Maintenance Notice

Due to necessary scheduled maintenance, the JMIR Publications website will be unavailable from Monday, December 24 through Wednesday, December 26 inclusive. We apologize in advance for any inconvenience this may cause you.

Who will be affected?

Advertisement

Citing this Article

Right click to copy or hit: ctrl+c (cmd+c on mac)

Published on 21.09.18 in Vol 6, No 9 (2018): September

Preprints (earlier versions) of this paper are available at http://preprints.jmir.org/preprint/11187, first published May 31, 2018.

This paper is in the following e-collection/theme issue:

    Original Paper

    The Effectiveness of Near-Field Communication Integrated with a Mobile Electronic Medical Record System: Emergency Department Simulation Study

    1Department of Emergency Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic Of Korea

    2Department of Computer Engineering, Seoul Digital University, Seoul, Republic Of Korea

    3Department of Information Strategy, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic Of Korea

    4Department of Digital Health, Samsung Advanced Institute for Health Sciences & Technology, Sungkyunkwan University, Seoul, Republic Of Korea

    Corresponding Author:

    Won Chul Cha, MD

    Department of Emergency Medicine

    Samsung Medical Center

    Sungkyunkwan University School of Medicine

    81, Irwon-ro

    Gangnam-gu

    Seoul, 06351

    Republic Of Korea

    Phone: 82 1053866597

    Fax:82 221487899

    Email:


    ABSTRACT

    Background: Improved medical practice efficiency has been demonstrated by physicians using mobile device (mobile phones, tablets) electronic medical record (EMR) systems. However, the quantitative effects of these systems have not been adequately measured.

    Objective: This study aimed to determine the effectiveness of near-field communication (NFC) integrated with a mobile EMR system regarding physician turnaround time in a hospital emergency department (ED).

    Methods: A simulation study was performed in a hospital ED. Twenty-five physicians working in the ED participated in 2 scenarios, using either a mobile device or personal computer (PC). Scenario A involved randomly locating designated patients in the ED. Scenario B consisted of accessing laboratory results of an ED patient at the bedside. After completing the scenarios, participants responded to 10 questions that were scored using a system usability scale (SUS). The primary metric was the turnaround time for each scenario. The secondary metric was the usability of the system, graded by the study participants.

    Results: Locating patients from the ED entrance took a mean of 93.0 seconds (SD 34.4) using the mobile scenario. In contrast, it only required a mean of 57.3 seconds (SD 10.5) using the PC scenario (P<.001). Searching for laboratory results of the patients at the bedside required a mean of only 25.2 seconds (SD 5.3) with the mobile scenario, and a mean of 61.5 seconds (SD 11.6) using the PC scenario (P<.001). Sensitivity analysis comparing only the time for login and accessing the relevant information also determined mobile devices to be significantly faster. The mean SUS score of NFC-mobile EMR was 71.90 points.

    Conclusions: NFC integrated with mobile EMR provided for a more efficient physician practice with good usability.

    JMIR Mhealth Uhealth 2018;6(9):e11187

    doi:10.2196/11187

    KEYWORDS



    Introduction

    Background

    An emergency department (ED) is often characterized by chaos and inefficiency [1]. It is where the severity of a patient’s injury or distress varies and changes unpredictably. The location of patients also changes based on their clinical process or test results, which may become available at different times. Physicians need to check changed locations and laboratory results frequently by walking back and forth between personal computer (PC) stations and patients’ beds. In a fast-paced ED, such interruptions cause physicians to waste a substantial amount of time and ultimately result in patient dissatisfaction [2]. As the installation of PCs to all bedsides is costly and ineffective regarding space utilization, better alternatives must be considered.

    Related Technologies

    The ED providers strive to improve the efficiency of the workflow by exploiting advanced technologies. With the emergence of electronic medical records (EMR), mobile EMR systems are receiving increasing attention as mobile devices (ie, mobile phones, tablets), and mobile apps are becoming more common [3,4]. The portability and near ubiquity of mobile EMR allow health care providers to access patient records wherever they are needed [5].

