evaluated a novel system, with functionality to link communication to patients, and created a system that aligned with the workflow of the clinicians. Messages were linked to the patient, not the sender or receiver, so other clinicians in the patient's circle of care could easily view the communication. Moreover, the system was designed to improve message response rates and allow for nonurgent messages.
Our communication system uses standard, commercially available components (smartphones, SMS), and relatively basic functionality (handover, secure messaging). Important findings are that the current system of paging can be transformed to a more efficient system that users will readily adopt. We found positive effects with components of the system. It appeared to improve efficiency and increase accountability. Accountability is crucial and moves from undocumented conversation to fully documented details of interactions. This can be used for both incident review and to review for quality improvement.
Using the system, physicians perceived that they were bothered by low‐value information, whereas nurses perceived a lack of response, and both found that the system was not ideal for complex messages. The mismatch between what physicians and nurses perceive as important has been attributed to their different timeframes and context.[18] For nurses with an upcoming change of shift, they wanted resolution of issues before handover. A physician on a different ward may not appreciate the context of a nurse having to directly interact with an irate family member. These different perceptions likely contributed to the lack of response to 8.6% of text messages. This is still better than other systems, such as paging, which can be as high as 33%.[10] For nonurgent items, clinicians would ideally check and clear items regularly from the system using a desktop computer, responding within the allotted timeframe. Unfortunately, this never became part of routine physician workflow, likely due to their busy workload, so many physicians would only respond when items became overdue. However, having a method to deal with nonurgent messages may have prevented some interruptions during protected educational times of trainees. The system was also not ideal for urgent or complex items. Complex items can be difficult to convey using the rarified communication medium of text messages.[19, 20] Urgent or complex issues are likely best resolved with a face‐to‐face or telephone conversation.
There are several limitations in our study that should be considered when interpreting the results. It is a study of usage and perceptions after implementation. Although more rigorous study is required to evaluate the effects, we see this as a first step in process improvement. Future research should measure the impact on improving patient care of this system and on patient outcomes such as adverse events. The study and intervention was limited to general internal medicine wards in 2 academic hospital settings where there are frequent rotations of medical personnel. The findings may not be generalizable to other hospital settings.
Future directions should be to further improve on the communication system and to educate and train staff on how to effectively communicate. Survey results showed that although users perceived increased efficiency, there was still significant opportunity to improve. One way to improve would be to have a mobile application in which physicians can easily review nonurgent items. Improvements could also be realized by educating clinicians on the use of the system and providing immediate feedback. Providing feedback to physicians on how well they respond could address nurses' issues around lack of timely response. By creating consensus between nurses and physicians on what is of high and low value to communicate could increase satisfaction for all users.
In summary, we present the usage and perceptions of a system designed to improve hospital communication. We found that there was high uptake, and that users perceived it to improve efficiency, collaboration, and accountability, but it may not be useful for communicating complex issues.
ACKNOWLEDGEMENTS
The authors acknowledge the nurses, physicians, residents, and other health professions on the general internal medicine ward for their patience and support as we continue to try to innovate. The authors also acknowledge the members of the information systems department (Shared Information Management Systems, University Health Network) who helped to support the Communication System, and the software developer, QRS, that helped to codevelop the software system.
Disclosures: The hospital was in a codevelopment agreement that has since terminated. No researcher or hospital received any funds from private industry for any purpose including personal or research. The authors report no conflicts of interest.
- . When conversation is better than computation. J Am Med Inform Assoc. 2000;7(3):277–286.
- , , , et al. Incidence of adverse events and negligence in hospitalized patients. Results of the Harvard Medical Practice Study I. N Engl J Med. 1991;324(6):370–376.
- , , , et al. Gaps in pediatric clinician communication and opportunities for improvement. J Healthc Qual. 2008;30(5):43–54.
- , , , , . Quality in Australian health care study. Med J Aust. 1996;164(12):754.
- , , , . Implementation and evaluation of an alphanumeric paging system on a resident inpatient teaching service. J Hosp Med. 2009;4(8):E34–E40.
- , , , et al. Demonstrating the BlackBerry as a clinical communication tool: a pilot study conducted through the Centre for Innovation in Complex Care. Healthc Q. 2008;11(4):94–98.
- , , , . The use of wireless email to improve healthcare team communication. J Am Med Inform Assoc. 2009;16(5):705–713.
- , , , , , . Smarter hospital communication: secure smartphone text messaging improves provider satisfaction and perception of efficacy, workflow. J Hosp Med. 2014;9(9):573–578.
- , , , , . Beyond paging: building a web‐based communication tool for nurses and physicians. J Gen Intern Med. 2009;24(1):105–110.
- , , , 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. 2013;20(4):766–777.
- , , , et al. Improving hospital care and collaborative communications for the 21st century: key recommendations for general internal medicine. Interact J Med Res. 2012;1(2):e9.
- , , , , , . 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. 2011;18(1):82–90.
- , , , et al. An evaluation of the use of smartphones to communicate between clinicians: a mixed‐methods study. J Med Internet Res. 2011;13(3):e59.
- , , , , . Organizational assessment in intensive care units (ICUs): construct development, reliability, and validity of the ICU nurse‐physician questionnaire. Med Care. 1991;29(8):709–726.
- . Impact of communication medium on task performance and satisfaction: an examination of media‐richness theory. Inform Manag. 1999;35:295–312.
- , , , et al. Effects of clinical communication interventions in hospitals: a systematic review of information and communication technology adoptions for improved communication between clinicians. Int J Med Inform. 2012;81(11):723–732.
- , , , et al. Provider‐to‐provider electronic communication in the era of meaningful use: a review of the evidence. J Hosp Med. 2013;8(10):589–597.
- , , , et al. Perceptions of urgency: defining the gap between what physicians and nurses perceive to be an urgent issue. Int J Med Inform. 2013;82(5):378–386.
- , , , , , . Short message service or disService: issues with text messaging in a complex medical environment. Int J Med Inform. 2014;83(4):278–284.
- , , . Instant messaging at the hospital: supporting articulation work? Int J Med Inform. 2013;82(9):753–761.
- . When conversation is better than computation. J Am Med Inform Assoc. 2000;7(3):277–286.
- , , , et al. Incidence of adverse events and negligence in hospitalized patients. Results of the Harvard Medical Practice Study I. N Engl J Med. 1991;324(6):370–376.
- , , , et al. Gaps in pediatric clinician communication and opportunities for improvement. J Healthc Qual. 2008;30(5):43–54.
- , , , , . Quality in Australian health care study. Med J Aust. 1996;164(12):754.
- , , , . Implementation and evaluation of an alphanumeric paging system on a resident inpatient teaching service. J Hosp Med. 2009;4(8):E34–E40.
- , , , et al. Demonstrating the BlackBerry as a clinical communication tool: a pilot study conducted through the Centre for Innovation in Complex Care. Healthc Q. 2008;11(4):94–98.
- , , , . The use of wireless email to improve healthcare team communication. J Am Med Inform Assoc. 2009;16(5):705–713.
- , , , , , . Smarter hospital communication: secure smartphone text messaging improves provider satisfaction and perception of efficacy, workflow. J Hosp Med. 2014;9(9):573–578.
- , , , , . Beyond paging: building a web‐based communication tool for nurses and physicians. J Gen Intern Med. 2009;24(1):105–110.
- , , , 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. 2013;20(4):766–777.
- , , , et al. Improving hospital care and collaborative communications for the 21st century: key recommendations for general internal medicine. Interact J Med Res. 2012;1(2):e9.
- , , , , , . 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. 2011;18(1):82–90.
- , , , et al. An evaluation of the use of smartphones to communicate between clinicians: a mixed‐methods study. J Med Internet Res. 2011;13(3):e59.
- , , , , . Organizational assessment in intensive care units (ICUs): construct development, reliability, and validity of the ICU nurse‐physician questionnaire. Med Care. 1991;29(8):709–726.
- . Impact of communication medium on task performance and satisfaction: an examination of media‐richness theory. Inform Manag. 1999;35:295–312.
- , , , et al. Effects of clinical communication interventions in hospitals: a systematic review of information and communication technology adoptions for improved communication between clinicians. Int J Med Inform. 2012;81(11):723–732.
- , , , et al. Provider‐to‐provider electronic communication in the era of meaningful use: a review of the evidence. J Hosp Med. 2013;8(10):589–597.
- , , , et al. Perceptions of urgency: defining the gap between what physicians and nurses perceive to be an urgent issue. Int J Med Inform. 2013;82(5):378–386.
- , , , , , . Short message service or disService: issues with text messaging in a complex medical environment. Int J Med Inform. 2014;83(4):278–284.
- , , . Instant messaging at the hospital: supporting articulation work? Int J Med Inform. 2013;82(9):753–761.
© 2014 Society of Hospital Medicine
Educational Impact of Smartphones
Medical residents are rapidly adopting smartphones. Recent statistics revealed that 85% of medical providers currently own a smartphone, and the majority use it in their clinical work.[1] Smartphone capabilities that include the use of text messaging, e‐mail, and mobile phone functions in the clinical setting may improve efficiency and quality of care by reducing the response time for urgent issues.[2] There is, however, increasing recognition that healthcare information technology can create unintended negative consequences. For example, studies have suggested that healthcare information technologies, such as the computerized physician order entry, may actually increase errors by creating new work, changing clinical workflow, and altering communication patterns.[3, 4, 5]
Smartphone use for clinical communication can have unintended consequences by increasing interruptions, reducing interprofessional relationships, and widening the gap between what nurses and physicians perceive as urgent clinical problems.[6] However, no studies have evaluated the impact of smartphones on the educational experience of medical trainees. Although previous studies have described the use of smartphones by trainees for rapid access to electronic medical resources,[7, 8, 9] we did not identify in our literature review any previous studies on the impact of using the smartphone's primary functionas a communication deviceon the educational experience of residents and medical students. Therefore, our study aimed to examine the impact of using smartphones for clinical communication on medical education.
METHODS
Design
The design of the study was qualitative research methodology using interview data, ethnographic data, and content analysis of text‐based messages.
Setting
From June 2009 to September 2010, we conducted a multisite evaluation study on general internal medicine (GIM) wards at 5 large academic teaching hospitals in the city of Toronto, Canada at St. Michael's Hospital, Sunnybrook Health Sciences Centre, Toronto General Hospital, Toronto Western Hospital, and Mount Sinai Hospital. Each hospital has clinical teaching units consisting typically of 4 medical teams. Each team includes 1 attending physician, 1 senior resident, 2 or more junior residents, and 2 to 4 medical students. Each hospital had 2 to 4 GIM wards in different geographic locations.
