THE USE OF INFORMATION TECHNOLOGY TO ENHANCE

PATIENT SAFETY AND NURSING EFFICIENCY

Dosen pengampu
Mahmud Ady Yuwanto, S. Kep., Ns., M. M., M.Kep.

Disusun Oleh:
Riska Amalia
20010167

PROGRAM STUDI ILMU KEPERAWATAN

FAKULTAS KESEHATAN
UNIVERSITAS dr. SOEBANDI
JEMBER
2021
A. Judul analisis artikel jurnal
The Use Of Information Technology To Enhance Patient Safety And Nursing Efficiency.

B. Rangkuman isi artikel jurnal

Masalah keselamatan pasien dan efisiensi keperawatan telah lama menjadi perhatian.
Memajukan peran informatika keperawatan dipandang sebagai cara terbaik untuk mengatasi hal
ini. Pengenalan NIS meningkatkan keselamatan pasien dan efisiensi keperawatan dan
meningkatkan kepuasan perawat dan pasien. Menurut perkiraan dari US Institute of Medicine,
rata-rata setiap pasien rawat inap cenderung mengalami kesalahan dalam pemberian obat. Shih
dkk. menemukan bahwa mayoritas dari 3.054 tenaga medis di 309 rumah sakit yang
berpartisipasi menunjukkan bahwa kesalahan dalam pemberian obat adalah insiden medis yang
paling sering terlihat. Penggunaan barcode untuk mengelola pemberian obat adalah metode lain
untuk mengurangi kesalahan dalam pemberian obat. Barcode memiliki kedekatan dan
pengulangan, sehingga ketika informasi diwujudkan dalam barcode, pengguna dapat
menggunakan pemindai barcode untuk mendapatkan informasi yang benar dan mengkonfirmasi
identifikasi dan mengurangi kesalahan manusia.
Tujuan dari penelitian ini adalah untuk menentukan apakah penggunaan, hasil, dan kepuasan
dengan sistem informasi keperawatan mobile (NIS) meningkatkan keselamatan pasien dan
kualitas asuhan keperawatan di sebuah rumah sakit di Taiwan. Efisiensi keperawatan dinilai
dari kepuasan perawat dengan sistem serah terima shift elektronik, kepuasan pasien terhadap
pelayanan keperawatan, kelengkapan catatan keperawatan, dan tingkat pergantian staf.
Sebuah studi sebelumnya menemukan bahwa kepuasan dengan pengurangan waktu serah
terima, pengurangan kertas, peningkatan akurasi serah terima, dan peningkatan kenyamanan
adalah karena efisiensi serah terima itu .
Secara umum, pengurangan besar dalam transkripsi manual dan input setelah adopsi NIS
meningkatkan kepuasan keperawatan. Selain itu, fungsi pengingat sistem menyediakan
mekanisme pencegahan kesalahan ketika perawat melakukan berbagai tugas, dan oleh karena
itu meningkatkan keselamatan pasien dan kualitas perawatan medis. Sebagai hasil dari
pengenalan NIS di sebuah rumah sakit di Taiwan, nilai-nilai fisiologis lebih mudah dibaca dan
diinterpretasikan, dibutuhkan lebih sedikit waktu untuk melengkapi catatan elektronik, jumlah
kesalahan dalam pemberian obat berkurang, barcode mengurangi jumlah kesalahan
pengambilan darah dan transportasi sampel patologis, kepuasan perawat terhadap pergantian
shift elektronik meningkat secara signifikan, terjadi penurunan pergantian perawat, dan
kepuasan pasien meningkat secara signifikan.

C. Kelebihan
Artikel jurnal ini memaparkan dan menjelaskan secara jelas dan lengkap mulai dari latar
belakang permasalahan keselamatan pasien dan efisiensi keperawatan. Penulisan artikel
jurnal teratur sesuai dengan kaidah penulisan artikel jurnal.

D. Kelemahan
Penulis menyebutkan Ada keterbatasan dalam penelitian ini yaitu data hanya diperoleh dari
satu rumah sakit dan enam atau 12 bulan setelah subsistem NIS online. Lebih banyak situs dan
periode tindak lanjut yang lebih lama direkomendasikan.

E. Saran dan solusi

Sebaikanya penelitian ini dikembangkan lagi, tidak hanya memperoleh data dari satu Rumah
Sakit saja supaya pembaca mengetahui bagaimana dampak atau efek dari penggunaan
teknologi informasi yang di gunakan pada beberapa rumah sakit apakah teknologi informasi
berhasil meningkatkan keselamatan pasien serta efisiensi perawatan atau tidak.
Technology and Health Care 25 (2017) 917–928 917
DOI 10.3233/THC-170848
IOS Press

The use of information technology to

enhance patient safety and nursing efficiency

Tso-Ying Leea,b,∗ , Gi-Tseng Suna , Li-Tseng Koua and Mei-Ling Yehb

a Nursing Department, Cheng Hsin General Hospital, Taipei, Taiwan
b Taipei Nursing and Health Science University, Taipei, Taiwan

