Understanding the triggers of senior managers’ IS investment decisions is important for both IS scholarship and IS practice. In this paper, the authors integrate insights from the behavioral theory of the firm and neoinstitutional theory to better understand how hospital senior managers make decisions about how and how much to invest in IS. Building on the understanding of IS investments as a form of organizational search, they argue that at least four recurring search mechanisms (i.e. problemitics search, institutionalized search, mimetic search and slack search) will influence the decision of the amount of resources allocated to IS.

Problemistic search is defined as search that is stimulated by a problem and is directed toward finding a solution to that problem. Slack search refers to a process that when resources are viewed as being in surplus, senior managers will seek to identify promising investment opportunities. Institutionalized search is commonly known as search conducted by organizations, which tends to be orderly, standardized, and somewhat independent of success or failure. Mimetic search can be conceived as a “search for information about what organizational characteristics are legitimated in their environmental niche. The first two search mechanism can be explained as behavioral search mechanism, while the last two are supported by neoinstitutional theory.

They draw on panel data from all 153 nonspecialist public hospital organizations in England. Findings from our dynamic panel data analyses suggest that hospital managers’ IS investment decisions are driven not only by the desire to improve organizational performance (problemistic search), but also by the need to achieve continuity in resource allocation (institutionalized search) and to signal compliance with external norms and expectations (mimetic search). The objective of making adequate use of uncommitted resources (slack search), however, was found to be salient as a trigger of IS investments only among hospital organizations with low regulative legitimacy.

The study tests how privacy protection affects the diffusion of electronic medical records based on the understanding of the network effect. There is empirical study of the role of network externalities in electronic banking adoption by banks. However, this study is the first to address how network effects and technology adoption decisions are affected by privacy protection. According to the conceptual model of hospital EMR adoption, when net gain from electronic records, defined relative to the alternative technology of paper records, is positive, a hospital will adopt EMRs.

As expected, state privacy protection of hospital medical information inhibits EMR adoption or the network effects of EMR, which is defined as positive externalities experienced by individual hospitals who adopt EMRs when other local hospitals have adopted electronic records. The finding shows that without hospital privacy protection, one hospital’s adoption increase the propensity of other hospitals in the local area to adopt by 7%, but with the privacy protection, no network effects are indicated. This study contributes in quantifying how a hospital’s decision to adopt EMRs is affected by whether state privacy protection restricts a hospital’s ability to disclose information.

The authors suggest for the future study to include the overall welfare effects of either EMRs or privacy laws, which is inferred from a trade-off between privacy protection and EMRs. For the further study, it is demanded to investigate the extent to which privacy protection can be optimized to minimize disruption to the diffusion and use of interdependent technology. I also agree that the positive spillovers from increase of security from protecting confidentiality and evaluations toward the effect of strategy toward adopting EMRs would increase the speed of diffusion.

Electronic Medical Records and Physician Productivity: Evidence from Panel Data Analysis

Bhargava and Mishra (2014) examined the impacts of electronic medical record (EMR) system implementation on physician productivity across different specialties and to uncover the dynamic and temporal nature of these impacts. The conceptual foundations of this study draw upon research from three strands of work: physician productivity, IT-enabled productivity, and task-technology fit (TTF). And Bargava and Mishra (2014) tried to fill the gap in the literature by examining two research questions: (1) How does physician productivity change over time as a result of EMR implementation? (2) Does this impact differ for physicians of different specialties?

Bhargave and Mishra (2014) collected pre and post-implementation productivity data on 87 primary care physicians (PCPs) over a 39-month period, yielding 3,186 physician (Internal medicine specialists (IMs), pediatricians (Peds), and family practitioners (FPs))-month observations at an academic medical center associated with a large public university in the western U.S. Also qualitative interview data and survey data were collected to obtain insights into the context of EMR use at this study.

