Skip to content

Advertisement

You're viewing the new version of our site. Please leave us feedback.

Learn more

BMC Medical Informatics and Decision Making

Open Access
Open Peer Review

This article has Open Peer Review reports available.

How does Open Peer Review work?

Critical factors influencing physicians’ intention to use computerized clinical practice guidelines: an integrative model of activity theory and the technology acceptance model

BMC Medical Informatics and Decision MakingBMC series – open, inclusive and trusted201616:3

https://doi.org/10.1186/s12911-016-0241-3

Received: 3 August 2015

Accepted: 8 January 2016

Published: 16 January 2016

Abstract

Background

With the widespread use of information communication technologies, computerized clinical practice guidelines are developed and considered as effective decision supporting tools in assisting the processes of clinical activities. However, the development of computerized clinical practice guidelines in Taiwan is still at the early stage and acceptance level among major users (physicians) of computerized clinical practice guidelines is not satisfactory. This study aims to investigate critical factors influencing physicians’ intention to computerized clinical practice guideline use through an integrative model of activity theory and the technology acceptance model.

Methods

The survey methodology was employed to collect data from physicians of the investigated hospitals that have implemented computerized clinical practice guidelines. A total of 505 questionnaires were sent out, with 238 completed copies returned, indicating a valid response rate of 47.1 %. The collected data was then analyzed by structural equation modeling technique.

Results

The results showed that attitudes toward using computerized clinical practice guidelines (γ = 0.451, p < 0.001), organizational support (γ = 0.285, p < 0.001), perceived usefulness of computerized clinical practice guidelines (γ = 0.219, p < 0.05), and social influence (γ = 0.213, p < 0.05) were critical factors influencing physicians’ intention to use computerized clinical practice guidelines, and these factors can explain 68.6 % of the variance in intention to use computerized clinical practice guidelines.

Conclusions

This study confirmed that some subject (human) factors, environment (organization) factors, tool (technology) factors mentioned in the activity theory should be carefully considered when introducing computerized clinical practice guidelines. Managers should pay much attention on those identified factors and provide adequate resources and incentives to help the promotion and use of computerized clinical practice guidelines. Through the appropriate use of computerized clinical practice guidelines, the clinical benefits, particularly in improving quality of care and facilitating the clinical processes, will be realized.

Keywords

Activity theoryAttitude to computersClinical practices guideline systemsEvidence-based medicineTechnology acceptance model

Background

Current status of clinical practice guidelines in Taiwan

Clinical practice guidelines (CPGs) are “systematically developed statements to assist practitioner and patient decisions about appropriate health care for specific clinical circumstances” [1]. Many studies have considered CPGs to be standardized clinical models developed on the basis of evidence-based medicine (EBM) for improving the quality of care and reducing variations in clinical practice at affordable costs [14]. Therefore, the effective use of CPGs can help health care professionals, particularly physicians, make appropriate health care decisions, thus improving the quality of care. In Taiwan, the concept of CPGs was introduced by the National Health Insurance Administration in 2003 to address the dilemma between medical quality and cost. Until 2008, there were 17 clinical guidelines developed under the assistance of various medical associations in Taiwan. However, each of these clinical guidelines was developed to provide guidelines for a single disease rather than complex problems of comorbidity and multimorbidity [58].

Even if a CPG is well developed, it might not be implemented appropriately [913]. Handler and Lackland [11], Mourad et al. [14], and Schnoor et al. [15] have indicated that implementing multiple strategies rather than a single strategy is more effective for the widespread use of CPGs. Feasible implementation strategies for promoting CPGs include education and training, leaflets or manuals, proactive reminder systems, and monitor and feedback of results. Numerous medical institutions suggest that physicians use paper-based clinical guidelines; however, those that are paper based are inconvenient and time consuming to read, and fail to help physicians making clinical decisions in a timely manner. To address the problems mentioned, computerized CPGs were developed, implemented, and integrated into existing hospital information systems (HIS) or electronic medical record (EMR) systems by selected Taiwanese hospitals [6]. Although it was confirmed that computerized CPG use significantly improves the process of care [16], the development of computerized CPGs in Taiwan is still at the early stage and acceptance level among major users (physicians) of computerized CPGs is not satisfactory. Physicians are the key providers of health care services and the principal users of computerized CPGs; thus, physicians’ intention to use computerized CPGs is critical for the success of computerized CPGs. To explore the critical factors influencing physicians’ intention to use computerized CPGs and to provide an understanding of the underlying hindrances to computerized CPG use, this study conducted a survey in three hospitals that have implemented computerized CPGs.

Studies on CPGs and computerized CPGs

Lyng [17] summarized four types of guidance that CPGs provide for health care professionals: decision support, process support, documentation support, and task support. Lyng [17], Grimshaw et al. [18], and Johnson and Turley [19] have mentioned that only a few CPGs are actively used, despite being proposed, recommended, and applied in clinical practice. One possible explanation is that CPG use may result in variations in the clinical behaviors of physicians, who can become resistant to their continued use [20, 21]. Larson [22] indicated that more attention should be paid to applying CPGs in clinical practice. Many studies have focused on promoting CPG use in clinical practice by emphasizing topics related to CPG development [2325], the establishment of interchange standards for computerized CPGs [26, 27], and computerized CPG design and implementation [2839]. Some of these studies have focused on solving comorbidity and multimorbidity problems [28, 29, 34, 39] by incorporating knowledge-based techniques such as semantic web or constraint logic. Although feasible suggestions have been proposed for solving comorbidity and multimorbidity problems, obstacles still hinder computerized CPG development. Furthermore, Isern and Moreno [37] indicated that computerized CPGs could benefit clinicians and patients, but the systems were not yet fully integrated and implemented into existing careflow management systems and thus not used in daily clinical practice. Lyng [17] argued that most computerized CPGs are technology driven [37], and only a few studies have focused on a user-center approach in the design of CPGs [30, 38].

In addition, some studies have investigated factors influencing the implementation of CPGs for health care professionals through a systematic metareview approach, to address managerial concerns about CPG use [4042]. Francke et al. [40] categorized the influencing factors of CPG use as follows: 1) characteristics of the guidelines (e.g., being easy to understand and implement, without requiring specific resources), 2) characteristics of the implementation strategies (e.g., effective strategies often have multiple components rather than a single strategy), 3) characteristics of professionals (e.g., awareness of the existence of a guideline and familiarity with its content), 4) characteristics of patients (e.g., comorbidity reduces the chance that guidelines are followed), and 5) characteristics of the environment (e.g., impediments such as a lack of support from peers or superiors as well as insufficient staff and time). Kortteisto et al. [41] indicated that attitudes toward the behavior, subjective norm, and perceived behavioral control were crucial factors associated with professionals’ intention to use CPGs. Flottorp et al. [42] identified seven domains influencing health care professional practices: guideline factors; individual health professional factors; patient factors; professional interactions; incentives and resources; capacity for organizational change; and social, political, and legal factors. As mentioned, these factors have provided insights into further research on computerized CPGs and the careful evaluation of the effects of their use.

Theoretical foundations

Activity theory

Activity theory was first proposed by Vygotsky for understanding the mental capabilities of an individual through in-depth analysis of the cultural and technical aspects of human actions [43]. Kuutti [44] argued that activity theory is a cross-disciplinary framework for studying different forms of human practice, factoring in the processes of context as a developmental process, at both the individual and social levels simultaneously, including the use of artifacts. According to Leontiev [45], the core of activity theory is activity, which is composed of a subject, object, actions, and operations. Nardi [46] argued that “activities are realized as individual and cooperative actions, and chains and networks of such actions, related to each other by the same overall object and motive.” In addition, activity theory focuses on social factors and the interaction between subjects and their environments and explains why the principle of tool mediation plays a central role in the activity approach. Tools are created and transformed during the development of an activity and vary depending on the culture. Therefore, tools constitute a means of accumulating and transmitting social knowledge.

