Jeff Helton
Jim Langabeer
Jami DelliFraine
Chiehwen Hsu

One of the best ways to measure whether an organization is realizing the benefits of its EHR investments may be to consider the extent to which the technologies promote greater labor productivity.


At a Glance

A study used FTE employees per adjusted occupied bed (FTE/AOB) as a measure to ascertain the effect of EHR investments on labor productivity. The study focused on three primary questions:
 

  • Do FTE/AOB decline as the number of EHR applications used in a hospital increases?  
  • Is impact on FTE/AOB greater with some EHR applications than with others?  
  • Do FTE/AOB decline over time, as the hospital continues to use the EHR application?  

Electronic health records (EHRs) are widely seen as being integral tools for "fixing" the U.S. healthcare system while they reduce costs and improve quality of delivering care in hospitals. Even so, hospitals have been slow to implement EHR technology, in large part because of the high cost, the hospitals' insufficient capital, and uncertain financial returns.a These factors have made it difficult for healthcare finance leaders to make the case for EHR technology investments, despite the findings of researchers in the general business world that investment in IT correlates with improved measures of organizational performance, such as cost per unit of output or overall profitability.b  

The problem is that, despite widespread assertions of the benefits of IT investments, not much is known about how these benefits are achieved. For example, improvements could derive from enhanced revenue recognition or reduced operating costs achieved through improved efficiency as measured by reduced staffing.

One way to understand how the financial benefits of EHR investments materialize is to analyze the impact of such investments on labor costs, given that these costs account for more than 50 percent of hospital operating costs.c However, such an analysis still poses challenges. Labor markets differ significantly across the United States, making it difficult to compare labor cost among U.S. hospitals. Add to this difficulty the challenge of comparing hospitals' results based on a correlation of labor costs with the amount of the investments. Not all hospitals pay the same amount for investments in technologies-even where the technologies are the same. The price can vary based on the hospital's use of bargaining position in the industry, group purchasing organizations, or other relationships with IT vendors.

Therefore, when seeking to understand how EHR investments provide ROI for hospitals, it may be best to focus less on costs and more on the extent to which these investments promote improved labor efficiency, which is also widely touted as a key benefit of investments in IT. For such an analysis, the most reliable basis for comparison may be productivity metrics, such as average FTE per unit of output.

With this premise in mind, the Fleming Center for Healthcare Management at the University of Texas School of Public Health in Houston conducted a study to examine the impact of increased use of EHR technology on labor efficiency in 2,865 general acute care hospitals across the United States. This study focused on staffing changes relative to the implementation of 13 specific EHR applications. These applications are listed in the sidebar below.

For purposes of the study, the productivity measure used was number of FTE per adjusted occupied bed (FTE/AOB). The study focused on three primary questions:

  • Do FTE/AOB decline as the number of EHR applications used in a hospital increases?
  • Is impact on FTE/AOB greater with some EHR applications than with others?
  • Do FTE/AOB decline over time, as the hospital continues to use the EHR application?

Effect of Increasing Number of EHR Applications

Our analysis found no strong correlation between the number of applications in use and average FTE/AOB. If anything, it appears that labor efficiency declines with greater numbers of EHR application in use: FTE/AOB actually were higher when more than nine of the study EHR systems were in use. This pattern is shown in the exhibit below.

Exhibit 1

f_helton_exhs1

Using a simple ordinary least squares trend analysis-where other factors noted such as ownership and case mix index could be factored in-we could find no relationship between FTE/AOB levels and the number of applications in use. The active environment of today's patient care units can create multiple interruptions to established workflows in the hospital setting that can disrupt productivity in hospitals that rely on manual health record processes. Unfortunately, based on our findings, this situation does not change with the addition of EHR technologies: After controlling for ownership, case mix, quality, location, and teaching status, we found that adding more EHR applications to automate more tasks does not seem to improve labor productivity, as measured by FTE/AOB. Indeed, other influences on staffing, such as service mix, position controls, or assessment of work processes for efficiencies, may have greater sway over improving performance on this staffing metric than simply adding more technology.

That FTE/AOB increased with the number of applications in use is actually not surprising. Hospitals that implement EHR resources without evaluating overall business practices can reasonably expect to see declines in efficiency. If underlying business processes are not aligned with the requirements of EHR technology, operating efficiency can suffer as staff perform tasks manually and then update EHR system records after the fact.d By failing to correct mismatches between technology and work processes, a hospital can delay or eliminate positive returns on EHR investments.

Relative Impact with Different EHR Applications

On analyzing FTE/AOB relative to specific EHR applications in use in each hospital, we found that only with clinical decision support, lab information, and quality/outcome management systems was FTE/AOB staffing below the average for all study hospitals by a statistically significant margin. By contrast, computerized provider order entry (CPOE) and physician documentation systems were associated with FTE/AOB staffing levels that were above average by statistically significant amounts (see the exhibit below).

