Content area
A conventional Nursing Information System (NIS), which supports the role of nurse in some areas, is typically deployed as an immobile system. However, the traditional information system can't response to patients' conditions in real-time, causing delays on the availability of this information. With the advances of information technology, mobile devices are increasingly being used to extend the human mind's limited capacity to recall and process large numbers of relevant variables and to support information management, general administration, and clinical practice. Unfortunately, there have been few studies about the combination of a well-designed small-screen interface with a personal digital assistant (PDA) in clinical nursing. Some researchers found that user interface design is an important factor in determining the usability and potential use of a mobile system. Therefore, this study proposed a systematic approach to the development of a mobile nursing information system (MNIS) based on Mobile Human-Computer Interaction (M-HCI) for use in clinical nursing. The system combines principles of small-screen interface design with user-specified requirements. In addition, the iconic functions were designed with metaphor concept that will help users learn the system more quickly with less working-memory. An experiment involving learnability testing, thinking aloud and a questionnaire investigation was conducted for evaluating the effect of MNIS on PDA. The results show that the proposed MNIS is good on learning and higher satisfaction on symbol investigation, terminology and system information.[PUBLICATION ABSTRACT]
J Med Syst (2012) 36:11391153 DOI 10.1007/s10916-010-9576-y
ORIGINAL PAPER
A Mobile Nursing Information System Based on Human-Computer Interaction Designfor Improving Quality of Nursing
Kuo-Wei Su & Cheng-Li Liu
Received: 5 February 2010 /Accepted: 16 August 2010 /Published online: 9 September 2010 # Springer Science+Business Media, LLC 2010
Abstract A conventional Nursing Information System (NIS), which supports the role of nurse in some areas, is typically deployed as an immobile system. However, the traditional information system cant response to patients conditions in real-time, causing delays on the availability of this information. With the advances of information technology, mobile devices are increasingly being used to extend the human minds limited capacity to recall and process large numbers of relevant variables and to support information management, general administration, and clinical practice. Unfortunately, there have been few studies about the combination of a well-designed small-screen interface with a personal digital assistant (PDA) in clinical nursing. Some researchers found that user interface design is an important factor in determining the usability and potential use of a mobile system. Therefore, this study proposed a systematic approach to the development of a mobile nursing information system (MNIS) based on Mobile Human-Computer Interaction (M-HCI) for use in clinical nursing. The system combines principles of small-screen interface design with user-specified requirements. In addition, the iconic functions were designed with metaphor concept that will help users learn the system more quickly with less working-memory. An experiment involving
learnability testing, thinking aloud and a questionnaire investigation was conducted for evaluating the effect of MNIS on PDA. The results show that the proposed MNIS is good on learning and higher satisfaction on symbol investigation, terminology and system information.
Keywords Human-computer interaction .
Nursing information system . Personal digital assistant . Small-screen interface design
Introduction
Mobile technology in clinical environments
With the advances of information technology, most medical institutions have adopted information technology (IT) to improve their efficiency and quality. Computerization for daily operation procedure is accepted by the medical field at present, but with the leading edge that mobile devices provide, there have been some changes in the nursing practice over the years. In recent years, mobile devices have been brought increasingly into the medical field. Nurses and physicians are highly mobile occupations due to the fact that they move frequently between wards, offices, conference rooms, and outpatient clinics. Stolworthy and Suszka-Hildebrandt (2001) indicated that handheld computers, or personal digital assistants (PDAs), offer the ideal solution for accessing and managing the cornucopia of information being produced in the healthcare industry each year [1]. Li et al., (2005) indicated mobile nursing technology refers to a combination of portable computers and wireless communication and networks which offers portability and mobile access to nurses needing information at the point of care [2]. Numerous studies have demonstrated widespread adoption of handheld computers in
K.-W. SuDepartment of Information Management,National Kaohsiung First University of Science and Technology, No. 2 Jhuoyue Rd., Nanzih,Kaohsiung, Taiwan 811, Republic of China
C.-L. Liu (*)
Department of Management and Information Technology, Vanung University,No. 1 Vannung Rd., Chungli,Taoyuan, Taiwan 320, Republic of Chinae-mail: [email protected]
1140 J Med Syst (2012) 36:11391153
health care, and many information technology leaders and executives have agreed that handheld computers will have a significant role in the future of health care [3, 4].
As mentioned above, PDAs are being increasingly used in healthcare. PDAs have the ability to allow clinicians to enter and access data both remotely and at the point of care [5]. The PDA provides a new method of information management for nursing personnel via a mobile nursing system. P. Yu and H. Yu (2004) mentioned that the PDA gives promise for pervasive computing to enable providing the right information to the right health care provider about the right patient at the right time and at the right place [6]. A PDA can be used as a medical decision support system (MDSS) to obtain a profile of the patients medication and to check for inappropriate drugs and drug combinations, and to reduce medication errors [7]. Additionally mobile information and communication systems incorporated into the clinical routine have the potential to greatly improve communication, facilitate information access, eliminate redundant documentation, and increase the quality of patient care in the long run [8]. Therefore, a few domestic and international hospitals are already using PDAs to improve medical personnel operation. However, the interface designer for such mobile systems faces a great challenge: because of the small-screen of the PDA, size constraints make it harder for the user to interact with the content.
Mobile human-computer interaction (M-HCI)
Human-computer interaction (HCI) is concerned with investigating the relationship between people and computer systems and applications. Users/operators generally dont understand the computers inner world of bits, bytes, files, RAM, etc., but they understand the computer through its interface, and the text and images that appears on the screen. Hence, a popular saying in the computer world is that for the user: The Interface is the System [9]. A designer should concern himself with understanding the users, their various capabilities and expectations and with how these can be taken into consideration in the mobile system or application design [10]. The consideration of mobile human-computer interaction (M-HCI) on user interface could follow HCI design concepts includes metaphors, cognition, navigation, appearance and usability should be considered [11]. Briefly, understanding who the users are, what tasks they are doing, and where they are using the system should be a critical component of HCI investigations. It is extremely important thus, when designing usable interactive systems, to consider who is going to be using the systems and where they are going to be used. Therefore, how to design a mobile nursing information system (MNIS) on PDA that meets specific user requirements is the emphasis of the present research.
