Longitudinal changes in the operating efficiency of public safety-net hospitals.
Article Type: Report
Subject: Strategic planning (Business) (Analysis)
Trauma centers (Management)
Authors: Helton, Jeffrey R.
Langabeer, James R., II
Pub Date: 05/01/2012
Publication: Name: Journal of Healthcare Management Publisher: American College of Healthcare Executives Audience: Trade Format: Magazine/Journal Subject: Business; Health care industry Copyright: COPYRIGHT 2012 American College of Healthcare Executives ISSN: 1096-9012
Issue: Date: May-June, 2012 Source Volume: 57 Source Issue: 3
Topic: Event Code: 200 Management dynamics Canadian Subject Form: Trauma centres Computer Subject: Company business management
Geographic: Geographic Scope: United States Geographic Code: 1USA United States
Accession Number: 292009074
Full Text: EXECUTIVE SUMMARY

Government-operated trauma facilities fill an important role as safety nets in our health system, providing care to millions of individuals who lack health insurance. Because these hospitals are often the most financially constrained, continuous improvement in operating efficiency seems to be a necessary component of their organizational strategy. In this study, we analyze the longitudinal changes in efficiency of a large sample of government-operated safety-net hospitals from 2004 to 2008. Employing an analytical tool called data envelopment analysis, our findings suggest that as a group these hospitals have become more efficient over time, improving by 2.1 percent over the five-year study period.

INTRODUCTION

Approximately 46 million people younger than age 65, or 25 percent of the LIS population in that age group, do not have continuous access to health insurance coverage (Hsieh, Clement, and Bazzoli 2010). The number of uninsured and medically indigent persons has been growing annually in recent years (Andrews et al. 2007). While current health reform efforts may reduce this number by as much as half, a substantial patient population will remain without resources to pay for needed healthcare services (KFF 2010; Ringel et al. 2010). Generally, care for the indigent and uninsured is provided by hospitals under the Emergency Medical Treatment and Active Labor Act (EMTALA) mandate requiring hospitals to treat patients with acute needs without consideration of their ability to pay (Cunningham 2008). Some hospitals provide a large proportion of their services to persons who do not have resources to pay for care. These hospitals are commonly referred to as safety-net hospitals (Zwanziger and Khan 2008).

Many of the facilities making up the nation's safety net are trauma centers operated by local government entities.

Providing care to persons who lack resources to pay for such care places a financial burden on safety-net organizations, even those that are government operated and receive an external subsidy to fund this community benefit mission (Meyer et al. 1999). The pressures of providing care to the increasing ranks of those who cannot pay combined with reimbursement reductions should spur greater efficiency in provider organizations, especially safety-net trauma facilities (Hsieh, Clement, and Bazzoli 2010).

The extent to which the level of efficiency has changed in government-operated safety-net hospitals across the United States in recent years is a question that, if answered, could assist administrators and policymakers in identifying opportunities to improve. This study evaluated changes in the operational efficiency of the nation's government-operated trauma centers during the years 2004-2008. It was conducted using a sophisticated data benchmarking technique known as data envelopment analysis (DEA). The primary contribution of this exploratory study is that it is the first to analyze longitudinal trends in the efficiency of government-operated safety-net hospitals.

BACKGROUND ON SAFETY NETS AND OPERATING EFFICIENCY

The prevailing employer-based health insurance model is in a state of decline as increasing numbers of businesses cease to offer health coverage as a fringe benefit (Cunningham 2008). At the same time, government programs such as Medicaid are facing difficulty maintaining funding levels; as a result, they must institute tightened eligibility criteria. These changes are limiting or reducing the number of persons on public health insurance programs, further increasing the population of uninsured residents (Fronstin 2005). Enactment of the Patient Protection and Affordable Care Act is expected to reduce the number of uninsured by as much as half after 2014 (KFF 2010; Ringel et al. 2010). However, with many remaining uninsured people continuing to seek uncompensated hospital care beyond that date, the policy issue of efficiency in delivering care by safety-net hospitals is one that cannot be ignored.