Received 16 February 2017

Accepted 18 July 2017

Abstract.
BACKGROUND: Issues in patient safety and nursing efficiency have long been of concern. Advancing the role of nursing
informatics is seen as the best way to address this.
OBJECTIVES: The aim of this study was to determine if the use, outcomes and satisfaction with a nursing information system
(NIS) improved patient safety and the quality of nursing care in a hospital in Taiwan.
METHOD: This study adopts a quasi-experimental design. Nurses and patients were surveyed by questionnaire and data re-
trieval before and after the implementation of NIS in terms of blood drawing, nursing process, drug administration, bar code
scanning, shift handover, and information and communication integration.
RESULTS: Physiologic values were easier to read and interpret; it took less time to complete electronic records (3.7 vs. 9.1
min); the number of errors in drug administration was reduced (0.08% vs. 0.39%); bar codes reduced the number of errors in
blood drawing (0 vs. 10) and transportation of specimens (0 vs. 0.42%); satisfaction with electronic shift handover increased
significantly; there was a reduction in nursing turnover (14.9% vs. 16%); patient satisfaction increased significantly (3.46 vs.
3.34).
CONCLUSIONS: Introduction of NIS improved patient safety and nursing efficiency and increased nurse and patient satisfac-
tion. Medical organizations must continually improve the nursing information system if they are to provide patients with high
quality service in a competitive environment.

Keywords: Nursing information system, patient safety, nurse efficiency, bar codes, shift handover

1. Introduction

Patient safety is an issue that has received increasing attention from medical organizations. The 2012
study of Andel et al. noted that mortality due to preventable medical errors was the sixth largest cause of
death in the United States, and preventable medical errors were also the cause of increased medical ex-
penditures [1]. Similarly, in an analysis of medical errors reported by Taiwan’s patient safety notification
system from 2005–2012, the Taiwan Joint Commission on Hospital Accreditation discovered that med-
ical errors causing major injury or death accounted for close to 3% of all such errors, and affected more
than 1,400 persons each year [2]. As a consequence, regardless of their professional role or the medical

∗
Corresponding author: Tso-Ying Lee, Deputy Director of the Nursing Department, Cheng Hsin General Hospital, No.
45 Cheng Hsin Street, Beitou District, Taipei 11219, Taiwan. Tel.: +886 2 28264400 6002; Fax: +886 2 28264574; E-mail:
ch4006@chgh.org.tw.

0928-7329/17/$35.00
c 2017 – IOS Press and the authors. All rights reserved
918 T.-Y. Lee et al. / NIS for safety and efficiency

situation, all healthcare professionals must share responsibility for ensuring that patients receive safe
care.
Nursing efficiency is also an important issue. Clinical nursing tasks include physical care, measure-
ment of vital signs [3], medication administration, blood drawing, and the maintenance of nursing
records. The content of nursing records includes the status of patient health and various assessments,
nursing activities and implementation of physicians’ orders, and appraisal of various tasks in patient
care. As a result, nursing records account for a large proportion of patient records. The keeping and
management of records not only has an impact on the quality of care, but also provides legal protec-
tion for both patients and health professionals. Nevertheless, one half of all nurses must stay at work
for 1–2 hours after the end of their shifts, and the chief reason for this is the need to complete nurs-
ing records [4]. The content of handwritten nursing records is not only repetitive, but also includes a
broad scope of descriptive information. Because of this, when time is limited, handwritten records often
contain errors and omissions, are difficult to read, and may lead to nursing errors or disputes [5].
According to estimates from the US Institute of Medicine, on average, each hospitalized patient is
likely to experience an error in medication administration [6]. Shih et al. found that a majority of the
3,054 medical personnel at 309 participating hospitals indicated that errors in medication administration
were the most commonly-seen medical incidents [7]. In addition to drafting clear technical standards for
medication administration, ensuring that nurses do not commit errors also requires training and audits,
and the names and photographs of drugs should be placed online for nurses’ reference. The use of
barcodes to manage medication administration is another method of reducing errors [8,9].
Specimen collection provides a basis for the correct diagnosis of hospitalized patients, and nurses play
an extremely important role in that process. This process is very complicated, and one study reported that
46% of errors occurred during collection [10]; this leads to a high rejection percentage and the frequent
need to collect new samples. Specimens may also be lost during transportation. Specimen errors not
only increase the clinical workload and the cost of medical care, but may also delay treatment and risk
patient safety [11]. Starting in 1970, medical organizations began to adopt barcode technology, especially
for such clinical purposes as blood transfusions, specimen collection, and medication administration.
Barcodes possess immediacy and repeatability, so when information is embodied in barcodes, users are
able to employ barcode scanners to obtain correct information and confirm identification, which reduces
human error [12].
Handing over their shift is one of the important, time-consuming tasks that nurses must perform every
day. By correctly handing over their shift, nurses can ensure that their colleagues receive the latest
information so that the continuity and integrity of care is maintained [13]. If the information transmitted
at shift handover is incomplete or in error, then this can potentially harm patients, and may also result in
complaints or a lack of trust among team members. When nurses hand over verbally or in writing, the
average shift handover time is 83.6 minutes, shift handover satisfaction percentage is 48.5% [14], and
shift handover completion percentage is 41% [15]. This approach to shift handover leads to poor nursing
efficiency and may compromise patient safety.
In order to fully implement patient safety, some hospitals in Taiwan have established hospital-wide
quality and patient safety management systems, such as by establishing hospital-wide patient safety
and quality management frameworks [2], establishing quality and patient safety management operating
models, and employing projects to reduce risks to patients’ safety [16]. In order to improve operating
procedures and provide an effective safety management model, various safety management functions,
including communications among personnel, incident notification, and risk management, can be imple-
mented via an information platform [17].
 T.-Y. Lee et al. / NIS for safety and efficiency 919

Internationally, improving nursing informatics is seen as the best way to improve hospital system per-
formance, collaboration, and the satisfaction of hospital personnel and patients as well [18–21]. The aim
of this study was to determine if the use, outcomes, and satisfaction with a mobile nursing information
system (NIS) improved patient safety and the quality of nursing care in a hospital in Taiwan.