Communication Systems
To make it easier for nurses and other health professionals to communicate with the physician teams, all sites centralized communication to 1 team member, who acts as the single point of contact on behalf of their assigned team in the communication of patient‐related issues. We facilitated this communication through a shared device (either a pager or a smartphone). The senior resident typically carried the shared device during the day and the on‐call junior resident at night and on the weekends. Two hospitals provided smartphones to all residents, whereas a third site provided smartphones only to the senior residents. The standard processes of communication required that physicians respond to all calls and text messages. At the 3 sites with institutional smartphones, nurses could send text messages with patient information using a Web‐based system. We encrypted data sent to institutional smartphones to protect patient information.
Data Collection
Using a mixed‐methods ethnographic approach, we collected data using semistructured interviews, ethnographic observations, and content analysis of text messages. The original larger study focused primarily on examining the overall clinical impact of smartphone use.[10] For our current study, we analyzed the data with a focus on evaluating the impact of smartphones on the educational experience of medical trainees on the GIM teaching service. The respective institutions' research ethics boards approved the study.
Interviews
We conducted semistructured interviews with residents, medical students, attending physicians, and other clinicians across all of the sites to examine how clinicians perceived the impact of smartphones on medical education. We used a purposeful sampling strategy where we interviewed different groups of healthcare professionals who we suspected would represent different viewpoints on the use of smartphones for clinical communication. To obtain diverse perspectives, we snowball sampled by asking interviewees to suggest colleagues with differing views to participate in the interviews. The interview guide consisted of open‐ended questions with additional probes to elicit more detailed information from these frontline clinicians who initiate and receive communication. One of the study investigators (V.L.) conducted interviews that varied from 15 to 45 minutes in duration. We recorded, transcribed verbatim, and analyzed the interviews using NVivo software (QSR International, Doncaster, Victoria, Australia). We added additional questions iteratively as themes emerged from the initial interviews. One of the study investigators (V.L.) encouraged participants to speak freely, to raise issues that they perceived to be important, and to support their responses with examples.
Observations
We observed the communication processes in the hospitals by conducting a work‐shadowing approach that followed individual residents in their work environments. These observations included 1‐on‐1 supervision encounters involving attending staff, medical students, and other residents, and informal and formal teaching rounds. The observation periods included the usual working day (from 8 am to 6 pm) as well as the busiest times on call, typically from 6 pm until 11 pm. We sampled different residents for different time periods. We adopted a nonparticipatory observation technique where we observed all interruptions, communication interactions, and patterns from a distance. We defined workflow interruptions as an intrusion of an unplanned and unscheduled task, causing a discontinuation of tasks, a noticeable break, or task switch behaviour.[11] Data collection included timing of events and writing field notes. One of the study investigators (V.L.) performed all the work‐shadowing observations.
E‐mail
To study the volume and content of messages, we collected e‐mail communications between January 2009 and June 2009 from consenting residents at the 2 hospitals that provided smartphones to all GIM residents. E‐mail information included the sender, the receiver, the time of message, and the message content. To look at usage, we calculated the average number of e‐mails sent and received. To assess interruptions on formal teaching sessions, we paid particular attention to e‐mails received and sent during protected educational timeweekdays from 8 am to 9 am (morning report) and 12 pm to 1 pm (noon rounds). We randomly sampled 20% of all e‐mails sent between residents for content analysis and organized content related to medical education into thematic categories.
Analysis
We used a deductive approach to analyze the interview transcripts by applying a conceptual framework that assessed the educational impact of patient safety interventions.[12] This framework identified 5 educational domains (learning, teaching, supervision, assessment, and feedback). Three study investigators mapped interview data, work‐shadowing data, and e‐mail content to themes (V.L., B.W., and R.W.), and grouped data that did not translate into themes into new categories. We then triangulated the data to develop themes of the educational impact of smartphone communication by both perceived use and actual use, and subsequently constructed a framework of how smartphone communication affected education.
RESULTS
We conducted 124 semistructured interviews with residents, medical students, attending physicians, and other clinicians across all the sites to examine how clinicians perceived the impact of smartphones on medical education. We work‐shadowed 40 individual residents for a total of 196 hours (Table 1). We analyzed the 13,714 e‐mails sent from or received to 34 residents. To analyze e‐mail content, we reviewed 1179 e‐mails sent among residents.
| Methods | Sites | |||||
|---|---|---|---|---|---|---|
| St. Michael's Hospital | Sunnybrook Health Sciences Centre | Toronto General Hospital | Toronto Western Hospital | Mount Sinai Hospital | All Hospitals | |
| ||||||
| Work‐shadowing residents | ||||||
| Hours | 60 hours | 35 hours | 57 hours 55 minutes | 27 hours 46 minutes | 15 hours | 196 hours |
| No. of residents | 12 | 7 | 12 | 6 | 3 | 40 |
| Interviews with clinicians | ||||||
| Physicians | 10 | 5 | 13 | 5 | 33 | |
| Medical students | 5 | 4 | 1 | 1 | 11 | |
| Nurses | 9 | 11 | 15 | 14 | 49 | |
| Other health professionsa | 7 | 10 | 8 | 6 | 31 | |
| Total | 31 | 30 | 37 | 26 | 124 | |
We found that 2 key characteristics of smartphone use for clinical communication, namely an increase in connectedness leading to an increase in interruptions, impacted 3 educational domains: teaching, supervision, and professionalism (Figure 1).
Increased Connectedness
As a communication device, smartphones increase the ability to receive and respond to messages through voice, e‐mail, and text messaging. Not surprisingly, with the improved ability and mobility to communicate, medical trainees perceived being more connected with their team members, who included other residents, medical students, and attending staff as well as with other clinical services and professions. These smartphone communication activities appeared to be pervasive, occurring on the wards, at the bedside, while in transit, and in teaching sessions (Box 1: increased connectedness).
Box
Increased connectedness
I've used the Blackberry system and it's nice to be able to quickly text each other little messages especially for meeting times because then you don't have to page them and wait by the phone. So that's been great for in the team. (Interview Resident 3)
It's incredibly useful for when you're paging somebody else. Often times I'll be consulting with another physician on a patient and I'll say This is my BlackBerry. Call me back after you've seen the patient' or Call me back when you have a plan' or, you know, whatever. So that's extremely valuable which we never had with pages and no one would ever page you for that because it was too much of a pain. (Interview Resident 1)
My personal experience has been that if you need to speak to a more senior individual it's much easier to contact them via the BlackBerry. (Interview Medical Student 1)
At 7:25 pm, MD11 returns to the patient's room and continues examining her. While in the patient's room, I could see her talking on the BlackBerrys. I asked her later what calls she had while in the room. It turns out she had 3 phone calls and 2 texts. Two of the calls were from the radiation oncologists and 1 call from the pathologist. She also received 1 text on the Team BlackBerry and 1 text on the Senior's BlackBerry from the pharmacist. (Field Notes, Work-shadowing MD11)
Interruptions
The increased connectedness caused by smartphone use led residents to perceive an increase in the frequency of interruptions. The multitude of communication and contact options made available by smartphones to health providers created an expansive network of connected individuals who were in constant communication with each other. Instead of the difficulties associated with numeric paging and waiting for a response, nurses typically found it easier to call directly or send a text message to residents' smartphones. From the e‐mail analysis, residents received, on a daily basis, on average 25.7 e‐mails, (median, 20; interquartile range [IQR]: 1428) to the team smartphone and sent 7.5 e‐mails (median, 6; IQR: 410). During protected educational time, each resident received an average of 1.0 e‐mail (median, 1; IQR: 01) between 8 am and 9 am and an average of 2.3 e‐mails (median, 2; IQR: 13) during 12 pm to 1 pm (Figure 2). Each of these communication events, whether a phone call, e‐mail, or text‐message, led to an interruption (Box 2). Given that smartphones made it easier for nurses to contact residents, some residents attributed the increase in interruptions to a reduction in the threshold for nurses to communicate.
Box
Increased interruptions
The only negative I can think of is just the incredible number of communications that you get, you know, text messages and e‐mails and everything else. So just the, the number can sometimes be overwhelming. (Interview Resident 1)
Some of [the nurses] rely a little bit more on the BlackBerry so that they will tend to call you a bit more frequently for things that maybe sometimes they should try to find answer for themselves (Interview Resident 2)
And now with the option of being able to, if you really needed to, call them and talk to them directly, I think that kind of improves communication. They're easier to find. (Interview Nurse 4)
Supervision
Smartphone communication appeared to positively impact trainee supervision. Increased connectedness between team members allowed junior trainees to have access and rapidly communicate with a more experienced clinician, which provided them with greater support. Residents found smartphones particularly useful in situations where they felt uncomfortable or where they did not feel competent. Some of these instances related to procedural competence, with residents feeling more comfortable knowing they have rapid access to support (Box 3: increased support).
Box
Supervision
Increased support
It makes me feel more comfortable in the sense that I can instantly make a call or a text and have a question answered if I need an answer. Or if it were an emergency having the ability to talk on the phone and be talked through an emergency situation, or a procedure for example like if you were in a remote area or the physician was in a remote area and you were in hospital and you would need some of that guidance or counselling, there's no substitution. (Interview Medical Student 1)
I'm ready can u dblchk [sic] that I landmarked correctly. (Email from Junior to Senior)
MD3 returns to the patient's room to do a paracentesis with [junior resident]. He calls on his BB to [senior resident] to inform her that they are starting and then hangs up. [Senior resident] arrives at the patient's room. (Field Notes, Workshadowing MD3)
Decreased autonomy
The difference with the Blackberry is they're more likely to say By the way, this happened. Should I do this?' And I write back Yes', No.' If they didn't have that contact like I said they probably would have done something and then because they're making a decision on their own they could very easily have spent the time to research whatever to figure whether that was the right thing to do before doing it. Now they have an outlet where they can pass an idea off of me and then have me make, it's easier for me to make a decision for them. So that can negatively impact education. (Interview Attending 1)
What do I do for a high phosphate?(Email from Junior to Senior)
Hey Pt X's k is 5.5. Was going to shift her. What do u think? (Email from Junior to Senior)
You probably saw the hb 92. Let's give prbc asap while he's on HD.(Email from Staff to Residents on the team)
hb‐ hemoglobin, prbc packed red blood cells, HD ‐ hemodialysis
Hi. Just checking the bloodwork. What is happening to ms X? [sic] Creatinine rising still. Is a foley in? Urology reconsulted? (Email from Attending Staff to Junior Resident)
On the other hand, supervisors perceived that the easy rapid access afforded by smartphone use lowered the threshold for trainees to contact them. In some instances, these attending physicians felt that their trainees would text them for advice when they could have looked up the information themselves. As a result, the increased reliance on the attending physician's input prior to committing to a management plan decreased the trainee's autonomy and independent decision making (Box 3: decreased autonomy). In addition to trainee requests for increased staff involvement, smartphone use made it easier for attending physicians to initiate text messages to their residents as well. In some instances, staff physicians adopted a more hands‐on approach by directing their residents on how to manage their patients. It is unclear if trainees perceived this taking over of care as negatively influencing their education.