Lack of access to care has been discussed as a barrier to the uninsured population using mainstream community providers (e.g., primary care physicians). Greenwald, Keefe, and DiCamillo (2004) note that the majority of the uninsured receive care sporadically or in inappropriate settings--such as hospital emergency departments--because they lack insurance and thus have limited access. Often, the hospitals where services are sought are government-operated trauma centers serving as safety-net providers.

The Institute of Medicine (2000) first raised the question of the risk of financial viability of government-operated trauma centers when it noted that adoption of Medicaid managed care contracts might adversely affect providers serving not only large proportions of Medicaid patients but the uninsured as well. In its definition of facilities at risk for financial harm from such changes, it said that the greatest risk was posed to the "core safety-net providers." These providers were characterized as hospitals that "either by legal mandate or explicitly adopted mission, offer care to patients regardless of their ability to pay for services" and "a substantial share of their patient mix are uninsured, Medicaid, and other vulnerable patients." Prominent among the providers included in that analysis were publicly funded trauma centers. At the same time, local government entities struggle to balance funding the healthcare safety net for citizens, other service priorities, and pressures by constituents to minimize community tax burdens (Rollins 2004). Recent declines in the national economy exacerbate the challenge of funding other needed public services such as education and police and fire protection from static or decreasing tax revenues.

Maintaining and improving efficiency is a significant political challenge to the indigent care delivery system. An Urban Institute study (Meyer et al. 1999) indicates that tax revenues to fund indigent healthcare are provided with the mandate that care will be given efficiently, making the greatest use of all resources provided for indigent care. Bovbjerg and Ullman (2001) note the challenge to providers to control costs in order to either maintain market competitiveness or operate within available funding. Logic dictates that hospitals facing static or decreasing reimbursements and increasing demand for services from the indigent and uninsured seek to improve efficiency. What is unknown is the extent to which facilities that receive governmental subsidies were insulated from market incentives to improve efficiency.

Langabeer (2008) broadly defines efficiency as "performing tasks with minimal waste and resource consumption." In the context of a hospital operation, common inputs include available patient beds, full-time equivalent employees (FTEs), direct operating expenses, value of supplies, and capital investment. Common output measures are patient days, admissions, discharges, outpatient visits, emergency department or clinic visits, surgeries, or some measure that combines outpatient and inpatient encounters such as an adjusted patient day (Chang and Troyer 2009). The challenge to hospital managers is to achieve and maintain an optimal balance of inputs that maximize output while holding quality per unit of output constant.

Hospitals can take one of two actions to improve efficiency when resources are constrained: reduce capacity or reduce direct inputs while seeking to increase outputs (Langabeer 2008). In the hospital setting, input change could be evidenced by reduction of inpatient bed capacity or available emergency treatment rooms, reductions in staff, negotiation of lower supply costs, or implementation of simplified work processes. In settings where input reductions were implemented but outputs remained static or increased, efficiency increased. Measurements of efficiency are meaningful when compared to a benchmark such as the same measure calculated for other organizations or for the same organization in a prior time frame. A common technique to benchmark operational efficiency is DEA.

Efficiency Measurement Using DEA

DEA is a nonparametric, deterministic technique that uses a linear programming model to evaluate multiple input and output measures and determine a relative efficiency measure for comparisons among entities (called decision-making units, or DMUs). As a non-parametric technique, DEA affords a simpler approach to analyzing the complexity of relationships between inputs such as beds or FFEs in a hospital and the outputs they create. Compared with other benchmarking techniques such as ratio analysis, linear regression, total factor productivity, or stochastic frontier analysis, DEA is more amenable to an analysis considering multiple inputs and multiple outputs (Ozcan 2008). An extension of DEA is the Malmquist productivity index, which uses a linear algebra technique to assess DEA efficiency measures across time periods, allowing for longitudinal analysis.