Objectives of this study
The PDA holds a significant role in the medical field as previously described. Indeed, PDA has inherent hardware limitations. However, the screen size and other main advantages of a PDAthat it is small and highly portable probably will not change. Unfortunately, there have been few studies about the combination of a well-designed small-screen interface with a PDA device in clinical nursing. Determining how to appropriately display information and utilize colors or icons to help user learning on limited small screen is a great challenge. The objectives of this study were the following:
(1) To determine which nursing tasks could be appropriately performed on a PDA and, by referring to principles of M-HCI and stated user requirements, develop a prototype of a MNIS which has a friendly for nurses and support for clinical nursing.
(2) To verify that the MNIS is easy-to-use, with high acceptance, and is helpful to nurses in clinical nursing by usability analysis.
Systematic approach to the development of MNIS
Initial analysis
The initial analysis was a reasoning process for identifying specification of the context of use and of the users requirements in consideration of the small-screen interface design with PDA. The study proceeds with two semi-structured interviews with 5 ward nurses. The first interview adopted workflow analysis, job analysis and interview methods to clearly understand nurses daily work and focused on investigation of the current status in order to define the usage context specification. Therefore, the study investigated a private hospital in the south of Taiwan, which has been striving for several years to improve the quality of medical care through a desktop medical information system. Thoroughly understanding the operation, functions, content and environment of the system, the accurate and necessary medical information could be considered to construct into the MNIS on PDA. There are four main questions to be considered: the daily work procedures and environment, the frequency of medication errors, the usability of desktop medical information system and user expectations. Meetings lasted approximately two and a half hours and were documented with audio and paper records. Table 1 describes a cycle of nursing workflow in general.
After understanding the current status of nursing practice, next, we discuss the user requirement analysis and determine what solutions are appropriate for problems presented. The second interview focused on findings in the documents and on
J Med Syst (2012) 36:11391153 1141
Table 1 Nursing workflow
No. Work Content
1 Check Check equipment and whether amounts of all materials are sufficient or not.
2 Shift Process shift report to illustrate patients matters needing attention.
Determine whether drugs on the patients order match the prescription list.
3 Measure vital signs and physical assessment
Measure temperature, respiratory rate, pulse, blood pressure and patients status are according to the prescribed order.
4 Drug delivery Deliver drugs and sign according to the order.
5 Record Record physical assessment and vital signs on the TPR (Temperature-Pulse-Respiratory rate) Sheet.
Recording the facts (process and improvement) if patient occurs special statuses.
6 Returned drugs Check remaining drugs and complete out checking sheet.Then put these on the returned drugs area.
7 Input/Output Record and measure the patients input and output values.
8 Valuation Count all consumed materials and disposed items.
10 Nursing regulation Arrange the environment, giving appropriate nursing interventions for the patient, assist the physician in diagnosing the patient, etc.
users responses in order to confirm what additional functions should be built into MNIS and find appropriate tasks whose functions can be accomplished on a PDA-based system. In addition, there was a need to understand some issues related to operation of the desktop NIS. Hence, the study prepared drafts version of the MNIS with information from relevant research so that nurses could understand the design concept underlying the MNIS. The meeting lasted approximately three hours and was documented with audio and paper records. The results of this interview, combined with those of the first interview, show that there are basic and extended functions should be considered in MNIS design for improving operational efficiency and reduce medication errors of nursing:
(1) Basic functions:
& Generally, the common use of nursing forms, from patient admission to discharge, can be divided into those for nursing history and for general records.
& A complete nursing system should possess some important functions for reducing the workload during clinical nursing activities as shown in Fig. 1. However, some functions marked with a * represents which can be proceeded on the desktop system and do not require immediate attention in MNIS design.
(2) Extended functions:
& The nurses hope to query drug information and patients blood results while working. This information is needed often but is usually difficult to remember. Hence, access to this information was an added function on the MNIS, enabling nurses to query the information at any time from anywhere.
Interface design based on M-HCI
When basic and extended functions were considered in MNIS design, then, a prototype MNIS based on M-HCI for use in clinical nursing was developed. In this study, the considerations of M-HCI on MNIS interface design include metaphors, cognition, navigation, appearance and usability. The use of a metaphor provides knowledge about what actions are possible, how to accomplish tasks, and can connect a virtual world with a users real world in way that utilizes common knowledge of
Fig. 1 The nursing processes and related nursing forms
1142 J Med Syst (2012) 36:11391153
long-term memory for easily mapping the users learning information and reducing the size of working memory [12]. For example, the deletion operation has often been metaphorized with a trash can icon. Therefore, a metaphor applied in MNIS must consider a nurses cognition base, so that they can directly associate each symbol with a specific function without memorization [13]. A cognitive model is used to describe how a users knowledge of a system is stored in long-term memory, and how it is influenced by the frequency with which the system is used, or by specific interactions or contexts are experienced [14]. A defective user interface is due mainly to a lack of understanding of the users cognitive needs. Based on cognitive theories, part of the reason for lack of retention is hypothesized to be the users inability to form a mental picture, or schema, of the information presented via a computer screen [15]. Therefore, additional basic research is needed to understand the users, their work activities, and their reasoning processes in order to adequately address their cognitive needs [16]. The conclusions of initial analysis described in the preceding section could be used as user cognitive needs. In information systems, navigation is the process of goal-directed locating and traversing hyperlinked information. In general, mobile devices arent aesthetically pleasing enough and navigation is troublesome [17]. Navigation or travel would be better in adaptive menu and familiar metaphors. In addition, consideration of appearance is an important task to determine how to condense information onto the small-screen size of PDA. P. Yu and H. Yu (2004) indicated that small screen size of PDA is hard to support viewing of large tables/electronic health records in a single screen and PDA should not be used for performing complicated tasks such as graphics processing or to support a database [18]. Mohageg and Wagner (2000) proposed five design considerations for information applications on the small screen; one of them was to
simplify, which puts the focus on important functions, relegating others to other platforms [19]. Finally, Usability is defined as both an attribute of software quality as well as a higher design objective [20]. Jones and Marsde (2006) mentioned that usability is also about whether the device works in harmony with the things around it [21]. They said that poor usability can come when there is a lack of thought into how the mobile device might integrate with a users other resources and the way they wish to carry out their work and broader lives. In general, usability consists of some kinds of attributes such as learnability, efficiency and satisfaction [22]. Learnability is the system should be easy to learn so that users can rapidly begin accomplishing work with the system. Efficiency is the system should be efficient so that users can achieve a high level of productively after learning to use the system. Satisfaction is the system should be pleasant to use, evidenced by users being subjectively satisfied when using it.