Teaching
Medical teams also frequently used smartphones to communicate the location and timing of educational rounds. We observed instances where residents communicated updated information relating to scheduled rounds, as well as for informing team members about spontaneous teaching sessions (Box 4: communicating rounds). Despite this initial benefit, staff physicians worried that interruptions resulting from smartphone use during educational sessions lowered the effectiveness of these sessions for all learners by creating a fragmented learning experience (Box 4: fragmented learning). Our data indicated that residents carrying the team smartphones received and sent a high number of e‐mails throughout the day, which continued at a similar rate during the protected educational time (Figure 2). Additionally, some of the teaching experiences that traditionally would occur in a face‐to‐face manner appeared to have migrated to text‐based interactions. It is unclear whether trainees perceive these text‐based interactions as more or less effective teaching encounters (Box 4: text‐based teaching).
Box
Teaching
Communicating rounds
One is that they can more efficiently communicate about the timing and location of education rounds in case they forget or sort of as an organizer for them (Interview Attending 3)
Physical Exam rounds is at 1:00 outside the morning report room. K. has kindly volunteered! If you miss us then the exam will be on the 3rd floor in room X. Pt X. See you there (Email from chief medical resident to trainees)
Fragmented learning
Because Blackberry is there, it's something that is potentially time occupying and can take the attention away from things and this is true of any Blackberries. People who have Blackberries they always look at their Blackberries so, you know, there are times when I'm sitting face to face with people and residents are looking at their Blackberries. So it's another way that they can be distracted. (Interview Attending 1).
I've seen that be an issue. I've certainly seen them losing concentration during a teaching session because they're being Blackberried, getting Blackberry messages. (Interview Attending 3)
2:06 Team meeting with Attending in a conference room.
2:29 Team BlackBerry (BB) beeps. Senior glances at BB. She dials a number on the Team BB. Speaks on the Team BB and turns to [Junior resident] to inform her that the family is here. She returns to the caller. Senior then hangs up and resumes to her teaching.
2:35 Attending's BB rings. She takes a look and ends the BB call.
2:39 Senior's BB rings. Senior picks up and talks about a patient's case and condition. Senior turns to [junior resident] and asks a question. Team members resume talking among themselves.
2:46 Senior hangs up on the phone call.
2:49 Team discusses another patient's condition/case.
2:57 Junior resident uses her BB to text.
3:02 Team BB beeps. It is a message about a patient's case.
3:05 Meeting ends. (Field Note excerpts, Work‐shadowing MD6)
Text‐based teaching
The resident would get very frustrated with how many questions we have once we've started. Like if three different medical students or four different medical students or four different places all texting him with, oh by the way, what does this stand for?, and he's responding to each of them individually then he has to answer it four different times as opposed to just in person when he can get us all together in a group and it's actually a learning experience. If questions are answered in an email, it's not really helpful for the rest of us. (Interview, medical student SB1)
That would be a great unifying diagnosis, but there may be some underlying element of ROH/NASH also I would hold off on A/C as we do not know if he has varices. Will need to review noncontrast CT ?HCC. Thx (Email from Consulting Staff to Junior)
A/C anticoagulation, CT computed tomography, HCC‐ hepatocellular carcinoma, NASH non‐alcoholic steatohepatitis, ROH alcohol
Professionalism
Our data revealed that smartphone interruptions occurred during teaching rounds and interactions with patients and with other clinical staff. Often these interruptions involved messages or phone calls pertaining to clinical concerns or tasks that nurses communicated to the residents via their smartphone (Box 5). Yet, by responding to these interruptions and initiating communications on their smartphones during patient care encounters and formal teaching sessions, trainees were perceived by other clinicians who were in attendance with them as being rude or disrespectful. Attending staff also tended to role model similar smartphone behaviors. Although we did not specifically work‐shadow attending staff, we did observe frequent usage of their personal smartphones during their interactions with residents.
Box
Professionalism
I don't like it when I see them checking messages when you're trying to talk to them. I think you're losing some of that communication sort of polite behaviour that maybe we knew a little bit more before all this texting and Blackberry. (Interview Allied Health 5)
I think that the etiquette of the Blackberry can be offensive, could be offensive especially with some of our older patients (Interview Allied Health 6)
Senior walks out of the patient's room while typing on the BlackBerry. She finishes typing and returns to the room at 5:36. Senior looks at her BlackBerry and starts typing inside the room in front of the patient. She paused to look at the patient and the resident doing the procedure [paracentesis]. She resumes texting again and walks out of the room at 5:38. Another resident walks out and Senior speaks with the resident. Senior returns to the room and speaks with the patient. She asks the patient if he has ever gotten a successful tap before. Senior looks at her BlackBerry and starts typing. (Field NotesWork‐shadowing MD2)
I think it is almost completely negative in terms of its medical education [Any positive] factors are grossly outweighed by the significant disruptions to their ability to concentrate and participate in the educational session. And I think almost to some extent it's an implicit permission that gets granted to the house staff to disrupt their own teaching experience and disrupt others around them because everybody is doing it because everybody is being Blackberried. So it almost becomes the new social norm and while that may be a new social norm I'm not sure that that's a good thing How big is the negative impact? That's much harder to say. It's probably not a big impact on top of the endless other disruptions in the day to teaching, but it is measurable because it's a new factor so it's observable by me on top of all the other factors which have been there for years. (Interview Attending 3)
2:10‐Attending goes to the whiteboard to teach research methods to the team. Spotted Medical student#1 looking at his IPhone and typing.
2:15‐Med student#1 using the calculator function on his IPhone.
2:20‐Attending glances at his BB quickly.
2:28‐Attending resumes discussion of the patients' cases (Field notes, Work‐shadowing MD7).
DISCUSSION
The educational impacts of smartphone use for communication appear to center on increased connectedness of medical trainees and increased interruptions, which have positive and negative impacts in the areas of teaching, supervision, and professionalism. Smartphone communication provided potential educational benefits through (1) safer supervision with rapid access to help and (2) easier coordination of teaching sessions. Threats to the educational experience included (1) a high level of interruptions to both teachers and learners, which may reduce the effectiveness of formal and informal teaching; (2) replacement of face‐to‐face teaching with texting; (3) a potential erosion of autonomy and independence due to easy access to supervisors and easy ability for supervisors to take over; and (4) professionalism issues with difficulties balancing between clinical service demands and communication during patient and interprofessional encounters.
This study is the first to describe the intersection of clinical communication with smartphones and medical education. A recent study found that residents reported high use of smartphones during rounds for patient care as well as personal issues.[13] We have previously described the perceived impacts of smartphones on clinical communication, which included improved efficiency but concerns for increased interruptions and threats to professionalism.[6] We also observed that sites that used smartphones had increased interruptions compared to those with just pagers.[10] We have also described the content of e‐mail messages between clinicians and found that all e‐mails from nurses to physicians involved clinical care, but e‐mail exchanges between physicians were split between clinical care (60.4%), coordination within the team (53.5%), medical education (9.4%), and social communication (3.9%).[14] This study adds to the literature by focusing on the impacts of smartphone use to medical education and describing the perceived and observed impacts. This study provides a further example of how healthcare information technology can cause unintended consequences on medical education and appear to relate to the linked impacts of increased connectivity and the increased interruptions.[3] In essence, the trainee becomes more global, less local. Being more global translates to increased connections with people separated in physical space. Yet, this increased global connectedness resulted in the trainee being less local, with attention diverted elsewhere, taking away from the quality of patient interactions and interactions with other interprofessional team members. It also reduces the effectiveness of educational sessions for all participants. Although the level of supervision and autonomy are independently related, being too connected to supervisors may lower trainee autonomy by reducing independent thinking around patient issues.[15] It may also move teaching and learning from face‐to‐face conversations to text‐based messages. Although there have been existing tensions between service delivery and medical education, increased connectedness may tilt the balance toward the demands of service delivery and efficiency optimization at the expense of the educational experience. Finally, smartphone use appeared to create an internal tension among trainees, who have to juggle balancing professional behaviors and expectations in their dual role as learner and care provider; it would be educationally unprofessional to interrupt a teaching session and respond to a text message. However, failing to respond to a nurse who has sent a message and is expecting a response would be clinically unprofessional.
To address these threats, we advocate improving systems and processes to reduce interruptions and provide education on the tensions created by increased connectedness. Smarter communication systems could limit interrupting messages to urgent messages and queue nonurgent messages.[16] They could also inform senders about protected educational time. Even more sophisticated systems could inform the sender on the status of the receiver. For example, systems could indicate if they are available or if they are busy in an educational session or an important meeting with a patient and their family. Processes to reduce interruptions include interprofessional consensus on what constitutes an urgent issue and giving explicit permission to learners to ignore their smartphones during educational sessions except for critical communications purposes. Finally, education around smartphone communication for both learners and teachers may help minimize threats to learner autonomy, to face‐to‐face teaching, and to professionalism.[17]
Our study has several limitations. We derived this information from a general study of the impact of smartphones on clinical communication. Our study can be seen as hypothesis generating, and further research is warranted to validate these findings. There may be limits to generalizability as all sites adopted similar communication processes that included centralizing communications to make it easier for senders to reach a responsible physician.
