Does the meaningful use of electronic health records improve patient outcomes?

• Author: Deepa Wani,Manoj Malhotra
• DOI: http://doi.org/10.1016/j.jom.2018.06.003
• Date: 01 May 2018
• Published date: 01 May 2018

Contents lists available at ScienceDirect

Journal of Operations Management

journal homepage: www.elsevier.com/locate/jom

Does the meaningful use of electronic health records improve patient

outcomes?

Deepa Wani

a,

∗, Manoj Malhotra

b

a

University of Texas, San Antonio, USA

b

Case Western Reserve University, USA

ARTICLE INFO

Accepted by: S. de Treville.

Keywords:

Electronic health records

Healthcare

Meaningful use of technology

Task complexity

Length of stay

ABSTRACT

While Electronic Health Records (EHRs) hold the promise of improving patient outcomes, past research on their

efficacy has yielded inconsistent results. In this study, we overcome several drawbacks of past research by

examining not just partial versus full adoption, but the impact of meaningful assimilation of EHRs as mandated

by the HITECH (Health Information Technology for Economic and Clinical Health) Act. Detailed patient-level

data from acute-care hospitals in California, coupled with relevant data from several other sources, is used to

conduct our analysis. After controlling for self-selection bias, our results show that overall length of stay (LOS) is

reduced by 3%, on average, for all patients who undergo treatment at hospitals that are meaningful-use as-

similated, relative to patients at hospitals that have fully adopted EHRs. The magnitude of this reduction is

increased for patients with greater comorbidity complexity and greater coordination complexity. In addition, we

find an overall decrease in readmission. We do not find such benefits among full adopters of EHRs. Thus, our

study provides empirical evidence that instead of merely adoption, assimilation of EHRs at a hospital-wide level

can improve the efficiency with which patients are treated, with benefits from such an assimilation being far

more pronounced for patients with a greater degree of health complexities. These are important findings, be-

cause hospitals are struggling to deliver quality care to their sickest patients without severe cost overruns.

Recommendations from our study point to a path forward in meeting this challenge.

1. Motivation

The importance and benefits of information technology (IT) in im-

proving the efficiency and quality of customer-facing operations has

been highlighted in previous literature (Froehle and Roth, 2004). Firms

in industries such as telecommunications, retail, etc. have seen some

benefits as a result of widespread use of IT throughout their organiza-

tions (Bower, 2005). Despite this enhancement, there is increasing

pressure to justify how such investments create value for businesses

(Zhu et al., 2006). As noted by Fichman and Kemerer (1999) (p. 256),

“A new technology may be introduced amid great enthusiasm and enjoy

widespread initial acquisition, but nevertheless still fails to be thor-

oughly deployed among many acquiring firms.”As a result, scholars

have started investigating the “innovation diffusion”aspect of IT, which

goes beyond just adoption and focuses on topics related to factors af-

fecting the usage and value of IT (Armstrong and Sambamurthy, 1999;

Fichman, 2000;Purvis et al., 2001;Chatterjee et al., 2002;Zhu and

Kraemer, 2005;Liang et al., 2007;Sodero et al., 2013). This debate on

the efficacy of IT is even more acute in healthcare, which comprises

nearly 20% of the United States' gross domestic product (GDP) (Berwick

and Hackbarth, 2012). While healthcare has adopted various informa-

tion technologies, none have received the kind of scrutiny that elec-

tronic health records (EHRs) have received. Researchers generally agree

that EHRs, if used correctly, have the potential to transform healthcare

delivery through the use of evidence-based medical guidelines and ef-

ficient coordination of patient treatment and care (Jha et al., 2009a,b;

Blumenthal and Tavenner, 2010). Despite EHRs' potential to improve

the efficiency and effectiveness of care, its adoption had been notor-

iously slow among US hospitals, with less than 10% of US hospitals

reporting a comprehensive EHR system across all clinical units in 2009

(Jha et al., 2009a,b). In order to overcome barriers and accelerate the

adoption of EHRs, the Health Information Technology for Economic

and Clinical Health (HITECH) Act was introduced in 2009 (US

Department of Health and Human Services, 2009). Under this Act, the

government committed $27 billion to incentivize hospitals and clin-

icians to not only adopt, but also meaningfully use, EHRs. Through this

HITECH Act, the government also set a high bar for healthcare provi-

ders to improve quality through the use of scientifically supported

https://doi.org/10.1016/j.jom.2018.06.003

Received 31 March 2017; Received in revised form 13 May 2018; Accepted 6 June 2018

∗

Corresponding author.