System implementation
The design of MNIS was not only based on HCI concepts but also on the results of analysis in a real environment. The study assumes that the MNIS works under a wireless local area network (WLAN). Nurses carry PDA with MNIS to perform paperless clinical nursing, and inputting records right away on the PDA via the WLAN upon receiving patient-related information. Medical personnel can retrieve the newest information from the desktop system in the nursing station, clinical laboratory or general practitioners (GPs) office. Figure 2 shows the conceptual model of the PDA solution. The hardware choice was a Dopod P100 Pocket PC. The structure of the MNIS includes three layers: identification, main functions and sub-functions, as shown in Fig. 3. The contents and operating conception are based
Fig. 2 The conceptual model of the PDA solution
WLAN
Nurse
Patient
Clinical Laboratory
Nursing Station Server GP' s Office
J Med Syst (2012) 36:11391153 1143
Layer 1 Layer 2 Layer 3
Fig. 3 The structure of the MNIS
on the desktop NIS of the hospital and on users habits. The summaries are as follows.
(1) The first layer is identification. The first screen is the front page of the system where nurses must input their username and password. If correct, nurses must then check whether their unit number is correct. The unit number should be modified if incorrect before the system is ready for use.
(2) The second layer shows the main functions of MNIS, which include accessing the general nursing record, nursing history, drug searches and blood test results, help, and logout. The general nursing record deals with daily health status of the patient. The nursing history records the patients background and first-time assessment data. The functions of drug searches and blood test results function help nurses query such information at the bedside. Help assists users when they have questions about operation of the system.
(3) The third layer includes sub-functions of the second layer and it is here that a user can fully complete a task.
Presentation of the MNIS interfaces
The main menu of the MNIS is shown in Fig. 4. There are five functions including in the main menu: general nursing
record, nursing history, drug search and blood test results, help, and logout. There are five iconic symbols designed in the system as shown on the left side, with characters and colors, so that novices can easily learn to use them and to economize on screens use. The use of color depends on the characteristics of icons to help nurses easy to discriminate.
When user clicks the function button in main menu, the screen will transfer to the sub-menu of the function. For example, Fig. 5(a) shows the sub-menu of the general nursing record: (1)vital signs; (2)physical assessment; (3) nursing planning; (4)nursing instruction, and (5)nursing record. The color of these buttons is blue, as its main-menu button (in Fig. 4). In addition, the sub-menu has a reminder at the right-top corner. The advantages of this design include that the user can know what kind of function they clicked and hence reduce the amount of working memory needed. If the user chooses vital signs from the general nursing record, the top of the blue area will show General Nursing Record and the top of the white area will show Vital Signs (see Fig. 5(b)). The main screen of each sub-menu is divided into three parts as in Fig. 5(c). The first part and third parts are shown on every work page. The first part includes the location, a patient combo box edit button, and a back function. The top of the blue area shows the main menu are where the user is (general nursing record, nursing history, drug search and blood test results, and help) and the top of the white area shows the sub-menu. The second part is a green area which lists the contents of the function. The third part is a shortcut bar which lets the user come and go freely. Even with a good system the user may forget the meaning of the symbols. A click of the house symbol will allow the user to return to this main-menu at any time. If a nurse would like to enter another main function, s/he can click the icon of the main function on the shortcut bar. Finally, the system offers appropriate feedback with audio when needs user pay attention as shown in Fig. 7, including: (a)login error; (b)unit number error; (c)deletion request; (e)deletion success and a save success (f).
General Nursing Record
Metaphor
Fig. 4 Main menu of MNIS
Nursinghistory
Drug Search and Blood Test Results
Help
Logout
1144 J Med Syst (2012) 36:11391153
General Nursing Record
Back
Vital signs
Edit
Reminder
Physical assessment
Nursing planning
Patient list
Nursing instruction
Nursing record
(a) Sub-menu (b) Vital signs
Diagnosis
Start time
Finish time
Name of nursing plan
Plan in finish
Content of nursing plan
(c) Physical assessment (d) Nursing planning
Instruction items
Handling
Save
(e) Nursing instruction (f) Nursing record
Fig. 5 The contents of the general nursing record
A summary of MNISs design with principles of M-HCI is as follows:
(1) Metaphor
& Pictorial realism: as shown in Fig. 4, the metaphor was designed to look like the variable that it represents.
(2) Cognition
1. Reduce short-term memory load
& Iconic menu: as shown in Fig. 4, characters of commands are metaphorical by iconic symbols that are easy to recollect the function menu in layer 2 or layer 3.
J Med Syst (2012) 36:11391153 1145
(a) Sub-menu
(b) Drug search
(c) Results of drug (d) Blood test results
Fig. 6 The contents of the drug search and blood test results function
& Shortcut bar: as shown in bottom of Fig. 5(c), the shortcut bar is consolidated in working pages. Therefore, users spend little time and effort to search or type information from page to page.
& Reminder: as shown in Fig. 5(a) and 6(a), the bigger iconic symbols remind the user of his or her location and of the functions meaning.
