Yoo et al.

BMC Medical Informatics and Decision Making (2018) 18:80

https://doi.org/10.1186/s12911-018-0656-0

RESEARCH ARTICLE Open Access

Real-time location system-based asset

tracking in the healthcare field: lessons
learned from a feasibility study
Sooyoung Yoo1†, Seok Kim1†, Eunhye Kim1, Eunja Jung2, Kee-Hyuck Lee1 and Hee Hwang1*

Abstract
Background: Numerous hospitals and organizations have recently endeavored to study the effects of real-time
location systems. However, their experiences of system adoption or pilot testing via implementation were not
shared with others or evaluated in a real environment. Therefore, we aimed to share our experiences and insight
regarding a real-time location system, obtained via the implementation and operation of a real-time asset tracking
system based on Bluetooth Low Energy/WiFi in a tertiary care hospital, which can be used to improve hospital efficiency
and nursing workflow.
Methods: We developed tags that were attached to relevant assets paired with Bluetooth Low Energy sensor beacons,
which served as the basis of the asset tracking system. Problems with the system were identified during implementation
and operation, and the feasibility of introducing the system was evaluated via a satisfaction survey completed by end
users after 3 months of use.
Results: The results showed that 117 nurses who had used the asset tracking system for 3 months were moderately
satisfied (2.7 to 3.4 out of 5) with the system, rated it as helpful, and were willing to continue using it. In addition, we
identified 4 factors (end users, target assets, tracking area, and type of sensor) that should be considered in the
development of asset tracking systems, and 4 issues pertaining to usability (the active tag design, technical
limitations, solution functions, and operational support).
Conclusions: The successful introduction of asset tracking systems based on real-time location in hospitals
requires the selection of clear targets (e.g., users and assets) via analysis of the user environment and implementation
of appropriate technical improvements in the system as required (e.g., miniaturization of the tag size and improvement
of the sensing accuracy).
Keywords: Asset tracking, Efficiency, Healthcare, Real-time locating system, System implementation

Background commodities in medical sites, and when their shifts end,

The appropriate placement and supply of commodities they transfer responsibility for the equipment in their de-
in hospitals not only improves the quality of patient partments and wards to nurses on the next shift. There-
treatment but also influences the outcomes of treatment fore, the effective management of commodities and a
in emergency situations. Therefore, commodity man- reduction in the time required to transfer them between
agement could be considered an exceedingly import- nurses would not only increase the time available to
ant task that should not be neglected by hospitals [1–4]. nurses for the provision of patient care but also improve
Nurses are generally tasked with the management of care quality [5–9].
One method of improving commodity management
* Correspondence: [email protected]; [email protected] involves real-time location systems (RTLSs), which are
†
Sooyoung Yoo and Seok Kim contributed equally to this work. used to manage assets efficiently in other industries.
1
Office of eHealth Research and Businesses, Seoul National University
Bundang Hospital, 166, Gumi-ro, Bundang-gu, Seongnam-si 436-707, South
Interest in the introduction of these systems to the
Korea healthcare field is increasing, [10] albeit gradually [5, 11].
Full list of author information is available at the end of the article

© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Yoo et al. BMC Medical Informatics and Decision Making (2018) 18:80 Page 2 of 10

