The "new economics" of clinical quality improvement: the case of community-acquired pneumonia.

EXECUTIVE SUMMARY

Hospitals and health systems have developed substantial infrastructure, at significant expense, to improve care quality and support the collection and distribution of quality metrics. Yet providers often have little understanding of what return, if any, they have earned on the investment because they typically view quality improvement efforts simply as a cost of doing business. After analyzing data from 10,512 patients with community-acquired pneumonia, we found that better performance on two quality measures was associated with shorter length of stay and improved financial performance. For example, a one-day decrease in the time until patients were shifted from intravenous to oral antibiotics was associated with a 0.8-day reduction in length of stay and a nearly 60 percent increase in margins. Providers can adapt the methods we used to derive these findings to identify other quality metrics that simultaneously increase care quality and generate economic value. To derive maximum clinical and financial benefit, however, providers must ensure that clinical quality staff members are adequately supported and skilled to set priorities and to implement effective initiatives.

**********

Ensuring the quality of care delivered to sick and injured patients is central to the mission of all hospitals and health systems. Yet providers waste millions of dollars annually on efforts to improve clinical quality. Too often, the investments yield neither benefits to patients (Dedier et al. 2001; Saint et al. 2003; Stone et al. 2005) nor proven economic return (HFMA 2005; Leatherman et al. 2003). It need not be this way. When properly focused, efforts to improve care quality can both enhance patient outcomes and generate economic value for providers. When such efforts are bolstered with strong senior executive support and an execution-oriented skill set, they can give providers a sustainable competitive advantage--both clinically and financially.

Obtaining these results is not easy. The process requires providers to think more strategically about their efforts to improve quality--that is, where they should focus and which skills are necessary to enable performance improvement. However, given the centrality of quality to organizational mission and the tight margins most providers face, we believe the journey is well worth undertaking. In this article, we demonstrate how providers' investments in quality, when properly focused and executed, can simultaneously improve patient care and generate attractive financial returns. Although our examples focus on community-acquired pneumonia (CAP), our recommendations could be applied in a variety of hospital settings. Successful implementation, however, requires that providers have appropriate metric selection and solid frontline skills and resources in place to sustain improvement.

TODAY'S APPROACH TO QUALITY

Hospitals and health systems have developed--at significant and increasing expense--intricate processes and substantial infrastructure to improve the quality of care and support the collection, verification, and distribution of information related to quality of care. Why do providers invest in these efforts if they are not producing direct financial benefit and if the evidence for clinical benefit is not always clear? In addition to wanting to ensure care quality, today's providers must respond to political and regulatory pressures as well as to the growing demands of increasingly active consumers. The Joint Commission (2006), for example, requires providers to report compliance with an evolving set of core measures and publishes the results on its website. Several other groups also publicly rate providers (IPRO 2006; HHS 2006; U.S. News & World Report 2005). Increasingly, many providers are viewing pay-for-performance programs as an important, albeit indirect, source of value (Safavi 2006).

These forces have combined to make quality improvement (QI) (1) a top priority for providers. Yet several factors currently impede QI efforts, including a lack of executive leadership, a lack of organized focus on concomitant financial outcomes, discrepancies between who must bear the cost of certain treatments and who benefits (financially or otherwise) from those treatments, and inappropriate frontline staff skill sets.

Lack of Leadership

Most senior hospital and health system executives have remained largely uninvolved in QI efforts (Meurer et al. 2004). Typically, they take a hands-off approach to clinical matters, leaving decisions about patient care to physicians and other health professionals. Perhaps as a result, it has been regulatory agencies, medical societies, and payers--not providers--that have led the debate over which quality metrics to measure. (Payers may be the most sophisticated in their ability to assess the economic consequences of QI agendas, but their perspective and goals are often quite different from those of providers [Safavi 2006].) Because most senior executives have remained uninvolved in QI efforts, they tend to regard those efforts as an unavoidable cost of doing business, not as a strategic investment. They rarely demand the types of financial returns that similar investments in efficiency improvements or supply chain management might yield. Even more rarely do they attempt to measure the extent of such returns (HFMA 2005).

Yet the absence of senior executive involvement may send an inadvertent signal that such efforts do not have strong organizational support. Indeed, lack of leadership may partially explain why many physicians neither participate in QI activities nor modify their behavior in response to change initiatives (Audet et al. 2005; Bradley et al. 2003).