& Location: as shown in Figs. 5 and 6, function name was shown on the top of each working page, so that it reminds users where they are, users dont need any effort to remember where they are.
& Consistency: as shown in Fig. 5, the same information and buttons were presented consistently on all screens to facilitate recognition (e.g., edit, back, save, delete, and shortcut bar) to reduce users search and identification time.
2. Support internal locus of control
& Edit button: as shown in Fig. 5(b), the system allows users to modify the patient list clicking the edit button.
& Feedback: as shown in Fig. 7, the system offers feedback messages when users are performing certain tasks.
3. Design dialogs to yield closure
& As shown in Fig. 7, giving confirmative and affirmative dialogues when tasks are completed or when a deletion action will occur is important for users.
(3) Navigation
1. Adaptive menu and familiar metaphors
& Functional menu: as shown in Fig. 4, menu was designed in structure according to daily nursing procedures.
1146 J Med Syst (2012) 36:11391153
(a) Login error
(c) Delete or not
(e) Save success
(b) Unit number error
(d) Delete success
Fig. 7 Notice message of system feedback
& Familiar metaphors: as shown in Fig. 4, iconic symbols were designed with characters and colors, so that novices can easily learn to use them and to economize on screens use.
2. Goal-directed locating and traversing hyperlink
& Function button: as shown in Figs. 4 and 5, a click of the function button in main menu, the screen will transfer to the sub-menu of the function. In addition, the sub-menu has a reminder metaphor of main function at the right-top corner.
& House symbol: as shown in Fig. 5, a click of the house symbol will allow the user to return to this main-menu at any time.
& Back button: as shown in Fig. 5(b), if the user enters an undesired page, it is easy to return to a previous page by clicking the back button.
& Shortcut bar: as shown in bottom of Fig. 5(c), if the user enters an undesired page, it is easy to return to the sub-menu page by clicking the icon button. Users can navigate freely without worry.
3. Offer informative feedback
& As shown in Fig. 7, the system offers appropriate feedback messages. If the system continuously or obsessively informs the user about what it is doing, users will become impatient with system.
4. Permit easy reversal of actions
& Back button: as shown in Fig. 5(b), it is easy to return to a previous by clicking the back button.
& Edit button: as shown in Fig. 5(b), the system allows users to modify patient list by clicking the edit button when they selected their patients.
(4) Appearance
1. Strive for consistanc
& Layout: as shown in Figs. 5 and 6, buttons (e.g., edit, back, save, delete, and shortcut bar), the users name, and page description maintain a fixed location.
& Font and Size: as shown in Figs. 4, 5 and 6, the
English font is Times New Roman and the Chinese font is MingLiU. The font size of reading text is 10, 12 for sub-titles, 14 for menu buttons, and 20 for menu titles. Differentiable sizes can be distinguished by users.& Color: as shown in Figs. 4, 5 and 6, no more than 7 colors are used for easy discrimination.
2. Prevent errors
& Highlight: as shown in Fig. 5(b), highlight of patient list can catch the users attention, so that the number of errors will decrease.
& Notice message: as shown in Fig. 7(c), the system offers notice when users perform a deletion action in order to prevent errors.
(5) Usability
& Iconic menu: as shown in Fig. 4, characters of commands are metaphorical by iconic symbols, which are namely metaphors used to achieve a system mapping to the users conceptual model of the information.
& Shortcut bar: as shown in bottom of Figs. 5 and 6.
Not every button can be incorporated into the shortcut bar. The buttons are available to directly access important functions that users may frequently need.& Highlight: as shown in Figs. 5 and 6, the study uses various ways to attract the users attention, including highlights by use of color, icon, font, font size, and boldface typesetting.& Reminder: as shown in Figs. 5(a) and 6(a), the bigger iconic symbols enhance the users memory and help them know where they are.& Location: as shown in Figs. 5 and 6, function name was shown on the top of each working page, so that it reminds users where they are. To avoid disorientation, users are given directive reminders.& Consistency: as shown in Fig. 5, the same information and buttons were presented consistently on all screens to facilitate recognition (e.g., edit, back, save, delete, and shortcut bar). Hence, users can easily gain system comprehension even with only partial knowledge of how the system works.
J Med Syst (2012) 36:11391153 1147
Table 2 The sequence of task scenarios in experiment
Targets Tasks
A. Identification 1) Input username and passwords, then modify the unit number
2) Select patients from the patient list page
B. Nursing History 1) Complete patient information2) Complete nursing assessment3) Complete admission nursingC. General Nursing Record 1) Complete vital signs2) Complete physical assessment3) Complete nursing instructionD. Drug Search and Blood Test Results
1) Search for drug information on AMOBITAL SODIUM/inhalants
2) Search blood test resultsE. Patient List 1) Modify self patient list
Although many principles of interface design are addressed, the prototype of MNIS is still necessary to evaluate the usability for users in order to design a satisfactory system. Next section is to explore more usability problems from experimental design.
Experimental design for measure usability of MNIS
Usability analysis
The purposes of experiment were to determine problems of usability stemming from interface design, to obtain objective and subjective data which can be used as a source of quantifiable and qualitative information for verifying the systems usability. In general, there are three types of usability evaluation methods, which include testing, inspection and inquiry [23]. The inspection and inquiry methods need usability specialists, software developers and other professionals to examine the system. However, the specialists or professionals are not easy to find or may be not adaptive. Therefore, in this study, we inspected usability of the prototype of MNIS with usability testing. Here, usability testing was implemented in laboratory settings, and involved thinking aloud and questionnaires as forms of subjective assessment (user attitude, such as satisfaction and attractiveness) and objective assessment (user performance, such as learnability test). Thinking aloud can be particularly useful if you have mobile devices you want to evaluate in lab settings [10]. In this process, the users voice and browse route were recorded; meanwhile, the study also utilized usability analysis software LOGSQare, which records user activities and screen shots, to analyze user behavior. The QUIS was developed by a multi-disciplinary team of researchers in the Human-Computer Interaction Lab (HCIL) at the University of Maryland at College Park and was designed to estimate users subjective satisfaction with specific aspects of the user interface. It consists of six constructs: participant demographics, screen, symbol investigation, terminology and system information, learning, and overall user reactions and uses a 5 point Likert scale (1 being poor and 5 being excellent). Learnability test is a proper aspect of usability testing which can be used to evaluate how easily the prototype can be learned and how quickly the user can start to use the system [24]. In addition, there were a series of task scenarios to implement the usability testing for exploring participants NIS operating experience.