2. Methods

2.1. Study design

A pre- and post-test quasi-experimental design was used.

2.2. Components of the NIS

2.2.1. Physiological value input and automatic drawing and calculation

This system interface includes blood pressure (BP), temperature (Temp), heart rate (HR), respiratory
rate (RR), central venous pressure (CVP), blood oxygen concentration (SpO2), input/output (I/O), an-
tibiotic use, special examinations, coma scale (GCS), pain scale (Pain), stool frequency (Stool), and
blood glucose (Sugar) measurements as input. Following data input, the system automatically produces
records, draws curve diagrams, and performs calculations; it also possesses a query function.

2.2.2. Nursing process

These systems include the GOLDEN nursing assessment system, North American nursing diagnosis
(NANDA) system, nursing record process (SOAPIE), and focus record method (FOCUS) systems, and
may incorporate specialized medical terminology phrase libraries.

2.2.3. Medication administration (including chemotherapy) (barcode point-of-care)

Drugs ordered by physicians are imported into the nursing drug administration system, after which
the system automatically displays the appearance, effect, and side effects of the drugs. When drugs
are administered, a nurse scans the drug’s barcode and also scans the patient’s wrist band to confirm
identification. The chemotherapy administration system also contains a tracking feature that can display
such information as the drug prescription, transportation, drugs present in patients’ rooms, previous drug
administration, and drug administration.

2.2.4. Specimen barcode scanning

When a physician orders an examination, management processes include printing a barcode and at-
taching it to a test tube, scanning the barcode and the patient’s bracelet, specimen collection, and barcode
reconfirmation.

2.2.5. Electronic shift handover

Physicians’ prescription, nursing care plans, tests, examinations, consultations, surgery, rehabilitation,
and various lines are linked with the shift handover system. An electronic whiteboard at the nursing
station provides links to physicians on duty, hospital room status, and special notes concerning patients.
This replaces conventional shift handover tasks involving handwritten notes on a whiteboard.

2.2.6. Integrated information and communications

These systems integrate hospital communications systems and NISs. The nursing manager can input
primary nurses’ cell phone numbers and the bed numbers of the patients they care for into the NIS
920 T.-Y. Lee et al. / NIS for safety and efficiency

system. This allows patients to contact primary nurses directly through their cell phones by pressing a
button in their rooms.

2.3. Participants

Convenience sampling was used to enroll nurses at a 1037-bed general hospital in Taiwan. Random
sampling was not employed, but in order to facilitate sampling, a questionnaire survey was administered
to all nurses who were working on a randomly selected day. Respondent nurses had to have worked at
the hospital for at least three months and to have had at least three years of nursing experience. A total
of 100 nurses completed the pre-test, and 101 completed the posttest. Since responses were anonymous,
these were not necessarily the same nurses. Respondent patients were required to have been hospitalized
for at least three days. Patient satisfaction surveys were collected in March and September. A total of
639 patients completed the pre-test, and 615 patients completed the posttest.

2.4. Study instruments

We monitored quality of care and nursing efficiency. Quality of care measures are determined by Tai-
wan’s Joint Commission on Accreditation of Hospitals (JCAH) and collected by each hospital’s Quality
Management Center. These include the number of near misses, medication error rate, and the specimen
error rate. Nursing efficiency was assessed by nurse satisfaction with the electronic shift handover sys-
tem, patient satisfaction with nursing service, completion of nursing records, and staff turnover rate. A
previous study found that satisfaction with reduced handover time, paper reduction, increased accuracy
of the handover, and increased convenience was due to the efficiency of that handover [22].
A structured questionnaire was used for data collection and was based on a literature review, our own
clinical experience, and recommendations by experts in the field. The questionnaire included sections
on:
1. Nurses’ satisfaction with the electronic shift handover system
Eight questions scored on Likert scales: very satisfied (four points), satisfied (three points), unsat-
isfied (two points), and very unsatisfied (one point). The content validity index (CVI) was 0.93.
Cronbach’s α was 0.92 for internal consistency.
2. Patients’ satisfaction with nursing services
Eleven questions scored on Likert scales: very satisfied (four points), satisfied (three points), un-
satisfied (two points)’, very unsatisfied (one point). The content validity index (CVI) was 0.91.
Cronbach’s α was 0.87for internal consistency reliability. This study employed only one indica-
tor connected with information and communications as a reference item – “Speed of response and
assistance after a nurse has been called”.