Recently, activity theory has been extended to the fields of the human-machine interface [47], learning [48, 49], e-commerce [50], intensive care patient discharge [51], and computerized diabetes disease management [52]. Most of the cited studies have applied activity theory as a practical and descriptive framework for facilitating the understanding of the interrelationships among activity systems through descriptive and qualitative research methodologies. Sun et al. [53] suggested that user behavior is closely related to the context in which the user is placed (e.g., social environment). Activity theory regards the information system as a manmade artifact or tool and is applied to analyze the interactive relationships between human behavior and social associations [46]. In the context of the health care industry (environment), a computerized CPG (tool) is viewed as a type of manmade information tool or a clinical decision-making system used by physicians (subjects) to obtain real-time information for completing clinical activities and decisions, to improve the quality of clinical decisions, and to ensure patient safety (objects). In addition, the use of computerized CPGs involves learning adaptive processes that entail numerous activities. Therefore, activity theory was adopted as the research framework of this study to provide insights into the potential reasons for and factors influencing computerized CPG use according to the dimensions of the environment, tool, and subject; these factors are critical for realizing the benefits of computerized CPGs (object). The identified factors found in prior CPG studies can be divided into three characteristics mentioned in activity theory for further validation in the context of computerized CPGs. Yusof et al. [54] proposed a similar research framework (HOT-fit) of activity theory and argued that successful healthcare information technologies (HIT) implementation depends on a good fit among human, organizational, and technical elements. Both activity theory and HOT-fit emphasize the interaction among the subject (human), environment (organization), and tool (technology).

Technology acceptance model

The technology acceptance model (TAM), developed by Davis [55], is one of the most used theoretical models for predicting and explaining whether users will accept new information technology (IT)/information systems (IS). The model has argued that people’s attitudes toward behaviors and subjective criteria determine their behavioral intentions toward technology applications, which consequently affect their own behavior [55]. Perceived usefulness and perceived ease of use are two major factors influencing user IT acceptance (intention/attitude/use), and the two factors are influenced by many external variables. Although numerous TAM-related studies have investigated various groups and application systems, Wu et al. [56] indicated that the TAM focuses more on technology aspects without considering the effects on human and organizational factors in the adoption process. Yarbrough and Smith [57] indicated that TAM constructs generally hold in a physician-specific context, but the perceived ease of use component of the model does not prove to be consistently related to either attitudes or perceived usefulness. Yarbrough and Smith [57] argued that one limitation of the TAM is its inability to consider the influence of external variables and barriers to technology acceptance. They further suggested customizing the inclusion of variables (personal characteristics, organizational characteristics, and IS characteristics) to enhance the model’s accuracy.

Recently, some studies have proposed an extended TAM, incorporating considerations of human, organizational, and technology factors based on the TAM, for investigating critical factors influencing both user (i.e., health care professional) acceptance of HIS [54] and pain management decision support systems [58]. The two cited studies have significantly improved (84.4 and 64 %, respectively) on the variance of user acceptance. By contrast, some studies have investigated physicians’ acceptance of health care systems by using the unified theory of acceptance and use of technology (UTAUT) model or modified UTAUT model [5961]. The UTAUT model identifies four key constructs (performance expectancy, effort expectancy, social influence, and facilitating conditions) as direct determinants of user intention and behavior, and four moderators (gender, age, experience, and voluntariness of user) [62]. However, the variance of physicians’ user acceptance (intention) of the TAM, UTAUT, or revised UTUAT model was approximately 20–47 %. For example, Venkatesh et al. [61] indicated that the original UTUAT predicted only approximately 20 % of the variance in intention, and the modified UTAUT predicted 44 %. This indicated that some predictors of physicians’ intention remain unknown and further investigation is necessary. Our study explored the critical factors influencing physicians’ intention to use computerized CPGs on the basis of the extended TAM concept rather than the original UTAUT.

Methods

Research framework

We propose an integrated research model for exploring and understanding critical factors influencing physicians’ intention to use computerized CPGs by incorporating activity theory (three dimensions of factors) with TAM concepts (intention as dependent variable). The investigated variables of the model are derived from results obtained in previous TAM and CPG studies. The model is composed of three independent dimensions: tool (technology) characteristics (i.e., complexity, perceived usefulness, compatibility, and perceived ease of use), subject (human) characteristics (i.e., attitude and task uncertainty), and environment (organization) characteristics (i.e., social influence and organizational support). These dimensions with eight variables are considered potential critical factors (independent variable) influencing physicians’ intention to use computerized CPGs. Relevant studies on the model are summarized in Table 1, and the research model is depicted in Fig. 1.
Table 1

Relevant studies to the dimensions of the research model

Dimension (variable)

Francke et al. [40]

Flottorp et al. [42]

Kortteisto et al. [41]

UTAUT related studies [5961]

Tool (Technology)

 Complexity

Easy to understand, can easily be tried out, and do not require specific resources

Guideline factors

  

 Perceived usefulness

   

Performance expectancy

Compatibility

Effective strategies often have multiple components than single strategy

   

 Perceived ease of use

   

Effort expectancy

Subject (Human)

 Attitude

Awareness of the existence of the guideline and familiarity with its content

Individual health professional factors

Attitude toward the behavior

 

 Task uncertainty

 

Individual health professional factors

  

Environment (Organization)

 Social influence

A lack of support from peers or superiors

Professional interactions

Subjective norm

Social influence

Organizational support

Insufficient staff and time

Incentives and resources, capacity for organizational change, and social, political, and legal factors

Perceived behavior control

Facilitating conditions

Fig. 1

Results of model validity regarding factors affecting physicians’ intention to use computerized CPGs

The relationships between activity theory and the three dimensions of factors investigated in this study are summarized as follows: 1) The ultimate goals (objects) of computerized CPG use are to improve the quality of care and to provide timely decision support; however, these goals are achievable only when the computerized CPGs are implemented and used widely by their major users (physicians); 2) physicians’ intention to use computerized CPGs is a surrogate for evaluating the successful implementation of computerized CPGs (user behavior); 3) the three dimensions (tool, subject, and environment/organization) of activity theory are critical for achieving expected goals obtained from computerized CPG use; 4) a tool is a characteristic of computerized CPGs, and four factors (complexity, perceived usefulness, compatibility, and perceived ease of use) are identified; 5) subjects are characteristics of the major users (physicians) of computerized CPGs, and two factors (attitude and task uncertainty) are identified; and 6) the environment/organization is the context in which users (physicians) are placed, and two factors (social influence and organizational support) are identified.

The tool characteristics of the research model are defined as follows: (1) Complexity is the degree to which the content and presentation of computerized CPGs is difficult to understand and read [40, 63]; (2) perceived usefulness represents users’ subjective beliefs about the benefits of using HIT to achieve job goals in medical practice [64, 65]; (3) compatibility refers to the degree to which the system is consistent with the user’s value, demand, and experience [65, 66]; and (4) perceived ease of use represents the degree to which users believe that using HIT is free of effort [57, 65]. Some studies [40, 42] have indicated that computerized CPGs are difficult to understand and that using them requires certain resources; thus, computerized CPGs have a low likelihood of being adopted by users. Thus, the high complexity of computerized CPGs and the substantial effort and time spent by users in learning how to use them inhibit computerized CPG use. Perceived usefulness was found to be a critical factor influencing the user acceptance of systems among various health care professionals [5861, 65, 67]. When physicians perceive a higher degree of system usefulness (i.e., real-time information and benefits in improving quality and physicians’ training and education), they have a more positive attitude and are willing to accept computerized CPGs.