Operational efficiency assessments using DEA have been widely documented in the management literature since introduction of the technique by Charnes, Cooper, and Rhodes in 1978 (Ozcan 2008). Among the DEA applications noted were studies of efficiency in police departments (Barros 2006), e-commerce (1o Storto 2009), brewing (Day, Lewin, and Li 1995), airlines (Greer 2009), sawmills (Helvoight and Grosskopf 2005), agriculture (Constantin, Martin, and Rivera 2009), public transit (Karlaftis 2003), and manufacturing (Golany, Hackman, and Passy 2006). While these studies represent only a brief cross-section of the literature on DEA, they support the broad applicability of DEA to efficiency measurement.

A variety of studies in healthcare organizations used DEA to develop performance benchmarks and compare relative efficiency in the industry. These studies used some metric of capacity quantified by beds licensed for operation as a proxy for capacity or capital investment. Operating inputs such as labor and nonlabor components vary among studies, but some form of a noncapital input was consistently noted in the literature. Examples include FFEs, labor expenses, and nonlabor expenses. Outputs have been quantified as patient days, discharges, or outpatient visits. Following are a few examples of DEA use in the healthcare literature.

Harrison and Sexton (2006) used DEA to examine the comparative efficiency of church-operated hospitals across a four-year period to determine the impact of nonprofit tax subsidies on efficiency. Langabeer and Ozcan (2009) used the technique in a longitudinal evaluation of the relative efficiency of cancer specialty hospitals. Sikka, Luke, and Ozcan (2009) used DEA to assess efficiency in hospital clusters.

When Hsieh, Clement, and Bazzoli (2010) used it in a study, they noted that a safety-net hospital in the local market was a favorable influence on the efficiency of other hospitals in a local market, positing that the safety-net facilities in a market took on the most complicated cases and freed up other facilities to operate more efficiently. Safety-net facilities thus seem to be an important part of the healthcare delivery system.

Zwanziger and Khan (2008) characterized a large proportion of hospitals serving in a safety net role as trauma facilities operated by local governments. While that study acknowledges other facilities served a safety net role, it notes they did so to the extent mandated by EMTALA and not as a core mission. The authors raised the question of whether public facilities may need to adopt a "no margin, no mission" approach, including an increased focus on improved operational efficiency. However, no measure of the extent to which such approaches were operationalized among government-operated safety-net facilities has been noted. That assessment is the primary goal of this exploratory work.

Research Question and Hypotheses

Tax subsidies to government-operated hospitals serving many of the uninsured may be eroding due to constriction in the broader economy. The result appears to be a challenge to these safety-net providers to render greater levels of care with limited financial resources (Hernandez et al. 2009). One logical response is to improve efficiency in producing healthcare outputs by public safety-net facilities. Thus, this exploratory study seeks to address the specific question, are government-operated safety-net hospitals becoming more or less efficient over time?

Given the environmental challenges noted here for public trauma centers serving as safety-net hospitals, the hypothesis to be evaluated by this study is as follows:

[H.sub.0]--Government safety-net facilities have become more efficient over time.

A Malmquist DEA technique was used in an input-oriented model to differentiate high- and low-efficiency performance among public safety-net facilities.

METHODS

Prior studies using the Malmquist DEA technique included measures of capital, labor, and nonlabor resources as inputs to the production of inpatient and outpatient care outputs. The other studies evaluated here generally used licensed beds as a proxy for capital investment and available production capacity in the absence of greater balance sheet detail of capital asset values in publicly available data sources. Labor inputs were measured with FFEs, and nonsalary operating expenses were used as an index of nonlabor inputs. Outputs evaluated in the prior studies included patient days and outpatient visits. That framework is followed here.

The American Hospital Association (AHA) Annual Survey Database was the source used to select facilities included in this study and to learn their bed capacity and outpatient visit totals for the years 2004-2008. The Centers for Medicare & Medicaid Services Hospital Cost Report Information System (HCRIS) database was also used as a data source for the study. Financial measures were taken from the HCRIS database because that source consists of audited financial data. Financial data submitted in the AHA annual survey is not subject to audit and so was considered less preferable for this work. FI'Es, nonsalary expense, and discharge values were obtained for study hospitals from the HCRIS database for each hospital for each year in the study. Variables used and their sources are summarized in Exhibit 1.