2.5. Data collection

A pre-test was administered before each subsystem was implemented. The posttest was administered
after the NIS subsystems had been completed, and data were collected from 2011 to 2014. The data
were based on the results of quality monitoring indicators of the Quality Management Committee of the
 T.-Y. Lee et al. / NIS for safety and efficiency 921

Table 1
Indicators of patient safety and nursing efficiency before and after system introductions
Items Pre-test Post-test 1 Post-test 2
Mean score of nursing record completions 94.7 for reading 32 charts 97.4 for reading 35 charts –
Mean time needed to write each nursing record 9.1 to write 32 charts 3.7 to write 35 charts –
(min.)
Drug administration near miss events 0.0307% (69/224392) 0.0458% (102/222415) –
Chemotherapy drug administration error 0.39% (12/3085) 0.08% (1/1214) 0.04% (1/2573)
(number/person-day)
Blood drawing specimen error events (error 0.004% (10/239013) 0% (0/290017) 0% (0/287007)
events/total events)
Specimens spending more than 2 hours in transit 21% (2724/12903) 12% (1578/13047) –
(error events/total events)
Pathology specimen errors percentage (error 0.42% (48/11335) 0.01% (13/11335) –
events/total events)
Nurse turnover over the year (number leaving 16% (119/742) 14.9% (108/727) 13.7% (102/743)
staff/total staff)

Department of Nursing. Our hospital has established a quality monitoring division in each department.
For example, nursing records are audited for completeness ever six months, 30 records per audit. The rate
of correct administration of chemotherapy is audited every month, 30 records per audit. The accuracy
rate for transportation of pathological specimens is audited every month, 30 records per audit. Medical
errors such as blood samples from the wrong patient, administration of incorrect medication, and near
miss events are monitored by the Quality Center in accordance with JCAH requirements. This study
obtained IRB approval (CHGH-IRB 105E-10).

2.6. Data analysis

The data were analyzed using SPSS 21.0 statistical software. The data for registered nurses’ satisfac-
tion with NIS were summarized as n (%) for each item. Difference between post-test 1 and post-test 2
were compared using Pearson Chi-square/or Fisher’s exact test. Statistical assessments were two-tailed
and considered significant at p < 0.05.

2.7. Results

After complete adoption of the NIS, the system included six subsystems.

2.7.1. Physiological value input and automatic drawing and calculation

This system was completed in December 2011. Following the computerization and implementation of
automatic drawing, numerical values were clear, neat, and easy to interpret. The system automatically
calculates inputs and outputs, which makes human errors in calculation much less likely. For example,
blood glucose curves can be compared with text records, making it easy to identify changes in a patient’s
blood glucose level.

2.7.2. Nursing process

This system was completed in July 2012. In June 2011, the average completion percentage of written
nursing records was 94.7% (n = 32), and it took an average of 9.1 minutes to complete handwritten
records. In December 2012, the average nursing record completion percentage was 97.4% (n = 35), and
it took an average of 3.7 minutes to complete electronic records. The record format and content are neat
and easy to read (Table 1).
922 T.-Y. Lee et al. / NIS for safety and efficiency

2.7.3. Drug administration (including chemotherapy)

This system was completed in September 2012, and includes a drug identification function. During
the one-year period prior to completion, there were 69/224392 (0.0307%) near misses defined as nurses
identifying the wrong drug or the wrong dose or the wrong patient before administration and 102/222415
(0.0458%) during the one-year period following completion of the system. This indicates that the system
design effectively prevented more errors in drug administration. With regard to the system’s chemother-
apy drug barcode scanning function, the average error percentage was 0.39% (12/3085) during the first
nine months of 2012 before the system was completed. After the system was completed, the error per-
centage dropped to 0.08% (1/1214) from October to December 2012. After system optimization, the
error percentage was 0.04% (1/2573) during the first nine months of 2013 (Table 1).

2.7.4. Specimen barcode scanning

The ordinary specimen barcode scanning system was completed in December 2012. Before the system
was completed, human mistakes led to 10 cases of blood drawing errors during 2012. After the barcode
identification system was implemented, blood drawing errors fell to zero during 2013, and there was
only one such error in 2014. The pathology specimen barcode system was completed in December
2014. Prior to completion, the average error rate for transportation of pathology specimens was 0.42%
(48/11544) during 2014. After the system was completed, the error percentage fell to zero. The fact that
it is no longer necessary to print out forms saves approximately USD$ 685 in paper costs each year, and
the reduction in repeated blood drawings saves approximately USD$ 370 each year. The percentage of
specimens spending more than 2 hours in transit also fell from 21% (2724/12903 to 12% (1578/13047)
after this system was implemented (Table 1).

2.7.5. Electronic shift handover

The electronic shift handover system was completed in December 2013. In March 2014, a survey of
100 nurses found that average overall satisfaction with the electronic shift handover system was 2.83 ±
0.64 out of a possible 4 points. The system was then optimized on the basis of user recommendations so
that it could present testing and consultation results in a comprehensive and systematic manner on the
shift handover screen. In December 2014, a survey of 101 nurses found that the average satisfaction score
rose significantly to 3.27 ± 0.59 points (p < 0.001). Since the shift handover operating system consist-
ing of nursing station electronic whiteboards was completed in December 2014, this system has saved
nurses 15.8 minutes of report writing time each day. (Results not shown) Table 2 shows the registered
nurses’ satisfaction with NIS. Satisfaction (extremely satisfied + satisfied) with this system increased
after system optimization except for “reduction in paper expenses” (all p-values < 0.05).