In healthcare, RTLS solutions are used mainly to track active tags, and web-based asset management application
medical staff, patients, and assets. The introduction of during the 13-month period.
these solutions has been shown to provide hospitals with The asset tracking system tracked the locations of as-
certain benefits [5, 7, 10, 12–16] including cost reduction, sets throughout every ward in the hospital using the
improvements in medical care quality and work processes, existing WiFi infrastructure; however, we used separate
and increased patient satisfaction. BLE beacons to provide location measurements with
Recently, the healthcare sector has adopted the bar- greater accuracy in the emergency room and the storage
code, radio-frequency identification (RFID), WiFi, and areas of the wards. The location of each asset and the
Bluetooth sensing technology in the development of status of the active tag battery could be checked using
tracking systems. Barcode technology, in particular, has the hospital information system, into which the asset
been widely adopted by nurses to reduce medication ad- management system was integrated. Figure 1 shows the
ministration errors and the related costs and to improve overall architecture of the BLE/WiFi-based asset track-
patient safety [17]. Furthermore, previous studies have ing system.
shown that these systems are effective in reducing costs, The asset management application consisted of 2
improving the quality of medical services, enhancing applications: 1 for users (i.e., nurses) and 1 for system
work processes, and boosting patient satisfaction, by administrators. In the user application, the types and po-
tracking medical professionals and patients [6, 8, 10, sitions of the assets on each floor were depicted as a
12–16, 18, 19]. RTLS technology is based mainly on bird’s eye view of the entire hospital, with the detailed
the use of WiFi and RFID, although some research locations marked clearly on a floor plan (see Fig. 2). The
has examined whether Bluetooth could be used as the user application also provided detailed location informa-
basis of RTLSs [5, 6, 20]. Researchers in the healthcare tion according to the asset status (i.e., onsite, offsite, or
field have demonstrated strong interest in localization sys- borrowed) for each department. In contrast, the admin-
tems; however, these systems are still in their infancy. istrator application provided information regarding the
Given the differences between sensors in terms of track- registration and management of target assets, the assets’
able distance, cost, and accuracy, people should consider most recent locations and location histories, whether as-
the pros and cons of sensors to ensure selection of the sets were onsite or offsite, and battery power levels for
most appropriate product [21]. In addition, the user inter- the tags. In addition, administrators could check the
face, security, and effects of interference need to be con- status of specific assets and all assets in a registered de-
sidered [22, 23]. Therefore, system users should share partment using the search filter.
their experiences of implementation with others to aid
medical institutions that are willing to adopt the technol- Active tags and BLE beacons
ogy in developing the most suitable systems [6, 11]. The active tags and BLE beacons were designed in con-
Therefore, this study aimed to share considerations sideration of the characteristics of the medical environ-
and insights and to provide information regarding the ment and RTLS-based tracking system. Because the
adoption of an RTLS in a medical environment, obtained active tags were used to track highly mobile assets, they
via the implementation and operation of a real-time were battery operated and able to transmit their own
Bluetooth Low Energy (BLE)/WiFi-based asset tracking location information based on data received from the
system in a tertiary general university hospital. beacon. Battery capacity was prioritized during the pro-
duction stage, with the aim of maximizing the conveni-
ence of the tags. We ultimately selected tags that were
Methods 66 × 40 × 25 mm in size and required 2 AA batteries
Development of the BLE/WiFi-based asset tracking system (3000 mAh, 3.0 V). We attached tags to a total of 400
The development of the asset tracking system initially instruments, including 4 iPads and 21 oxygen holders in
involved obtaining the definition of the asset manage- the emergency room, 200 infusion pumps in the emer-
ment process by various medical professionals including gency room and 3 intensive care units (ICUs), and 160
a medical doctor, 4 nurses, a medical informatics profes- patient monitors and 15 bladder scanners in all depart-
sor, and 2 staff members from the asset management de- ments. The tag locations were selected to minimize
partment at the tertiary general university hospital to interference with the equipment and prioritized accord-
which the authors are affiliated. Thirty developers ultim- ing to improvements in nursing workflow and relocation
ately participated in the development process, which frequency.
took place over the course of 13 months, from November The beacons relied on BLE, which is a low-power,
2013 to November 2014. For approximately 2 months low-cost solution that has been used widely in recent
thereafter, we stabilized the system and performed tests to years. They operated with replaceable batteries to obvi-
improve the software used to develop the BLE beacon, ate the need for separate power supplies. The beacons
Yoo et al. BMC Medical Informatics and Decision Making (2018) 18:80 Page 3 of 10

Fig. 1 Overall architecture of the asset tracking system

Fig. 2 Interface of the asset tracking system for users (Nurses)

Yoo et al. BMC Medical Informatics and Decision Making (2018) 18:80 Page 4 of 10

were 120 × 120 × 30 mm in size, with an estimated bat- with responses provided on a 5-point scale. The ques-
tery life expectancy of a year. In total, 98 BLE beacons tionnaire data were analyzed via frequency analysis and
were created; of these, 53, 31, 4, and 9 were positioned the Kruskal-Wallis test, using SAS 9.3 (SAS Institute,
in the emergency room, the wards, the ICUs, and other Cary, NC, USA) to determine whether the results were
hospital areas, respectively. The beacons were installed affected by the participants’ demographic characteristics.
on the ceiling of each location.