In conclusion, we have provided a summary of the impact of rapidly emerging information technology on the educational experience of medical trainees and identified both positive and negative impacts. Of note, the negative impacts appear to be related to being more global and less local and high interruptions. Further research is required to confirm these unintended consequences as well as to develop solutions to address them. Educators should be aware of these findings and the need to develop curriculum to address and manage the negative impacts of smartphone use in the clinical training environment.
Acknowledgments
Disclosure: Nothing to report.
- , . Smartphone app use among medical providers in ACGME training programs. J Med Syst. 2012;36:3135–3139.
- , , , et al. The use of smartphones for clinical communication on internal medicine wards. J Hosp Med. 2010;5:553–559.
- , , , , , . 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. 2011;18(1):82–90.
- , , , , . Types of unintended consequences related to computerized provider order entry. J Am Med Inform Assoc. 2006;13(5):547–556.
- , , , . “e‐Iatrogenesis”: the most critical unintended consequence of CPOE and other HIT. J Am Med Inform Assoc. 2007;14(3):387–388.
- , , , et al. An evaluation of the use of smartphones to communicate between clinicians: a mixed‐methods study. J Med Internet Res. 2011;13(3):e59.
- . Smartphones in clinical practice, medical education, and research. Arch Intern Med. 2011;171(14):1294–1296.
- , , , . Use of handheld computers in medical education. A systematic review. J Gen Intern Med. 2006;21(5):531–537.
- , , , . A review and a framework of handheld computer adoption in healthcare. Int J Med Inform. 2005;74(5):409–422.
- , , , 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 [published online ahead of print January 25, 2013]. J Am Med Inform Assoc. doi:10.1136/amiajnl‐2012‐001160.
- , , , , . Hospital doctors' workflow interruptions and activities: an observation study. BMJ Qual Saf. 2011;20(6):491–497.
- , , , et al. Computerised provider order entry and residency education in an academic medical centre. Med Educ. 2012;46:795–806.
- , , , . Smartphone use during inpatient attending rounds: Prevalence, patterns and potential for distraction. J Hosp Med. 2012;7(8):595–599.
- , , , et al. Understanding interprofessional communication: a content analysis of email communications between doctors and nurses. Applied Clinical Informatics. 2012;3:38–51.
- , , , . Preserving professional credibility: grounded theory study of medical trainees' requests for clinical support. BMJ. 2009;338:b128.
- , , , , . Beyond paging: building a web‐based communication tool for nurses and physicians. J Gen Intern Med. 2009;24(1):105–110.
- , . Distracted doctoring: smartphones before patients? CMAJ. 2012;184:1440.
Medical residents are rapidly adopting smartphones. Recent statistics revealed that 85% of medical providers currently own a smartphone, and the majority use it in their clinical work.[1] Smartphone capabilities that include the use of text messaging, e‐mail, and mobile phone functions in the clinical setting may improve efficiency and quality of care by reducing the response time for urgent issues.[2] There is, however, increasing recognition that healthcare information technology can create unintended negative consequences. For example, studies have suggested that healthcare information technologies, such as the computerized physician order entry, may actually increase errors by creating new work, changing clinical workflow, and altering communication patterns.[3, 4, 5]
Smartphone use for clinical communication can have unintended consequences by increasing interruptions, reducing interprofessional relationships, and widening the gap between what nurses and physicians perceive as urgent clinical problems.[6] However, no studies have evaluated the impact of smartphones on the educational experience of medical trainees. Although previous studies have described the use of smartphones by trainees for rapid access to electronic medical resources,[7, 8, 9] we did not identify in our literature review any previous studies on the impact of using the smartphone's primary functionas a communication deviceon the educational experience of residents and medical students. Therefore, our study aimed to examine the impact of using smartphones for clinical communication on medical education.
METHODS
Design
The design of the study was qualitative research methodology using interview data, ethnographic data, and content analysis of text‐based messages.
Setting
From June 2009 to September 2010, we conducted a multisite evaluation study on general internal medicine (GIM) wards at 5 large academic teaching hospitals in the city of Toronto, Canada at St. Michael's Hospital, Sunnybrook Health Sciences Centre, Toronto General Hospital, Toronto Western Hospital, and Mount Sinai Hospital. Each hospital has clinical teaching units consisting typically of 4 medical teams. Each team includes 1 attending physician, 1 senior resident, 2 or more junior residents, and 2 to 4 medical students. Each hospital had 2 to 4 GIM wards in different geographic locations.
Communication Systems
To make it easier for nurses and other health professionals to communicate with the physician teams, all sites centralized communication to 1 team member, who acts as the single point of contact on behalf of their assigned team in the communication of patient‐related issues. We facilitated this communication through a shared device (either a pager or a smartphone). The senior resident typically carried the shared device during the day and the on‐call junior resident at night and on the weekends. Two hospitals provided smartphones to all residents, whereas a third site provided smartphones only to the senior residents. The standard processes of communication required that physicians respond to all calls and text messages. At the 3 sites with institutional smartphones, nurses could send text messages with patient information using a Web‐based system. We encrypted data sent to institutional smartphones to protect patient information.
Data Collection
Using a mixed‐methods ethnographic approach, we collected data using semistructured interviews, ethnographic observations, and content analysis of text messages. The original larger study focused primarily on examining the overall clinical impact of smartphone use.[10] For our current study, we analyzed the data with a focus on evaluating the impact of smartphones on the educational experience of medical trainees on the GIM teaching service. The respective institutions' research ethics boards approved the study.
Interviews
We conducted semistructured interviews with residents, medical students, attending physicians, and other clinicians across all of the sites to examine how clinicians perceived the impact of smartphones on medical education. We used a purposeful sampling strategy where we interviewed different groups of healthcare professionals who we suspected would represent different viewpoints on the use of smartphones for clinical communication. To obtain diverse perspectives, we snowball sampled by asking interviewees to suggest colleagues with differing views to participate in the interviews. The interview guide consisted of open‐ended questions with additional probes to elicit more detailed information from these frontline clinicians who initiate and receive communication. One of the study investigators (V.L.) conducted interviews that varied from 15 to 45 minutes in duration. We recorded, transcribed verbatim, and analyzed the interviews using NVivo software (QSR International, Doncaster, Victoria, Australia). We added additional questions iteratively as themes emerged from the initial interviews. One of the study investigators (V.L.) encouraged participants to speak freely, to raise issues that they perceived to be important, and to support their responses with examples.
Observations
We observed the communication processes in the hospitals by conducting a work‐shadowing approach that followed individual residents in their work environments. These observations included 1‐on‐1 supervision encounters involving attending staff, medical students, and other residents, and informal and formal teaching rounds. The observation periods included the usual working day (from 8 am to 6 pm) as well as the busiest times on call, typically from 6 pm until 11 pm. We sampled different residents for different time periods. We adopted a nonparticipatory observation technique where we observed all interruptions, communication interactions, and patterns from a distance. We defined workflow interruptions as an intrusion of an unplanned and unscheduled task, causing a discontinuation of tasks, a noticeable break, or task switch behaviour.[11] Data collection included timing of events and writing field notes. One of the study investigators (V.L.) performed all the work‐shadowing observations.
E‐mail
To study the volume and content of messages, we collected e‐mail communications between January 2009 and June 2009 from consenting residents at the 2 hospitals that provided smartphones to all GIM residents. E‐mail information included the sender, the receiver, the time of message, and the message content. To look at usage, we calculated the average number of e‐mails sent and received. To assess interruptions on formal teaching sessions, we paid particular attention to e‐mails received and sent during protected educational timeweekdays from 8 am to 9 am (morning report) and 12 pm to 1 pm (noon rounds). We randomly sampled 20% of all e‐mails sent between residents for content analysis and organized content related to medical education into thematic categories.
Analysis
We used a deductive approach to analyze the interview transcripts by applying a conceptual framework that assessed the educational impact of patient safety interventions.[12] This framework identified 5 educational domains (learning, teaching, supervision, assessment, and feedback). Three study investigators mapped interview data, work‐shadowing data, and e‐mail content to themes (V.L., B.W., and R.W.), and grouped data that did not translate into themes into new categories. We then triangulated the data to develop themes of the educational impact of smartphone communication by both perceived use and actual use, and subsequently constructed a framework of how smartphone communication affected education.
RESULTS
We conducted 124 semistructured interviews with residents, medical students, attending physicians, and other clinicians across all the sites to examine how clinicians perceived the impact of smartphones on medical education. We work‐shadowed 40 individual residents for a total of 196 hours (Table 1). We analyzed the 13,714 e‐mails sent from or received to 34 residents. To analyze e‐mail content, we reviewed 1179 e‐mails sent among residents.
| Methods | Sites | |||||
|---|---|---|---|---|---|---|
| St. Michael's Hospital | Sunnybrook Health Sciences Centre | Toronto General Hospital | Toronto Western Hospital | Mount Sinai Hospital | All Hospitals | |
| ||||||
| Work‐shadowing residents | ||||||
| Hours | 60 hours | 35 hours | 57 hours 55 minutes | 27 hours 46 minutes | 15 hours | 196 hours |
| No. of residents | 12 | 7 | 12 | 6 | 3 | 40 |
| Interviews with clinicians | ||||||
| Physicians | 10 | 5 | 13 | 5 | 33 | |
| Medical students | 5 | 4 | 1 | 1 | 11 | |
| Nurses | 9 | 11 | 15 | 14 | 49 | |
| Other health professionsa | 7 | 10 | 8 | 6 | 31 | |
| Total | 31 | 30 | 37 | 26 | 124 | |
We found that 2 key characteristics of smartphone use for clinical communication, namely an increase in connectedness leading to an increase in interruptions, impacted 3 educational domains: teaching, supervision, and professionalism (Figure 1).
Increased Connectedness
As a communication device, smartphones increase the ability to receive and respond to messages through voice, e‐mail, and text messaging. Not surprisingly, with the improved ability and mobility to communicate, medical trainees perceived being more connected with their team members, who included other residents, medical students, and attending staff as well as with other clinical services and professions. These smartphone communication activities appeared to be pervasive, occurring on the wards, at the bedside, while in transit, and in teaching sessions (Box 1: increased connectedness).