2.7.6. Information and communications integration

The information and communications integration system was completed in March 2014. In a survey
of 603 hospitalized patients taken in March 2014, before the system was in use, the patients expressed
an average satisfaction of 3.39 ± 0.5 out of 4 points on the item “Nurses respond rapidly and provide
necessary assistance when call bells are used.” In September 2014, after the system was completed, in
a survey of 615 patients, the respondents expressed an average satisfaction of 3.46 ± 0.5. (Result not
shown), and the rate of satisfaction increased after system optimization (p = 0.04). In addition to this
item, patients’ satisfaction on items “Friendly attitude on the part of nurses” and “Nurses pay attention to
my privacy and provide the appropriate covering when performing treatment” increased after the system
optimization (both = 0.01) (Supplemental Table 1).
 T.-Y. Lee et al. / NIS for safety and efficiency 923

Table 2
Registered nurses’ satisfaction with NIS
Post-test 1 (n = 100) Post-test 2 (n = 101)
Item Extremely Satisfied Dissatisfied Extremely Extremely Satisfied Dissatisfied Extremely p-value
satisfied dissatisfied satisfied dissatisfied
1. Can shorten shift 10 63 23 4 32 54 14 1 0.001∗∗
handover time (10.00%) (63.00%) (23.00%) (4.00%) (31.68%) (53.47%) (13.86%) (0.99%)
2. Can reduce verbal 17 57 25 1 34 57 10 0 0.004∗∗
shift handover (17.00%) (57.00%) (25.00%) (1.00%) (33.66%) (56.44%) (9.90%) (0.00%)
transcription errors
3. Reduction in 7 62 30 1 31 61 9 0 0.001∗∗
errors due to (7.00%) (62.00%) (30.00%) (1.00%) (30.69%) (60.40%) (8.91%) (0.00%)
misheard
information
4. Reduction in paper 42 48 7 3 46 53 2 0 0.10
expenses (42.00%) (48.00%) (7.00%) (3.00%) (45.54%) (52.48%) (1.98%) (0.00%)
5. Enhancement of 17 56 26 1 46 47 8 0 0.001∗∗
the convenience of (17.00%) (56.00%) (26.00%) (1.00%) (45.54%) (46.53%) (7.92%) (0.00%)
shift handover
6. Enhancement of 11 61 26 2 51 41 9 0 0.001∗∗
the integrity of the (11.00%) (61.00%) (26.00%) (2.00%) (50.50%) (40.59%) (8.91%) (0.00%)
content at shift
handover
7. Enhancement of 11 71 16 2 44 51 6 0 0.001∗∗
patient care safety (11.00%) (71.00%) (16.00%) (2.00%) (43.56%) (50.50%) (5.94%) (0.00%)
8. Overall 11 63 24 2 35 58 8 0 0.001∗∗
satisfaction with (11.00%) (63.00%) (24.00%) (2.00%) (34.65%) (57.43%) (7.92%) (0.00%)
the electronic shift
handover system
∗
p < 0.05, ∗∗ p < 0.01, ∗∗∗ p < 0.001 compared with pre-system optimization.

Generally speaking, the major reduction in manual transcription and input after the adoption of a NIS
enhanced nursing satisfaction. Nurse turnover fell from 16% in 2011, before computerization, to 14.9%
in 2012, after computerization, and fell further to 13.7% in 2013 (Table 1). In addition, the system’s
reminder function provides a mistake-prevention mechanism when nurses are performing various tasks,
and therefore enhances patient safety and the quality of medical care. Finally, by making paper printouts
unnecessary, the system has saved the hospital a considerable amount of money each year.

3. Discussion

As a result of the introduction of NIS at a hospital in Taiwan, physiologic values were easier to read
and interpret, it took less time to complete electronic records, the number of errors in drug administration
was reduced, barcodes reduced the number of errors in blood drawing and transportation of pathological
samples, nurses’ satisfaction with electronic shift turnover increased significantly, there was a reduction
in nursing turnover, and patient satisfaction increased significantly.
Data input improved and this result was similar to that of Chang et al. [23]. Although mobile devices
are available to every health professional, internet coverage was available in only 85% of the hospital.
Corners in certain hospital wards had poor signal reception and a change in location was required to stay
924 T.-Y. Lee et al. / NIS for safety and efficiency