Setting asset tracking periods Results

The continuous transmission and reception of location End-user satisfaction
information is essential for real-time location confirm- The end user participants who completed the satisfac-
ation. However, because the active tags were battery tion questionnaire were primarily women, which is prob-
powered and attached to mobile equipment, real-time ably due to the nature of the nursing profession, and a
transmission of the location information caused a bat- large proportion were general duty nurses. In addition,
tery drainage problem. Accordingly, in consideration of the participants were evenly distributed between depart-
the battery life, we set each tag to transmit location in- ments (i.e., wards, ICUs, and the emergency room) and
formation 16 times per day within a different period, were primarily in their 20s and 30s (Table 1).
mainly during the nurses’ work shifts, to avoid the bur- The participants’ mean satisfaction ratings for the 17
den of frequent battery replacement. Specifically, tag items ranged from 2.7 (for the tag size) to 3.4 (for the
location signals were emitted at 30-min intervals provision of the information required for nursing work,
throughout the nurses’ work shifts and at 2-h intervals the need for the system, and intention regarding its fu-
during other hours. This setting was determined after ture use).
consulting nurses who worked in the field and consider- As shown in Table 2, the ICU nurses generally exhib-
ing the system’s operational efficiency according to the ited higher levels of satisfaction than the emergency
battery performance of the tags. The 16 specific times room nurses. In addition, statistically significant differ-
per day were finally set as 06:00, 06:30, 07:00, 09:00, ences were observed for all items in the expectancy ef-
11:00, 13:00, 13:30, 14:00, 14:30, 15:00, 17:00, 19:00, fects category, which pertained to the system’s expected
21:00, 21:30, 22:00, and 23:00. Tracking 16 times per day effects with constant use in the future, and for 3 of the 4
was a rule that we implemented upon introducing the items in the information quality category, which
system, but the tracking intervals can be modified to reflected the accuracy and display of information. In
provide optimal settings that reflect the users’ needs. other words, while the nurses’ acceptance of the system’s
practical functions differed between departments, they
Satisfaction questionnaire for the end users gave similar responses regarding whether the asset track-
Following the development of the asset tracking system, ing system provided functions that were necessary on
a survey was conducted to examine the perceived quality the ground, and they intended to use the system.
and usability of the pilot system and to determine the Table 3 shows the participants’ other subjective opin-
end users’ overall satisfaction over the course of ions (provided via free text in the questionnaire) of the
2 months starting from December 2014. Nurses were ways in which the asset tracking system could be im-
the end users for the asset tracking system. Although proved based on their 2-month pilot use. These opinions
the system was installed in the emergency room and all were classified according to whether they pertained to
wards and ICUs, the self-report satisfaction survey was location accuracy, the tag, asset tracking, or user educa-
conducted only in the 3 wards with the highest numbers tion. The most frequently observed comments con-
of users, 3 ICUs (i.e., surgical, internal medicine, and cerned location accuracy in the system. In addition, we
neonatal), and the emergency room. In total, 280 nurses classified the content of the categories into subcategories
completed the survey over the course of a month and found that comments regarding expansion of the
(February 16 to March 16, 2015). We received 244 range of target assets for tracking were expressed fre-
anonymous responses (response rate: 87.1%), and the quently. Issues regarding the size of the active tag were
final dataset contained data for 117 nurses following also noted.
data cleansing and the exclusion of data for nurses
who had not used the system (Fig. 3).
The survey questionnaire was developed based on the Discussion
Computer System Usability Questionnaire to assess the The following discussion considers the lessons that were
overall system, and the Questionnaire for User Interface learned during the introduction and pilot operation of
Satisfaction to assess the user experience. The question- the BLE/WiFi-based asset tracking system, based on the
naire consisted of 17 items divided between 5 categories, end users’ opinions.
Yoo et al. BMC Medical Informatics and Decision Making (2018) 18:80 Page 5 of 10

Fig. 3 Participant flow in the user satisfaction survey. * The insincere response means respondents having same answers for all questionnaires