Rogers [63] defined compatibility as the degree to which a newly developed system conforms to the organizational culture, individual experiences, and user requirements. Previous studies have indicated that compatibility is a critical factor affecting users’ willingness to adopt innovative technology [68, 69]. A newly developed system that highly conforms to the original organizational culture and user requirements is highly compatible and easy to accept. Hence, we assume that the compatibility of computerized CPGs with the original organizational system and physicians’ work experience, practice, and culture positively influences physicians’ willingness to use them. In addition, Venkatesh et al. [62] argued that IS is easy to use and can enhance performance. When physicians perceive that learning how to use CPGs is easy, they have a more positive attitude toward accepting computerized CPGs and a higher intention to use computerized CPGs. Therefore, computerized CPGs that have an easy-to-operate interface require little time and effort for learning, require no dedicated personnel for education and training, and have a high probability of being adopted [70]. Perceived ease of use was found to be a critical factor influencing the user acceptance of systems among various health care professionals [58, 65, 67].

The subject characteristics are defined as follows: (1) Attitude is an individual’s positive or negative perception of an object, an action, or a person [71], and (2) task uncertainty refers to the degree of uncertainty and unstructured conditions faced by physicians in clinical work [72, 73]. Saillour-Glenisson and Michel [74] have indicated that a lack of knowledge, familiarity, and recognition causes a negative physician attitude toward CPGs. They found that the degree of structure of the work environment is a crucial factor affecting the physicians’ CPGs use. Kortteisto et al. [41] found that health care professionals’ attitude toward a behavior was a critical factor associated with the intention to use CPGs. Therefore, we hypothesized that uncertainty and unstructured clinical work conditions positively influence physicians’ intention to use computerized CPGs.

The environment characteristics are defined as follows: (1) Social influence represents the degree to which an individual perceives that important others (i.e., colleagues or supervisors) believe that he or she should use a system [62, 75], and (2) organizational support refers to resources (time, money, and human) provided by health care institutions to physicians to alleviate stress during computerized CPG use [76]. Francke et al. [40] and Flottorp et al. [42] have argued that support from colleagues or supervisors is critical for facilitating CPG use. Moreover, Kortteisto et al. [41] determined that the subjective norm, a concept similar to social influence, is a crucial factor associated with health care professionals’ intention to use CPGs. Therefore, we hypothesized that a supervisor and colleague having a positive attitude toward computerized CPG use and actively providing a supporting environment positively influence physicians’ willingness to use computerized CPGs. Organizational resources include human resources, support provided by relevant facilities, and time for diagnosis and treatment. Francke et al. [40], Flottorp et al. [42], and Kortteisto et al. [41] have suggested that organizational resources are critical for CPG use. Hence, we hypothesized that medical institutions providing adequate human resources and sufficient time to reduce physicians’ workload positively influence physicians’ willingness to use computerized CPGs.

The dependent variable of this study is physicians’ intention to use computerized CPGs, and is operationalized as the strength of an individual’s willingness to take a certain action [62, 71]. The higher the strength of a behavioral intention is, the higher the likelihood that the action will be taken. Behavioral intention measures the degree of willingness to perform a certain action. In this study, physicians’ intention to use computerized CPGs is defined as physicians’ internal orientation toward and motivation to apply computerized CPGs. As discussed, we proposed eight hypotheses that can be divided into three categories.
  1. H1:
    Tool (technology) factors have a significant impact on physicians’ intentions to use computerized CPGs.
    1. H1a:

      The complexity of computerized CPGs affects physicians’ intention to use computerized CPGs.

       
    2. H1b:

      The perceived usefulness of computerized CPGs affects physicians’ intention to use computerized CPGs.

       
    3. H1c:

      The compatibility of computerized CPGs affects physicians’ intention to use computerized CPGs.

       
    4. H1d:

      The perceived ease of use of computerized CPGs affects physicians’ intention to use computerized CPGs.

       
     
  2. H2:
    Subject (human) factors have a significant impact on physicians’ intention to use computerized CPGs.
    1. H2a:

      Physicians’ attitude affects their intention to use computerized CPGs.

       
    2. H2b:

      The task uncertainty faced by physicians affects their intention to use computerized CPGs.

       
     
  3. H3:
    Environment (organization) factors have a significant impact on physicians’ intention to use computerized CPGs.
    1. H3a:

      Social influence affects physicians’ intention to use computerized CPGs.

       
    2. H3b:

      Organizational support affects physicians’ intention to use computerized CPGs.

       
     

Instrument and subjects

The initial research framework and corresponding questionnaires were developed according to a literature review and revised by a panel consisting of four experts (two directors of an EBM center, one attending physician practicing clinical medicine, and one professor with a master’s degree in medical information) to increase content validity. Expert validity was measured using a content validity index (CVI) of 0.8 as the criterion for item selection [77, 78]. All CVI values of measurement items, except for one item, exceeded 0.8, and the overall CVI was 0.96, indicating excellent expert validity. In addition, the experts provided necessary semantic revisions to ensure that the retained items were appropriate and consistent. The revised questionnaire was then pretested by five physicians who were not employees of the investigated hospitals to correct semantic misunderstandings before distribution to the respondents. To address potential ethical concerns, our study protocol and informed consent forms were reviewed and approved by the Institutional Review Board (IRB) of the Chi-Mei Medical Center in Taiwan before the surveys were distributed and collected. After receiving approval from the IRB of the target hospital (IRB-10206-006), research was conducted from 13 June 2013 to 14 June 2014. Study participants were voluntary and verbal consent (waiver documentation of consent). Responses were anonymous and untraceable to individual physician.

We collected data from physicians of the three investigated hospitals (one medical center and two regional hospitals) that have established EBM centers. The reason for the selection of the case hospitals is that EBM centers are the major units promoting clinical guidelines and providing clinical guidelines related to training and education courses for physicians in Taiwanese hospitals. The EBM is a candidate course on health care quality in physicians’ continuing education, and EBM has been a fundamental training course for postgraduate year 1 (PGY1) resident physicians since 2003. The PGY1 training courses are mandatory and require resident physicians to complete 3–6 h of training in using EBM databases and at least eight clinical case analysis reports on EBM topics in 2011 [5]. Furthermore, the investigated hospitals are pilot adopters of computerized CPGs in Taiwan, and they have implemented computerized CPGs by incorporating them into existing HIS and EMR systems to provide timely clinical decision support since 2011. Because this study investigated critical factors influencing physicians’ intention to use computerized CPGs rather than the actual use of computerized CPGs, all physicians of the investigated hospitals were potential respondents.

This study employed a survey methodology that involved using a 41-item structured questionnaire composed of the following two major parts: (1) a part for recording the respondents’ demographic data, and (2) a part evaluating factors influencing physicians’ intention to use computerized CPGs. The questionnaire items were rated using a 5-point Likert scale, with scores ranging from 1 (strongly disagree) to 5 (strongly agree).

The tool construct (technology characteristics) addressed complexity, perceived usefulness, compatibility, and perceived ease of use. Complexity was measured using four items adapted from previous studies [74, 76, 79]. Perceived usefulness was measured using five items adapted from previous studies [8082]. Compatibility was measured using four items adapted from Rogers [63] and Teng et al. [65]. Perceived ease of use was measured using three items adapted from prior studies [58, 65, 67]. The subject construct (human characteristics) comprised attitude and task uncertainty. Attitude was measured using six items adapted from previous studies [21, 74, 83]. Task uncertainty was measured using three items adapted from previous studies [74, 84]. The environment construct (organization characteristics) comprised social influence and organizational support. Social influence was measured using four items adapted from Venkatesh et al. [62, 75]. Organizational support was measured using three items adapted from Cabana et al. [21] and Simpson et al. [76]. Physicians’ intention to use computerized CPGs was measured using two items adapted from Bhattacherjee and Hikmet [85]. Detailed descriptions of the questionnaire are shown in the Appendix.

Data analysis

Structural equation modeling is an extension of several multivariate techniques (most notably multiple regression and factor analysis) that has been widely used in many research areas such as education, psychology, sociology, management, and health. ([86], p. 578). Hair et al. [86] summarized the reasons for the wide use of this technique twofold: (1) It provides a straightforward method for addressing multiple relationships simultaneously while enabling statistical efficiency, and (2) it assesses relationships comprehensively and provides a transition from exploratory to confirmatory analysis. We conducted confirmatory analysis to investigate the factors influencing physicians’ intention to use computerized CPGs. A structural equation modeling technique was used as the tool for further data analysis. The reliability and validity of the measurement model were assessed by conducting CFA with LISREL 8.5, and the maximum likelihood method was applied to estimate the parameters of the research model. A structural model was used to examine the causal model of the investigated model.