connected This affected the level of convenience; this result was the same as that reported by Tseng et
al. [24] and Lee [25].
It was discovered during development of the nursing record system that there were differences between
wards in the hospital. In addition to the database’s original standard NANDA classification, various
customized “commonly-used classifications” were added on the basis of a ward’s attributes during the
optimization process in order to facilitate the rapid determination of suitable nursing diagnoses by users.
Similarly, a medical terminology library reflecting a ward’s attributes was also added to save time and to
make it easier and more convenient for nurses to complete nursing records. As a result, satisfaction with
the system has increased. Development of the system took advantage of innovation diffusion theory
and the designers and users communicated with each other and came up with new ideas during the
process of system development; this ensured that the system’s information design met the need for
customization [26].
The design of the medication administration system allowed nurses to move a mouse cursor to a drug’s
location, and this causes the system to automatically display the appearance and effect of the drug. This
system allowed nurses to identify the drugs and provide consultation to patients with regard to the use
of the medication. It also improved work flow, a result similar to the observations of Early et al. [27]
and Huang ane Lee [28]. The fact that near misses in medication administration have increased since the
implementation of the system indicates that system design can effectively prevent errors. Furthermore,
the system uses sounds to draw nurses’ attention to errors in identification; this has increased nurses’
vigilance and increased the safety of medication administration. The enhanced efficiency resulting from
the system is consistent with the findings of Song et al. [8].
The design of the specimen barcode scanning system requires the signatures of two persons in order
to complete computerized tasks. This design has eliminated the difficulty in obtaining both signatures
which was often encountered when the tasks were performed manually. It also lessened the risk of error
during the process of specimen transportation. Because of computer-controlled procedures, both parties
must now sign before the system can proceed to the next step. Many problems with patient safety are
associated with procedures, and the computerization of operating procedures will significantly improve
patient safety [29]. Although the system was designed to be user-friendly, if users fail to follow standard
operating procedures, they may commit errors. Even with computerization, the hospital’s administration
must still continue to educate personnel and perform process audits to ensure effective handling. This
observation is similar to that of Early et al. [27] and Chen et al. [30].
The nurses had a favorable impression of the electronic shift handover system. Because this system is
fully linked with other systems and provides clear reports from the testing and examination systems, it
is extremely convenient and can provide information very rapidly. The nurses were uncomfortable with
fully paperless operations during the initial period after the system went online, however, and remained
in the habit of manual transcription of information when performing shift handover. At such times,
designers must give users support in the form of additional training and awareness; this will ensure
that users become more confident and gradually accept the new system [31]. In addition, nurses’ levels
of satisfaction with the level of convenience of the electronic shift handover system and the level of
completeness of the provided information were mostly above the level of somewhat satisfied, which is
consistent with the result of the study conducted by Lai et al. [32]. In the event of an unstable internet
connection or a system crash, however, data acquisition will be interrupted and an obstacle to work.
Therefore, maintaining internet stability via hardware is a very important supplementary measure.
With regard to information and communications integration, the increase in satisfaction with regard
to “Speed of response and assistance after a nurse has been called” was statistically significant. This
 T.-Y. Lee et al. / NIS for safety and efficiency 925

indicates that patients used this system to directly contact primary nurses, which increased the nurses’
service speed and efficiency. Unexpectedly, medical personnel including physicians, medical technolo-
gists, operating room personnel, and testing personnel also had quite positive assessments of the system.
This was because physicians and examination room, testing room, and operating room personnel also
need to contact primary nurses, and can use this system to quickly locate them. This system has not only
increased correct communication, but also boosted the efficiency of communication between different
areas of specialization. The system complies with the Taiwan Joint Commission on Hospital Accredita-
tion requirements for strengthened professional communication between different teams for the purpose
of increasing patient safety [33].
One problem with this system is that personnel must have the ability to immediately revert to a manual
operating mode if the system temporarily cannot be used due to power outage or a computer crash, and
to be able to use paper and pen to record measurement results and times, so that clinical work can
proceed normally. Nevertheless, hospital personnel have gotten into the habit of implementing various
clinical tasks with assistance from the information system, and would find it very difficult to re-accustom
themselves to manual tasks. As a consequence, in order to minimize disruption from system stoppage,
all hospital wards and units currently implement regular manual drills. In addition, unstable wireless
transmission may make it difficult to accomplish tasks.
There were limitations to this study. Data were obtained from only a single hospital and six or
12 months after subsystems of the NIS went online. More sites and longer follow-up periods are rec-
ommended. The respondent sample size was small and results may not be generalized. Because of the
sampling procedure, the same patient may have completed the questionnaire more than once, but this
is unlikely to have affected the results. Because the hospital’s Wi-Fi coverage was only 85% during
the period of system construction, this may have influenced their assessment of system efficiency. A
satisfaction survey for written shift handover was not conducted before completion of the electronic
handover system; therefore, this survey shows the difference in satisfaction only after development and
after optimization of the system.

4. Conclusions

This study provided solid evidence for the importance of NIS in the care of hospitalized patients. Pa-
tient safety was improved as was the level of satisfaction of both nurses and patients. The next step in the
process would be to link systems via a wireless network [34,35]. The rapid development of the internet
and its use in accessing a vast array of information has created a unique computing environment. Cloud
technology enables medical personnel to query information via the internet to enhance the immediacy
of medical applications. Medical organizations must continually improve if they are to provide patients
with high quality service in a competitive environment.

Acknowledgments

This study received financial and policy support from the hospital while it was underway. Although
the process of information system development was very protracted, the project team’s strong consensus
and mature cooperation enabled us to proceed. Although communication with computer engineers re-
quired large amounts of time during the initial period, differences in specializations training, and thought
processes led to different solutions and expectations. While some specialized medical terminology was
926 T.-Y. Lee et al. / NIS for safety and efficiency

unfamiliar to the computer engineers, the project team was ultimately able to achieve a high degree of
tacit understanding after a period of time and consensus-building. The computer engineers consequently
became important partners, and have helped create nursing value.

Conflict of interest

All authors declare that they have no conflict of interest.

Funding

This study was sponsored by Cheng Hsin General Hospital (Grant number: CHGH106-31).