Box
Increased connectedness
I've used the Blackberry system and it's nice to be able to quickly text each other little messages especially for meeting times because then you don't have to page them and wait by the phone. So that's been great for in the team. (Interview Resident 3)
It's incredibly useful for when you're paging somebody else. Often times I'll be consulting with another physician on a patient and I'll say This is my BlackBerry. Call me back after you've seen the patient' or Call me back when you have a plan' or, you know, whatever. So that's extremely valuable which we never had with pages and no one would ever page you for that because it was too much of a pain. (Interview Resident 1)
My personal experience has been that if you need to speak to a more senior individual it's much easier to contact them via the BlackBerry. (Interview Medical Student 1)
At 7:25 pm, MD11 returns to the patient's room and continues examining her. While in the patient's room, I could see her talking on the BlackBerrys. I asked her later what calls she had while in the room. It turns out she had 3 phone calls and 2 texts. Two of the calls were from the radiation oncologists and 1 call from the pathologist. She also received 1 text on the Team BlackBerry and 1 text on the Senior's BlackBerry from the pharmacist. (Field Notes, Work-shadowing MD11)
Interruptions
The increased connectedness caused by smartphone use led residents to perceive an increase in the frequency of interruptions. The multitude of communication and contact options made available by smartphones to health providers created an expansive network of connected individuals who were in constant communication with each other. Instead of the difficulties associated with numeric paging and waiting for a response, nurses typically found it easier to call directly or send a text message to residents' smartphones. From the e‐mail analysis, residents received, on a daily basis, on average 25.7 e‐mails, (median, 20; interquartile range [IQR]: 1428) to the team smartphone and sent 7.5 e‐mails (median, 6; IQR: 410). During protected educational time, each resident received an average of 1.0 e‐mail (median, 1; IQR: 01) between 8 am and 9 am and an average of 2.3 e‐mails (median, 2; IQR: 13) during 12 pm to 1 pm (Figure 2). Each of these communication events, whether a phone call, e‐mail, or text‐message, led to an interruption (Box 2). Given that smartphones made it easier for nurses to contact residents, some residents attributed the increase in interruptions to a reduction in the threshold for nurses to communicate.
Box
Increased interruptions
The only negative I can think of is just the incredible number of communications that you get, you know, text messages and e‐mails and everything else. So just the, the number can sometimes be overwhelming. (Interview Resident 1)
Some of [the nurses] rely a little bit more on the BlackBerry so that they will tend to call you a bit more frequently for things that maybe sometimes they should try to find answer for themselves (Interview Resident 2)
And now with the option of being able to, if you really needed to, call them and talk to them directly, I think that kind of improves communication. They're easier to find. (Interview Nurse 4)
Supervision
Smartphone communication appeared to positively impact trainee supervision. Increased connectedness between team members allowed junior trainees to have access and rapidly communicate with a more experienced clinician, which provided them with greater support. Residents found smartphones particularly useful in situations where they felt uncomfortable or where they did not feel competent. Some of these instances related to procedural competence, with residents feeling more comfortable knowing they have rapid access to support (Box 3: increased support).
Box
Supervision
Increased support
It makes me feel more comfortable in the sense that I can instantly make a call or a text and have a question answered if I need an answer. Or if it were an emergency having the ability to talk on the phone and be talked through an emergency situation, or a procedure for example like if you were in a remote area or the physician was in a remote area and you were in hospital and you would need some of that guidance or counselling, there's no substitution. (Interview Medical Student 1)
I'm ready can u dblchk [sic] that I landmarked correctly. (Email from Junior to Senior)
MD3 returns to the patient's room to do a paracentesis with [junior resident]. He calls on his BB to [senior resident] to inform her that they are starting and then hangs up. [Senior resident] arrives at the patient's room. (Field Notes, Workshadowing MD3)
Decreased autonomy
The difference with the Blackberry is they're more likely to say By the way, this happened. Should I do this?' And I write back Yes', No.' If they didn't have that contact like I said they probably would have done something and then because they're making a decision on their own they could very easily have spent the time to research whatever to figure whether that was the right thing to do before doing it. Now they have an outlet where they can pass an idea off of me and then have me make, it's easier for me to make a decision for them. So that can negatively impact education. (Interview Attending 1)
What do I do for a high phosphate?(Email from Junior to Senior)
Hey Pt X's k is 5.5. Was going to shift her. What do u think? (Email from Junior to Senior)
You probably saw the hb 92. Let's give prbc asap while he's on HD.(Email from Staff to Residents on the team)
hb‐ hemoglobin, prbc packed red blood cells, HD ‐ hemodialysis
Hi. Just checking the bloodwork. What is happening to ms X? [sic] Creatinine rising still. Is a foley in? Urology reconsulted? (Email from Attending Staff to Junior Resident)
On the other hand, supervisors perceived that the easy rapid access afforded by smartphone use lowered the threshold for trainees to contact them. In some instances, these attending physicians felt that their trainees would text them for advice when they could have looked up the information themselves. As a result, the increased reliance on the attending physician's input prior to committing to a management plan decreased the trainee's autonomy and independent decision making (Box 3: decreased autonomy). In addition to trainee requests for increased staff involvement, smartphone use made it easier for attending physicians to initiate text messages to their residents as well. In some instances, staff physicians adopted a more hands‐on approach by directing their residents on how to manage their patients. It is unclear if trainees perceived this taking over of care as negatively influencing their education.
Teaching
Medical teams also frequently used smartphones to communicate the location and timing of educational rounds. We observed instances where residents communicated updated information relating to scheduled rounds, as well as for informing team members about spontaneous teaching sessions (Box 4: communicating rounds). Despite this initial benefit, staff physicians worried that interruptions resulting from smartphone use during educational sessions lowered the effectiveness of these sessions for all learners by creating a fragmented learning experience (Box 4: fragmented learning). Our data indicated that residents carrying the team smartphones received and sent a high number of e‐mails throughout the day, which continued at a similar rate during the protected educational time (Figure 2). Additionally, some of the teaching experiences that traditionally would occur in a face‐to‐face manner appeared to have migrated to text‐based interactions. It is unclear whether trainees perceive these text‐based interactions as more or less effective teaching encounters (Box 4: text‐based teaching).
Box
Teaching
Communicating rounds
One is that they can more efficiently communicate about the timing and location of education rounds in case they forget or sort of as an organizer for them (Interview Attending 3)
Physical Exam rounds is at 1:00 outside the morning report room. K. has kindly volunteered! If you miss us then the exam will be on the 3rd floor in room X. Pt X. See you there (Email from chief medical resident to trainees)
Fragmented learning
Because Blackberry is there, it's something that is potentially time occupying and can take the attention away from things and this is true of any Blackberries. People who have Blackberries they always look at their Blackberries so, you know, there are times when I'm sitting face to face with people and residents are looking at their Blackberries. So it's another way that they can be distracted. (Interview Attending 1).
I've seen that be an issue. I've certainly seen them losing concentration during a teaching session because they're being Blackberried, getting Blackberry messages. (Interview Attending 3)
2:06 Team meeting with Attending in a conference room.
2:29 Team BlackBerry (BB) beeps. Senior glances at BB. She dials a number on the Team BB. Speaks on the Team BB and turns to [Junior resident] to inform her that the family is here. She returns to the caller. Senior then hangs up and resumes to her teaching.
2:35 Attending's BB rings. She takes a look and ends the BB call.
2:39 Senior's BB rings. Senior picks up and talks about a patient's case and condition. Senior turns to [junior resident] and asks a question. Team members resume talking among themselves.
2:46 Senior hangs up on the phone call.
2:49 Team discusses another patient's condition/case.
2:57 Junior resident uses her BB to text.
3:02 Team BB beeps. It is a message about a patient's case.
3:05 Meeting ends. (Field Note excerpts, Work‐shadowing MD6)
Text‐based teaching
The resident would get very frustrated with how many questions we have once we've started. Like if three different medical students or four different medical students or four different places all texting him with, oh by the way, what does this stand for?, and he's responding to each of them individually then he has to answer it four different times as opposed to just in person when he can get us all together in a group and it's actually a learning experience. If questions are answered in an email, it's not really helpful for the rest of us. (Interview, medical student SB1)
That would be a great unifying diagnosis, but there may be some underlying element of ROH/NASH also I would hold off on A/C as we do not know if he has varices. Will need to review noncontrast CT ?HCC. Thx (Email from Consulting Staff to Junior)
A/C anticoagulation, CT computed tomography, HCC‐ hepatocellular carcinoma, NASH non‐alcoholic steatohepatitis, ROH alcohol
Professionalism
Our data revealed that smartphone interruptions occurred during teaching rounds and interactions with patients and with other clinical staff. Often these interruptions involved messages or phone calls pertaining to clinical concerns or tasks that nurses communicated to the residents via their smartphone (Box 5). Yet, by responding to these interruptions and initiating communications on their smartphones during patient care encounters and formal teaching sessions, trainees were perceived by other clinicians who were in attendance with them as being rude or disrespectful. Attending staff also tended to role model similar smartphone behaviors. Although we did not specifically work‐shadow attending staff, we did observe frequent usage of their personal smartphones during their interactions with residents.
Box
Professionalism
I don't like it when I see them checking messages when you're trying to talk to them. I think you're losing some of that communication sort of polite behaviour that maybe we knew a little bit more before all this texting and Blackberry. (Interview Allied Health 5)
I think that the etiquette of the Blackberry can be offensive, could be offensive especially with some of our older patients (Interview Allied Health 6)
Senior walks out of the patient's room while typing on the BlackBerry. She finishes typing and returns to the room at 5:36. Senior looks at her BlackBerry and starts typing inside the room in front of the patient. She paused to look at the patient and the resident doing the procedure [paracentesis]. She resumes texting again and walks out of the room at 5:38. Another resident walks out and Senior speaks with the resident. Senior returns to the room and speaks with the patient. She asks the patient if he has ever gotten a successful tap before. Senior looks at her BlackBerry and starts typing. (Field NotesWork‐shadowing MD2)
I think it is almost completely negative in terms of its medical education [Any positive] factors are grossly outweighed by the significant disruptions to their ability to concentrate and participate in the educational session. And I think almost to some extent it's an implicit permission that gets granted to the house staff to disrupt their own teaching experience and disrupt others around them because everybody is doing it because everybody is being Blackberried. So it almost becomes the new social norm and while that may be a new social norm I'm not sure that that's a good thing How big is the negative impact? That's much harder to say. It's probably not a big impact on top of the endless other disruptions in the day to teaching, but it is measurable because it's a new factor so it's observable by me on top of all the other factors which have been there for years. (Interview Attending 3)
2:10‐Attending goes to the whiteboard to teach research methods to the team. Spotted Medical student#1 looking at his IPhone and typing.