Malmquist DEA models reviewed in the literature appeared to take a constant-returns-to-scale, input-oriented approach, having operated under the assumption that management is able to influence the allocation of inputs to the production of care and that returns to scale do not vary across a relevant range of production levels. That approach is consistent with work by Langabeer and Ozcan (2009) and was used in this analysis. DEA models typically are limited to calculations over a single period. Through use of the Malmquist index, a series of DEA efficiency scores can be compared across periods. This technique requires the user to calculate a value for the DEA efficiency in each of two periods and the efficiency frontier for each period. The efficiency frontier across periods is expressed mathematically as a product of both productivity and technological changes, as shown in Exhibit 2.

The efficiency change making up the left-hand side of the equation in Exhibit 2 measures the extent to which the units of input per unit of output have changed relative to the benchmark of all DMUs in the calculation (called the production frontier). A value greater than 1 for this component indicates an improvement in efficiency, a value less than 1 indicates lost efficiency, and a value equal to 1 indicates no efficiency change across the period. The technological change component on the right side of the formula evaluates the change in the production frontier (or a measure of best practice technology), with values interpreted in the same manner as those for efficiency (Chilingerian and Sherman 2004; Constantin, Martin, and Rivera 2009; Jacobs, Smith, and Street 2006). Given the null hypothesis in this study of improved efficiency in government-operated hospitals, the expected result of the Malmquist index when comparing 2004 with later years should be a value greater than 1.

The list of facilities operating as a Level I, II, or III trauma center during the years 2004 through 2008 as categorized in the AHA Annual Survey Database yielded 456 hospitals for evaluation in the study. The hospitals selected for analysis were segregated on the basis of ownership into two groups: A study population was made up of 101 trauma facilities operated by local government entities, and a control population was derived consisting of 355 privately operated trauma facilities.

Using a commonly available software package, DEAFrontier, a Malmquist index was calculated to determine the longitudinal change in efficiency from 2004 to 2008 for both the government and nongovernment facility groups. The results were compared using a t-test to identify the significance of any differences observed.

RESULTS

The government-operated safety net facilities in the United States as a group appear to have become slightly more efficient over the five-year period of 2004-2008, with a mean calculated Malmquist index value of 1.021 (interpreted as a 2.1 percent improvement in efficiency over this period). At the same time, safety net facilities operated by nongovernment entities achieved a similar mean improvement of 2.8 percent. The variance between the two groups was not significant to a t-test evaluation (p = 0.73). These findings are summarized in Exhibit 2.

Recalling that the Malmquist index value is made up of an efficiency component and a technical component poses an interesting view of the findings in Exhibit 3. While both groups achieved relatively similar improvements in overall efficiency, the means of achieving those outcomes were very different. Government-owned facilities improved efficiency in terms of resource use while the technical component slipped modestly. Thus, government facilities as a group appeared to reduce the gross number of inputs per unit of output significantly while allowing their technological frontier to decline. Conversely, nongovernment facilities were less aggressive in managing gross inputs and significantly advanced their technological frontier. T-tests of variances between the two groups for efficiency and technical components were significant (p < 0.0005).

Across both groups, an approximate 55/45 proportion of facilities improving/not improving efficiency was noted. A graphical depiction of the distribution of Malmquist index values is presented in Exhibit 4, and a tabular display of efficiency and technical components is shown in Exhibit 5.

Changes in the component ratios used to derive these results are particularly illustrative, as facilities demonstrating improved efficiency over the study period exhibited improvements in operating expenses (inflation adjusted, using Bureau of Labor Statistics price indices from 2009), FTE staffing, and discharges per bed. These observations are summarized in Exhibit 6.