References
[1] Andel C, Davidow SL, Hollander M, Moreno DA. The economics of health care quality and medical errors. J Health
Care Finance 2012; 39(1): 39-50.
[2] Taiwan Joint Commission on Hospital Accreditation: 2012 Annual statement of the Taiwan patient safety notification
system 2012. http://tpr.org.tw/index03.php?getid=year [Taiwan Joint Commission on Hospia1 Accreditation: Taiwan
Patient-safety reporting system Annual Report 2012. http://www.tpr.org.tw/index03.php?getid=year].
[3] Bauer JC. Automated, wireless, continuous, intelligent, and noninvasive systems for risk management and performance
improvement in vital signs monitoring. J Healthc Risk Manag 2013; 33(1): 12-16. Doi: 10.1002/jhrm.21113.
[4] Feng J, Lin P. Factors causing nurses to delay leaving work. Veterans General Hospital Nursin 2010; 27: 10-19. [article
in Chinese]
[5] Dal Sasso G, Barra D, Paese F, de Almeida SR, Rios GC, Marinho MM, et al. Computerized nursing process: Methodol-
ogy to establish associations between clinical assessment, diagnosis, interventions, and outcomes. Rev Esc Enferm USP
2013; 47(1): 238-245 [article in Spanish].
[6] Aspden P, Wolcott J, Bootman JL, Cronenwett LR. Preventing medication errors: Qualitychasmseries. 2007. Retrieved
from http://www.nap.edu/catalog/11623.html.
[7] Shih C, Hou S, Hsueh Y, Chung K, Su S, Liao H. Abnormal incident notification system and notification obstacles.
Taiwan Medical Journal 2007; 9: 63-70.
[8] Song L, Park B, Oh KM. Analysis of the technology acceptance model in examining hospital nurses’ behavioral inten-
tions toward the use of barcode medication administration. Comput Inform Nurs 2015; 33(4): 157-165.
[9] Holden RJ, Brown RL, Scanlon MC, Karsh BT. Modeling nurses’ acceptance of bar code medication administration
technology at a pediatric hospital. J Am Med Inform Assoc 2012; 19(6): 1050-1058.
[10] Becan-McBride K. Laboratory sampling: Does the process affect the outcome? J Intrav Nurs 1999; 22(3): 137-142.
[11] Liang H, Chang F, Wang S, Lin W. Application of barcode technology to the improvement of the effectiveness of medical
specimen transport and examination. Yuan-Yuan Nursing 2014; 8: 27-35. [article in Chinese]
[12] Yuan H, Weng S. Application of barcodes to outpatient HIS. Chinese Medical Equipment Journal 2009; 30: 46-47.
[article in Chinese]
[13] Penaforte MH, Martins MM. The visibility of hygiene self-care in nurse-to-nurse shift change reports. Revista Latino-
Americana Enfermagem 2011; 19(1): 131-139 [article in Portuguese].
[14] Yeh S, Tsai Y, Kang Y, Hsu H. Shortening average shift handover time at a surgical intensive care center. Yuan-Yuan
Nursing 2015; 9: 45-54. [article in Chinese]
[15] Lai L, Yang H, Chang H, Chen H, Chang M. Boosting the integrity of intensive care electronic nursing shift handover
procedures. Cheng Ching Medical Journal 2015; 11: 66-75.
[16] Huang H, Chen L, Hsu P, Szu-Fen Huang S, Lin C, Kang H, et al. Patient safety and quality improvement: From
management framework to improvement projects. Formosan Journal of Medicine 2014; 18: 558-565. Doi: 10.6320/
FJM.2014.18(5).09. [article in Chinese]
[17] Feng J, Yeh Y. New nursing perspectives – The evolution and development of nursing information. Journal of Nursing
2012; 61: 78-84. [article in Chinese]
[18] Peltonen LM, Topaz M, Ronquillo C, Pruinelli L, Sarmiento RF, Badger MK, et al. Nursing informatics research priori-
ties for the future: Recommendations from an international survey. Stud Health Technol Inform 2016; 225: 222-226.
 T.-Y. Lee et al. / NIS for safety and efficiency 927