Exhibit 2

f_helton_exhs2

The finding that clinical decision support systems are associated with improved efficiency seems reasonable in view of the typical hospital operation. Clinical staffs are called on to synthesize a large amount of data about a patient's current medical condition and medical history to render care safely. Addressing critical patient care issues such as food/drug interactions, drug sensitivities, and unusual diagnostic lab values can consume labor resources as organizations rely on staff experts to address questions from front-line caregivers. Availability of IT resources to assist staff in addressing such questions would logically translate to fewer labor hours used and thus greater labor efficiency.

It also makes sense that laboratory information systems are associated with greater labor efficiency, because these systems allow managers to schedule work efficiently, taking advantage of opportunities to prioritize less urgent work to times of lower activity (such as at night) when staff would otherwise be operating at lower rates of output. In addition, laboratory information systems allow hospitals to maintain quality control records that otherwise must be manually compiled and analyzed using additional labor resources.

A correlation between outcome/quality management systems and improved labor efficiency seems reasonable, as well. With these systems, hospitals can use evidence gathered from clinical research and patient outcome data to identify the best pathways for organizing patients' hospital stays. By establishing standard procedures for patient encounters, such pathways can help minimize variations in the delivery of care among patients, thereby improving efficiency. These systems can also monitor performance against internal and industry benchmarks, giving managers some stimulus for improvement.

It also is not surprising that average performance on the FTE/AOB metric tends to be poorer with use of CPOE and physician documentation systems. Researchers have pointed to evidence of a significant disconnect between the routine operation of a hospital patient care unit and use of CPOE applications.e To the extent that a hospital medical staff member is unable or unwilling to learn how to use such applications, hospital staff must devote time to helping physicians to complete order entry or care documentation tasks. Time spent by hospital employees to assist medical staff with these tasks diverts time away from other work tasks, representing lost efficiency.

Changes in Staffing Resulting from Extended Use of EHR Applications

On reviewing how long hospitals had been using each of the 13 EHR applications in their facilities, up to three years, we found that the applications in use for the longest time, on average, are case mix management, clinical data repository, clinical decision support, lab/pharmacy/radiology department information systems, and PACS. By contrast, we found that relatively few of the study hospitals had been using CPOE, physician documentation, and RFID tracking systems for as much as one, two, or three years, and most had not even yet installed these systems.

Notably, those systems with longer average longevity in hospitals-specifically, clinical decision support, outcome/quality management, and lab information systems-were those associated with lower staffing levels in terms of FTE/AOB. Although the statistical relationships were not as strong as in the prior analysis, the associations were still meaningful. Conversely, installed systems associated with poorer FTE/AOB performance-specifically, CPOE and physician documentation systems-tended to be in use for a shorter average time (1.98 years) than those with strong associations to better performance (2.62 years). Our analysis did not show any significant correlations between FTE/AOB and years in use with other EHR applications.

Exhibit 3

f_helton_exh3

Key Considerations for Finance Leaders

Based on our findings for 2,865 U.S healthcare facilities, the relationship between EHR use in hospitals and greater labor efficiency measured across all labor disciplines does not appear to be strong: Hospitals with a greater number of EHR applications in use do not appear to realize significant gains in labor efficiency as measured by FTE/AOB. However, our findings do suggest that some specific EHR applications are more effective than others in generating operational improvements, and some applications actually could reduce efficiency. Therefore, hospital finance leaders considering EHR investments should evaluate their strategic goals before selecting specific technologies, and they should set realistic goals in anticipating labor savings as a potential source for ROI.

As EHR technologies evolve, they might increasingly be redesigned in ways that contribute to greater labor efficiency, so additional analysis may be required to determine how such improvements play out. At this writing, 2008 is the latest year where all data elements required for this analysis was available. Nonetheless, our findings based on those data are likely to retain their significance today, as it is unlikely that hospitals have been able to fully implement the process changes necessary for them to reap efficiency improvements in the time since those data were published.

The key point to keep in mind is that EHR technologies can deliver value to hospitals by improving communication among operational entities, promoting more timely and accurate documentation of patient care interventions, and identifying opportunities to prevent delivery of unnecessary care, and none of these benefits can be measured with labor hours alone. Although improved labor productivity still may represent a long-term benefit from EHR investments, hospital finance leaders should perhaps look beyond this measure of ROI and consider setting goals for improvement in other performance metrics that can ultimately contribute to improved use of EHR applications.

The findings of our study should not be construed as casting EHR technology use in a negative light with regard to financial returns. Instead, they should demonstrate to financial executives the need to identify the value of other, nonlabor benefits that could help hospitals adapt to a challenging operational environment. Enhancement of quality may not yield immediate financial returns to hospitals, but there remains the promise of as-yet-unquantifiable gains from the long-term improvement in relationships within the healthcare delivery system and in the health status of local communities.

Objective evidence of the benefits to hospitals from EHR use may be mixed, but hospitals may still realize broader, as-yet-unseen benefits in the future, because interoperability of EHR systems among healthcare providers remains in its infancy.f As interoperability improvements through health information exchanges are achieved, improved efficiencies in the delivery of healthcare services may yet be observed. Also, incentive payments associated with the Physician Quality Reporting Initiative could be more easily earned through use of EHR technologies by helping physicians track data needed for required reporting. Hospital-physician affiliations would definitely have greater value through this sort of cooperative data exchange.