    Near-field communication (NFC) is widely used in various communication apps. In the field of health care, usage scenarios including patient identification [6], blood transfusion [7], drug administration [8], medical staff tracking [9], and medical record access [10] have already been proven. The wave of NFC technology in the health care field has been combined with the internet of things technologies [11]. Through a combination of mobile EMR systems, NFC technology can improve workflow using bedside technology [12].

    Study Objectives

    Numerous studies have investigated the qualitative and quantitative benefits of each technology separately [12-14]. However, the efficacy of the combined technologies for improved physician productivity in health care has not been investigated to date. More rigorous quantitative studies investigating usability estimates are required to develop and eventually adopt such systems in practice. Additionally, it is essential to determine a system’s effectiveness in clinical settings. This study aims to determine the effectiveness of an NFC-integrated mobile EMR system regarding physician turnaround time in an ED.


    Methods

    Study Setting

    This simulation study took place in an academic ED in Seoul, South Korea. The study was reviewed and approved by the Samsung Medical Center Institutional Review Board (IRB no. SMC 2018-01-144-001).

    The ED is part of a tertiary academic teaching hospital with approximately 9000 daily outpatients and 2000 inpatient beds [15]. The ED holds 69 treating beds. The number of annual visits is approximately 79,000. Although the ED is heavily equipped with PCs at each station (84 PCs total), there are no PCs at the bedsides. Most of the beds are not in private rooms but are open to stations except isolation beds.

    The hospital developed the proposed mobile EMR system. It operates on the institution’s EMR system, which was also developed internally. The overall system had a significant update in July 2016. Figure 1 shows a schematic of the architecture of the EMR system.

    The mobile EMR uses an Android app that gives physicians access to inpatient, outpatient, and emergency department information. Users can log into the system with their fingerprint and search for locations, clinical notes, vital signs, laboratory results, and medical images. The NFC function was implemented in April 2017. When a physician links the EMR mobile device to an NFC tag which contains information about its location, the mobile app is automatically initiated. Using fingerprint authentication, physicians can log in and view the EMR of patients at the location that corresponds with their NFC location. When tagging NFC tags at the entrance of a specific zone, the list of patients at the tagged NFC zone who are in charge of mobile device users is popped up. Figure 2 shows the access process.

    Study Participants

    Physicians who worked in the ED during the study were asked to participate. Physician participants were recruited between April 1 to April 20, 2018. Among the 35 ED physicians, 25 (71%) agreed to participate in this study.

    Study Scenarios and Sensitivity Analysis

    After a brief introduction, the physicians went through 2 sequential scenarios. The first scenario (scenario A) involved locating patients in the ED from the ED gate. Physicians were given the name of a patient and were required to locate them using either a PC EMR at the nearest site in the ED gate or mobile EMR. After locating the patient, the participant was guided to reach their bedside. The second scenario (scenario B) involved looking up a laboratory result from the bedside. Physicians were brought to a patient’s bedside and were required to determine a specific laboratory result using either a mobile device or PC interface. As there were no PCs at the bedside, physicians had to perform a few steps to identify available PCs and return with a report. The steps in each scenario are shown in Figure 3, and the flow of each scenario is shown in Figure 4.

    Physicians were randomly assigned to follow either scenario A or B using either a mobile device (mobile case) or a PC (PC case) as described in Multimedia Appendix 1. An independent observer recorded the activities with a camera and completed a case report form with time stamps during the process. Patients were not simulated. Real patients were accessed in the emergency department. However, since we used only partial patient data such as name, location, and laboratory data which was already available, the clinical condition did not influence the study’s outcome.

    Figure 1. Overall schematic description of the hospital information system architecture relationship at the Samsung Medical Center. DARWIN: data analytics and research window for integrated knowledge; CPOE: computerized physician order entry; MIS: management information system; MDM: master data management; CRM: customer relationship management.
    View this figure
    Figure 2. Usage scene when the mobile electronic medical records (EMR) communicate with the near-field communication (NFC) system and the display of the mobile EMR progression after tagging NFC. V/S: vital sign.
    View this figure
    Figure 3. Schematic view of simulation scenarios. (a) Locating the patient. (b) Looking up laboratory results for the patient. ED: emergency department; PC: personal computer.
    View this figure
    Figure 4. The flow of locating the patient and the scenario place in the emergency department. CT: computed tomography; NFC: near-field communication; PC: personal computer.
    View this figure

    We performed a sensitivity analysis using the data without considering movement intervals. This test was performed to determine whether or not there was a consistent outcome if the condition allowed for more available PCs, which are at the gate and the bedside.