[19] Peltonen LM, Alhuwail D, Ali S, Badger MK, Eler GJ, Georgsson M, et al. Current trends in nursing informatics: Results
from an international survey. Stud Health Technol Inform 2016; 225: 938-939.
[20] Topazz M, Ronquillo C, Peltonen LM, Pruinelli L, Sarmiento RF, Badger MK, et al. Advancing nursing informatics in
the next decade: Recommendations from an international survey. Stud Health Technol Inform 2015; 225: 123-127.
[21] Karimi F, Poo DC, Tan Y. Clinical information systems end user satisfaction: The expectations and needs congruencies
effects. J Biomed Inform 2015; 53: 342-354.
[22] Lin HC, Chen YC, Lin CH. Promotion of electronic medical records integrity of nursing handover in intensive care unit.
Tzu Chi Nursing Journal 2013; 12: 77. [article in Chinese]
[23] Chang FM, Lee TT, Lin KC. Exploring nurse usage effectiveness of mobile nursing station. Journal of Clinical Nursing
2013; 60: 32-40. [article in Chinese]
[24] Tseng HM, Chan PT, Chiang CY. Applying technology acceptance model to examine factors related to nurses’ accep-
tance for the use of mobile nursing station in hospitals. Journal of the Taiwanese Association for Medical Informatics
2009; 18: 23-38. [article in Chinese]
[25] Lee TT. Nursing Information Users’ experience of system in Taiwan one year after its implementation. Journal of Clinical
Nursing 2008; 7: 763-771. [article in Chinese]
[26] Lee TT, Shih YS. Using innovation diffusion theory to improve implementation of nursing information. Hu Li Za Zhi
2009; 56(3): 18-22.
[27] Early C, Riha C, Martin J, Lowdon KW, Harvey EM. Scanning for safety: An integrated approach to improved bar-code
medication administration. Comput Inform Nurs 2011; 29(3): 157-164.
[28] Huang HY, Lee TT. Impact of bar-code medication administration on nursing activity patterns and usage experience in
Taiwan. Comput Inform Nurs 2011; 29(10): 554-563.
[29] Teng C, Cheng Y, Chen S, Li Y, Chang S, Chen Y, et al. Applications and overall performance of an intelligent healthcare
innovation system. J Healthc Manag 2014; 15: 128-149.
[30] Chen HC, Huang SF, Liu TS, Ye YY, Shen YC, Tsai LH. Implementation of bar code technology to minimize labeling
errors of specimens collected by nurses. Cheng Vhing Medical Journal 2012; 8(1): 56-64. [article in Chinese]
[31] Vollmer AM, Prokosch HU, Evans S, Kuttler K. Evaluation of acceptance of nursing information system in a German
and American hospital. Stud Health Technol Infor 2016; 225: 222-226.
[32] Lai LC, Yang HC, Chang HJ, Chen HY, Chang MY. Effectiveness of a computerized handover system in an intensive
care unit. Cheng Vhing Medical Journal 2015; 11(3): 66-75. [article in Chinese]
[33] Joint Commission on Accreditation of Healthcare Organizations (January 1, 2010). 2010 National Patient Safety Goals.
Retrieved May 1 2016 from http://www.jointcommission.org/PatientSafety/NationalPatientSafetyGoals.
[34] Jarvis B, Johnson T, Butler P, O’Shaughnessy K, Fullam F, Tran L, et al. Assessing the impact of electronic health
records as an enabler of hospital quality and patient satisfaction. Acad Med 2013; 88(10): 1471-1477.
[35] Appari A, Eric Johnson M, Anthony DL. Meaningful use of electronic health record systems and process quality of care:
Evidence from a panel data analysis of U.S. acute care hospitals. Health Serv Res 2013; 48(2 Pt 1): 354-355.
928 T.-Y. Lee et al. / NIS for safety and efficiency

Supplement material

Supplemental Table 1
Patient satisfaction survey
Post-test 1 (n = 603) Post-test 2 (n = 615)
Item Extremely Satisfied Dissatisfied Extremely Extremely Satisfied Dissatisfied Extremely p-value
satisfied dissatisfied satisfied dissatisfied
1. Friendly attitude on 290 304 8 1 341 272 1 1 0.01∗
the part of nurses (48.09%) (50.42%) (1.32%) (0.17%) (55.45%) (44.23%) (0.16%) (0.16%)
2. Nurses respond rapidly 247 343 12 1 299 306 10 0 0.04∗
and provide necessary (40.96%) (56.88%) (1.99%) (0.17%) (48.62%) (49.76%) (1.62%) (0.00%)
assistance when call
bells are used
3. When I need assistance, 265 326 10 2 312 293 7 3 0.10
nurses do their utmost (43.94%) (54.06%) (1.67%) (0.33%) (50.73%) (47.64%) (1.13%) (0.50%)
to help me
4. Nurses use a language I 270 330 2 1 307 306 1 1 0.32
understand, and ac- (44.78%) (54.72%) (0.33%) (0.17%) (49.91%) (49.76%) (0.16%) (0.16%)
tively tell me guide-
lines for my period of
hospitalization
5. Nurses actively explain 245 348 9 1 281 327 7 0 0.24
bed and wheelchair (40.63%) (57.71%) (1.49%) (0.17%) (45.69%) (53.17%) (1.14%) (0.00%)
usage and safety in-
structions
6. Nurses pay attention to 264 328 9 2 322 289 3 1 0.01∗
my privacy and provide (43.79%) (54.39%) (1.49%) (0.33%) (52.36%) (47.00%) (0.48%) (0.16%)
appropriate covering
when performing treat-
ment
7. Nurses confirm my 294 303 5 1 322 286 7 0 0.40
identity before each (48.75%) (50.25%) (0.83%) (0.17%) (52.35%) (46.51%) (1.14%) (0.00%)
treatment session
8. Nurses provide a 276 319 6 2 307 303 4 1 0.44
detailed explanation of (45.77%) (52.90%) (0.99%) (0.34%) (49.92%) (49.27%) (0.65%) (0.16%)
the entire process in a
language I understand
before examination or
treatment
∗
Preparatory matters
9. Nurses actively explain 272 313 14 4 289 309 16 1 0.50
drug usage and effects (45.10%) (51.90%) (2.32%) (0.68%) (46.99%) (50.24%) (2.60%) (0.17%)
10. I can obtain a prompt 231 355 14 3 274 324 15 2 0.16
response or action (38.30%) (58.87%) (2.32%) (0.51%) (44.55%) (52.68%) (2.43%) (0.34%)
when I make a com-
plaint or recommenda-
tion
∗
p < 0.05, ∗∗ p < 0.01, ∗∗∗ p < 0.001 compared with pre-system optimization.