Because patient acuity affects resource intensity, we further analyzed whether case mix index (CMI) had any impact on our findings. Government and nongovernment facilities were stratified into two groups (those that gained versus lost efficiency), and two-tailed t-tests were used to compare the average CMI for each group. No significant difference was seen in CMI between government- and nongovernment-operated hospitals across the study period (p = 0.5269). A pairwise correlation was performed among the group of government-operated hospitals with improved efficiency where the Malmquist index value was correlated with the case mix, and this correlation was found to be nonsignificant (p = 0.1760) as well. Nongovernment hospitals with improved productivity were noted to have nonsignificant relationships with case mix (p = 0.8331), and facilities with reduced productivity also showed no CMI impact (p = 0.3036 for government and 0.3099 for nongovernment). Therefore, we conclude that changes in the acuity of patients as measured by CMI could not explain the changes in efficiency noted here.

[GRAPHIC 4 OMITTED]

DISCUSSION

Our analysis shows that government-operated safety-net hospitals experienced an overall improvement in efficiency from 2004 to 2008 in the magnitude of ~2 percent. More hospitals improved efficiency than lost efficiency (54 percent to 46 percent). Similar findings were noted in the group of nongovernment-operated facilities. Therefore, government-operated safety-net facilities appear to have responded to current market challenges by improving efficiency in a manner similar to privately operated organizations. Considering the decline in tax revenues that could subsidize operations in government-operated organizations, this finding seems positive from the policymaking perspective.

Bed size appeared to be a factor in efficiency between the groups of facilities. The mean bed size of government facilities with improved efficiency was 235, while that for facilities with reduced efficiency was 266, a statistically significant difference (p = 0.0255). Similarly, smaller bed size was noted in facilities operated by nongovernment entities; the average bed size associated with improvement was 343, versus 363 for those with efficiency declines (p = 0.0014). Hence, smaller organizations appeared to be more able to react positively to environmental changes during this time frame. Productivity of bed capacity (measured as discharges/bed) was important, as government-operated hospitals increased an average of 11 discharges per bed while those with declined productivity showed no change in this measure. Similar results were observed in the nongovernment group of hospitals.

The management of labor appears to be an important factor in the efficiency changes observed here. The results presented in Exhibit 4 show that facilities with efficiency increases had improvements in FTE measures of slightly greater magnitude than those for nonlabor operating expenses per discharge and nonlabor operating expenses per outpatient visit (inflation adjusted). Conversely, the facilities with efficiency declines showed increases in FTE-based metrics that were slightly larger in magnitude than those for the nonlabor expense metrics. Government-operated facilities seemed to take a broader approach to improving relative productivity, as the magnitude of change in nonlabor expense and FTE metrics were narrowly distributed, while nongovernment facilities appeared to apply proportionally greater reductions in staffing than in nonlabor expense.

Government-operated safety-net facilities seemed to be able to achieve improvements in relative efficiency comparable to those facilities operated by private entities, thus supporting acceptance of the study hypothesis. Considering possible erosion of tax subsidies to government facilities, the apparent response of managers in such organizations to the incentive to act in a more market-based fashion seems both effective and appropriate. Going forward, policymakers should seek to thoroughly understand the operational need for and impact of any bed capacity expansion, given the association between larger bed capacities and the relatively lower efficiency noted here. Also, managers in government-operated facilities may find additional efficiencies through closer assessment of existing staffing levels for possible adjustment, as the experience of nongovernment facilities has shown here.

LIMITATIONS AND FUTURE RESEARCH

Future research may seek to expand on some limitations of the present study. Inputs used in the DEA were limited to three general components validated from similar studies. However, these variables may not best explain the level of efficiency in this unique set of hospital facilities, and the conclusions here may be explored in future work that considers other operational inputs, such as physician medical staff, location, extent of nonoperating external subsidies, or service mix. Also helpful in future work would be to operationalize an objective measure of quality of care to better normalize hospital outputs. Finally, some measure of the financial impact of the large volumes of uninsured patients may provide greater understanding regarding the need for efficiency improvement in this unique subset of hospitals.