    Survey

    After completing all scenarios, physician participants responded to 10 questions using the system usability scale (SUS). The SUS is composed of a 5-point Likert scale rated from 1 (strongly disagree) to 5 (strongly agree) that investigates the usability of the NFC-integrated mobile EMR system [16]. The SUS score calculation formula is as follows:

    Measurement and Outcome

    The primary metric was the length of turnaround time for each scenario. The secondary metric was the usability of the system, as graded by the study physician participants. We collected demographic data from each participant and recorded the time intervals of each step of the process for both scenarios. We also analyzed time intervals among groups sorted by age, gender, and occupation. Afterward, the SUS questionnaires were collected and analyzed.

    Statistical Analysis

    Continuous variables are expressed in terms of mean and standard deviations (SD), whereas categorical variables are expressed in frequencies and percentages. The time mean difference was examined using a paired t-test. A value of P<.05 was considered to be statistically significant. As descriptive statistics could not confirm a normal distribution of participants between the 2 dependent groups divided by age, gender, and occupation, the Mann-Whitney U test was applied for time interval difference analysis.


    Results

    Main Outcome

    Among 25 physician participants, 14 (56%) were male, and 11 (44%) were female. The general characteristics of the participants are shown in Table 1.

    It required a mean of 93.0 seconds (SD 34.4) to locate the patient from the entrance of the ED in the PC case but only a mean of 57.3 seconds (SD 10.5) in the mobile case, which was significantly faster (P<.001). Accessing laboratory results at the patient’s bedside required a mean of only 25.2 seconds (SD 5.3) in the mobile case compared to a mean of 61.5 seconds (SD 11.6) in the PC case. These data were statistically significant (P<.001). A schematic comparison is shown in Figure 5.

    Sensitivity Analysis

    We compared the time required for login with the time for finding relevant information. Login using the mobile device EMR required a mean of 13.1 seconds (SD 2.9) for scenario A and a mean of 12.5 seconds (SD 2.1) for scenario B. Login by PC took longer with a mean of 36.2 seconds (SD 15.2) for scenario A and a mean of 30.5 seconds (SD 7.7) for scenario B. The differences in time were statistically significant (P<.001). Finding the location of patients after login required a mean of only 6.8 seconds (SD 3.6) using the mobile device, whereas it took a mean of 18.9 seconds (SD 16.9) using a PC. Accessing a specific laboratory test result required a mean of 12.8 seconds (SD 5.3) using the mobile device and a mean of 26.5 seconds (SD 8.0) using a PC. These data were statistically significant (P<.001). The results are shown in Table 2.

    Survey

    The mean SUS score of NFC-mobile EMR was 71.90 points. The results are shown in Table 3.

    Table 1. Characteristics of the physician participants.
    View this table
    Figure 5. Graphical view of main results. (a) Locating the patient. (b) Looking up laboratory results for the patient. NFC: near-field communication; PC: personal computer.
    View this figure
    Table 2. A comparison between the 2 scenarios of time spent on specific tasks.
    View this table
    Table 3. Score results (n=25) from the system usability scale (SUS) to assess the near-field communication mobile emergency medical record (NFC-mobile EMR).
    View this table

    Discussion

    Principal Findings

    This study aimed to improve physician efficiency by reducing the time spent walking to check patient information with the aid of the technological integration between NFC and mobile device EMR. To the best of our knowledge, this is the first study to examine the efficiency of this system and comparing it with the PC EMR. The mobile total turnaround time for performing tasks was significantly reduced in both scenarios. Sensitivity analysis showed that mobile device EMR incorporated with NFC was significantly faster than PC-integrated EMR regarding login time and accessing laboratory results.