E-mail addresses: deepa.wani@utsa.edu (D. Wani), malhotra@case.edu (M. Malhotra).

Journal of Operations Management 60 (2018) 1–18

Available online 22 June 2018

0272-6963/ © 2018 Elsevier B.V. All rights reserved.

T

decision support systems and sharing of data to reduce costs. The

adoption rates for basic EHRs climbed to 80% by 2015 (Office of the

National Coordinator for Health Information Technology, 2016) due to

the enactment of HITECH. It can be viewed as a mandate on IT adoption

and assimilation, i.e., “the extent to which the use of technology dif-

fuses across the organizational projects or work processes and becomes

routinized in the activities of those projects and processes”(Purvis

et al., 2001, p. 121). Does such a mandate on the meaningful use of

these systems really improve patient outcomes, given that past studies

have not found overwhelming support from the adoption of EHRs on

outcomes? This remains an open empirical question.

We use Information Processing Theory (IPT), along with relevant

literature on IT assimilation, to investigate whether a hospital-wide IT

assimilation, as measured by the meaningful use of EHRs arising from

the passage of the HITECH Act, has improved the effectiveness with

which hospitals treat patients. By creating and contrasting three groups

of hospitals –partial-EHR adopters, full-EHR adopters, and meaningful-

EHR assimilators, we fill a critical gap in past healthcare studies which

have almost exclusively focused on the adoption of EHRs, but not its

actual assimilation. See Appendix A for more details. In addition, while

previous studies have looked at the factors that impact adoption of IT

and subsequent performance in a voluntary environment, we in-

vestigate the benefits of IT assimilation in a mandated environment.

Another major limitation of previous healthcare studies is that they use

adoption as a metric for IT assimilation, but this measure is ambiguous

on the extent of adoption and use of EHRs within a given hospital.

Furthermore, IT adoption and assimilation studies, in general, have

relied on survey data or case studies to capture this important metric

(Armstrong and Sambamurthy, 1999;Fichman, 2000;Purvis et al.,

2001;Chatterjee et al., 2002;Premkumar et al., 2005;Gattiker and

Goodhue, 2005;Zhu and Kraemer, 2005;Liang et al., 2007;Sodero

et al., 2013) In contrast, by using a government dataset that provides a

list of IT assimilators using a well-defined set of measures, we overcome

a major limitation of self-reporting bias in the data. Limitations in

healthcare studies also arise from a narrow focus on specific medical

conditions, aggregate hospital-level data, and use of clinical outcomes

(e.g. mortality, readmissions) that may require a well-established

health information technology that a majority of hospitals lack (Jha

et al., 2009a,b). By considering a wider patient population, we seek to

shed light on whether the meaningful use of EHRs is beneficial to pa-

tients.

According to IPT, a key challenge for organizations is managing

uncertainty, one important form of which can be task complexity

(Galbraith, 1973,1977;Tushman and Nadler, 1978). As task com-

plexity increases, decision makers face greater cognitive loads and may

possibly seek tradeoffs between decision accuracy and decision-making

time (Johnson and Payne, 1985;Speier et al., 2003). Unlike other in-

dustries, such as telecommunications, manufacturing, retail, and e-

commerce, where products and services can be standardized to a great

extent, every patient is unique, which puts greater challenges of pro-

viding personalized care on healthcare providers. But previous health-

care studies have also not taken into account various types of patient

complexities in their analysis. This is important to investigate, because,

in a knowledge-intensive industry like healthcare, diagnosing a pa-

tient's condition and treating it effectively is a complex task due to the

fact there are currently about 13,600 diagnoses with 6000 drugs and

4000 procedures to treat these diagnoses (The New Yorker, 2011). In

addition, the US has adopted the International Classification of Diseases

(ICD) 10 codes since 2015, which permits tracking of up to 14,400

different codes associated with various diseases, causes, and symptoms

(Centers for Medicare and Medicaid Services, 2015). Comorbidities

such as hypertension, diabetes, obesity, etc. are increasing in the United

States, and clinicians have to take these factors into consideration while

designing effective treatment plans. Given the heterogeneity and com-

plexity that healthcare providers face, does the value of meaningful-

EHR assimilation increase or decrease with increasing patient

complexity? We conduct a thorough examination of the arguments that

associate meaningful use of EHRs with improved operating efficiency.