Lack of Organized Focus on Financial Outcomes

Currently, the QI agendas for most providers are set by stakeholders 

with few incentives to think through the financial implications of their decisions (Shulkin 2000). At most hospitals and health systems, the agenda is developed by the medical staff and QI directors (typically, nurses who have transitioned into an administrative role). These people select the metrics they are required to report or those they believe are the most clinically important. The financial impact (positive or negative) of their decisions is a secondary consideration.

Regulators, medical societies, and other interest groups use similar methods to set their QI agendas; again, the financial impact on providers is not a central concern. For example, both the Joint Commission (2002) and the Centers for Medicare & Medicaid Services (2005) have defined oxygenation assessment as a core measure for CAP patients. Because these groups require that the percentage of hospitalized CAP patients who undergo this assessment be monitored, providers must assign nurses not only to measure oxygenation levels but also to double-check and report on whether such measurements have been taken. Yet the Joint Commission (2002) has acknowledged that before the test was defined as a core measure, oxygenation assessment was being performed in 95 percent of hospitalized CAP patients. Thus, the requirement to report compliance with this assessment is unlikely to have a substantive effect on patient outcomes, but it does increase providers' costs.

Temporal Discrepancies

Providers must often shoulder the cost of interventions designed to improve patient outcomes, even though the benefits are not realized until after discharge and thus accrue to other stakeholders. For example, predischarge pneumococcal vaccination requires providers to pay for the cost of the vaccine plus the staff to deliver it. However, any subsequent financial rewards from pneumonia prophylaxis go to the payer, employer, or patient.

In general, it has been difficult to demonstrate that providers can derive financial benefit from QI efforts (Leatherman et al. 2003). We have found, however, that hospitals can derive financial benefit from QI efforts that reduce costs immediately (Table 1). Examples of such efforts include strategies to decrease the rate of uncompensated complications and, in many cases, to lower length of stay (LOS) (HFMA 2004). The best approach, therefore, is to focus on quality metrics that not only increase compliance with regulatory requirements but also provide immediate improvements in care quality and significant economic value to providers.

Inappropriate Skill Sets

Even if QI staff members are given strong incentives to focus on the financial implications of their decisions, they may not have the skills or resources to do so. Frontline QI personnel often have no training in the analyses needed to identify which problems are most significant, which quality metrics create the greatest financial value, or even which changes have the greatest effect on patient outcomes (Sullivan, Bretschneider, and McCausland 2003). In many cases, team members lack the ability to execute effective quality improvement efforts (Woodard 2005). As Barron and colleagues (2005, 225) have noted, performance improvement in healthcare requires "knowledge and tools that reside largely outside the training and experience of most clinicians and health care managers."

Frontline QI personnel also often lack the skills necessary to influence physicians' behavior (Shulkin 2000). Yet if improvement efforts are to succeed, a compelling case for change must be made to the hospital's physicians (Bradley et al. 2003). To assess the effectiveness of hospital quality staff in leading QI initiatives, we conducted a survey of 69 frontline QI directors at one healthcare system (the response rate was 80 percent). The directors reported that they spent, on average, only about 5 percent of their time conducting analyses to inform their choice of metrics (Table 2 and Figure 1). (In fairness, more than half of the metrics they track are Joint Commission requirements.) Fewer than one-third said they consistently use analytic techniques to understand processes, diagnose bottlenecks, and track improvement. More typically, the directors focus on basic compliance and administrative tasks (e.g., data gathering, validation, and distribution of quality metrics) or on "putting out fires." Only 10 percent of their time is spent on activities directly related to improving quality of care.

BUSINESS CASE FOR QI

Providers that want to improve their focus on financial outcomes should begin by using a more comprehensive evaluation of how QI investments can affect hospital economics. Four factors must be considered, as shown in the following equation:

Profit from quality improvement =

([DELTA] Volume * [DELTA] Price) - ([DELTA] Cost of QI infrastructure + [DELTA] Cost of inpatient care)

Historically, the primary lever that providers used to increase the impact of QI efforts was to minimize the cost of their QI infrastructure. Volume and price came into play only when payers, employers, or patients demanded greater (or lesser) access to care or looked less favorably on hospitals with poor perceived quality. More recently, providers have put greater emphasis on volume and price because of their concern about quality scorecards: the information in these scorecards may enable healthcare purchasers/consumers to divert volume to providers of choice, and it has given both purchasers and providers a new negotiating lever. For example, certain governmental organizations and large payers have established criteria that allow some providers to be defined as centers of excellence (OWH 2006; UHC 2006); providers with that designation often experience a boost in volume.