Results

Demographic data

The survey was distributed to 505 physicians of the three investigated hospitals, and 238 valid responses were returned, representing a response rate of 47.1 %. Voluntary participation by physicians and top management support might explain the relatively high response rate compared with the average rate (10–20 %) for physicians in Taiwan [87]. Most of the respondents were male (87.8 %), and all of them had a bachelor (73.9 %) or master’s (or higher) degree (26.1 %). Those aged between 31 and 50 years accounted for the highest proportion of the respondents (70.2 %). The major service departments of the respondents were other specialties (44.5 %), internal medicine (26.5 %), surgery (20.2 %), and obstetrics and pediatrics (8.6 %). Almost half of the respondents (52.9 %) had 10 or more years of experience in clinical practice at their current hospital. Most of the respondents (92.9 %) had previously used EBM databases, and 68.5 % had 3 years or more of experience in using CPGs. Approximately 62.6 % of the respondents had previously used CPGs, and 44.1 % had more than 1 year of experience of computerized CPG use. Therefore, most of the respondents were experienced users of computerized CPGs and EBM databases, and all of them had considerable experience in clinical practice. The representativeness of the respondents was thus demonstrated. The demographic data of the respondents are shown in Table 2.
Table 2

Participant demographic data

Measure

Category

No (#)

Percent (%)

Age

<30

16

6.7

31–40

80

33.6

41–50

87

36.6

51–60

48

20.2

>60

7

2.9

Gender

Male

209

87.8

Female

29

12.2

Education level

Bachelor

176

73.9

Master (or higher)

62

26.1

Department

Internal medicine

63

26.5

Surgery

48

20.2

Obstetrics and pediatrics

21

8.8

Others

106

44.5

Years of experience in clinical practice

<1

2

0.8

1–5

43

18.1

5–10

67

28.2

>10

126

52.9

Years of experience in using EBM

None

17

7.1

<1

23

9.7

1-3

35

14.7

>3

163

68.5

Years of experience in using CCPG

None

89

37.4

<1

44

18.5

1–3

58

24.4

>3

47

19.7

Key factors contributing to physicians’ intention to CPGs use

Measurement model

The model comprised nine variables (complexity, perceived usefulness, compatibility, perceived ease of use, attitude, task uncertainty, social influence, organizational support, and intention to use computerized CPGs) with a total of 34 items. Seven model-fit measures were used to evaluate the overall goodness fit of the model and collected data as follows: the ratio of the chi-square (χ 2) to degree-of-freedom (df); goodness-of-fit index (GFI); normalized fit index (NFI); nonnormalized fit index (NNFI); incremental fit index (IFI); comparative fit index (CFI); and root mean square error of approximation (RMSEA). The results of goodness-of-fit statistics are summarized in Table 3. The ratio of the chi-square (χ 2 = 691.01) to degree-of-freedom (df = 263.00) was 2.63 for the measurement model. The value of GFI was 0.81, which is slightly lower than the recommended value but acceptable; the values of the NFI (0.90), NNFI (0.90), IFI (0.90), and CFI (0.90) were equal to the suggested 0.9 benchmark [86]. The RMSEA was 0.076, which is within the acceptable levels of 0.05–0.08 [86]. Therefore, most of the fit indices met the recommended values, indicating a reasonable fit between the data and the proposed measurement model.
Table 3

Model evaluation - overall fit measurement

Fit indices

Recommended value

Value

χ2

N/A

691.03

d.f

N/A

263

χ2/d.f

≤3.0

2.63

Goodness-of-fit index (GFI)

0.9

0.81

Normalized fit index (NFI)

0.9

0.90

Non-normalized fit index (NNFI)

0.9

0.90

Incremental fit index (IFI)

0.9

0.90

Comparative fit index (CFI)

0.9

0.90

Root mean square error of approximation (RMSEA)

≤0.08

0.076

The results of Cronbach’s α, composite reliability (CR), convergent validity, and discriminant validity are shown in Table 4. The values of Cronbach’s α for all constructs were higher than the 0.7 threshold and ranged from 0.702 to 0.927. The internal consistency of the measurement model by evaluating CR of the constructs ranged from 0.762 to 0.913, exceeding the recommended value of 0.70. Fornell and Larcker [88] suggested that the value of average variance extracted (AVE) exceed 0.5 and be greater than each square correlation, to indicate adequate convergent validity and discriminant validity. As shown in Table 4, the AVE of the measurements ranged from 0.580 to 0.832, indicating excellent convergent validity. This study also revealed that the square roots of all AVE values were greater than the off-diagonal elements presented in the corresponding rows and columns, suggesting excellent discriminant validity. Therefore, this study demonstrated adequate reliability, convergent validity, and discriminant validity.
Table 4

Results of reliability and validity of the research model

Variable

CMP

PU

COM

PE

PA

TU

SI

OS

ITU

AVE

CR

Cronbach’s α

CMP

0.790

        

0.621

0.868

0.853

PU

0.577

0.762

       

0.580

0.805

0.821

COM

0.689

0.614

0.864

      

0.747

0.923

0.894

PE

0.735

0.576

0.805

0.829

     

0.687

0.869

0.835

PA

0.208

0.392

0.269

0.342

0.840

    

0.706

0.878

0.702

TU

0.100

0.096

0.090

0.005

0.261

0.835

   

0.698

0.820

0.789

SI

0.206

0.450

0.486

0.424

0.115

0.077

0.912

  

0.832

0.908

0.891

OS

0.353

0.521

0.501

0.482

0.301

0.117

0.602

0.893

 

0.798

0.922

0.927

ITU

0.359

0.636

0.485

0.489

0.386

0.122

0.548

0.696

0.868

0.753

0.859

0.846

The bold numbers in the diagonal row are square roots of the AVE. Off-diagonal elements are the correlations among constructs

Abbreviations: CMP complexity, PU perceived usefulness, COM compatibility, PE perceived ease of use, PA physicians’ attitude, TU task uncertainty, SI social influence, OS organizational support, ITU intention to use computerized CPGs

Structure model

The path coefficient and R2 value were estimated primarily to test the structural model. The path coefficient represents the magnitude and direction of the relation vector between the variables and is used to test their significance; the R2 value refers to the extent to which the exogenous (independent) variables explain the variance of the endogenous (dependent) variables (in percentages), representing the predictive power of the model. Both the path coefficient and R2 value showed the fit between the structural model and empirical data.

As shown in Fig. 1, four hypotheses were supported significantly in this study. The results indicated that the perceived usefulness of computerized CPGs (γ = 0.219, p < 0.05) in tool characteristics, attitude toward using computerized CPGs (γ = 0.451, p < 0.001) in subject characteristics, and social influence (γ = 0.213, p < 0.05) and organizational support (γ = 0.285, p < 0.001) in environment characteristics were critical factors influencing physicians’ intention to use computerized CPGs, supporting H1b, H2a, H3a, and H3b. These factors accounted for 68.6 % of the total explained variance in physicians’ intention to use computerized CPGs. However, the data showed that the hypotheses regarding complexity (H1a), compatibility (H1c), perceived ease of use (H1d) in tool factors, and task uncertainty (H2b) in subject factors were not significantly supported.

Discussion

The results show that the perceived usefulness of computerized CPGs, attitude toward using computerized CPGs, social influence, and organizational support are critical factors influencing physicians’ intention to use computerized CPGs. Consistent with Wu et al. [56], our study revealed that attitude is the most critical factor affecting physicians’ intention to use computerized CPGs. Cabana et al. [21] reported that identification with CPGs is the primary cause that hinders the promotion of computerized CPGs. Previous studies have suggested that a lack of knowledge, familiarity, and recognition indicates a negative physician attitude toward computerized CPGs [21, 74]. To improve physicians’ attitude toward computerized CPGs, hospital managers should expend substantial effort in computerized CPG training and education.