Bhargave and Mishra (2014) found that: First, the net impact of EMR is more benign on IMs than on Peds and FPs. Second, physician experience is not significantly related to productivity. Third, FPs and Peds experience a decrease in productivity compared to IMs in the learning phase.

The main contributions in the study: (1) Extends the research in both health informatics and IT productivity literatures. (2) Provides informed indications that the nature, direction, and magnitude of impacts induced by technological innovations. (3) Provides a rigorously derived dynamic pattern for describing the productivity impact of EMRs over time. (4) Demonstrates divergence in EMR impact across specialties, require further investigation and are of consequence to both health IT vendors and users.

Investing in Information Systems: On the Behavioral and Institutional Search Mechanisms Underpinning Hospitals’ IS Investment Decisions

While most of the literature on IS investment is focusing on the adoption and the value of IS investments, little literature inquiries the determinants of those investments. To this end, this paper conceptualizes senior managers’ decisions on IS investments in hospitals as four search mechanisms that jointly determine IS investment intensity of an organization.

Based on the behavioral theory of the firm (for the first two mechanisms) and the institutional theory (for the last two mechanisms), the four search mechanisms proposed are problemistic search, slack search, institutionalized search, and mimetic search, as described below. Problemistic search occurs when senior managers realize that there are problems to be solved, and is measured by the difference between hospitals’ current performance and expected performance. Slack search is motivated by financial surplus, which arises excess capital resource that allows managers to make investments; this is captured by the liquidity of hospitals’ financial performance. Institutionalized search is routine-based, temporally stable investments which is measured by the one-year lagged IS investment intensity of the organization. Finally, mimetic search is the imitation of the investment done by peer organizations, and is captured by the IS investment intensity of a reference group.

The paper empirically proves that problemistic search, institutional search and mimetic search are three mechanisms that directly lead to IS investment. It also proposes a moderating role of regulative legitimacy, and shows that slack search, institutional search and mimetic search indirectly determine IS investment, through the moderation of regulative legitimacy.

Blunting Damocles’ Sword: A Longitudinal Model of Healthcare IT Impact on Malpractice Insurance Premium and Quality of Patient Care

Motivation:

Prior studies on the business value of information technology (IT) mainly focus on the impact of IT investments

on productivity and firm profitability. Few have considered its implication on expected and actual

product or service quality. This paper fills this gap.

Research Question:

This study investigates the impact of past healthcare IT (HIT) expenditure on the malpractice insurance premium (MIP) and the moderating effect of past HIT expenditure on the relationship bet theyen past MIP and current quality of patient care.

Conclusion:

By examining hospitals’ IT spending and performance, they found evidence of a direct impact of HIT on MIP as well as on the quality of patient care. The evidence of lower readmissions and mortalities in association with past HIT expenditure validates Hypothesis H1 regarding the direction of impact of HIT on quality of patient care. The hypothesis was supported by the argument that IT is able to provide appropriate alerts during patient care. Readily available patient information facilitates good quality medical care, and coordination between staff members administering medical care is facilitated by HIT, leading to fewer errors in patient care. Although this result is consistent with conclusions in prior studies, the significance of the coefficients of past MIP in both models of patient care quality indicates that prior research had not satisfactorily accounted for ex ante risk. Thus, Hypothesis H3 contributes to healthcare literature by providing an explanation for the inconsistency among the findings of prior research.

Electronic Medical Records and Physician Productivity: Evidence from Panel Data Analysis

Bhargava and Mishra (2014) investigated the impact of an electronic medical record (EMR) system on the productivity of physicians. This study is of interest and significant for the following reasons: 1) there is ambiguous evidence on whether and how EMRs could facilitated efficient workflows for physicians, causing practically concerns but receiving little academic attentions. 2) The EMRs’ impact on physician productivity is also related to whether or not the technology fulfills its potential to curtail the rise in healthcare expenditures in U.S.