Participants
A laboratory experimental method was employed in this study. There were 12 female undergraduate students (aged 2023) in the nursing department participated. They are
students in the school of continuing education, practice and serve as the nurses assistants in the hospital in the daytime, and go back to school and have lessons in the evening. They were divided into N1 group and N2. In N1 group, they worked in the nursing station and helped to key in the clinical records into the desktop system (i.e. NIS) in most of the time, sometimes, helped nurses to deal with nursing care. In N2 group, they were assistants in the ward to take down the general nursing records on the nursing record sheet, but didnt need to key in materials into the NIS. Therefore, the N1 group was defined as experimental group (who have NIS operating experience), the N2 group was defined as control group (who havent NIS operating experience).
Environment and apparatus
The experiment was took place in a controlled room which was divided into three areas for learning, testing, and filling out questionnaire, respectively. Furthermore, two notebooks were prepared for tuition and for testing based on the LOGSQare software. A Dopod P100 PDA was used for learning. Participants used a Pocket PC simulator on a desktop during testing because so that their operating screens could be record by LOGSQare.
Experimental variables
In learnability test for objective evaluation of usability, the performance variable was time spent on a task, which is noted as the time to finish the task. The spending time of the two groups (N1 and N2) was tested to determine whether there was a difference between them or no? Whether the
1148 J Med Syst (2012) 36:11391153
Table 3 Spending time of each participant in each task (seconds)
Student Number Task A Task B Task C Task D Task E Total
a.1 a.2 b.1 b.2 b.3 c.1 c.2 c.3 d.1 d.2 e.1
N1 S01 30.4 22.4 40.7 53.0 34.7 59.0 33.6 16.9 22.6 17.0 29.6 359.9 S03 32.2 23.8 49.0 80.6 49.9 66.2 45.1 17.2 34.7 14.7 59.3 472.7 S06 32.5 26.1 46.7 69.0 40.2 83.5 31.7 20.4 51.5 36.1 56.7 494.4 S08 20.3 28.8 32.1 65.8 64.4 71.8 36.4 20.1 30.3 21.4 46.0 437.4 S09 44.1 31.1 30.0 70.5 52.0 69.7 89.9 21.2 35.0 20.0 37.2 500.7
S12 42.4 36.3 31.3 71.8 51.5 75.4 41.0 16.0 45.7 20.1 45.3 476.8 Mean 33.7 28.1 38.3 68.5 48.8 70.9 46.3 18.6 36.6 21.6 45.7 457.0 S.D. 8.7 5.1 8.3 9.1 10.3 8.3 21.9 2.2 10.5 7.5 11.3 52.5 N2 S02 19.5 37.0 30.2 58.3 39.4 67.3 47.6 41.1 83.0 19.5 54.5 497.4
S04 19.0 31.7 46.3 94.4 47.0 101.9 59.9 19.7 43.7 20.7 35.9 520.2 S05 46.2 42.8 44.8 40.6 95.4 82.9 67.5 22.1 31.8 16.4 24.6 515.1 S07 36.3 48.7 26.1 112.4 68.5 62.4 30.3 17.3 31.1 19.1 37.0 489.2 S10 38.1 27.6 37.1 76.3 44.9 70.4 52.0 24.4 23.8 38.9 34.6 468.1 S11 42.1 48.0 28.6 56.4 70.8 94.8 39.1 17.5 37.5 16.5 32.3 483.6 Mean 33.5 39.3 35.5 73.1 61.0 80.0 49.4 23.7 41.8 21.9 36.5 495.6 S.D. 11.6 8.7 8.6 26.7 21.2 15.9 13.6 9.0 21.3 8.5 9.9 19.6
MNIS was good on learning or not? Additionally, in subjective evaluation of usability, the satisfaction of end users was analyzed by QUIS.
Task scenarios
Andrews (2003) noted that tasks for testing were provided should be prioritized by frequency and importance, as well as selected from among those most frequent and critical [25]. The study goes through the nursing procedure and asks nurses for advice, so as to list appropriate task scenarios to test. When participants were doing the tasks, LOGSQare software recorded their operating screens in. Five targets and 11 tasks were tested. The sequence of task scenarios was listed in Table 2.
Experimental procedure
Before the experiment beginning, participants were informed the experimental purposes and procedure. There were two steps in the learning process: tuition via power point and self-learning with the PDA in the first experiment. Tuition included a picture of each screen, function, and icon. Participants also practiced for three minutes on their own before performing the tasks. In the meanwhile, they could ask any question about the system and receive answers.
In the testing stage, participants were encouraged to vocalize questions, but would receive no responses. This method is called Thinking Aloud. Audio, as well as the
mouse trajectory and screen contents, are recorded in the testing phase as important research information. If participants finished all tasks, they will complete a questionnaire for User Interaction Satisfaction (QUIS).
Results
Table 3 shows the overall performance data (i.e. spending time) for both the N1 group and N2. Since the sample size of the study is less 50, the Shapiro-Wilk was employed. The results show that all significances associated with the test of normality are >0.05, so the spending time is normally distributed. First, the independent samples t-test was performed to test whether NIS operating experience would affect participants learning ability. The result shows that the differences of spending time on five tasks between N1 and N2 are all insignificant (t (22) Task A=1.525, p=
0.071; t (34) Task B=0.676, p=0.252; t (34) Task C=0.659, p=0.257; t (22) Task D=0.431, p=0.335; t (10) Task E=1.502, p=0.08). Because the most operation function of MNIS was designed according to NIS. The participants have ever used the NIS would be easy to use the MNIS. However, the difference of spending time between N1 and N2 are not significant. That is, when the users were first introduced to MNIS, their adaptation to the system was much quicker. Additionally, subjective measurement on QUIS exhibits high reliability (range of Cronbachs alpha scores from 0.68 to 0.88). The results of the QUIS are summarized in
J Med Syst (2012) 36:11391153 1149
Table 4 Subjective satisfaction with the interface
Construct Measure Average S.D.