    As the familiarity of mobile device use could be different among the demographic groups, we compared the total time interval difference between PC and mobile cases. Multimedia Appendix 2 shows that the mobile case was consistently faster for all groups. However, there were significant differences in the time interval between age and occupation during scenario B. These findings are contrary to the general belief that the younger generation is more familiar with newer technology [17]. A further study on mobile device familiarity is needed because the simulation was done with a small sample size.

    We also evaluated usability with the SUS questionnaire. The SUS was used after the physician participant had an opportunity to use the system being evaluated. A score over 70 on the questionnaire (range 0-100) indicated that the NFC-integrated mobile device EMR was “acceptable,” and the adjective rating was “good” [18]. There was no significant statistical difference among groups based on age, gender, and occupation.

    Advantages and Disadvantages

    Various measures have been implemented to address ED inadequacies. Improving the ED work efficiency is one crucial in-hospital factor. Ideal physical structures for work have already been demonstrated [19]. Several studies have shown the positive effect of developing clinical guidelines and protocols for effective evaluation of efficiency [20,21]. Newer technologies such as radio frequency identification-integrated point-of-care testing [22], triage kiosks [23], and dashboards [24] have been well studied. Ubiquitous near patient access to EMR via NFC is determined to be useful in this regard. Compared to installing new structures in an already heavily equipped ED, implementing an NFC tag system is a relatively easy way to improve workflow regarding cost and space utilization.

    As most mobile EMR functions are more readily accessible with PCs, our study paid attention to superiority only available in mobile EMR. Mobile device systems outperform PC systems concerning mobility and personalization (at the provider level, and patient level). We have measured the turnaround time as the primary outcome of these merits. Thus, we have shown that physicians can gain access to information without physically moving the location of their patients.

    Portability of mobile EMR could be improved by incorporating accessibility through NFC. Our study revealed a statistically significant difference in login time which was more effective by mobile EMR than by PC EMR (Table 2). A previous study by Holden [25] demonstrated that the issue of accessibility to EMR such as system login and system response time could negatively impact the usability of mobile EMR. NFC integrated response and fingerprint login at the location of interest using a mobile device could be beneficial because the process is simplified and less time-consuming. This system also appears to reduce security concerns from failed logouts or departure without logging out, by using the individual’s mobile device.

    An increase in the length of time physicians spend at the bedside is likely to increase patient satisfaction [26]. With this bedside technology, the physician can show radiologic results or laboratory results to patients who cannot ambulate.

    Inconsistent loading time due to varying network coverage could be a disadvantage for this technology. For example, mobile devices without NFC function cannot be used. Physicians might routinely tend to use PC EMR because PC EMR covers mobile EMR. A previous study by Duhm et al [14] demonstrated that a physician usually underestimates actual time savings during their professional capacity. The results of this study make a compelling argument and provide preliminary evidence in support of adequately addressing this tendency, particularly concerning reduced workflow using mobile EMR with NFC functionality.

    However, to enhance emergency physician performance, a multidimensional approach is required, rather than a single tool. ED processes are complicated, with multiple steps from various providers often originating from outside the ED.

    Limitations

    First among the limitations of this study is that this investigation was conducted at a single center. Additional studies conducted at multiple centers or EDs are needed to improve the generalizability of our conclusions.

    Secondly, participants had different levels of familiarity with mobile devices and NFC tags. Only some participants were familiar with NFC because the system was built over a year ago, which might cause bias.

    Thirdly, each participant encountered various encumbrances because this study was conducted in an actual emergency room. For example, when attempting to locate a patient in the middle of a scenario, the nearest PC may have been occupied by another staff member, which led to the physician being forced to use a PC that was further away. Also, while moving to a patient’s bedside, there was an occasion when a participant was forced to stop because a moving stretcher cart or medical staff member blocked the aisle. In addition, some of the PCs used were comparatively slow. As mentioned above, unpredictable circumstances might influence the overall time measured for each scenario. As shown in Multimedia Appendix 3, the variability of turnaround time fluctuated. However important, these events could not be systemically quantified.