Perceived usefulness refers to users’ subjective beliefs about the benefits of using HIT to achieve job goals in medical practice [64, 65]. Stoddard et al. [89] found that when clinical guidelines are flexible and can be adjusted according to the situation to improve the quality of medical decisions, physicians are more willing to use the guidelines. Consistent with previous studies investigating health care professionals’ acceptance of HIT [5861, 65, 67], our study demonstrated that perceived usefulness is a significant factor influencing physicians’ intention to use computerized CPGs. This finding implies that when physicians consider that actual benefits can be obtained through computerized CPG use, they have a higher intention to use such systems. We suggest that hospital managers establish a dedicated EBM team for promoting and educating physicians about the benefits of computerized CPGs to improve the quality of medical decisions, job performance, and cost effectiveness.

Our study showed that support from colleagues and supervisors positively affects physicians’ intentions to use computerized CPGs; this is consistent with the findings of Davis and Taylor-Vaisey [79] and Kortteisto et al. [41]. Thus, support from colleagues and supervisors is encouraged for facilitating the widespread use of computerized CPGs in a health care institute. Organizational support refers to resources (time, money, and human resources) provided by health care institutions to physicians to alleviate stress during computerized CPG use. Cabana et al. [21] and Simpson et al. [76] have reported that deficiency in organizational resources (e.g., personnel and time) increases workloads and personal work stress, leading to poor results of clinical guideline implementation. This finding is consistent with our results. Therefore, hospitals managers can alleviate physicians’ stress during computerized CPG use by providing sufficient personnel, time, and equipment, thereby increasing physicians’ intention to use computerized CPGs.

Although several critical factors influencing physicians’ intention to use computerized CPGs were identified, complexity, compatibility, perceived ease of use, and task uncertainty had no significant effect on physicians’ intention to CPG use. Simpson et al. found that higher patient disease complexity may lower the rate of computerized CPG adoption. A computerized CPG that accounts for various clinical conditions simultaneously has not been developed. In a follow-up interview with physicians, the experts expressed that physicians consider various conditions when making clinical decisions and cannot completely accept every suggestion provided by computerized CPGs. Thus, computerized CPGs may provide insufficient support when physicians are faced with complex decision-making situations. In addition, when clinical guidelines become increasingly comprehensive and interdisciplinary, their complexity inevitably increases.

Our study derived a result inconsistent with those of previous studies, in that compatibility was a critical factor affecting users’ willingness to adopt innovative technology [68, 69]. In 2009, the Ministry of Health and Welfare added EBM to the evaluation criteria for the Evaluation Standards and Measurement Criteria for Teaching Hospitals. EBM practices and the writing of medical records were included in the training of teachers, medical interns, and resident physicians. EBM learning and application were also incorporated into training for general medical skills. Thus, to manage and improve the quality of medical care, EBM standards should be used to determine whether the quality meets the required standards. In addition, regular reviews should be conducted. Therefore, knowledge regarding EBM is now included in the basic training for medical interns and general medical education; hence, the number of physicians who have read and understood the clinical guidelines has gradually increased. Moreover, several developed clinical guidelines have been combined with EMR and HIS to assist physicians in clinical decision-making processes. This can provide a possible explanation for why compatibility does not significantly affect physicians’ intention to use computerized CPGs.

Although perceived ease of use was found to be a critical factor influencing the user acceptance of systems among various health care professionals [58, 65, 67], perceived ease of use had no significant influence in our study. Moreover, Bhattacherjee and Hikmet ([85], p. 734) found that perceived ease of use nonsignificantly influenced physicians’ intention to use HIT. They argued that the effect of perceived ease of use can be mediated by other factors such as perceived usefulness [55] and perceived technology control [90]. Therefore, perceived ease of use may not be influential at the postimplementation stage of CPG use.

Task uncertainty refers to the frequency at which physicians encounter uncertainties and disorganization-related problems. Raymond and Bergeron [91] indicated that when encountering task uncertainty, system users exhibited increased satisfaction toward personalized decision-making systems. Although much attention has been paid to clinical guidelines in assisting physicians’ decision-making processes, task uncertainty was not a critical factor significantly influencing physician’ intention to use computerized CPGs. In practice, this finding may be attributed to three reasons: (1) When providing clinical care, physicians frequently encounter sudden, unexpected problems; (2) not all hospital departments have guidelines that can be followed to manage such problems; and (3) clinical care work can involve unprecedented diseases or complications, and treating patients generally requires that physicians from other disciplines participate and share their expertise. Although comprehensive and fully integrated cross-department clinical guidelines facilitate addressing aforementioned problems, few effective complex cross-departmental clinical guidelines have been developed. In addition, ANOVA analysis was conducted to investigate the effects of age and computerized CPG experience on physicians’ intention to use computerized CPGs. The results demonstrated that age and computerized CPG experience were significant demographic factors influencing physicians’ intention to use computerized CPGs. Although age and user experience were found to be significant to user intention in this study, future research can focus on exploring the moderate effect between the investigated factors and user intention.

The findings of this study are subject to five major limitations. First, we examined three hospitals with EBM centers belonging to one medical group in Southern Taiwan, to investigate organizational concerns (culture differences and the establishment of EBM centers), thus potentially restricting the generalization of the findings to other medical institutions. Second, data were collected from experienced users of EBM databases and computerized CPGs. The results from the respondents were based on users’ perceptions, experiences, and understanding. Thus, the data collected may not be adequately objective. Third, we analyzed only cross-sectional data collected by physicians during one period. The data should be carefully interpreted regarding the effect of time. Fourth, multimorbidity is one of the major causes inhibiting the development of CPGs. Because the 17 developed CPGs in Taiwan focus on a single disease rather than multiple conditions of patients, the findings might be not adequately applied in computerized CPGs for addressing multimorbidity problems. Finally, computerized CPG use in the three investigated hospitals is voluntary; thus, tracing individual physicians is difficult. This study investigated physicians’ intention to use computerized CPGs rather than actual use, because we could not directly identify the actual users of computerized CPGs. Two conditions may have caused this problem: (1) Some medical specialists (departments) do not have access to developed computerized CPGs, and (2) some physicians are computerized CPG users but do not consider themselves actual computerized CPG users because they may violate or neglect the suggestions from computerized CPGs. Future studies should pay particular attention to the mentioned problems.

Conclusions

Crucial findings

This study investigated critical factors influencing physicians’ intention to use computerized CPGs through an integrated model derived from activity theory and the TAM. This model comprising tool (technology) characteristics (complexity, usefulness, compatibility, and ease of use), subject (human) characteristics (attitude and task uncertainty), and environment (organization) characteristics (social influence and organizational support) was proposed and validated. The results showed that attitudes toward using computerized CPGs (γ = 0.451, p < 0.001), organizational support (γ = 0.285, p < 0.001), perceived usefulness of computerized CPGs (γ = 0.219, p < 0.05), and social influence (γ = 0.213, p < 0.05) were critical factors influencing physicians’ intention to use computerized CPGs, and these factors can explain 68.6 % of the variance in intention to use computerized CPGs. Managers should pay considerable attention to these identified factors and provide adequate resources and incentives to facilitate the promotion and use of computerized CPGs. The appropriate use of computerized CPGs will enable realizing clinical benefits, particularly by improving the quality of care and facilitating clinical processes. Computerized CPGs, which incorporate the advantages of both IT and CPGs, are considered effective means of facilitating the use of CPGs and improving the quality of care at affordable costs [23, 31]. This study is among the few computerized CPG studies that have emphasized management concerns. The findings of this study are useful in formulating appropriate strategies for increasing user participation at various system implementation stages (computerized CPG design, implementation, and postimplementation).