2:15‐Med student#1 using the calculator function on his IPhone.
2:20‐Attending glances at his BB quickly.
2:28‐Attending resumes discussion of the patients' cases (Field notes, Work‐shadowing MD7).
DISCUSSION
The educational impacts of smartphone use for communication appear to center on increased connectedness of medical trainees and increased interruptions, which have positive and negative impacts in the areas of teaching, supervision, and professionalism. Smartphone communication provided potential educational benefits through (1) safer supervision with rapid access to help and (2) easier coordination of teaching sessions. Threats to the educational experience included (1) a high level of interruptions to both teachers and learners, which may reduce the effectiveness of formal and informal teaching; (2) replacement of face‐to‐face teaching with texting; (3) a potential erosion of autonomy and independence due to easy access to supervisors and easy ability for supervisors to take over; and (4) professionalism issues with difficulties balancing between clinical service demands and communication during patient and interprofessional encounters.
This study is the first to describe the intersection of clinical communication with smartphones and medical education. A recent study found that residents reported high use of smartphones during rounds for patient care as well as personal issues.[13] We have previously described the perceived impacts of smartphones on clinical communication, which included improved efficiency but concerns for increased interruptions and threats to professionalism.[6] We also observed that sites that used smartphones had increased interruptions compared to those with just pagers.[10] We have also described the content of e‐mail messages between clinicians and found that all e‐mails from nurses to physicians involved clinical care, but e‐mail exchanges between physicians were split between clinical care (60.4%), coordination within the team (53.5%), medical education (9.4%), and social communication (3.9%).[14] This study adds to the literature by focusing on the impacts of smartphone use to medical education and describing the perceived and observed impacts. This study provides a further example of how healthcare information technology can cause unintended consequences on medical education and appear to relate to the linked impacts of increased connectivity and the increased interruptions.[3] In essence, the trainee becomes more global, less local. Being more global translates to increased connections with people separated in physical space. Yet, this increased global connectedness resulted in the trainee being less local, with attention diverted elsewhere, taking away from the quality of patient interactions and interactions with other interprofessional team members. It also reduces the effectiveness of educational sessions for all participants. Although the level of supervision and autonomy are independently related, being too connected to supervisors may lower trainee autonomy by reducing independent thinking around patient issues.[15] It may also move teaching and learning from face‐to‐face conversations to text‐based messages. Although there have been existing tensions between service delivery and medical education, increased connectedness may tilt the balance toward the demands of service delivery and efficiency optimization at the expense of the educational experience. Finally, smartphone use appeared to create an internal tension among trainees, who have to juggle balancing professional behaviors and expectations in their dual role as learner and care provider; it would be educationally unprofessional to interrupt a teaching session and respond to a text message. However, failing to respond to a nurse who has sent a message and is expecting a response would be clinically unprofessional.
To address these threats, we advocate improving systems and processes to reduce interruptions and provide education on the tensions created by increased connectedness. Smarter communication systems could limit interrupting messages to urgent messages and queue nonurgent messages.[16] They could also inform senders about protected educational time. Even more sophisticated systems could inform the sender on the status of the receiver. For example, systems could indicate if they are available or if they are busy in an educational session or an important meeting with a patient and their family. Processes to reduce interruptions include interprofessional consensus on what constitutes an urgent issue and giving explicit permission to learners to ignore their smartphones during educational sessions except for critical communications purposes. Finally, education around smartphone communication for both learners and teachers may help minimize threats to learner autonomy, to face‐to‐face teaching, and to professionalism.[17]
Our study has several limitations. We derived this information from a general study of the impact of smartphones on clinical communication. Our study can be seen as hypothesis generating, and further research is warranted to validate these findings. There may be limits to generalizability as all sites adopted similar communication processes that included centralizing communications to make it easier for senders to reach a responsible physician.
In conclusion, we have provided a summary of the impact of rapidly emerging information technology on the educational experience of medical trainees and identified both positive and negative impacts. Of note, the negative impacts appear to be related to being more global and less local and high interruptions. Further research is required to confirm these unintended consequences as well as to develop solutions to address them. Educators should be aware of these findings and the need to develop curriculum to address and manage the negative impacts of smartphone use in the clinical training environment.
Acknowledgments
Disclosure: Nothing to report.
Medical residents are rapidly adopting smartphones. Recent statistics revealed that 85% of medical providers currently own a smartphone, and the majority use it in their clinical work.[1] Smartphone capabilities that include the use of text messaging, e‐mail, and mobile phone functions in the clinical setting may improve efficiency and quality of care by reducing the response time for urgent issues.[2] There is, however, increasing recognition that healthcare information technology can create unintended negative consequences. For example, studies have suggested that healthcare information technologies, such as the computerized physician order entry, may actually increase errors by creating new work, changing clinical workflow, and altering communication patterns.[3, 4, 5]
Smartphone use for clinical communication can have unintended consequences by increasing interruptions, reducing interprofessional relationships, and widening the gap between what nurses and physicians perceive as urgent clinical problems.[6] However, no studies have evaluated the impact of smartphones on the educational experience of medical trainees. Although previous studies have described the use of smartphones by trainees for rapid access to electronic medical resources,[7, 8, 9] we did not identify in our literature review any previous studies on the impact of using the smartphone's primary functionas a communication deviceon the educational experience of residents and medical students. Therefore, our study aimed to examine the impact of using smartphones for clinical communication on medical education.
METHODS
Design
The design of the study was qualitative research methodology using interview data, ethnographic data, and content analysis of text‐based messages.
Setting
From June 2009 to September 2010, we conducted a multisite evaluation study on general internal medicine (GIM) wards at 5 large academic teaching hospitals in the city of Toronto, Canada at St. Michael's Hospital, Sunnybrook Health Sciences Centre, Toronto General Hospital, Toronto Western Hospital, and Mount Sinai Hospital. Each hospital has clinical teaching units consisting typically of 4 medical teams. Each team includes 1 attending physician, 1 senior resident, 2 or more junior residents, and 2 to 4 medical students. Each hospital had 2 to 4 GIM wards in different geographic locations.
Communication Systems
To make it easier for nurses and other health professionals to communicate with the physician teams, all sites centralized communication to 1 team member, who acts as the single point of contact on behalf of their assigned team in the communication of patient‐related issues. We facilitated this communication through a shared device (either a pager or a smartphone). The senior resident typically carried the shared device during the day and the on‐call junior resident at night and on the weekends. Two hospitals provided smartphones to all residents, whereas a third site provided smartphones only to the senior residents. The standard processes of communication required that physicians respond to all calls and text messages. At the 3 sites with institutional smartphones, nurses could send text messages with patient information using a Web‐based system. We encrypted data sent to institutional smartphones to protect patient information.
Data Collection
Using a mixed‐methods ethnographic approach, we collected data using semistructured interviews, ethnographic observations, and content analysis of text messages. The original larger study focused primarily on examining the overall clinical impact of smartphone use.[10] For our current study, we analyzed the data with a focus on evaluating the impact of smartphones on the educational experience of medical trainees on the GIM teaching service. The respective institutions' research ethics boards approved the study.
Interviews
We conducted semistructured interviews with residents, medical students, attending physicians, and other clinicians across all of the sites to examine how clinicians perceived the impact of smartphones on medical education. We used a purposeful sampling strategy where we interviewed different groups of healthcare professionals who we suspected would represent different viewpoints on the use of smartphones for clinical communication. To obtain diverse perspectives, we snowball sampled by asking interviewees to suggest colleagues with differing views to participate in the interviews. The interview guide consisted of open‐ended questions with additional probes to elicit more detailed information from these frontline clinicians who initiate and receive communication. One of the study investigators (V.L.) conducted interviews that varied from 15 to 45 minutes in duration. We recorded, transcribed verbatim, and analyzed the interviews using NVivo software (QSR International, Doncaster, Victoria, Australia). We added additional questions iteratively as themes emerged from the initial interviews. One of the study investigators (V.L.) encouraged participants to speak freely, to raise issues that they perceived to be important, and to support their responses with examples.
Observations
We observed the communication processes in the hospitals by conducting a work‐shadowing approach that followed individual residents in their work environments. These observations included 1‐on‐1 supervision encounters involving attending staff, medical students, and other residents, and informal and formal teaching rounds. The observation periods included the usual working day (from 8 am to 6 pm) as well as the busiest times on call, typically from 6 pm until 11 pm. We sampled different residents for different time periods. We adopted a nonparticipatory observation technique where we observed all interruptions, communication interactions, and patterns from a distance. We defined workflow interruptions as an intrusion of an unplanned and unscheduled task, causing a discontinuation of tasks, a noticeable break, or task switch behaviour.[11] Data collection included timing of events and writing field notes. One of the study investigators (V.L.) performed all the work‐shadowing observations.
E‐mail
To study the volume and content of messages, we collected e‐mail communications between January 2009 and June 2009 from consenting residents at the 2 hospitals that provided smartphones to all GIM residents. E‐mail information included the sender, the receiver, the time of message, and the message content. To look at usage, we calculated the average number of e‐mails sent and received. To assess interruptions on formal teaching sessions, we paid particular attention to e‐mails received and sent during protected educational timeweekdays from 8 am to 9 am (morning report) and 12 pm to 1 pm (noon rounds). We randomly sampled 20% of all e‐mails sent between residents for content analysis and organized content related to medical education into thematic categories.
Analysis
We used a deductive approach to analyze the interview transcripts by applying a conceptual framework that assessed the educational impact of patient safety interventions.[12] This framework identified 5 educational domains (learning, teaching, supervision, assessment, and feedback). Three study investigators mapped interview data, work‐shadowing data, and e‐mail content to themes (V.L., B.W., and R.W.), and grouped data that did not translate into themes into new categories. We then triangulated the data to develop themes of the educational impact of smartphone communication by both perceived use and actual use, and subsequently constructed a framework of how smartphone communication affected education.