Although many in our industry may have assumed that adoption of EHR technology will invariably lead to improvements in staffing efficiency, our analysis to a certain extent belies that assumption-and we believe it gives hospital executives a basis to better manage operational expectations when making these significant technology investments. The significance of such findings is perhaps best summed up by Beverly Bell and Kelly Thornton in their article in the February 2011 issue of hfm ("From Promise to Reality: Achieving the Value of an EHR"): "[T]he reason for an EHR implementation is simple: It is the right thing to do for patients and is a must for the hospital's future." EHR technology may not address labor inefficiency, but it can still deliver considerable value in improving the delivery of patient care.


Jeffrey Helton, PhD, FHFMA, CMA, CFE, is assistant professor, University of Texas School of Public Health, Houston, and a member of HFMA's Texas Gulf Coast Chapter (jeffrey.r.helton@uth.tmc.edu).

Jim Langabeer, PhD, CMA, is associate professor, University of Texas School of Public Health, Houston (james.r.langabeer@uth.tmc.edu).

Jami DelliFraine, PhD, is assistant professor, University of Texas School of Public Health, Houston (jami.l.dellifraine@uth.tmc.edu).

Chiehwen (Ed) Hsu, PhD, is associate professor, University of Texas School of Public Health, Houston (chiehwen.e.hsu@uth.tmc.edu).


Footnotes

a. DesRoches, C., et al., "Electronic Health Records' Limited Successes Suggest More Targeted Uses," Health Affairs, April 2010.

b. Brynjolfsson, E., and Hitt, L., "Beyond Computation: Information Technology, Organizational Transformation and Business Performance." Journal of Economic Perspectives, Fall 2000; and Weill, P., The Relationship Between Investment in Information Technology and Firm Performance: A Study of the Valve Manufacturing Sector, Information Systems Research, 1992.

c. Kazahaya, G., "Harnessing Technology to Redesign Labor Cost Management Reports," hfm, April 2005.

d. Ash, J., Berg, M., and Coiera, E., "Some Unintended Consequences of Information Technology in Health Care: The Nature of Patient Care Information System-related Errors," Journal of the American Medical Informatics Association, March-April 2004.

e. Koppel, R., et al., "Role of Computerized Physician Order Entry Systems in Facilitating Medication Errors," JAMA, March 9, 2005.

f. Jha,A.,et al.,"Use of Electronic Health Records in US Hospitals," New England Journal of Medicine, April 16, 2009.


Sidebar

About the Study  

This analysis took an approach that is different from others currently available in the industry. Instead of measuring the dollars invested in technology, the study sought to eliminate the influence of any variances in EHR pricing among hospitals by focusing on the types of EHR technologies hospitals are using based on data from an annual study conducted by the Dorenfest Institute and the Health Information and Management Systems Society (HIMSS). In their 2009 survey, Dorenfest and HIMSS list 13 core applications commonly identified as EHR technologies:

  • Case mix management systems
  • Clinical data repositories
  • Clinical decision support systems
  • Computerized provider order entry (CPOE)
  • Electronic medication administration records
  • Laboratory information systems
  • Nursing/clinical documentation systems
  • Outcome/quality management systems
  • Pharmacy information systems
  • Physician documentation systems
  • Picture archive and communication systems (PACSs)
  • Radio Frequency Identification (RFID) patient tracking
  • Radiology information systems

Efficiency in this study was measured by FTE employees per adjusted occupied bed (FTE/AOB) calculated from data in filed Medicare cost reports for that same year. Total nonphysician paid hours in a hospital (as noted on Worksheet S-3, Part III) were used to obtain the total paid FTE for each hospital. Patient days were obtained from Worksheet S-3 Part II of the cost report and were adjusted for the proportion of total patient revenues to inpatient revenues noted on Worksheet G-2 of the same cost report. The FTE/AOB value was calculated for 2,865 facilities where HIMSS data and cost report data could be matched based on the hospital's Medicare provider number for fiscal years ending in 2008-the latest year for which complete data were available.

Other influences on observed FTE/AOB performance exist in hospitals, and those factors were controlled for in the analysis. The other factors considered in the analysis included:

  • Ownership (for-profit, not-for-profit, government)
  • Teaching status
  • Rural location
  • Multihospital system membership
  • Case mix index
  • Quality of care (measured by Agency for Healthcare Research and Quality [AHRQ] patient safety indicators [PSIs])
  • Minimum nurse-to-patient staffing ratio (as in California)

Case mix index and PSI data were obtained from the Centers for Medicare & Medicaid Services website. Values for the other variables considered here were obtained from Worksheets S-2 and S-3 of the cost reports filed for each hospital in the analysis.
 

Publication Date: Wednesday, February 01, 2012

Login Required

If you are an existing member, please log in below. Username and password are required.

Username:

Password:

Forgot User Name?
Forgot Password?

If you are not an HFMA member and would like to access portions of our content for 30 days, please fill out the following.

First Name:

Last Name:

Email:

   Become an HFMA member instead