1.1 Characters on the computer screen are easy to read 4.33 0.491.2 Highlights on the screen are helpful to you (e.g.,
colors, icons, boldfaces, and front sizes) 4.33
1.3 Information and hint on the screen are sufficient to
operate 4.08
1.4 Information Arrangement on the screen is
consistent 4.42
1.5 The levels of system structure are clear(main menu sub-menu content) 4.08
1.6 Can go to any screen easily 4 0.741.7 The current location is easy to know 4 0.74
Screen
0.78
0.79
0.51
0.51
0.67
0.51
0.51
0.39
0.90
0.79
Total 4.18 0.66
2.1 Back to Main Menu is easy to associate with the symbol
4.58
2.2 General Nursing Record is easy to associate withthe symbol 4.42
2.3 Nursing History is easy to associate with the symbol 4.58
2.4 Drug Search and Blood Test Results is easy toassociate with the symbol 4.83
2.5 Help is easy to associate with the
symbol
4.42
Symbols Investigation
2.6 Logout is easy to associate with
The symbol
4.42
2.7 The size of symbols is easy to see 4.33 0.652.8 Iconic functions help to learn the system 4.50 0.52
4.51
4.31
4.19
Terminology and System Information
Total 0.58
4.1 Learning how to operate system is easy 4.25 0.454.2 Remembering function names and usages are easy 4.25 0.75Learning 4.3 The number of steps of each task is acceptable 4.08 0.51
Total 0.58
5.1 The system is helpful to you. 4.25 0.455.2 The aesthetic of the interface design is satisfactory. 4.33 0.895.3 You are willing to use the system in the future. 4.33 0.495.4 The system is satisfactory after using. 4.17 0.58
0.51
0.493.3 System gives appropriate feedback (save success!) 4.50 0.67
Total 0.63
3.1 Use of computer and nursing terminology are
consistent on the system 4.08
3.2 Use of nursing terminology conforms with your
specialty 4.33
Overall User Reactions
4.27 0.61
(Questionnaire Category: Strongly Agree=5; Agree=4; No opinion=3; Disagree=2; Strongly Disagree=1;
S. D.=Standard Deviation)
Total
Table 4. The percentage indicates that the 12 participants responses and all averages of the assessable items are all above 4 points, which is between agree and strongly agree. However, the item 1.6 and 1.7 on the screen construct still
can be improved. Afterwards, the subjective satisfaction of the two groups is summarized in Table 5. The results reveal the two groups are all over 4 points and the N2 group shows a little satisfaction higher than the N1 group. Though N2 has
1150 J Med Syst (2012) 36:11391153
Table 5 Subjective satisfaction of the two groups
Construct NIS Experience Part 1 Part 2 Part 3 Part 4 Part 5
N1 N2 N1 N2 N1 N2 N1 N2 N1 N2
Mean 4.17 4.19 4.50 4.52 4.28 4.33 4.17 4.22 4.21 4.33 S.D. 0.41 0.63 0.40 0.44 0.40 0.44 0.59 0.40 0.60 0.26
not used the desktop NIS, they gave good feedback about MNIS such as screen layout, symbols, use of terminology and learning. It means that the MNIS is good on learning and using for persons have no NIS operating experience. In general, according to objective and subjective data, there are no significant differences on overall spending time for the two groups and the subjective satisfaction results reveal that the averages of all constructs have over 4 points on the scale. Once again, the results prove the system is satisfactory for persons who have experience on desk NIS or none. Especially, it is easy-to-learn for persons have no experience on desk NIS.
Further, this study got the full and meaningful questions using the participants own knowledge in the operating process from their voice and experimental screen. With their responses, the study can not only understand their perspectives, but these important responses will be the basis for improving the MNIS prototype. Some analyzed suggestions were summarized from the thinking aloud and LOGSQare software as follows. (S = Subject; the numeral is the serial number)
(1) Screen
& S7: The layout and color is clear. & S12: Low system speed.
(2) Symbol Investigation
& S4: It is a little difficult to associate symbols the first time, but familiar later.
& S6: It is better if the function name appears when the cursor moves to icon.
& S12: The designs of iconic functions and search functions (drug information and blood test results) are very good.
(3) Terminology and System Information
& S3: The terminology need be clearer in function ofPhysical assessment.
(4) Learning
& S6: I dont know the Edit buttons meaning.& S7: The system is very easy to use with low difficulty. & S12: The system is very easy to understand and may decrease the time required to perform nursing work.
(5) Overall User Reactions
& S9: I feel that the screen layout is consistent and the system is easy to use.
& S11: The system almost shows nursing information on the screen for the nurse to click. Although it is convenient in nursing practice, it may reduce the nurses competence.
& S12: The PDA is necessary to apply in the nursing field. The MNIS can reduce our workload.
The results showed that suggestions in Terminology and system information S3 and Learning S6 are problems of description of function. It means that the recognition of user on function definition or button name should be proved clearly again. The suggestion in screen S12 is problem of information technique. It is not discussed temporarily in this study. Nevertheless, other suggestions are mostly satisfied.
