Health information technology and patient outcomes: the role of information and labor coordination

• Author: Jeffrey S. McCullough,Robert Town,Stephen T. Parente
• Date: 01 February 2016
• Published date: 01 February 2016
• DOI: http://doi.org/10.1111/1756-2171.12124

RAND Journal of Economics

Vol.47, No. 1, Spring 2016

pp. 207–236

Health information technology and patient

outcomes: the role of information and labor

coordination

Jeffrey S. McCullough∗

Stephen T. Parente∗∗

and

Robert Town∗∗∗

Health information technology (IT) adoption, it is argued, will dramaticallyimprove patient care.

We study the impact of hospital IT adoption on patient outcomes focusing on the role of patient

and organizational heterogeneity. We link detailed hospital discharge data on all Medicare fee-

for-service admissions from 2002–2007 to detailed hospital-level IT adoption information. For

all IT-sensitive conditions, we find that health IT adoption reduces mortality for the most complex

patients but does not affect outcomes for the median patient. Benefits fromhealth IT are primarily

experienced by patients whose diagnoses requirecross-specialty care coordination and extensive

clinical information management.

1. Introduction

The US health care system is often criticized as fragmented and uncoordinated. The con-

sequences of this poorly integrated system are stark, leading to a large number of preventable

medical errors and wasteful resource allocation. The cost of medical errors are famously large

with some estimates attributing over 44,000 deaths annually to inpatient hospital errors (IOM,

1999). Faced with a health care system that appears far from optimized, many health policy

analysts have placed health information technology (IT) at the center of their reform initiatives

∗University of Michigan; jeffreysmccullough@gmail.com.

∗∗University of Minnesota; stephen.parente@gmail.com.

∗∗∗University of Pennsylvania and NBER; rjxtown@gmail.com.

We gratefully acknowledgeassistance from The Rober t Wood Johnson Foundation’s Changes in Health Care Financing

& Organization (grant no. 64845) and the Agency for Health Research and Quality.We have received helpful comments

from David Dranove, Leila Agha, seminar participants at the University of Southern California, Ohio State University,

the American Health Economics Conference, and the NBER Summer Institute. We also thank the Health Information

Management Systems Society (HIMSS) Analytics for use of their data and the Minnesota Supercomputing Institute (MSI)

for computing resources. The views expressed herein are those of the authors and do not necessarily reflect the views of

the National Bureau of Economic Research.

C2016, The RAND Corporation. 207

208 / THE RAND JOURNAL OF ECONOMICS

(IOM, 1999, 2001; Buntin et al., 2011).1Recent improvements in health IT and the growth of

IT investments in other sectors of the economy suggest that health IT can dramatically improve

the practice of medicine. The recent Health Information Technology for Economic and Clinical

Health (HITECH) Act (2009) commits an estimated $34 billion in subsidies for private health care

providers to adopt and utilize health IT. Although health IT’s potential is clear, these technologies

are not without their critics, and the ultimate economic and clinical impact of widespread health

IT adoption remains uncertain.2,3

Weanalyze the impact of hospitals’ adoptions of different types of health IT on outcomes for

millions of Medicare patients. Our data span a period of rapid IT adoption. The volume, quality,

and detail of our data allow us to examine heterogeneity in IT’s impact across organizational and

patient domains that, in turn, helps illuminate mechanisms through which health IT acts.

This study contributes to a growing economic literature on the value and impact of IT.

Information technologies have reduced the costs of coordinating production, communication,

and information processing, leading to a substantial increase in productivity in a wide variety

of industries and settings (Stiroh, 2002; Brynjolfsson and Hitt, 2003). Information technology’s

benefits are not, however, evenly distributed across organizations as the gains from these tech-

nologies depend upon organizational and labor complementarities (Bresnahan, Brynjolfsson, and

Hitt, 2002). For example, Bartel, Ichniowski, and Shaw (2007) found that IT-enabled manu-

facturing equipment increased productivity, dramatically increased product customization, and

increased machine operator skill levels. Bloom, Sadun, and Reenen (2012b) found that US-based

firms operating in the UK earned higher returns from IT investments than other firms. These

high returns were a consequence of US firms’ organizational structures that complemented IT

investments. Garicano and Heaton (2010) found that the impact of police department IT adoption

complements particular organizational and management practices. Similarly, Autor, Katz, and

Krueger (1998) uncovered strong evidence of skill-biased IT and labor complementarity. Other

studies emphasized the role of task-level complementarities between IT and labor inputs (Autor,

Levy, and Murnane, 2003; Baker and Hubbard, 2003; Acemoglu and Autor, 2011; Bloom et al.,

2012a).

Health care is an interesting environment for studying the roles of organizational and

task-based complementarities in IT. Each patient’s diagnoses presents its own set of information

challenges. Some diagnoses require extensive information management, whereas others need

cross-provider care coordination. In addition, guideline adherence and medical order communi-

cations are commonly noted barriers to good quality hospital care. Health IT, it is argued, is an

effective vehicle to reduce the barriers to high-quality care by capturing, managing, analyzing,

and disseminating this information.

Electronic Medical Records (EMRs) provide the foundation for health information systems.

EMRs systematically collect patients’ health information, replacing traditional medical charts.

Computerized provider order entry (CPOE) allows providers to electronically enter orders for

services and medications. Through direct order entry,CPOE reduces oppor tunities for miscommu-

nication between disparate care providers.These technologies also serve as a platform for decision

support functions, which may reduce prescribing errors and improve clinical guideline compli-

ance. Our analysis primarily (but not exclusively) focuses on these core clinical applications.

Health IT’s value may depend on the organizational context, how organizations apply the

technology, and the information content of tasks for which the technology is used. EMR sys-

tems may automate standardized treatment guidelines and implement rules-based procedures

to prevent common errors. This combination of technological and organizational capabilities is

commonly referred to as “Meaningful Use”—these guideline mechanisms are emphasized in the

1In his January 3, 2009 radio address, President Obama stated, “Wewill update and computerize our health care

system to cut red tape, prevent medical mistakes, and help reduce health care costs by billions of dollars each year.”

2Black et al. (2011), for example, review the literature and suggest that the IT and quality relationship is unclear.

3There are many acronyms in this article; therefore, an acronym dictionary is provided in the Appendix.

C

The RAND Corporation 2016.