Theoretical and practical implications

From a theoretical perspective, our study integrated the concepts of activity theory and the TAM to address the research gap created by previous IS/IT studies focusing excessively on technical-economic perspectives; thus, we propose a more comprehensive model for effectively explaining the phenomena relating to changes in health care IT use behavior, as suggested by Kuutti [44]. Evaluation based on activity theory perspective can provide insights into the potential reasons for and factors influencing computerized CPG use according to the subject, environment, and tool dimensions, and these factors are critical for realizing computerized CPG benefits (objects). Thus, activity theory can be applied to elucidate the interaction among individuals, tools, and the environment (at the organizational, societal, and other levels). This study formulated an integrative research model for investigating critical factors influencing HIT adoption according to activity theory and TAM results. The proposed research model can be extended and used to identify critical factors influencing the adoption of HIT and clinical decision support systems in the health care industry.

The increasing emphasis on EBM has transformed the concept of EBM from a mere topic in academic research to a method applied in clinical practice. Empirical evidence-based decision models can be used to assist decision makers in establishing policies based on priorities and the efficient use of limited resources; thus, these models are vital decision-making tools [82]. In practice, implementing computerized CPG influences both the medical care procedures and patient care habits of physicians; therefore, the identified key factors contribute to physicians’ willingness to use computerized CPGs. By understanding the critical factors, further analysis can be conducted and effective strategies can be formulated to facilitate the widespread use and promotion of computerized CPGs.

Previous studies have found that introducing computerized CPGs may change physicians’ clinical behaviors and result in resistance to computerized CPGs use [20, 21]. Our study showed that physicians’ attitude toward computerized CPGs is the most crucial factor affecting their intention to use computerized CPGs. In addition, our study demonstrated that perceived usefulness is a significant factor influencing physicians’ intention to use computerized CPGs. This finding indicates that when physicians consider that actual benefits are obtained through computerized CPG use, they have a higher intention to use the system. Effectively educating physicians and advocating the computerized CPG use could reinforce the usefulness of clinical guidelines, improve medical decision adequacy, enable clinical cost effectiveness, and ultimately improve physicians’ intention to use computerized CPGs.

Our study confirmed that social influence and organizational support are critical for physicians’ intention to use computerized CPGs. Supervisor support and colleague recommendations regarding computerized CPG use are helpful for enhancing physicians’ intention to use CPGs. The results demonstrated that providing physicians with sufficient resources (time, money, and human resources) facilitates alleviating stress during computerized CPGs use. Therefore, hospital managers can alleviate physicians’ stress from using computerized CPGs by providing sufficient personnel, time, and equipment, thereby increasing physicians’ intention to use computerized CPGs.

Declarations

Acknowledgements

Sincere thanks and recognition are given to the National Science Council of Taiwan for funding this study (NSC 101-2410-H-041 -005).

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Authors’ Affiliations

(1)
Department of Hospital and Health Care Administration, Chia-Nan University of Pharmacy and Science
(2)
Department of Information Management, Chia-Nan University of Pharmacy and Science