Scope of asset tracking in the ICUs, and 265 assets in 33 categories in the
Medical institutions that intend to develop and intro- emergency room. Furthermore, nurses are required to
duce RTLSs are required to define the end users, target confirm the locations of all assets managed by their
assets, tracking areas, and sensor types via thorough dis- departments during handover to nurses on other
cussion prior to implementation [14, 20]. In the present shifts [14, 20].
study, nurses were selected as the end users of the sys- The selection of the target assets for the pilot system’s
tem. Most nurses in South Korea are responsible for a implementation involved the creation of a list of all types
higher number of patients relative to nurses in other ad- of equipment managed by the nurses, which resulted in
vanced countries; nevertheless, they undertake various the selection of the most meaningful 5 assets (Fig. 4).
duties in addition to patient care. This additional work Because of their high levels of mobility between depart-
includes the management of assets between work shifts ments, tracking the locations of these instruments was
in order to ensure continuous care. A considerable expected to increase nursing efficiency. The end-user
amount of time is spent performing these duties. For ex- satisfaction survey indicated that the asset tracking sys-
ample, in the present study, nurses managed 44 assets in tem was more helpful in managing/transferring/locating
11 categories on the wards, 118 assets in 11 categories the assets than the manual management, with a satisfac-
tion score that was above average. Considering the lim-
Table 1 Participants’ demographic characteristics ited technology and limited tracking items used during
Classification n % the pilot, it seems that nursing work efficiency will be
Sex Male 4 3.4 improved with the wider adoption of the system. Some
Female 113 96.6
examples of the practical use by the nurses in this study
are as follows.
Position Head nurse 8 6.8
Nurse 109 93.2 “In the ward, it is not easy to trace the movement of
Department Ward 32 27.4 patients who need various tests (MRI, CT, etc.), many
Intensive care unit 45 38.5 procedures (emergency hemodialysis, angiography,
Emergency room 40 34.2 etc.), and consultation. However, using the system, we
Age group 20s 59 50.4
were able to easily identify and collect the assets that
the patient needed.”
30s 52 44.4
40s 6 5.1 “When it is urgent to inject an emergency patient
Work experience < 1 year 28 23.9 with many medications, such as fluid, dopamine,
1–3 years 38 32.5 dobutamine, Lasix, and albumin through an
3–5 years 12 10.3 infusion pump, an asset tracking system can be
> 5 years 39 33.3
used to quickly locate available devices. As such,
medical staff can provide intervention to patients
Total 117 100.0
more quickly.”
Yoo et al. BMC Medical Informatics and Decision Making (2018) 18:80 Page 6 of 10

Table 2 Comparison of the mean satisfaction ratings according to the department (Intensive Care Unit, Emergency Room, and Ward)
Survey items M (SD) P
Total ER ICU Ward
Information quality 1 Does the asset tracking system provide the information 3.4 (0.785) 3.3 (0.716) 3.5 (0.661) 3.3 (0.991) 0.201
necessary to perform nursing work?
2 Is the information provided by the asset tracking 2.9 (0.928) 2.4 (0.781) 3.3 (0.793) 2.9 (1.014) < .0001*
system accurate?
3 Is the information provided by the asset tracking system 2.9 (0.844) 2.6 (1.005) 3.2 (0.712) 2.9 (0.641) 0.006*
expressed using appropriate terminology and in an
appropriate format?
4 Does the asset tracking system provide up-to-date 2.9 (0.908) 2.6 (0.868) 3.3 (0.733) 2.8 (1.019) 0.002*
information appropriate for the current situation?
System quality 1 Is the asset tracking system easy to use? 3.1 (0.902) 3.0 (0.974) 3.2 (0.773) 3.1 (0.976) 0.297
2 Are you satisfied with the speed of the asset tracking 3.0 (0.904) 2.6 (0.838) 3.2 (0.735) 3.0 (1.078) 0.010*
system?
3 Do you need a separate user manual for the asset 3.6 (0.993) 3.5 (1.132) 3.7 (0.905) 3.6 (0.948) 0.824
tracking system?
4 Are you satisfied with using the asset tracking system 3.2 (0.886) 3.1 (0.841) 3.3 (0.879) 3.2 (0.954) 0.371
through the electronic medical record?
Active tag 1 Are you satisfied with the size of the tag affixed to 2.7 (0.992) 2.6 (0.979) 2.7 (0.953) 2.9 (1.070) 0.712
the device?
2 You can replace the battery by pushing open the top 3.0 (0.895) 2.7 (0.905) 2.9 (0.915) 3.3 (0.745) 0.014*
part of the tag cover. Do you think this method of
opening and closing the tag is appropriate?
Expectancy effects 1 Is the asset tracking system more helpful for asset 3.2 (0.928) 2.9 (0.810) 3.5 (0.894) 3.4 (1.008) 0.007*
management than the previous method (i.e., manual
management)?
2 Is the asset tracking system more helpful for transferring 3.2 (0.952) 2.9 (0.778) 3.4 (0.957) 3.3 (1.085) 0.033*
the nursing instruments than the previous method
(manual management)?
3 Is the asset tracking system more helpful for locating 3.3 (0.972) 2.8 (0.813) 3.5 (0.894) 3.5 (1.078) 0.001*
mobile equipment than the previous method
(manual management)?
4 Do you think the equipment will be used more efficiently 3.3 (0.927) 3.0 (0.832) 3.5 (0.894) 3.5 (0.983) 0.017*
with the asset tracking system than with the previous
method (manual management)?
Overall satisfaction 1 Are you satisfied with the overall use of the asset 3.1 (0.834) 2.7 (0.716) 3.3 (0.780) 3.3 (0.902) 0.002*
tracking system?
2 Do you need the asset tracking system for your 3.4 (0.884) 3.1 (0.883) 3.5 (0.815) 3.4 (0.946) 0.083
nursing work?
3 Do you intend to continue using the asset tracking 3.4 (0.811) 3.2 (0.747) 3.5 (0.815) 3.6 (0.833) 0.068
system?
*p < .05; comparisons were performed between 3 departments. ER emergency room, ICU intensive care unit