Discussion
Mainly a PDA was believed to improve patient care and save time for nurses because it was regarded as a convenient medium to provide and access electronic medical records at any time and any place [4]. Rodrguez et al., (2009) found that nurses were significantly more satisfied with the PDA-based system than with the paper-based system with every interaction and system aspect evaluated on their study [26]. A PDA has the potential to be a part of nurses routine care planning. However, the small-screen of the PDA, size constraints make it harder for the user to interact with the content. In recent years there are some researches discussed the design on the small-screen. Zhao et al., (2001) suggested that type of information display significantly affect user performance [27]. When the display was in color, participants performed better. Participants preferred less information and they rated displays with intermediate amounts of information most comfortable. Churchill and Hedberg (2008) reported that design for one step interaction, minimize scrolling, design for short contact time and design to match the task would be effective design of learning objects on the small-screen. However, these studies are mostly focused on display of the small-screen [28]. At this stage availability of relevant guidelines and a
J Med Syst (2012) 36:11391153 1151
solid literature in relation to consider the appropriately performed on a PDA in clinical nursing including specification and requirements on users and M-HCI consideration is very limited. In this study, we integrated the method of initial analysis and the principles of M-HCI to develop a prototype of MNIS which has a friendly for nurses and support for clinical nursing. The initial analysis including workflow analysis, job analysis and interview methods was a reasoning process for confirming what additional functions should be built into MNIS in consideration of the small-screen interface. Here we found that the initial analysis is very important. Via initial analysis, the designer can clearly understand nurses daily work and focused on investigation of the current status in order to define the usage context specification and easily confirm appropriate tasks whose functions can be accomplished on a PDA-based system. Thus, the essential information can be mastered thoroughly to define the major function on a small-screen with PDA. Then, the concepts of M-HCI (metaphors, cognition, navigation, appearance and usability) can be adapted to develop MNIS.
The results by usability testing indicated that our view supported by the applications of the principles of the MHCI have an agreeable usability design besides satisfying the users need of content and functions. According to the results, some findings of this study are worth summarizing:
(1) Based on the results of objective assessment, the T-test showed that there is no significant difference between N1 (NIS operating experience users) and N2 (non-experience users). It means that the MNIS is no difference on learning and using for persons between have NIS operating experience and no. The MNIS is good on learning and using for persons have no NIS operating experience.
(2) Based on the results of the subjective assessment, the questionnaires showed that the degree of satisfaction is higher on symbol investigation, terminology and system information, and overall user reactions to the constructs. It means that when users first use a new system, the recognition and comprehension of symbol and constructs of information and function would be important for users to understand how to use the system and what function in the system.
(3) On the whole, participants felt satisfied after using the MNIS, but items 1.6 (ability to go to any screen easily) and 1.7 (the current location is easy to know) in QUIS have lower satisfaction levels which may lead us to reconsider how to improve navigation in the future.
(4) The difference of N1 group (S.D.=0.6) on Part 5 (overall user reflections) of QUIS was larger than the
N2 group (S.D.=0.26) and the standard deviation of total spending time (S.D.=52.5) was also larger than
N2 group (S.D.=19.6). The study infers that there are
different transformation and comprehension capabilities from the N1 group. It means that past experience might not be help user in studying the new things.
Because every desktop NIS of hospital is different, and because of differences in transforming information from a larger screen to a smaller screen, different reactions may arise for these reasons as well.
Based on related studies, the user interface is an important aspect of the system design that can affect whether users are willing to use the system or not. The results show that the small-screen design of the MNIS prototype is approved by users through objective and subjective assessments. Further, nurses indicated that in addition to in the general ward, the emergency room should be considered for implementation of a similar system. However, it should be noted that nurses who work in emergency room must care for more patients than nurses who work in general ward and that their work involves more multifarious information to deal with.
Conclusions
The study developed a mobile medical system MNIS on PDA, the experimental results indicate that the subjective satisfaction on each construct was significant. It means the system is easy-to-use, easy-to-learn and artistically pleasing for nurses. Therefore, the designer must take into account not only the technological aspects of the system, but also the users cognitive model, so that the system can result in successful interaction with the user. Some of the practical and academic contributions of this study are as follows:
(1) This study develops a prototype MNIS to support clinical nursing. Nurses can not only carry a PDA during the daily clinical routine, but can perform paperless recording of nursing information. In addition, they can perform self-learn about drug information via the drug search and blood test results function embedded in the MNIS.
(2) According to related literatures, PDA-based systems promote operational efficiency, reduce nurses workload, and can reduce medical errors [29, 30]. Generally speaking, the MNIS improves (a) operational efficiency; (b) reduces medication errors; (c) increases patients safety because of the reduction of medication errors and increase time for nurses to care for patients; and (e) improves overall healthcare quality as a result.
(3) Very little of the related research focuses on the combination of a well-designed interface and mobile medical system. In additional to concepts from M-HCI, iconic functions are considered in the design process. The fruitful results offer a new innovation regarding the design of a traditional medical system.
1152 J Med Syst (2012) 36:11391153
(4) Use of a mobile device solving the main problem of information inconvenience during the clinical routine is a good way. However, researchers must think about what information and tasks are needed for nurses. The adopted methodologies of this study serve as a pattern for related research. This study adopted HCI concepts to analyze the current and users requirements for discovering problems, and incorporated concepts of M-HCI for creating a friendly user interface in the development of MNIS.
The focus of this study was on small-screen interface design with M-HCI guidelines. The study identified several aspects to be considered for further research as follows:
(1) The MNIS is a prototype system and has much room for improvement. The next stage is to improve the MNIS based on users suggestions and then obtain a deeper analysis from usability experts and wider-ranging participants.
(2) Push the MNIS into a real environment and popularize for an emergency room or others that most need information improvement. The study may reveal a difference in usability for a controlled laboratory experiment and a real environment. Undoubtedly, that the MNIS can be used well in practice is the ultimate goal.
(3) Because of considering the differences among participants on age and experience, participants were chosen in consider with the same age level. The probationers with 2023 years old are the most in the investigated hospital and so considered to be subjects in this study. However, there are only six persons with 2023 years old have NIS operating experience, in considering equal number of participants, there are also six persons in the control group (N2 group). Future studies with more hospitals to investigate the effects of MNIS would be considered.