    Finally, the usability assessment for NFC-mobile EMR via SUS could be overrated because responses were filled out immediately after performing scenarios, which in most cases, resulted in the superiority of NFC-mobile EMR. Further studies could investigate usability over a more extended period of the physician’s working practice.

    Conclusion

    NFC-integrated mobile EMR is effective for reducing the turnaround time of physicians when practicing in the field and has excellent usability.

    Acknowledgments

    The authors would like to acknowledge the support provided by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (grant number 2018R1C1B6002877).

    Conflicts of Interest

    None declared.

    Multimedia Appendix 1

    Randomly allocated scenario quest for each participant. PC: personal computer.

    PDF File (Adobe PDF File), 19KB

    Multimedia Appendix 2

    Comparison of time spent among age, gender, and occupation for each scenario.

    PDF File (Adobe PDF File), 24KB

    Multimedia Appendix 3

    Graphical comparison between mobile and personal computer scenario time of all participants.

    PPTX File, 51KB

    References

    1. Berger E. Breaking point: Report calls for Congressional rescue of hospital emergency departments. Ann Emerg Med 2006 Aug;48(2):140-142. [Medline]
    2. Weigl M, Beck J, Wehler M, Schneider A. Workflow interruptions and stress atwork: a mixed-methods study among physicians and nurses of a multidisciplinary emergency department. BMJ Open 2017 Dec 22;7(12):e019074 [FREE Full text] [CrossRef] [Medline]
    3. Kim J, Lee Y, Lim S, Lee J. How Are Doctors Using Mobile Electronic Medical Records? An In-Depth Analysis of the Usage Pattern. Stud Health Technol Inform 2017;245:1231. [Medline]
    4. Yeo K, Lee K, Kim J, Kim T, Choi Y, Jeong W, et al. Pitfalls and Security Measures for the Mobile EMR System in Medical Facilities. Healthc Inform Res 2012 Jun;18(2):125-135 [FREE Full text] [CrossRef] [Medline]
    5. Burdette SD, Herchline TE, Oehler R. Surfing the web: practicing medicine in a technological age: using smartphones in clinical practice. Clin Infect Dis 2008 Jul 01;47(1):117-122. [CrossRef] [Medline]
    6. Aguilar A, van der Putten W, Kirrane F. Positive patient identification using RFID and wireless networks. 2006 Nov Presented at: 11th Annual Conference and Scientific Symposium (HISI); November 2006; Dublin, Ireland.
    7. Gutierrez A, Levitt J, Reifert D, Raife T, Diol B, Davis R, et al. Tracking blood products in hospitals using radio frequency identification: Lessons from a pilot implementation. VOXS 2013 May 31;8(1):65-69. [CrossRef]
    8. Houliston B. Integrating RFID technology into a drug administration system. In: Journal of Applied Computing and Information. 2005 Presented at: HIC 2005 and HINZ 2005: Proceedings; 2005; Brunswick East, Vic.: Health Informatics Society of Australia p. 453-461.
    9. Halamka J. Early experiences with positive patient identification. J Healthc Inf Manag 2006;20(1):25-27. [Medline]
    10. Ajami S, Rajabzadeh A. Radio Frequency Identification (RFID) technology and patient safety. J Res Med Sci 2013 Sep;18(9):809-813 [FREE Full text] [Medline]
    11. Lee BM, Ouyang J. Intelligent Healthcare Service by using Collaborations between IoT Personal Health Devices. IJBSBT 2014 Feb 28;6(1):155-164. [CrossRef]
    12. Köstinger H, Gobber M, Grechenig T, Tappeiner B, Schramm W. Developing an NFC based patient identification and ward round system for mobile devices using the android platform. In: Point-of-Care. 2013 Jan 16 Presented at: 2013 IEEE Point-of-Care Healthcare Technologies (PHT); 16-18 Jan. 2013; Bangalore, India p. 2013.