CONCLUSION

Despite the advent of health reforms that seek to expand access to health insurance in the United States, safety-net hospitals in the country face increasing market pressures to improve operational efficiency. As the number of uninsured persons is expected to remain significant after full enactment of the Patient Protection and Affordable Care Act of 2010, the demand on these facilities to serve this population will remain as well. Government-operated safety-net hospitals have achieved efficiency improvements from 2004 to 2008 of approximately 2 percent. These facilities were able to demonstrate relative efficiency gains comparable to privately operated safety net providers by managing both FTE inputs and nonlabor operating expenses while also improving use of available operating capacity. Future opportunities for managers and policymakers rest in the close scrutiny of available bed capacity and FTE staffing, as lower levels in both appear to be associated with greater relative efficiency. Generally, these results should be encouraging as efforts to address the need for hospital care for the nation's uninsured residents continues to evolve.

REFERENCES

Andrews, R., D. Stull, B. Friedman, and R. Houchens. 2007. Serving the Uninsured: Safety-Net Hospitals, 2003. Rockville, MD: Agency for Healthcare Research and Quality.

Barros, C. 2006. "The City and the Police Force: Analysing Relative Efficiency in City Police Precincts with Data Envelopment Analysis." International Journal of Police Science & Management 9 (2): 164-82.

Bovbjerg, R., and F. Ullman. 2001. "Health Insurance and Health Access: Reengineering Local Safety Nets." Journal of Legal Medicine 22: 247-62.

Bureau of Labor Statistics. 2009. Producer Price Index Detailed Report for December 2008. Washington, DC: US Department of Labor, Bureau of Labor Statistics.

Chang, C., and J. Troyer. 2009. "The Impact of TennCare on Hospital Efficiency." Health Care Management Science 12: 201-16.

Chilingerian, J., and H. Sherman. 2004. "Health Care Applications: From Hospitals to Physicians, from Productive Efficiency to Quality Frontiers." In Handbooh on Data Envelopment Analysis, edited by W. Cooper, L. Seiford, and J. Zhu, 481-538. Boston: Kluwer Academic.

Constantin, P., D. Martin, and E. Rivera. 2009. "Cobb-Douglas, Translog Stochastic Production Function and Data Envelopment Analysis in Total Factor Productivity in Brazilian Agribusiness." Journal of Operations and Supply Chain Management 2 (2): 20-34.

Cunningham, P. 2008. "The Healthcare Safety Net: What Is It, What Good Does It Do, and Will It Still Be There When We Need It?" Harvard Health Policy Review, 5-15.

Day, D., A. Lewin, and H. Li. 1995. "Strategic Leaders or Strategic Groups: A Longitudinal Data Envelopment Analysis of the U.S. Brewing Industry." European Journal of Operational Research 80:619-38.

Fronstin, P. 2005. "Sources of Health Insurance and Characteristics of the Uninsured: Analysis of the March 2005 Current Population Survey." Employee Benefit Research Institute. Published November 2005. www.ebri.org/ pdf/briefspdf/EBRI_IB_11-20051.pdf.

Golany, B., S. Hackman, and U. Passy. 2006. "An Efficiency Measurement Framework for Multi-stage Production Systems." Annals of Operations Research 145: 51-68.

Greenwald, H., S. Keefe, and M. DiCamillo. 2004. "The Importance of Public Sector Health Care in an Underserved Population." Journal of Health and Human Services Administration 27 (2): 142-57.

Greet, M. 2009. "Is It the Labor Unions' Fault? Dissecting the Causes of the Impaired Technical Efficiencies of the Legacy Carriers in the United States." Transportation Research 43: 779-89.

Harrison, J., and C. Sexton. 2006. "The Improving Efficiency Frontier of Religious Not-for-Profit Hospitals." Hospital Topics 84 (1): 2-10.

Helvoight, T., and S. Grosskopf. 2005. "Productivity Growth, Technical Efficiency, and Returns to Scale in the Washington State Sawmill Industry." International Journal of Information Technology & Decision Making 4 (3): 477-90.

