Electronic Medical Record implementation in Hospital: 1 An empirical investigation of individual and organizational determinants.

The implementation of hospital-wide Electronic Medical Records (EMRs) is still an unsolved quest 29 for many hospital managers. Despite past studies acknowledged the numerous advantages of EMRs, hospital 30 professionals such as physicians and nurses have been found indifferent, when not resistant, to the adoption of 31 EMRs. This study combines institutional and individual factors to explain which determinants can trigger or 32 inhibit the EMRs implementation in an hospital setting, and which variables managers can exploit to guide 33 behaviours. Data have been collected through a survey administered to physicians and nurses 34 in an Italian University Hospital in Rome. A total of 114 high-quality responses had been received. Results 35 show that both, physicians and nurses, expect many benefits from the use of EMR. In particular, it is believed 36 that the EMR will have a positive impact on: quality, efficiency and effectiveness of care; handover 37 communication between healthcare workers; teaching, tutoring and research activities; greater control of your 38 own business. Data show an interplay between individual and institutional determinants: normative factors 39 directly affect perceived usefulness (C = 0.30 **), perceived ease of use (C = 0.26 **) and intention to use 40 EMR (C = 0.33 **). Regulative factors directly affect only intention to use EMR (C = -0.21 **). Control 41 variables have no impact on other variables in the model. The analysis carried out shows that the key 42 determinants of the intention to use EMR are the normative ones (peer influence). Therefore, the Management 43 can leverage on power users to motivate, generate and manage change.


Introduction 48
The implementation of hospital-wide Electronic Medical Records (EMRs) is still an unsolved quest 49 for many hospitals [1,2]. Despite a significant body of evidence about the numerous advantages of EMRs [e.g., 50 2-9], many initiatives of EMRs implementation still fall far behind expectations and hospitals appear unable 51 to fully capture the opportunities offered by EMRs in terms of improvement of the organization of hospital 52 healthcare delivery, performance monitoring and support to clinical research and trials. As a result of these 53 failures, hospital managers and professionals developed different narratives to justify these negative outcomes. 54 On the one hand, hospital managers complained that physicians and nurses were indifferent, when not resistant, 55 to the adoption of EMRs [2,10]. This happened because hospitals are professional organizations, where 56 hospital professionals have wide jurisdiction on decision-making and practices and, as consequence, hospital 57 managers have limited authority to mandate innovation and change of established practices and behaviours 58 influence of hospital managers [16,17] on the basis of perceived usefulness and ease of use. On the other hand, 104 we grounded on Institutional Theory, whose claim is that professionals are strongly subject to institutional 105 forc ac b a a a d c d a ac c a a a 106 expects from them [18]. I , c a a d decision to engage with an EMR is not entirely 107 based on rational thinking, but it is affected by the influence of the overarching structures, rules, social norms 108 and culture in which they are embedded [19,20]. By combining these theories that have been adopted so far 109 to offer opposite explanations , this study investigated the interplay between institutional and individual 110 factors, thus offering novel insights on the determinants of hospital a EMR . In 111 this view, the main goal of this study is pointing-out how and to what extent individual and organizational 112 determinants might trigger or inhibit EMR implementation in hospitals, and whether an interplay does exist 113 between them. 114

115
Theoretical background 116 In order to evaluate the potential interplay between individual and institutional variables, a research 117 framework has been created ( Fig. 1). As anticipated above, the framework integrates into a coherent view of 118 two theories that belong to two different and still isolated bodies of literature: 119 -The Technology Acceptance Model (TAM), from Information Science, that has been widely used in the 120 last decades in healthcare to understand what leads professionals or patients to accept or reject Information 121 Technology [21]; 122 -The Institutional Theory, from Public Management, that has been largely adopted in the last decades to 123 assess how institutional factors shape professionals behaviours [22][23][24]. 124

125
The TAM was introduced for the first time by Davis in 1989 [21]. The main problem raised by the 126 author was to understand what leads people to accept or reject Information Technology. In this regards, two 127 main variables have been identified: the perceived usefulness and the ease of use. Perceived usefulness 128 a degree to which a person believes that using a particular system would enhance his or her job a d, a a degree to which a person believes that using a system would be 131 [21,25] and induces the potential users to use a certain technology since it requires low energy 132 expenditure while it may bring advantages. The first one induces an individual to use technology as it allows 133 to obtain better results in his work; the ease of use, on the other hand, stimulates potential users to use a certain 134 technology since many advantages are supported with low energy expenditure. 135

136
The Institutional Theory refers to a line of organizational research that recognize the significant 137 organizational effects that are associated with the increase of cultural and social forces. According to Scott 138 Coherently to past researches about user acceptance of new technologies [26,27], we considered age and job 155 seniority as key control variables. Additionally, to narrow the knowledge gap about how hospital professionals 156 belonging to either different profession (e.g., physicians vs. nurses) or different speciality (e.g., cardiology vs. orthopaedics) might be interested to use an EMR, we included clinical speciality and profession as control 158 variables. Figure 1 offers a synoptic view of our research framework, where the independent variable (i.e., the 159 intention to use an EMR) is explained by individual factors from TAM (i.e., perceived usefulness and perceived 160 ease of use) as well as by institutional factors from Institutional Theory (i.e., regulative factors that refer to the 161 degree of adhesion to a a a a , and normative factors that explain the peer influence among 162 hospital colleagues. Control variables have been also displayed. is mid-size (around 300 beds), many-disciplines, teaching and private. These characteristics are salient for our 184 study because they have been found to increase the complexity of EMR implementation. Being teaching hospital, there is more room for divergent goals between professionals and managers, thus creating the correct 186 setting where to investigate the interplay between individual and organizational factors. Being many-187 discipline, there is room to study the potential conflict among professionals from different disciplines with 188 respect to the implementation of an EMR. Finally, being mid-size, CBM is a valid setting to observe the 189 potential divergence between nurses and doctors about EMR. These considerations persuaded us that CBM 190 could be assumed as an exemplary case to be investigated. To increase the generalizability of our results, a 191 mixed-method approach has been applied for both data collection and analysis. Structured interviews have 192 been organized with hospital managers and professionals to understand the organizational context and deep-193 diving in the available secondary sources of information that have been made available (e.g., internal reports, 194 meeting minutes, press release, etc.). This qualitative analysis has been coupled with a more quantitative one

Determinants of current behaviours 220
Our data show that both physicians and nurses expect many benefits from the use of EMR. In

238
Our study sought to better clarify the relationship between organizational and individual determinants 239 of the intention to use EMRs in a hospital setting by nurses and physicians. Previous studies [2,5,8-240 10,12,13,33] have focused mainly on either the barriers or the facilitators that might impact on the 241 implementation of EMRs, but, to the best of our knowledge, it has never been deepened if and how 242 organizational and individual factors interact and affect jointly hospital a motivation to use EMRs. 243 Our findings show that the main determinants of the intention to use EMRs are the normative ones (peer 244 as reported by Gastaldi et al. [34] in the absence of coercive mechanisms, institutional pressures toward EMR 261 use are primarily normative (i.e., the organization has binding expectations about EMR use, with which 262 professionals comply with social obligations) and/or mimetic (i.e., EMRs fit with existing taken-for-granted 263 beliefs and logics of action, so professionals should support their adoption) [32]; therefore, as already 264 mentioned, it is possible to leverage some power users in order to manage change. 265