“In the past, we had to check all of the assets we were Location vs. positioning
managing, including the quantity and the location of RTLSs use both access points (APs) and beacons,
the assets at the end of the shift. However, after using which are typically placed indoors, to ensure track-
the asset tracking system, we can check the location ing accuracy. In the present study, locations were
and quantity at once by looking at the dashboard, tracked using an existing WiFi infrastructure that
which can save a lot of time.” covered all of the hospital wards, while the location
positioning of greater accuracy in the emergency and
The main reasons for equipment to be excluded were storage rooms was measured via BLE beacons. Be-
as follows: it was not needed by the end users, the tag cause errors begin to occur at a distance of 10 m for
influenced the usability of the asset, poor return on in- WiFi or 5 m for BLE, it is necessary to choose an
vestment, and it did not fulfill the criteria for the track- appropriate type of sensor for each target asset and
ing area (Table 4). area [21, 24].
Yoo et al. BMC Medical Informatics and Decision Making (2018) 18:80 Page 7 of 10

Table 3 Participants’ subjective opinions

Major classification n Sub-classification n Opinion
Locating/positioning 30 Detailed locations 7 I wish that the system showed more detailed locations in the ward
Location accuracy 18 The item is not in the location indicated
The location information is unreliable
Location errors occur depending on the strength of the WiFi signal
Tracking period 5 Real-time location identification would be nice
Active tag 18 Tag loss 7 The tags fall off the assets due to weak adhesion, and having to find the
lost tags is burdensome
Tag size 9 I wish the tag was smaller
Battery cover 1 The battery cover falls off easily
Battery life 1 The battery drains too quickly
Target assets 21 Tracking additional assets 10 I wish we could add tags to other equipment
I wish that all the equipment in the asset categories (e.g., patient monitors)
was included in the systema
Asset identification method 11 In addition to the asset identifier being automatically assigned to each
asset for the system to recognize, a manual method of managing asset
identifiers would be nice
Education 4 Need for user education 4 An explanation of the purpose and use of the system is needed
Comments were provided via free text in the questionnaire
a
Some assets (e.g., patient monitors) used by some departments were not tracked in the study

The system had several technical limitations. For ex- number of assets. Therefore, with respect to the tags, it
ample, the most important aspect of asset tracking is the would be necessary to decide whether to miniaturize
ability to determine the precise location of a piece of them for greater usability or fit them with large-capacity
equipment. However, the sensors do not always provide batteries for greater operational efficiency. The sensing
accurate locations in indoor tracking due to interference period for location identification would also require ad-
from other radio waves. Indeed, in the present study, justment. This could be achieved using either signals
users occasionally reported that assets were located on a generated by the tags or call signals from the beacon
different floor or in a different room to that shown in and AP; however, the provision of real-time measure-
the tracking application; this is likely to have occurred ments would require the generation of signals by the
because of the AP signal strength. In addition, we noted tags every 5 to 10 s, which would be fatal for battery life.
a problem with the batteries in the tags. As mentioned In the present study, signal generation was limited to 16
earlier, batteries were used in the tags and beacons to times per day and focused on the nurses’ shifts. How-
ensure their portability; however, periodical battery re- ever, because the emergency room typically requires
placement for the beacons (once per year) and tags frequent movement of equipment, the generation of lo-
(every 100 days) would be difficult because of the sheer cation information only 16 times per day might have