Hernandez, G., L. Fomos, V. Mika, K. Urbansky, and R. Villareal. 2009. "One Regional Health System's Innovative Steps to Deal with the Uninsured." Journal of Healthcare Finance 36 (1): 70-84.

Hsieh, H., D. Clement, and G. Bazzoli. 2010. "Impacts of Market and Organizational Characteristics on Hospital Efficiency and Uncompensated Care." Health Care Management Review 35 (1): 77-87.

Institute of Medicine. 2000. America's Health Care Safety Net: Intact but Endangered. Washington, DC: National Academies Press.

Jacobs, R., P. Smith, and A. Street. 2006. Measuring Efficiency in Health Care. New York: Cambridge University Press.

Karlaftis, M. 2003. "Investigating Transit Production and Performance: A Programming Approach." Transportation Research 37: 225-40.

Kaiser Family Foundation (KFF). 2010. "The Uninsured: A Primer." Kaiser Family Foundation. Accessed February 14. www.kff.org/ uninsured/7451.cfm.

Langabeer, 1. 2008. Healthcare Operations Management. Sudbury, MA: Jones and Bartlett.

Langabeer, J., and Y. Ozcan. 2009. "The Economics of Cancer Care: Longitudinal Changes in Provider Efficiency." Health Care Management Science 12: 192-200.

Lo Storto, C. 2009. "A Distributed Cognition Framework to Compare E-commerce Websites Using Data Envelopment Analysis." Proceedings of World Academy of Science 40: 71-79.

Meyer, l., M. Legnini, E. Fatula, and L. Stepnick. 1999. The Role of Local Governments in Financing Safety Net Hospitals: Houston, Oakland, and Miami. Washington, DC: Urban Institute.

Ozcan, Y. 2008. Health Care Benchmarking and Performance Evaluation. New York: Springer Science + Business Media.

Ringel, J., F. Girosi, A. Cordova, C. Price, and E. McGlynn. 2010. "Analysis of the Patient Protection and Affordable Care Act (H.R. 3590)." Rand. Accessed January 17, 2011. www.rand.org/pubs/research_briefs/ RB9514.html.

Rollins, G. 2004. "Taxing Matters: Gaining Public Support for the Safety Net." Safety Net 18 (1): 11-13.

Sikka, V., R. Luke, and Y. Ozcan. 2009. "The Efficiency of Hospital-Based Clusters: Evaluating System Performance Using Data Envelopment Analysis." Health Care Management Review 34 (3): 251-61.

Zwanziger, J., and N. Khan. 2008. "Safety-Net Hospitals." Medical Care Research and Review 65 (4): 478-95.

PRACTITIONER APPLICATION

Max Ludehe, MS, FACHE, CEO, North Texas Community Hospital, Bridgeport

Measuring efficiency remains a critical task for today's healthcare manager, especially in that group of hospitals known as the safety net. Hospitals continue to be challenged to provide at least medically necessary emergency care under the Emergency Medical Treatment and Active Labor Act for a growing population of uninsured and underinsured patients. Yet at the same time, revenues to support hospital missions are being reduced, threatening the very survival of some organizations. Efficient operations are a must for hospitals in the role as a safety-net provider in the healthcare market.

Performance comparisons among organizations have been common in hospital management for years as a way to differentiate acceptable performance from a performance with opportunity for improvement. However, much of that performance benchmarking has been based on a subjective comparison of common performance metrics against industry "norms." As many senior executives have seen in discussions with our management teams, the simple comparison of performance metrics often shows just how different one hospital is from another. Benchmarking in the customary manner is limited in its power to help managers identify real performance deficits.

The study's authors have used a much more sophisticated technique to create performance benchmarks among a similar group of hospitals. The data envelopment analysis technique used in the study ranks hospital performance on common metrics of full-time equivalent employees (FFEs) per discharge, FFEs per outpatient visit, operating expense per outpatient visit, and operating expense per discharge. In addition, their analytical technique creates a meaningful index that combines the performance ranking for each hospital on each metric into an overall composite efficiency score. The comparisons among hospitals on overall efficiency are therefore meaningful to management.