References

  1. Field MJ, Lohr KN. Guidelines for clinical practice: from development to use. Washington, D.C: National Academy Press; 1992.Google Scholar
  2. Andrews EJ, Redmond HP. A review of clinical guidelines. Br J Surg. 2004;91(8):956–64.PubMedView ArticleGoogle Scholar
  3. de Jong JD, Groenewegen PP, Spreeuwenberg P, Schellevis F, Westert GP. Do guidelines create uniformity in medical practice? Soc Sci Med. 2010;70(2):209–16.PubMedView ArticleGoogle Scholar
  4. Grando A, Peleg M, Glasspool D. A goal-oriented framework for specifying clinical guidelines and handling medical errors. J Biomed Inform. 2010;43(2):287–99.PubMedView ArticleGoogle Scholar
  5. Chen CF. Glocalization: the outlook for Taiwan evidence based health care. J Nurs. 2014;61(6):12–6.Google Scholar
  6. Chen C, Chen K, Hsu CY, Chiu WT, Li YC. A guideline-based decision support for pharmacological treatment can improve the quality of hyperlipidemia management. Comput Methods Programs Biomed. 2010;97(3):280–5.PubMedView ArticleGoogle Scholar
  7. Jian WS, Hsu CY, Hao TH, Wen HC, Hsu MH, Lee YL, et al. Building a portable data and information interoperability infrastructure-framework for a standard Taiwan Electronic Medical Record Template. Comput Methods Programs Biomed. 2007;88(2):102–11.PubMedView ArticleGoogle Scholar
  8. Wu CT, Chung KP, Miao NF, Wang CH, Hsu YH, Chang KJ, et al. Exploring factors of physicians’ willingness on implementation of clinical practice guidelines-focus on characters of clinical practice guidelines. J Healthc Manag. 2005;6(2):153–72.Google Scholar
  9. Giguere A, Legare F, Grimshaw J, Turcotte S, Fiander M, Grudniewicz A, et al. Printed educational materials: effects on professional practice and health care outcomes. Cochrane Database Syst Rev. 2012;10, CD004398.PubMedGoogle Scholar
  10. Grimshaw JM, Eccles MP, Lavis JN, Hill SJ, Squires JE. Knowledge translation of research findings. Implement Sci. 2012;7:50.PubMedPubMed CentralView ArticleGoogle Scholar
  11. Handler J, Lackland DT. Translation of hypertension treatment guidelines into practice: a review of implementation. J Am Soc Hypertens. 2011;5(4):197–207.PubMedView ArticleGoogle Scholar
  12. Larisch A, Reuss A, Oertel W, Eggert K. Does the clinical practice guideline on Parkinson’s disease change health outcomes? A cluster randomized controlled trial. J Neurol. 2011;258(5):826–34.PubMedView ArticleGoogle Scholar
  13. Mitchell C, Dwyer R, Hagan T, Mathers N. Impact of the QOF and the NICE guideline in the diagnosis and management of depression: a qualitative study. Brit J Gen Pract. 2011;61(586):e279–89.View ArticleGoogle Scholar
  14. Mourad SM, Hermens RPMG, Liefers J, Akkermans RP, Zielhuis GA, Adang E, et al. A multi-faceted strategy to improve the use of national fertility guidelines; a cluster-randomized controlled trial. Hum Reprod. 2011;26(4):817–26.PubMedView ArticleGoogle Scholar
  15. Schnoor M, Meyer T, Suttorp N, Raspe H, Welte T, Schafer T, et al. Development and evaluation of an implementation strategy for the German guideline on community-acquired pneumonia. Qual Saf Health Care. 2010;19(6):498–502.PubMedGoogle Scholar
  16. Damiani G, Pinnarelli L, Colosimo SC, Almiento R, Sicuro L, Galasso R, et al. The effectiveness of computerized clinical guidelines in the process of care: A systematic review. BMC Health Serv Res. 2010. doi:10.1186/1472-6963-10-2.PubMedPubMed CentralGoogle Scholar
  17. Lyng KM. From clinical practice guidelines, to clinical guidance in practice -impacts for computerization. Int J Med Inform. 2013;82(12):358–63.View ArticleGoogle Scholar
  18. Grimshaw J, Eccles M, Tetroe J. Implementing clinical guidelines: current evidence and future implications. J Contin Educ Health Prof. 2004;24 Suppl 1:S31–7.PubMedView ArticleGoogle Scholar
  19. Johnson CM, Turley JP. The significance of cognitive modeling in building healthcare interfaces. Int J Med Inform. 2006;75(2):163–72.PubMedView ArticleGoogle Scholar
  20. Bouaud J, Seroussi B, Falcoff H, Venot A. Design factors for success or failure of guideline-based decision support systems: an hypothesis involving case complexity. AMIA Annu Symp Proc. 2006;71–5.Google Scholar
  21. Cabana MD, Rand CS, Powe NR, Wu AW, Wilson MH, Abboud PA, et al. Why don’t physicians follow clinical practice guidelines? a framework for improvement. JAMA. 1999;282(15):1458–65.PubMedView ArticleGoogle Scholar
  22. Larson E. Status of practice guidelines in the United States: CDC guidelines as an example. Prev Med. 2003;36(5):519–24.PubMedView ArticleGoogle Scholar
  23. Shahar Y, Young O, Shalom E, Galperin M, Mayaffit A, Moskovitch R, et al. A framework for a distributed, hybrid, multiple-ontology clinical-guideline library, and automated guideline-support tools. J Biomed Inform. 2004;37(5):325–44.PubMedView ArticleGoogle Scholar
  24. Goud R, Hasman A, Strijbis AM, Peek N. A parallel guideline development and formalization strategy to improve the quality of clinical practice guidelines. Int J Med Inform. 2009;78(8):513–20.PubMedView ArticleGoogle Scholar
  25. Tu SW, Campbell JR, Glasgow J, Nyman MA, McClure R, McClay J, et al. The SAGE guideline model: achievements and overview. J Am Med Inform Assoc. 2007;14(5):589–98.PubMedPubMed CentralView ArticleGoogle Scholar
  26. Boxwala AA, Peleg M, Tu S, Ogunyemi O, Zeng QT, Wang D, et al. GLIF3: a representation format for sharable computer-interpretable clinical practice guidelines. J Biomed Inform. 2004;37(3):147–61.PubMedView ArticleGoogle Scholar
  27. Wang D, Peleg M, Tu SW, Boxwala AA, Ogunyemi O, Zeng Q, et al. Design and implementation of the GLIF3 guideline execution engine. J Biomed Inform. 2004;37(5):305–18.PubMedView ArticleGoogle Scholar
  28. Abidi S. A conceptual framework for ontology based automating and merging of clinical pathways of comorbidities. In: Knowledge management for health care procedures. Berlin: Springer; 2009. p. 55–66.View ArticleGoogle Scholar
  29. Abidi S, Cox J, Shepherd M, Abidi SR. Using OWL ontologies for clinical guidelines based comorbid decision support, 45th Hawaii international conference on system sciences. 2012. p. 3030–8.Google Scholar
  30. Butler T, Feller J, Pope A, Emerson B, Murphy C. Designing a core IT artefact for knowledge management systems using participatory action research in a government and a non-government organization. J Strategic Inf Syst. 2008;17(4):249–67.View ArticleGoogle Scholar
  31. Colombet I, Aguirre-Junco AR, Zunino S, Jaulent MC, Leneveut L, Chatellier G. Electronic implementation of guidelines in the EsPeR system: a knowledge specification method. Int J Med Inform. 2005;74(7–8):597–604.PubMedView ArticleGoogle Scholar
  32. Dufour JC, Fieschi D, Fieschi M. Coupling computer-interpretable guidelines with a drug-database through a web-based system - the PRESGUID project. BMC Med Inform Decis Mak. 2004;4:2.PubMedPubMed CentralView ArticleGoogle Scholar
  33. Fraccaro P, Casteleiro MA, Ainsworth J, Buchan I. Adoption of clinical decision support in multimorbidity: A systematic review. JMIR Med Inform. 2015;3(1):1–18.View ArticleGoogle Scholar
  34. Jafarpour B, Abidi S. Merging disease-specific clinical guidelines to handle comorbidities in a clinical decision support setting. Artif Intell Med. 2013;7885:28–32.View ArticleGoogle Scholar
  35. Roshanov PS, Fernandes N, Wilczynski JM, Hemens BJ, You JJ, Handler SM, et al. Features of effective computerised clinical decision support systems: meta-regression of 162 randomised trials. BMJ. 2013;346:f657.PubMedView ArticleGoogle Scholar
  36. Verhoeven F, Steehouder MF, Hendrix RMG, Van Gemert-Pijnen JEWC. From expert-driven to user-oriented communication of infection control guidelines. Int J Hum-Comput St. 2010;68(6):328–43.View ArticleGoogle Scholar
  37. Isern D, Moreno A. Computer-based execution of clinical guidelines: a review. Int J Med Inform. 2008;77(12):787–808.PubMedView ArticleGoogle Scholar
  38. Lyng KM, Pedersen BS. Participatory design for computerization of clinical practice guidelines. J Biomed Inform. 2011;44(5):909–18.PubMedView ArticleGoogle Scholar
  39. Wilk S, Michalowski W, Michalowski M, Farion K, Hing MM, Mohapatra S. Mitigation of adverse interactions in pairs of clinical practice guidelines using constraint logic programming. J Biomed Inform. 2013;46(2):341–53.PubMedView ArticleGoogle Scholar
  40. Francke AL, Smit MC, de Veer AJE, Mistiaen P. Factors influencing the implementation of clinical guidelines for health care professionals: a systematic meta-review. BMC Med Inform Decis Mak. 2008;8:38.PubMedPubMed CentralView ArticleGoogle Scholar
  41. Kortteisto T, Kaila M, Komulainen J, Mäntyranta T, Rissanen P. Healthcare professionals’ intentions to use clinical guidelines: A survey using the theory of planned behaviour. Implement Sci. 2010;5:51.PubMedPubMed CentralView ArticleGoogle Scholar
  42. Flottorp S, Oxman A, Krause J, Musila N, Wensing M, Godycki-Cwirko M, et al. A checklist for identifying determinants of practice: A systematic review and synthesis of frameworks and taxonomies of factors that prevent or enable improvements in healthcare professional practice. Implement Sci. 2013;8:35.PubMedPubMed CentralView ArticleGoogle Scholar
  43. Vygotsky LS. Psychology of Art. Cambridge: MIT Press; 1971.Google Scholar