RESULTS
We conducted 124 semistructured interviews with residents, medical students, attending physicians, and other clinicians across all the sites to examine how clinicians perceived the impact of smartphones on medical education. We work‐shadowed 40 individual residents for a total of 196 hours (Table 1). We analyzed the 13,714 e‐mails sent from or received to 34 residents. To analyze e‐mail content, we reviewed 1179 e‐mails sent among residents.
| Methods | Sites | |||||
|---|---|---|---|---|---|---|
| St. Michael's Hospital | Sunnybrook Health Sciences Centre | Toronto General Hospital | Toronto Western Hospital | Mount Sinai Hospital | All Hospitals | |
| ||||||
| Work‐shadowing residents | ||||||
| Hours | 60 hours | 35 hours | 57 hours 55 minutes | 27 hours 46 minutes | 15 hours | 196 hours |
| No. of residents | 12 | 7 | 12 | 6 | 3 | 40 |
| Interviews with clinicians | ||||||
| Physicians | 10 | 5 | 13 | 5 | 33 | |
| Medical students | 5 | 4 | 1 | 1 | 11 | |
| Nurses | 9 | 11 | 15 | 14 | 49 | |
| Other health professionsa | 7 | 10 | 8 | 6 | 31 | |
| Total | 31 | 30 | 37 | 26 | 124 | |
We found that 2 key characteristics of smartphone use for clinical communication, namely an increase in connectedness leading to an increase in interruptions, impacted 3 educational domains: teaching, supervision, and professionalism (Figure 1).
Increased Connectedness
As a communication device, smartphones increase the ability to receive and respond to messages through voice, e‐mail, and text messaging. Not surprisingly, with the improved ability and mobility to communicate, medical trainees perceived being more connected with their team members, who included other residents, medical students, and attending staff as well as with other clinical services and professions. These smartphone communication activities appeared to be pervasive, occurring on the wards, at the bedside, while in transit, and in teaching sessions (Box 1: increased connectedness).
Box
Increased connectedness
I've used the Blackberry system and it's nice to be able to quickly text each other little messages especially for meeting times because then you don't have to page them and wait by the phone. So that's been great for in the team. (Interview Resident 3)
It's incredibly useful for when you're paging somebody else. Often times I'll be consulting with another physician on a patient and I'll say This is my BlackBerry. Call me back after you've seen the patient' or Call me back when you have a plan' or, you know, whatever. So that's extremely valuable which we never had with pages and no one would ever page you for that because it was too much of a pain. (Interview Resident 1)
My personal experience has been that if you need to speak to a more senior individual it's much easier to contact them via the BlackBerry. (Interview Medical Student 1)
At 7:25 pm, MD11 returns to the patient's room and continues examining her. While in the patient's room, I could see her talking on the BlackBerrys. I asked her later what calls she had while in the room. It turns out she had 3 phone calls and 2 texts. Two of the calls were from the radiation oncologists and 1 call from the pathologist. She also received 1 text on the Team BlackBerry and 1 text on the Senior's BlackBerry from the pharmacist. (Field Notes, Work-shadowing MD11)
Interruptions
The increased connectedness caused by smartphone use led residents to perceive an increase in the frequency of interruptions. The multitude of communication and contact options made available by smartphones to health providers created an expansive network of connected individuals who were in constant communication with each other. Instead of the difficulties associated with numeric paging and waiting for a response, nurses typically found it easier to call directly or send a text message to residents' smartphones. From the e‐mail analysis, residents received, on a daily basis, on average 25.7 e‐mails, (median, 20; interquartile range [IQR]: 1428) to the team smartphone and sent 7.5 e‐mails (median, 6; IQR: 410). During protected educational time, each resident received an average of 1.0 e‐mail (median, 1; IQR: 01) between 8 am and 9 am and an average of 2.3 e‐mails (median, 2; IQR: 13) during 12 pm to 1 pm (Figure 2). Each of these communication events, whether a phone call, e‐mail, or text‐message, led to an interruption (Box 2). Given that smartphones made it easier for nurses to contact residents, some residents attributed the increase in interruptions to a reduction in the threshold for nurses to communicate.
Box
Increased interruptions
The only negative I can think of is just the incredible number of communications that you get, you know, text messages and e‐mails and everything else. So just the, the number can sometimes be overwhelming. (Interview Resident 1)
Some of [the nurses] rely a little bit more on the BlackBerry so that they will tend to call you a bit more frequently for things that maybe sometimes they should try to find answer for themselves (Interview Resident 2)
And now with the option of being able to, if you really needed to, call them and talk to them directly, I think that kind of improves communication. They're easier to find. (Interview Nurse 4)
Supervision
Smartphone communication appeared to positively impact trainee supervision. Increased connectedness between team members allowed junior trainees to have access and rapidly communicate with a more experienced clinician, which provided them with greater support. Residents found smartphones particularly useful in situations where they felt uncomfortable or where they did not feel competent. Some of these instances related to procedural competence, with residents feeling more comfortable knowing they have rapid access to support (Box 3: increased support).
Box
Supervision
Increased support
It makes me feel more comfortable in the sense that I can instantly make a call or a text and have a question answered if I need an answer. Or if it were an emergency having the ability to talk on the phone and be talked through an emergency situation, or a procedure for example like if you were in a remote area or the physician was in a remote area and you were in hospital and you would need some of that guidance or counselling, there's no substitution. (Interview Medical Student 1)
I'm ready can u dblchk [sic] that I landmarked correctly. (Email from Junior to Senior)
MD3 returns to the patient's room to do a paracentesis with [junior resident]. He calls on his BB to [senior resident] to inform her that they are starting and then hangs up. [Senior resident] arrives at the patient's room. (Field Notes, Workshadowing MD3)
Decreased autonomy
The difference with the Blackberry is they're more likely to say By the way, this happened. Should I do this?' And I write back Yes', No.' If they didn't have that contact like I said they probably would have done something and then because they're making a decision on their own they could very easily have spent the time to research whatever to figure whether that was the right thing to do before doing it. Now they have an outlet where they can pass an idea off of me and then have me make, it's easier for me to make a decision for them. So that can negatively impact education. (Interview Attending 1)
What do I do for a high phosphate?(Email from Junior to Senior)
Hey Pt X's k is 5.5. Was going to shift her. What do u think? (Email from Junior to Senior)
You probably saw the hb 92. Let's give prbc asap while he's on HD.(Email from Staff to Residents on the team)
hb‐ hemoglobin, prbc packed red blood cells, HD ‐ hemodialysis
Hi. Just checking the bloodwork. What is happening to ms X? [sic] Creatinine rising still. Is a foley in? Urology reconsulted? (Email from Attending Staff to Junior Resident)
On the other hand, supervisors perceived that the easy rapid access afforded by smartphone use lowered the threshold for trainees to contact them. In some instances, these attending physicians felt that their trainees would text them for advice when they could have looked up the information themselves. As a result, the increased reliance on the attending physician's input prior to committing to a management plan decreased the trainee's autonomy and independent decision making (Box 3: decreased autonomy). In addition to trainee requests for increased staff involvement, smartphone use made it easier for attending physicians to initiate text messages to their residents as well. In some instances, staff physicians adopted a more hands‐on approach by directing their residents on how to manage their patients. It is unclear if trainees perceived this taking over of care as negatively influencing their education.
Teaching
Medical teams also frequently used smartphones to communicate the location and timing of educational rounds. We observed instances where residents communicated updated information relating to scheduled rounds, as well as for informing team members about spontaneous teaching sessions (Box 4: communicating rounds). Despite this initial benefit, staff physicians worried that interruptions resulting from smartphone use during educational sessions lowered the effectiveness of these sessions for all learners by creating a fragmented learning experience (Box 4: fragmented learning). Our data indicated that residents carrying the team smartphones received and sent a high number of e‐mails throughout the day, which continued at a similar rate during the protected educational time (Figure 2). Additionally, some of the teaching experiences that traditionally would occur in a face‐to‐face manner appeared to have migrated to text‐based interactions. It is unclear whether trainees perceive these text‐based interactions as more or less effective teaching encounters (Box 4: text‐based teaching).
Box
Teaching
Communicating rounds
One is that they can more efficiently communicate about the timing and location of education rounds in case they forget or sort of as an organizer for them (Interview Attending 3)
Physical Exam rounds is at 1:00 outside the morning report room. K. has kindly volunteered! If you miss us then the exam will be on the 3rd floor in room X. Pt X. See you there (Email from chief medical resident to trainees)
Fragmented learning
Because Blackberry is there, it's something that is potentially time occupying and can take the attention away from things and this is true of any Blackberries. People who have Blackberries they always look at their Blackberries so, you know, there are times when I'm sitting face to face with people and residents are looking at their Blackberries. So it's another way that they can be distracted. (Interview Attending 1).
I've seen that be an issue. I've certainly seen them losing concentration during a teaching session because they're being Blackberried, getting Blackberry messages. (Interview Attending 3)
2:06 Team meeting with Attending in a conference room.
2:29 Team BlackBerry (BB) beeps. Senior glances at BB. She dials a number on the Team BB. Speaks on the Team BB and turns to [Junior resident] to inform her that the family is here. She returns to the caller. Senior then hangs up and resumes to her teaching.
2:35 Attending's BB rings. She takes a look and ends the BB call.
2:39 Senior's BB rings. Senior picks up and talks about a patient's case and condition. Senior turns to [junior resident] and asks a question. Team members resume talking among themselves.
2:46 Senior hangs up on the phone call.
2:49 Team discusses another patient's condition/case.
2:57 Junior resident uses her BB to text.
3:02 Team BB beeps. It is a message about a patient's case.
3:05 Meeting ends. (Field Note excerpts, Work‐shadowing MD6)
Text‐based teaching
The resident would get very frustrated with how many questions we have once we've started. Like if three different medical students or four different medical students or four different places all texting him with, oh by the way, what does this stand for?, and he's responding to each of them individually then he has to answer it four different times as opposed to just in person when he can get us all together in a group and it's actually a learning experience. If questions are answered in an email, it's not really helpful for the rest of us. (Interview, medical student SB1)
That would be a great unifying diagnosis, but there may be some underlying element of ROH/NASH also I would hold off on A/C as we do not know if he has varices. Will need to review noncontrast CT ?HCC. Thx (Email from Consulting Staff to Junior)
A/C anticoagulation, CT computed tomography, HCC‐ hepatocellular carcinoma, NASH non‐alcoholic steatohepatitis, ROH alcohol
Professionalism
Our data revealed that smartphone interruptions occurred during teaching rounds and interactions with patients and with other clinical staff. Often these interruptions involved messages or phone calls pertaining to clinical concerns or tasks that nurses communicated to the residents via their smartphone (Box 5). Yet, by responding to these interruptions and initiating communications on their smartphones during patient care encounters and formal teaching sessions, trainees were perceived by other clinicians who were in attendance with them as being rude or disrespectful. Attending staff also tended to role model similar smartphone behaviors. Although we did not specifically work‐shadow attending staff, we did observe frequent usage of their personal smartphones during their interactions with residents.