To answer our research questions, we conducted a longitudinal

study using patient-level data from 2010 to 2013 from all acute-care

hospitals in California, and a new dataset from the Medicare EHR

Incentive Program (with data on meaningful-use hospitals available

from 2011 onwards). Our econometric model of patient length of stay

and readmission is based on multiple datasets, including detailed pa-

tient data containing important patient-level controls and patient-spe-

cific conditions which help us account for a multitude of factors that

may affect outcomes. One of the key challenges in measuring the effect

of EHR assimilation on patient outcomes is the presence of self-selection

bias, which requires us to model a hospital's decision to pursue mean-

ingful-use attestation sooner, rather than later. It is possible that hos-

pital factors associated with earlier adoption of such EHRs may play a

role in earlier attestation of meaningful use of EHRs. Financial in-

centives also progressively diminish if hospitals delay attestation for

meaningful use of EHRs. Without controlling for this endogenous se-

lection process, the impact of EHR assimilation on outcomes may be

biased. Our analysis explicitly deals with the endogeneity inherent in

self-selecting to attest for meaningful use sooner. Our model and related

analysis thus offer a new perspective on this issue that has captured the

attention of healthcare providers, policymakers, and academicians over

the last few years. More specifically, to the best of our knowledge, this

is one of the first healthcare studies to empirically use objective sec-

ondary data to show that meaningful-EHR assimilation in a mandatory

environment can transform healthcare delivery through the use of built-

in, evidence-based medical guidelines, efficient coordination of patient

treatment and care, and reduction of a health care provider's cognitive

load when working on complex tasks.

The remainder of our paper is organized as follows: In Section 2,we

discuss the problem background on the meaningful use of EHRs, fol-

lowed by a review of the relevant literature in Section 3. Our hy-

potheses are described in Section 4. Data description and the econo-

metric model used in this paper are in Sections 5 and 6, respectively.

We present our results, robustness test, and post-hoc tests in Section 7,

and conclude with a discussion of implications for research and practice

in Section 8.

2. Background on the meaningful use of electronic health records

The HITECH Act was passed in October 2010 to encourage hospitals

to not just adopt EHRs, but also use them meaningfully. Full details of

the meaningful use program can be found at Healthit.gov (2013), but

we provide a brief summary below.

The EHR initiative is rolled out in three stages. In the first stage,

which is the focus of this paper, hospitals have to “successfully attest to

demonstrating meaningful use of certified EHRs to qualify for an in-

centive payment scheme through the Medicare EHR program ad-

ministered by the Centers for Medicare and Medicaid Services (CMS)”

(Healthit.gov, 2013). Hospitals must demonstrate use of EHRs at the

hospital-wide level to receive financial incentives. This use includes

capturing patient information electronically in a standardized format,

using patient information to track key clinical conditions, integrating

test and imaging results and using decision support tools, commu-

nicating the information to all providers for the purposes of care co-

ordination, initiating reporting of key clinical quality measures, and,

finally, using the information to engage families and patients in their

care. For the successful attestation of first stage of EHRs, which mea-

sures how well EHRs have been assimilated, hospitals are also required

to maintain a current list of diagnoses, maintain active medication and

allergy lists, implement drug-drug and drug-allergy checks, record vital

statistics and demographics, enter medication orders electronically for

at least 80% of their patients, and provide electronic copies of health

records and discharge instructions for at least 50% of patients. Based on

the certification requirements, EHR depends not only on demonstrating

D. Wani, M. Malhotra Journal of Operations Management 60 (2018) 1–18

2