    13. Free C, Phillips G, Felix L, Galli L, Patel V, Edwards P. The effectiveness of M-health technologies for improving health and health services: a systematic review protocol. BMC Res Notes 2010 Oct 06;3:250 [FREE Full text] [CrossRef] [Medline]
    14. Duhm J, Fleischmann R, Schmidt S, Hupperts H, Brandt SA. Mobile Electronic Medical Records Promote Workflow: Physicians' Perspective From a Survey. JMIR Mhealth Uhealth 2016 Jun 06;4(2):e70 [FREE Full text] [CrossRef] [Medline]
    15. Cha WC, Ahn KO, Shin SD, Park JH, Cho JS. Emergency Department Crowding Disparity: a Nationwide Cross-Sectional Study. J Korean Med Sci 2016 Aug;31(8):1331-1336 [FREE Full text] [CrossRef] [Medline]
    16. Jordan P, Thomas B, Weerdmeester B, McClelland I. SUS-A quick and dirty usability scale. In: Usability Evaluation in Industry. Bristol, PA, USA: Florence, Kentucky, U.S.A.: Taylor & Francis; 1996:189-194.
    17. McAlearney AS, Schweikhart SB, Medow MA. Doctors' experience with handheld computers in clinical practice: qualitative study. BMJ 2004 May 15;328(7449):1162 [FREE Full text] [CrossRef] [Medline]
    18. Bangor A, Kortum P, Miller J. Determining what individual SUS scores mean: adding an adjective rating scale. J Usability Studies 2009 May;4(3):114-123.
    19. Zusman E. Form facilitates function: innovations in architecture and design drive quality and efficiency in healthcare. Neurosurgery 2010 Jun;66(6):N24. [CrossRef] [Medline]
    20. Butti L, Bierti O, Lanfrit R, Bertolini R, Chittaro S, Delli CS, et al. Evaluation of the effectiveness and efficiency of the triage emergency department nursing protocol for the management of pain. J Pain Res 2017;10:2479-2488 [FREE Full text] [CrossRef] [Medline]
    21. Wright S, Trott A, Lindsell C, Smith C, Gibler W. Evidence-based emergency medicine. Creating a system to facilitate translation of evidence into standardized clinical practice: a preliminary report. Ann Emerg Med 2008 Jan;51(1):80-6, 86.e1. [CrossRef] [Medline]
    22. Rao A, Dighe A. Radiofrequency Identification and Point-of-Care Testing. Point of Care: The Journal of Near-Patient Testing & Technology 2004;3(3):130-134. [CrossRef]
    23. Ackerman S, Tebb K, Stein J, Frazee B, Hendey G, Schmidt L, et al. Benefit or burden? A sociotechnical analysis of diagnostic computer kiosks in four California hospital emergency departments. Soc Sci Med 2012 Dec;75(12):2378-2385. [CrossRef] [Medline]
    24. Swartz J, Cimino J, Fred M, Green RA, Vawdrey DK. Designing a clinical dashboard to fill information gaps in the emergency department. AMIA Annu Symp Proc 2014;2014:1098-1104 [FREE Full text] [Medline]
    25. Holden RJ. Physicians' beliefs about using EMR and CPOE: in pursuit of a contextualized understanding of health IT use behavior. Int J Med Inform 2010 Feb;79(2):71-80 [FREE Full text] [CrossRef] [Medline]
    26. Lin CT, Albertson GA, Schilling LM, Cyran EM, Anderson SN, Ware L, et al. Is patients' perception of time spent with the physician a determinant of ambulatory patient satisfaction? Arch Intern Med 2001 Jun 11;161(11):1437-1442. [Medline]


    Abbreviations

    ED: emergency department
    EMR: emergency medical record
    NFC: near-field communication
    PC: personal computer
    SUS: system usability scale


    Edited by G Eysenbach; submitted 31.05.18; peer-reviewed by S Lim, J Wu; comments to author 09.07.18; revised version received 31.07.18; accepted 31.08.18; published 21.09.18

    ©Kwang Yul Jung, Taerim Kim, Jaegon Jung, JeanHyoung Lee, Jong Soo Choi, Kang Mira, Dong Kyung Chang, Won Chul Cha. Originally published in JMIR Mhealth and Uhealth (http://mhealth.jmir.org), 21.09.2018.

    This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR mhealth and uhealth, is properly cited. The complete bibliographic information, a link to the original publication on http://mhealth.jmir.org/, as well as this copyright and license information must be included.