Fig. 4 Five types of asset targeted for tracking. Locations at which the active tags were attached to the 5 types of equipment selected as asset
tracking targets
Yoo et al. BMC Medical Informatics and Decision Making (2018) 18:80 Page 8 of 10

Table 4 Reasons for excluding assets from the real-time tracking

Asset Reason for exclusion
Patient tracking Little need for tracking at the site (e.g., psychiatric patients, neonates)
Laryngoscopy handle Reduces the usability of intubation devices (i.e., it is difficult to affix the tag because it could hinder visibility)
Difficulty in managing intubation devices (e.g., disinfection following each use)
Sandbag and ice pack While necessary, the return on investment is low (i.e., the tag cost is high relative to the unit cost for these items)
Operating room items The mobility of expensive medical instruments in the operating room is markedly low, so there is little need
for tracking
Wheelchairs While necessary, they were excluded because they are not for indoor use only (e.g., they are used outdoors
frequently at medical institutions)

been insufficient. The fact that the number of assets the end user to use the system while checking patient re-
managed by the emergency room was higher, relative to cords. However, after 3 months of pilot use, the end users
those managed by other departments, could also explain required a function to modify the names of assets since we
the low satisfaction levels observed in the study. displayed the names that were managed by the logistics de-
This configuration increased the difficulty in providing partment with a read-only mode for end users. This func-
real-time identification of asset locations, which was in- tion could be integrated into the interface easily and would
convenient for the end users. The limitations concerning probably increase user satisfaction. Notably, the emergency
accuracy and battery life coincide with technical limita- room nurses’ satisfaction levels were considerably lower
tions noted in other studies [5, 11, 14, 20, 25] and are ex- than those of nurses at the other sites, despite the higher lo-
pected to improve gradually. However, accurate location cation accuracy in the emergency room, which contained
positioning using BLE may not be easily solved due to the 53 beacons, while each storage room in the ICUs and wards
frequency hopping of the BLE beacon [26]. Therefore, contained a single beacon. It seems that the emergency
each site will be required to develop solutions to these is- room nurses had a higher expectation and a higher need
sues as they arise. In addition, given the recent expansion initially, resulting in low satisfaction due to the accuracy
of services based on wireless communication, and consid- problem. Okoniewska et al. also suggested that the RTLS
ering the interference caused by each signal, a more suit- system should improve the accuracy of location tracking
able method should be used to track assets [22, 23, 27]. and visualization in response to the converging opinions of
nurses [20]. Furthermore, regarding the patients’ satisfac-
Design of the active tags tion with the patient tracking system, Stubig et al. found
We developed the BLE/WiFi-based tags because the tar- that their satisfaction improved when the medical staff also
get assets did not include a signal-sensing function. used the location information compared to when their
However, this increased the size of the tags, which re- location information was only tracked [11].
duced their adhesive power and led to the loss of some
tags. This occurred because we ultimately selected
non-rechargeable AA batteries during the production Education and operational support
process to increase the duration of the battery replace- Users reported that sufficient support would be needed
ment cycle (i.e., lithium polymer [26 × 35 × 15 mm] and to improve the approachability and efficiency of the sys-
alkaline or manganese [AA size of 66 × 40 × 25 mm] tem. Educational efforts and a larger management work-
batteries last for 20 and 100 days, respectively), and to force would be major factors in providing this support
ensure greater management efficiency. Although tag loss [20]. In addition, it would be necessary to supply suffi-
disrupted the continuity of asset tracking and incurred cient information regarding the system operation and to
ongoing costs, the prioritization of battery capacity over provide appropriate education to ensure its smooth use.
tag size was unavoidable considering the long-term man- In addition, workforce support is required to ensure
agement efficiency. Because of this inverse relationship timely battery replacement (for the beacon and tags). In
between the efficiency of the tag size and management, the present study, end users were not required to replace
the needs of the site and the organization’s policy stan- the batteries themselves because it was unnecessary dur-
dards should be established and reflected in the produc- ing the 3-month pilot period, and the study was con-
tion and introduction of the active tags. ducted during the maintenance period of the system
development. However, these issues will require consid-
System functionality and usability eration in future; therefore, end users should be pro-
The interface used in the asset tracking system was inte- vided with the necessary information regarding battery
grated with the hospital information system, which allowed replacement [20].
Yoo et al. BMC Medical Informatics and Decision Making (2018) 18:80 Page 9 of 10