References
1. Stolworthy, Y., and Suszka-Hildebrandt, S. E., Mobile information technology at the point-of-care. Medscape Nurses. 3(1), 2001. Available at http://www.medscape.com/viewarticle/408421
Web End =http://www.medscape.com/viewarticle/408421 . Accessed November 25, 2006.
2. Li, Y. C., Chang, I C., Hung, W. F., and Fu, H. K., The critical factors affecting hospital adoption of mobile nursing technologies. Proceedings of the 38th Hawaii International Conference on System Sciences (HICSS 05). 2005.
3. Lu, Y. C., Xiao, Y., Sears, A., and Jacko, J. A., A review and a framework of handheld computer adoption in healthcare. Int. J. Med. Informat. 74:409422, 2005.
4. Berglund, M., Nilssona, C., Revaya, R., Petersson, G., and Nilsson, G., Nurses and nurse students demands of functions and usability in a PDA. Int. J. Med. Informat. 76:530537, 2007.
5. Carroll, A. E., Saluja, S., and Tarczy-Hornoch, P., Development of a Personal Digital Assistant (PDA) Based Client/Server NICU Patient Data and Charting System. Proceedings of the AMIA 2001 Symposium, 100104. 2001.
6. Yu, P., and Yu, H., Lessons Learned from the Practice of Mobile Health Application Development, Proceedings of the 28th Annual International Compute Software and Applications conference (COMPSAC 04), 5859. 2004.
7. Johansson, P. E., Peterssona, G. I., and Nilssona, G. C., Personal digital assistant with a barcode readerA medical decision support system for nurses in home care. Int. J. Med. Informat. 79:234242, 2010.
8. Ammenwerth, E., Buchauer, A., Bludau, B., and Haux, R., Mobile information and communication tools in the hospital. Int. J. Med. Informat. 57:2140, 2000.
9. Green, M., Error and injury in computers & medical devices, Visual Expert Human Factors 2004. Available at http://www.visualexpert.com/Resources/compneg.html
Web End =http://www. http://www.visualexpert.com/Resources/compneg.html
Web End =visualexpert.com/Resources/compneg.html . Accessed December 3, 2006.
10. Love, S., Understanding mobile human-computer interaction. Elsevier Butterworth-Heinemann, Amsterdam, 2005.
11. Marcus, A., Dare we define user-interface design? Interactions 9:1924, 2002.
12. Wozny, L. A., The application of metaphor, analogy, and conceptual models in computer systems. Interact. Comput. 1:273283, 1989.
13. Liu, C. M., A Study of the Development Model for Iconic Function InterfacePatient Care Record System. Tainan, Taiwan: National Cheng Kung University; 2001. Thesis.
14. Javaux, D., A method for predicting errors when interacting with finite state systems. How implicit learning shapes the users knowledge of a system. Reliab. Eng. Syst. Saf. 75:147165, 2002.
15. Chalmers, P. A., The role of cognitive theory in human-computer interface. Comput. Hum. Behav. 19:593607, 2003.
16. Patel, V., and Kushniruk, A. W., Interface design for health care environments: the role of cognitive science. Proceedings of AMIA 1998 Symposium, 121125. 1998.
17. Olsson, A., and Svanteson, S., User Intelligence Will Make Mobile Solutions Fly, 2001 Whitepaper, Stockholm. Available at http://www.hci.uu.se/~jg/UCD2001/Olsson.pdf
Web End =http://www.hci.uu.se/~jg/UCD2001/Olsson.pdf . Accessed January 21, 2007.
18. Yu, P., and Yu, H., Lessons learned from the practice of mobile health application development. Proceedings of the 28th Annual International Compute Software and Applications conference (COMPSAC 04). 5859, 2004.
19. Mohageg, M., and Wagner, A., Design considerations for information appliances. In: Bergman, E. (Ed.), Information Appliances and Beyond. Morgan Kaufmann Press, San Francisco, CA, 2000.
20. Folmer, E., and Bosch, J., Architecting for usability: A survey. J. Syst. Softw. 70:6178, 2004.
21. Jones, M., and Marsde, G., Mobile interaction design. Wiley, New Jersey, 2006.
22. Nielsen, J., Usability engineering. Academic, London, 1993.23. Hom, J., The usability methods toolbox. Available at http://jthom.best.vwh.net/usability/
Web End =http://jthom. http://jthom.best.vwh.net/usability/
Web End =best.vwh.net/usability/ . Accessed June 15, 2009.
24. Folmer, E., and Bosch, J., Architecting for usability: a survey. J. Syst. Softw. 70:6178, 2004.
25. Andrews, K., 2003 Web Usability: Theory and Practice, FH Joanneum Summer Business School. Available at http://dmt.fh-joanneum.at/kd3/mm/pdf/sbs/sbs_webusab.pdf
Web End =http://dmt.fh http://dmt.fh-joanneum.at/kd3/mm/pdf/sbs/sbs_webusab.pdf
Web End =joanneum.at/kd3/mm/pdf/sbs/sbs_webusab.pdf . Accessed January 10, 2007.
26. Rodrguez, N. J., Borges, J. A., Crespo, G., Prez, C., Martinez,C., Coln-Rivera, C. R., and Ardn, A., Users can do better with PDAs than paper: A usability study of PDA-based vs. paper-based nursing documentation systems. Lect. Notes Comput. Sci. 5616:395403, 2009.
J Med Syst (2012) 36:11391153 1153
27. Zhao, C., Zhang, T., Zhang, K., User interface design for the small screen display. Proceedings of the Human Factors and Ergonomics Society 45th Annual Meeting. 2001.
28. Churehill, D., and Hedberg, J., Learning object design considerations for small-screen handheld devices. Comput. Educ. 50:881893, 2008.
29. Garvin, R., Otto, F., and Mcrae, D., Using handheld computers to document family practice resident procedure experience. Fam. Med. 32:115118, 2000.
30. Carroll, A. E., and Christakis, D. A., Pediatricians use of and attitudes about personal digital assistants. Pediatrics 113:238242, 2004.
Springer Science+Business Media, LLC 2012