  44. Kuutti K. Activity Theory as a potential framework for human-computer interaction research. In: Nardi BA, editor. Context and consciousness: activity theory and human computer Interaction. Cambridge: MIT Press; 1995. p. 17–44.Google Scholar
  45. Leontiev AN. Problems of the development of the mind. Moscow: Progress; 1981 (Russian edition published 1959).Google Scholar
  46. Nardi BA. Studying context a comparison of activity theory situated action models and distributed cognition. In: Nardi BA, editor. Context and consciousness: activity theory and human-computer interaction. Cambridge: MIT Press; 1995. p. 35–52.Google Scholar
  47. Tuikka T. Remote concept design from an activity theory perspective, In proceedings of the 2002 ACM conference on computer supported cooperative work. 2002. p. 186–95.Google Scholar
  48. Liaw SS, Huang HM, Chen GD. An activity-theoretical approach to investigate learners’ factors toward e-learning systems. Comput Hum Behav 2007. 2007;23(4):1906–20.View ArticleGoogle Scholar
  49. Lim CP, Hang D. An activity theory approach to research of ICT integration in Singapore schools. Comput Educ. 2003;41(1):49–63.View ArticleGoogle Scholar
  50. Beaudry A, Carillo KD. The customer-centered B2C literature through the lens of activity theory: a review and research agenda. Commun Assoc Inf Syst. 2006;17:428–503.Google Scholar
  51. Lin F, Chaboyer W, Wallis M, Miller A. Factors contributing to the process of intensive care patient discharge: an ethnographic study informed by activity theory. Int J Nnurs Stud 2013. 2013;50(8):1054–66.Google Scholar
  52. Boisen E, Bygholm A, Cavan D, Hejlesen OK. Copability, coping, and learning as focal concepts in the evaluation of computerised diabetes disease management. Int J Med Inform. 2003;70(2):353–63.PubMedView ArticleGoogle Scholar
  53. Sun J, Koong KS, Poole MS. Critical success factors for context-aware mobile communication systems. Int J Mobile Communication. 2009;7(3):290–307.View ArticleGoogle Scholar
  54. Yusof MM, Kuljis J, Papazafeiropoulou A, Stergioulas LS. An evaluation framework for health information systems: human, organization and technology-fit factors (HOT-fit). Int J Med Inform. 2008;77(6):386–98.PubMedView ArticleGoogle Scholar
  55. Davis FD. Perceived usefulness, perceived ease of use, and user acceptance of information technology. Mis Quart. 1989;13(3):319–40.View ArticleGoogle Scholar
  56. Wu IL, Li JY, Fu CY. The adoption of mobile healthcare by hospital’s professionals: an integrative perspective, Decis. Support Syst. 2011;51:587–96.View ArticleGoogle Scholar
  57. Yarbrough AK, Smith TB. Technology acceptance among physicians: a new take on TAM. Med Care Res Rev. 2007;64(6):650–72.PubMedView ArticleGoogle Scholar
  58. Hsaio JL, Wu WC, Chen RF. Factors of accepting pain management decision support systems by nurse anesthetists. BMC Med Inform Decis Mak. 2013;13:16.View ArticleGoogle Scholar
  59. Chang IC, Hwang HG, Hung WF, Li YC. Physicians’ acceptance of pharmacokinetics-based clinical decision support systems. Expert Syst Appl. 2007;33(2):296–303.View ArticleGoogle Scholar
  60. Esmaeilzadeh P, Sambasivan M, Kumar N, Nezakati H. (2015). Adoption of clinical decision support systems in a developing country: Antecedents and outcomes of physician’s threat to perceived professional autonomy. Int J Med Inform. 2015;84(8):548–60.PubMedView ArticleGoogle Scholar
  61. Venkatesh V, Sykes T, Zhang X. Just what the doctor ordered: a revised UTAUT for EMR system adoption and use by doctors, 44th Hawaii international conference on system sciences (HICSS). 2014. p. 1–10.Google Scholar
  62. Venkatesh V, Morris MG, Davis G, Davis F. User acceptance of information technology: toward a unified view. Mis Quart. 2003;27(3):425–78.Google Scholar
  63. Rogers EM. Diffusion of innovation. 5th ed. New York: Free Press; 2003.Google Scholar
  64. Wakefield DS, Halbesleben JRB, Ward MM, Qiu Q, Brokel J, Crandall D. Development of a measure of clinical information systems expectations and experiences. Med Care. 2007;45(9):884–90.PubMedView ArticleGoogle Scholar
  65. Chen RF, Hsaio JL. An investigation on physicians’ acceptance of hospital information systems: a case study. Int J Med Inform. 2012;81:810–20.PubMedView ArticleGoogle Scholar
  66. Teng JTC, Grover V, Guttler W. Information technology innovations: general diffusion patterns and its relationships to innovation characteristics. IEEE Trans Eng Manage. 2002;49(1):13–27.View ArticleGoogle Scholar
  67. Aldosari B. User acceptance of a picture archiving and communication system (PACS) in a Saudi Arabian hospital radiology department. BMC Med Inform Decis. 2012;12:44.View ArticleGoogle Scholar
  68. Craig VS, France B, Christie LC. Factors influencing the adoption of web based shopping: the impact of trust. ACM SIGMIS Database. 2004;35(2):32–49.View ArticleGoogle Scholar
  69. Tung FC, Chang SC, Chou CM. An extension of trust and TAM model with IDT in the adoption of the electronic logistics information system in HIS in the medical industry. Int J Med Inform. 2008;77(5):324–35.PubMedView ArticleGoogle Scholar
  70. Vogelzang M, Zijlstra F, Nijsten MW. Design and implementation of GRIP: a computerized glucose control system at a surgical intensive care unit. BMC Med Inform Decis Mak. 2005;5:38.PubMedPubMed CentralView ArticleGoogle Scholar
  71. Fishbein M, Ajzen I. Belief attitude, intention, and behavior: an introduction to theory and research. Reading: Addison Wesley; 1975.Google Scholar
  72. Alexander LW, Randolph WA. The fit between technology and structure as a predictor of performance in nursing subunits. Acad Manage J. 1985;28(4):844–59.PubMedView ArticleGoogle Scholar
  73. Blili S, Raymond L, Rivard S. Impact of task uncertainty, end-user involvement, and competence on the success of end-user computing. Inform Manage. 1998;33(3):137–53.View ArticleGoogle Scholar
  74. Saillour-Glenisson F, Michel P. Individual and collective facilitators of and barriers to the use of clinical practice guidelines by physicians: a literature review. Rev Epidemiol Sante Publique. 2003;51(1 Pt 1):65–80.PubMedGoogle Scholar
  75. Venkatesh V, Morris MG. Why don’t men ever stop to ask for directions? Gender, social influence, and their role in technology acceptance and usage behavior. Mis Quart. 2000;24(1):115–39.View ArticleGoogle Scholar
  76. Simpson SH, Marrie TJ, Majumdar SR. Do guidelines guide pneumonia practice? a systematic review of interventions and barriers to best practice in the management of community-acquired pneumonia. Respir Care Clin N Am. 2005;11(1):1–13.PubMedView ArticleGoogle Scholar
  77. Hilton A, Skrutkowski M. Translating instruments into other languages: development and testing processes. Cancer Nurs. 2002;25(1):1–7.PubMedView ArticleGoogle Scholar
  78. Petrick JF. Development of a multi-dimensional scale for measuring the perceived value of a service. J Leis Res. 2002;34(2):119–34.Google Scholar
  79. Davis DA, Taylor-Vaisey A. Translating guidelines into practice: a systematic review of theoretic concepts, practical experience and research evidence in the adoption of clinical practice guidelines. CMAJ. 1997;157(4):408–16.PubMedPubMed CentralGoogle Scholar
  80. Warner BW, Kulick RM, Stoop MM, Mehta S, Stephan M, Kotagal UR. An evidence-based clinical pathway for acute appendicitis decreases hospital duration and cost. J Pediatr Surg. 1998;33(9):1371–5.PubMedView ArticleGoogle Scholar
  81. Wolff M, Bower DJ, Marbella AM, Casanova JE. US family physicians’ experiences with practice guidelines. Fam Med. 1998;30(2):117–21.PubMedGoogle Scholar
  82. Flores G, Lee M, Bauchner H, Kastner B. Pediatricians attitudes, beliefs, and practices regarding clinical practice guidelines: a national survey. Pediatrics. 2000;105:496–501.PubMedView ArticleGoogle Scholar
  83. Davis FD. User acceptance of information technology: system characteristics, user perception and behavioral impacts. Int J Man Mach Stud. 1993;38:319–40.View ArticleGoogle Scholar
  84. Goodhue DL, Thompson RL. Task-technology fit and individual performance. Mis Quart. 1995;19(2):213–36.View ArticleGoogle Scholar
  85. Bhattacherjee A, Hikmet N. Physicians’ resistance toward healthcare information technology: a theoretical model and empirical test. Eur J Inform Syst. 2007;16(6):725–37.View ArticleGoogle Scholar
  86. Hair JF, Anderson RE, Tatham RL, Black WC. Multivariate data analysis. 6th ed. New Jersey: Prentice Hall; 1998.Google Scholar
  87. Chen DR. Psychosocial work environment and self-rated health: a comparative study of DC and ERI models in a sample of medical specialists. Taiwan J Public Health. 2006;25:93–106.Google Scholar
  88. Fornell C, Larcker DF. Evaluating structural equation models with unobservable variables and measurement error. J Mark Res. 1981;18(1):39–50.View ArticleGoogle Scholar
  89. Stoddard JJ, Grossmen JM, Rudell LS. Physicians more likely to face guality incentives than incentives that may restrain care. Issue Brief Cent Stud Health Syst Change. 2002;48:1–4.PubMedGoogle Scholar
  90. Chau PYK, Hu PJ. Examining a model of information technology acceptance by individual professionals: an exploratory study. J Manage Inform Syst. 2002;18(4):191–229.Google Scholar
  91. Raymond L, Bergeron F. Personal DSS success in small enterprises. Inform Manage. 1992;22(5):301–8.View ArticleGoogle Scholar

Copyright

© Hsiao and Chen. 2016

Advertisement