Limitations Korea. 2Nursing Department, Seoul National University Bundang Hospital,

The main limitations of the present study were that it 166, Gumi-ro, Bundang-gu, Seongnam-si 436-707, South Korea.

was conducted in a single medical institution, and the Received: 27 November 2017 Accepted: 17 August 2018
study design excluded the general outpatient environ-
ment. In addition, the tracking method was limited to
WiFi and Bluetooth, and only some of the assets were
References
tracked, which did not allow maximization of the bene- 1. Castro L, Lefebvre E, Lefebvre LA. Adding intelligence to mobile asset
fits of tracking the assets used by nurses. Future research management in hospitals: the true value of RFID. J Med Syst. 2013;37(5):9963.
should be conducted in consideration of the issues 2. Gastaldi L, Mangiaracina R, Miragliotta G, Perego A, Tumino A. Measuring
the benefits of tracking medical treatment through RFId. Int J Product Perform
highlighted in the present study. Moreover, future stud- Manag. 2015;64(2):175–93.
ies should not only involve the collection of data regard- 3. Roper KO, Sedehi A, Ashuri B. A cost-benefit case for RFID implementation
ing end users’ subjective experiences but also include in hospitals: adapting to industry reform. Facilities. 2015;33(5/6):367–88.
4. Sato K, Kuroda T, Seiyama A. Visualization and quantitative analysis of nursing
usability tests and time-motion studies. staff trajectories based on a location system. In: Innovation in medicine and
healthcare 2015: Springer; 2016. p. 25–35.
Conclusions 5. Fisher JA, Monahan T. Evaluation of real-time location systems in their hospital
contexts. Int J Med Inform. 2012;81(10):705–12.
The results of the study demonstrated that BLE/WiFi- 6. Hopkins CW. Measuring leadership rounding performance in a hospital
based asset tracking systems are needed in medical orga- environment using real time location systems. Greenville: East Carolina
nizations and that they are helpful for nursing work. University; 2015.
7. Jones TL, Schlegel C. Can real time location system technology (RTLS) provide
However, several technical and managerial issues were useful estimates of time use by nursing personnel? Research in nursing &
identified and should be addressed to ensure effortless health. 2014;37(1):75–84.
use of the system and improve user convenience. In 8. Prathiba SA, Jain D: Hospital Management System Using RFID. 2015.
9. Stubig T, Suero E, Zeckey C, Min W, Janzen L, Citak M, Krettek C, Hufner T,
addition, the development and introduction of an asset Gaulke R. Improvement in the workflow efficiency of treating non-emergency
tracking system in consideration of the lessons learned outpatients by using a WLAN-based real-time location system in a level I
from the present study is required. trauma center. J. Am. Med. Inform. Assoc. 2013;20(6):1132–6.
10. Ward DR, Ghali WA, Graham A, Lemaire JB. A real-time locating system
Abbreviations observes physician time-motion patterns during walk-rounds: a pilot
AP: Access point; BLE: Bluetooth Low Energy; ICU: Intensive care unit; study. BMC medical education. 2014;14:37.
RFID: Radio-frequency identification; RTLS: Real-time location system 11. Stubig T, Zeckey C, Min W, Janzen L, Citak M, Krettek C, Hufner T, Gaulke R.
Effects of a WLAN-based real time location system on outpatient contentment
Funding in a level I trauma center. Int J Med Inform. 2014;83(1):19–26.
This work was supported by the Information Technology Research and 12. Cagle R, Darling E, Kim B. Identifying healthcare activities using a real-time
Development program at Seoul National University Bundang Hospital location system. J. Med. Pract. Manage. 2014;30(2):128–30.
and SK Telecom. 13. Fernandez-Llatas C, Lizondo A, Monton E, Benedi JM, Traver V. Process mining
methodology for health process tracking using real-time indoor location