Drawing on prior literature on physician productivity, IT productivity and task-technology fit theory, the authors examined how two types of EMRs use (information review and information enter) could affect physician productivity, and how those effects are different in specialties (internal medicine specialists (IMs), pediatricians (Peds), and family practitioners (FPs)) across implementation process (pre-entry, learning and stable phases).

They used a panel data set comprising 87 physicians with different specialties in 12 primary care clinics of an academic healthcare system in the western U.S. They employed the Arellano-Bond system GMM estimation technique on their data set with 3186 physician-month productivity observations over 39 months. They found that productivity drops sharply after EMRs implementation and recovers partly over the next few months. The longer-term impact depends on physician specialty. And the impact of EMRs is more benign on IMs than on Peds and FPs.

They postured that the fit provided by an EMR system to the task requirements of physicians of various specialties may be key to disentangling the productivity dynamics. Their finding revealed that, on one hand, EMR systems do not produce immediate productivity gain that could lead to substantial savings in healthcare; at the same time, EMRs do not cause a major productivity loss on a sustained basis, as many physicians fear.

Chan and Ghose (2014) tenaciously examine one phenomenon: the positive impact of Craigslist’s entry on the increase in HIV incidence. The authors theoretically explain the phenomenon with a few underlying mechanisms. First, people tend to prefer an instant satisfaction to a delayed reward, a choice that an online intermediary can easily support. Second, an online intermediary can provide the means to lower the cost of information seeking and to find a broader (potential) audience who may not be available otherwise.

Based on the natural experiment to identify the entry effect of Craigslist in the period from 1999 to 2008, the authors find that the entry does lead to more cases of HIV. This simple-but-powerful finding is supported through various control variables, robustness checks, and falsification tests. As for control variables, the empirical analysis includes demographic features, socioeconomic indicators, and the Internet availability along with alternative models to accommodate fixed effects of states and years. More importantly, the empirical test is compared with other sample data of non-physical contact diseases to catch a spurious effect. An alternative test also addresses the possibility of (if any) pre-entry effect. In order to secure exogeneity of Craigslist entry, hierarchical duration models are employed to see whether the control variables in the main model can predict the entry. The presence of HIV incidence is included to check the reverse causality. Throughout all the models, the entry of Craigslist is consistently significant and positive to the increase of HIV incidence.

All in all, the authors evidently support a substantial economic impact of an information intermediary on issues such as outbreak of diseases and surge of healthcare costs.

Blunting Damocles’ Sword: A Longitudinal Model of Healthcare IT Impact on Malpractice Insurance Premium and Quality of Patient Care

Menon and Kohli (2013) examine the impact of IT investments on actual product or service quality. Specifically, this paper studies the relationship between the past healthcare IT (HIT) expenditure and the malpractice insurance premium (MIP). It mainly wants to investigate two problems (1) whether past HIT expenditures affects MIP and the quality of patient care and (2) whether this expenditure moderates the relationship between past MIP and the quality of patient care.

The authors propose that HIT can incur lower MIP and improve the patient care quality because it can efficiently monitor, control, and reduce information asymmetry between the hospital and insurer. Also, HIT’s recordkeeping and monitoring capabilities induce employees and physicians to align their behaviors with service delivery prescribed, which result in higher quality.

Empirically, the paper develops a longitudinal model including a panel data set for 66 general medical and surgical hospitals. It uses readmission and mortality rate as the proxy of the quality of patient care and adopts the depreciation expenses across information processing units as the measure of HIT. The regression model uses the random effects to rule out the impact of hospital-specific and time-specific factors. In addition, the model uses lags of dependent variables as instruments to account for the endogeneity of the independent variables.

The paper found that past HIT expenditures improve the quality of patient care and that it is negatively associated with current MIP. They also found a negative moderating effect of past HIT expenditures on the link between past MIP and current service quality. This paper contributes to the business value of IT by conceptualizing the impact of HIT on ex-ante risk as an expectation and future operations.