Box
Professionalism
I don't like it when I see them checking messages when you're trying to talk to them. I think you're losing some of that communication sort of polite behaviour that maybe we knew a little bit more before all this texting and Blackberry. (Interview Allied Health 5)
I think that the etiquette of the Blackberry can be offensive, could be offensive especially with some of our older patients (Interview Allied Health 6)
Senior walks out of the patient's room while typing on the BlackBerry. She finishes typing and returns to the room at 5:36. Senior looks at her BlackBerry and starts typing inside the room in front of the patient. She paused to look at the patient and the resident doing the procedure [paracentesis]. She resumes texting again and walks out of the room at 5:38. Another resident walks out and Senior speaks with the resident. Senior returns to the room and speaks with the patient. She asks the patient if he has ever gotten a successful tap before. Senior looks at her BlackBerry and starts typing. (Field NotesWork‐shadowing MD2)
I think it is almost completely negative in terms of its medical education [Any positive] factors are grossly outweighed by the significant disruptions to their ability to concentrate and participate in the educational session. And I think almost to some extent it's an implicit permission that gets granted to the house staff to disrupt their own teaching experience and disrupt others around them because everybody is doing it because everybody is being Blackberried. So it almost becomes the new social norm and while that may be a new social norm I'm not sure that that's a good thing How big is the negative impact? That's much harder to say. It's probably not a big impact on top of the endless other disruptions in the day to teaching, but it is measurable because it's a new factor so it's observable by me on top of all the other factors which have been there for years. (Interview Attending 3)
2:10‐Attending goes to the whiteboard to teach research methods to the team. Spotted Medical student#1 looking at his IPhone and typing.
2:15‐Med student#1 using the calculator function on his IPhone.
2:20‐Attending glances at his BB quickly.
2:28‐Attending resumes discussion of the patients' cases (Field notes, Work‐shadowing MD7).
DISCUSSION
The educational impacts of smartphone use for communication appear to center on increased connectedness of medical trainees and increased interruptions, which have positive and negative impacts in the areas of teaching, supervision, and professionalism. Smartphone communication provided potential educational benefits through (1) safer supervision with rapid access to help and (2) easier coordination of teaching sessions. Threats to the educational experience included (1) a high level of interruptions to both teachers and learners, which may reduce the effectiveness of formal and informal teaching; (2) replacement of face‐to‐face teaching with texting; (3) a potential erosion of autonomy and independence due to easy access to supervisors and easy ability for supervisors to take over; and (4) professionalism issues with difficulties balancing between clinical service demands and communication during patient and interprofessional encounters.
This study is the first to describe the intersection of clinical communication with smartphones and medical education. A recent study found that residents reported high use of smartphones during rounds for patient care as well as personal issues.[13] We have previously described the perceived impacts of smartphones on clinical communication, which included improved efficiency but concerns for increased interruptions and threats to professionalism.[6] We also observed that sites that used smartphones had increased interruptions compared to those with just pagers.[10] We have also described the content of e‐mail messages between clinicians and found that all e‐mails from nurses to physicians involved clinical care, but e‐mail exchanges between physicians were split between clinical care (60.4%), coordination within the team (53.5%), medical education (9.4%), and social communication (3.9%).[14] This study adds to the literature by focusing on the impacts of smartphone use to medical education and describing the perceived and observed impacts. This study provides a further example of how healthcare information technology can cause unintended consequences on medical education and appear to relate to the linked impacts of increased connectivity and the increased interruptions.[3] In essence, the trainee becomes more global, less local. Being more global translates to increased connections with people separated in physical space. Yet, this increased global connectedness resulted in the trainee being less local, with attention diverted elsewhere, taking away from the quality of patient interactions and interactions with other interprofessional team members. It also reduces the effectiveness of educational sessions for all participants. Although the level of supervision and autonomy are independently related, being too connected to supervisors may lower trainee autonomy by reducing independent thinking around patient issues.[15] It may also move teaching and learning from face‐to‐face conversations to text‐based messages. Although there have been existing tensions between service delivery and medical education, increased connectedness may tilt the balance toward the demands of service delivery and efficiency optimization at the expense of the educational experience. Finally, smartphone use appeared to create an internal tension among trainees, who have to juggle balancing professional behaviors and expectations in their dual role as learner and care provider; it would be educationally unprofessional to interrupt a teaching session and respond to a text message. However, failing to respond to a nurse who has sent a message and is expecting a response would be clinically unprofessional.
To address these threats, we advocate improving systems and processes to reduce interruptions and provide education on the tensions created by increased connectedness. Smarter communication systems could limit interrupting messages to urgent messages and queue nonurgent messages.[16] They could also inform senders about protected educational time. Even more sophisticated systems could inform the sender on the status of the receiver. For example, systems could indicate if they are available or if they are busy in an educational session or an important meeting with a patient and their family. Processes to reduce interruptions include interprofessional consensus on what constitutes an urgent issue and giving explicit permission to learners to ignore their smartphones during educational sessions except for critical communications purposes. Finally, education around smartphone communication for both learners and teachers may help minimize threats to learner autonomy, to face‐to‐face teaching, and to professionalism.[17]
Our study has several limitations. We derived this information from a general study of the impact of smartphones on clinical communication. Our study can be seen as hypothesis generating, and further research is warranted to validate these findings. There may be limits to generalizability as all sites adopted similar communication processes that included centralizing communications to make it easier for senders to reach a responsible physician.
In conclusion, we have provided a summary of the impact of rapidly emerging information technology on the educational experience of medical trainees and identified both positive and negative impacts. Of note, the negative impacts appear to be related to being more global and less local and high interruptions. Further research is required to confirm these unintended consequences as well as to develop solutions to address them. Educators should be aware of these findings and the need to develop curriculum to address and manage the negative impacts of smartphone use in the clinical training environment.
Acknowledgments
Disclosure: Nothing to report.
- , . Smartphone app use among medical providers in ACGME training programs. J Med Syst. 2012;36:3135–3139.
- , , , et al. The use of smartphones for clinical communication on internal medicine wards. J Hosp Med. 2010;5:553–559.
- , , , , , . 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. 2011;18(1):82–90.
- , , , , . Types of unintended consequences related to computerized provider order entry. J Am Med Inform Assoc. 2006;13(5):547–556.
- , , , . “e‐Iatrogenesis”: the most critical unintended consequence of CPOE and other HIT. J Am Med Inform Assoc. 2007;14(3):387–388.
- , , , et al. An evaluation of the use of smartphones to communicate between clinicians: a mixed‐methods study. J Med Internet Res. 2011;13(3):e59.
- . Smartphones in clinical practice, medical education, and research. Arch Intern Med. 2011;171(14):1294–1296.
- , , , . Use of handheld computers in medical education. A systematic review. J Gen Intern Med. 2006;21(5):531–537.
- , , , . A review and a framework of handheld computer adoption in healthcare. Int J Med Inform. 2005;74(5):409–422.
- , , , 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 [published online ahead of print January 25, 2013]. J Am Med Inform Assoc. doi:10.1136/amiajnl‐2012‐001160.
- , , , , . Hospital doctors' workflow interruptions and activities: an observation study. BMJ Qual Saf. 2011;20(6):491–497.
- , , , et al. Computerised provider order entry and residency education in an academic medical centre. Med Educ. 2012;46:795–806.
- , , , . Smartphone use during inpatient attending rounds: Prevalence, patterns and potential for distraction. J Hosp Med. 2012;7(8):595–599.
- , , , et al. Understanding interprofessional communication: a content analysis of email communications between doctors and nurses. Applied Clinical Informatics. 2012;3:38–51.
- , , , . Preserving professional credibility: grounded theory study of medical trainees' requests for clinical support. BMJ. 2009;338:b128.
- , , , , . Beyond paging: building a web‐based communication tool for nurses and physicians. J Gen Intern Med. 2009;24(1):105–110.
- , . Distracted doctoring: smartphones before patients? CMAJ. 2012;184:1440.
- , . Smartphone app use among medical providers in ACGME training programs. J Med Syst. 2012;36:3135–3139.
- , , , et al. The use of smartphones for clinical communication on internal medicine wards. J Hosp Med. 2010;5:553–559.
- , , , , , . 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. 2011;18(1):82–90.
- , , , , . Types of unintended consequences related to computerized provider order entry. J Am Med Inform Assoc. 2006;13(5):547–556.
- , , , . “e‐Iatrogenesis”: the most critical unintended consequence of CPOE and other HIT. J Am Med Inform Assoc. 2007;14(3):387–388.
- , , , et al. An evaluation of the use of smartphones to communicate between clinicians: a mixed‐methods study. J Med Internet Res. 2011;13(3):e59.
- . Smartphones in clinical practice, medical education, and research. Arch Intern Med. 2011;171(14):1294–1296.
- , , , . Use of handheld computers in medical education. A systematic review. J Gen Intern Med. 2006;21(5):531–537.
- , , , . A review and a framework of handheld computer adoption in healthcare. Int J Med Inform. 2005;74(5):409–422.
- , , , 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 [published online ahead of print January 25, 2013]. J Am Med Inform Assoc. doi:10.1136/amiajnl‐2012‐001160.
- , , , , . Hospital doctors' workflow interruptions and activities: an observation study. BMJ Qual Saf. 2011;20(6):491–497.
- , , , et al. Computerised provider order entry and residency education in an academic medical centre. Med Educ. 2012;46:795–806.
- , , , . Smartphone use during inpatient attending rounds: Prevalence, patterns and potential for distraction. J Hosp Med. 2012;7(8):595–599.
- , , , et al. Understanding interprofessional communication: a content analysis of email communications between doctors and nurses. Applied Clinical Informatics. 2012;3:38–51.
- , , , . Preserving professional credibility: grounded theory study of medical trainees' requests for clinical support. BMJ. 2009;338:b128.
- , , , , . Beyond paging: building a web‐based communication tool for nurses and physicians. J Gen Intern Med. 2009;24(1):105–110.
- , . Distracted doctoring: smartphones before patients? CMAJ. 2012;184:1440.
© 2013 Society of Hospital Medicine