Availability of data and materials systems. Sensors (Basel). 2015;15(12):29821–40.
The datasets analyzed during the current study are available from the 14. Kamel Boulos MN, Berry G. Real-time locating systems (RTLS) in healthcare: a
corresponding author on reasonable request. condensed primer. Int J Health Geogr. 2012;11:25.
15. Li J: Just-in-time Management in Healthcare Operations. 2015.
Authors’ contributions 16. Wamba SF, Ngai EW. Importance of the relative advantage of RFID as enabler
SY and S designed the study and wrote the manuscript. EK, EJ, and KL of asset management in the healthcare: results from a Delphi study. In:
contributed to the data categorization and analysis. H contributed to the System Science (HICSS), 2012 45th Hawaii International Conference on:
discussion and reviewed the manuscript. All authors read and approved 2012: IEEE; 2012. p. 2879–89.
the final manuscript. 17. Voshall B, Piscotty R, Lawrence J, Targosz M. Barcode medication administration
work-arounds: a systematic review and implications for nurse executives. J. Nurs.
Ethics approval and consent to participate Adm. 2013;43(10):530–5.
This study was approved by the institutional review board at Seoul National 18. Dimitrov DV. Medical internet of things and big data in healthcare. Healthcare
University Bundang Hospital, South Korea (IRB no. X-1603/340–903). Consent informatics research. 2016;22(3):156–63.
was not required in this study because the analyzed data were from an existing 19. Baek M, Koo BK, Kim BJ, Hong KR, Kim J, Yoo S, Hwang H, Seo J, Kim D, Shin
anonymous questionnaire that was obtained during the pilot operation. K. Development and utilization of a patient-oriented outpatient guidance
system. Healthcare informatics research. 2016;22(3):172–7.
Consent for publication 20. Okoniewska B, Graham A, Gavrilova M, Wah D, Gilgen J, Coke J, Burden J,
Not applicable. Nayyar S, Kaunda J, Yergens D, et al. Multidimensional evaluation of a radio
frequency identification wi-fi location tracking system in an acute-care hospital
Competing interests setting. J. Am. Med. Inform. Assoc. 2012;19(4):674–9.
The authors declare that they have no competing interests. 21. Dardari D, Closas P, Djurić PM: Indoor tracking: theory, methods, and
technologies. IEEE Transactions on Vehicular Technology 2015, 64(4):
1263–1278.
Publisher’s Note 22. Yearp A, Newell D, Davies P, Wade R, Sahandi R. Wireless remote patient
Springer Nature remains neutral with regard to jurisdictional claims in monitoring system: effects of interference. In: Innovative Mobile and Internet
published maps and institutional affiliations. Services in Ubiquitous Computing (IMIS), 2016 10th International Conference
on: 2016: IEEE; 2016. p. 367–70.
Author details 23. Frisby J, Smith V, Traub S, Patel VL. Contextual computing: a Bluetooth based
1
Office of eHealth Research and Businesses, Seoul National University approach for tracking healthcare providers in the emergency room. J Biomed
Bundang Hospital, 166, Gumi-ro, Bundang-gu, Seongnam-si 436-707, South Inform. 2017;65:97–104.
Yoo et al. BMC Medical Informatics and Decision Making (2018) 18:80 Page 10 of 10

24. Yoo S, Jung SY, Kim S, Kim E, Lee KH, Chung E, Hwang H. A personalized
mobile patient guide system for a patient-centered smart hospital: lessons
learned from a usability test and satisfaction survey in a tertiary university
hospital. Int J Med Inform. 2016;91:20–30.
25. No-authors. Asset tracking systems. Health devices. 2006;35(11):397–419.
26. Faragher R, Harle R. An analysis of the accuracy of bluetooth low energy for
indoor positioning applications. In: Proceedings of the 27th international
technical meeting of the satellite division of the Institute of Navigation (ION
GNSS+‘14): 2014; 2014. p. 201–10.
27. Hurstel J, Guivarc'h M, Pommel L, Camps J, Tassery H, Cohen S, Bukiet F. Do
cell phones affect establishing electronic working length? J Endod. 2015;
41(6):943–6.