For years hospitals trying to evaluate efficiency have focused on staffing as a key metric to differentiate efficient operations from inefficient operations. While the authors show that staffing remains an important element in an efficient hospital operation, the nonlabor components of the operation are also important. Within safety-net hospitals, it appears that objective evaluation of available bed capacity is equally critical to efficient operations. Helton and Langabeer's article should serve as a reminder to management practitioners that studying all components of the organization's operation is critical to achieving efficient operation.

As managers in today's healthcare market, we have a vested interest (even with pending healthcare reforms) in the survival of the healthcare safety net. Its place in the overall healthcare delivery system is vital to meet a need for the many patients in our communities without resources to pay for care. The observation that efficiency improved in that segment of the industry is encouraging. At the same time, we can learn a great deal from the experiences of these hospitals in managing efficiency over time. The authors have identified some important areas of concern for healthcare organization managers with respect to efficiency.

Jeffrey R. Helton, PhD, University of Texas School of Public Health, Houston, and James R. Langabeer II, PhD, Fleming Center for Healthcare Management, University of Texas School of Public Health

For more information on the concepts in this article, please contact Dr. Helton at jeffrey.r.helton@uth.tmc.edu.
EXHIBIT 2

Efficiency Frontier Across Periods

[M.sub.0] = Period - 1 / Period - 2 x [[Period - 2 /
Period - 1_on_Period - 2 x Period - 2_on_Period - 1 /
Period - 2].sup.1/2]

Efficiency change               Technological change


EXHIBIT 1
Variables and Sources

Variable                      Input/Output   Source

Licensed Beds                 Input          AHA Annual Survey Database
FTEs                          Input          HCRIS database
Nonlabor Operating Expenses   Input          HCRIS database
Discharges                    Output         HCRIS database
Outpatient Visits             Output         AHA Annual Survey Database

EXHIBIT 3
Summary of Malmquist Index Component Values

                                     Mean        Mean        Mean
                                  Efficiency   Technical   Malmquist
Group                               Change      Change     Index(Mo)

Government-Operated Facilities      1.025        0.998       1.021
(N = 101)

Nongovernment-Operated Facilities   0.824        1.260       1.028
(N = 355)

EXHIBIT 5
Summary of Malmquist Index Component Values

                      Efficiency Increased

                                                              Mean
                         No.         Mean        Mean      Malmquist-
                      (% total)   Efficiency   Technical      Index
       Group                        Change      Change     ([M.sub.O])

Government-              55
Operated Facilities     (54%)       1.119        1.023        1.139
Nongovernment-           199
Operated Facilities     (56%)       0.899        1.262        1.125

                      Efficiency Decreased

                                                              Mean
                         No.         Mean        Mean       Malmquist
                      (% total)   Efficiency   Technical   Index (M.)
       Group                        Change      Change     ([M.sub.O])

Government-              46
Operated Facilities     (46%)       0.912        0.968        0.880
Nongovernment-           156
Operated Facilities     (44%)       0.727        1.257        0.904

EXHIBIT 6
Summary of Inputs per Output and Efficiency Measures

                               % [DELTA]   % [DELTA]
                Improved/      Nonlabor    Non-Labor   % [DELTA]
                Declined       Expense/    Expense/      FTE/
Group           Productivity   Discharge   OP Visit    Discharge

Government      Improved         -7.4        -5.1        -7.2
                Declined          7.8        14.6         7.9
Nongovernment   Improved         -1.2        -7.5       -13.0
                Declined          8.0         8.0         9.6

                  % [DELTA]      % [DELTA]
                FTE/Oupatient   Discharges/
Group               Visit           Bed

Government          -9.4           15.1
                    14.7            0.0
Nongovernment       -7.0           12.9
                    10.3            0.0
Gale Copyright: Copyright 2012 Gale, Cengage Learning. All rights reserved.