Furthermore, providers may qualify for additional reimbursement under pay-for-performance programs if they can demonstrate compliance with the programs' guidelines (Alexander 2005; O'Hare 2005; Safavi 2006). Many providers may have overlooked the fourth lever--the cost of inpatient care--as a way to profit from QI efforts. Yet we believe this lever is very important.

Data Analysis

To investigate this hypothesis, we examined a database containing clinical and administrative records for 10,512 primary admissions for adult CAP; all patients had been admitted to a 97-hospital healthcare system between June 2002 and December 2003. The healthcare system we studied includes a variety of small (25-50 bed) to large (400-600 bed) community hospitals as well as five academic medical centers. The hospitals have significant autonomy with regard to the focus of, and approach used in, their clinical quality programs. However, the system imposes a few structural requirements (e.g., each hospital must employ a director of clinical QI and participate in Leapfrog initiatives). Furthermore, a corporate (systemwide) clinical quality department provides training and information to the QI staff at each institution.

To identify appropriate patients, we used the Joint Commission's definition of pneumonia and specific demographic and clinical criteria--for example, all patients had to be at least 18 years old, they had to be discharged alive within 19 days of admission (the 95th percentile of LOS in our dataset), and those who were readmitted to the same hospital within 14 days were excluded. Because LOS has such a strong impact on inpatient costs, many of our analyses focused on CAP metrics that might reduce this variable. Both simple t-test and multivariate analyses were used to investigate relationships among the various parameters. The linear multivariate regressions included the standard Joint Commission CAP metrics captured during hospitalization--for example, whether blood cultures were drawn before antibiotic administration, whether the first dose of antibiotics was given within four hours of admission, whether the antibiotics were appropriate, and whether an oxygenation assessment was completed.

All of these metrics were entered as dichotomized variables, Joint Commission CAP metrics associated with discharge (e.g., pneumococcal vaccination) were not included because they are unlikely to affect LOS. Two metrics were included that were not from the Joint Commission: time to ambulation and time to conversion from intravenous to oral antibiotic administration; both of these metrics were measured in days. All of the analyses were adjusted for sex, age, pneumonia severity index, and admission source. They were conducted using STATA (version 8, College Station, Texas).

Evidence of Clinical Benefit

Four of the six CAP measures investigated were not associated with any change in mean LOS: rapidity of initial antibiotic administration, antibiotic selection consistent with guidelines, assessment of oxygenation, and performance of blood cultures. However, two metrics--time to oral antibiotic conversion and time to ambulation--had a statistically significant association with hospitalization duration. Each one-day reduction in time until patients were converted from intravenous to oral antibiotics was associated with a 0.8-day (95 percent confidence interval [CI], 0.78-0.81) decrease in LOS; each one-day improvement in getting patients to ambulate was associated with a 0.2-day (95 percent CI, 0.18-0.21) LOS decrease.

Other investigators have also demonstrated that these two clinical practices are associated with decreased LOS, and neither appears to increase the risk of adverse outcomes (Mundy et al. 2003; Rhew et al. 1998). Indeed, our findings suggest that both practices are associated with a lower complication rate. Furthermore, patients report high levels of satisfaction with earlier antibiotic conversion (Ramirez et al. 1999).

As the American College of Chest Physicians has noted, "Hospitalization is not without hazard" (Ramsdell et al. 2005, 1759). Among the risks the ACCP cites for hospitalized CAP patients are deconditioning, pressure ulcers, and bacterial colonization with virulent or drug-resistant organisms. These risks are lowered with reduced LOS. Reduced LOS has no impact on the subsequent rates of complications, readmissions, or death as long as CAP patients are clinically stable at discharge and have no severe coexisting illnesses (Mundy et al. 2003; Ramirez et al. 1999; Rhew et al. 1998).

Evidence of Financial Benefit

We then calculated the potential cost savings associated with reduced LOS from earlier ambulation or antibiotic conversion (Table 3). To perform this calculation, we had to determine the direct variable costs that would be saved by eliminating each of these excess days. The provider system we examined uses a cost-accounting system that assigns direct and indirect (variable and fixed) cost dollar values for each charge line item appearing on a patient's bill, using a relative-value-unit methodology that is standard in the provider industry. For CAP patients, the average variable cost across all inpatient days was $876 per day.

We reasoned, however, that the final day of hospitalization might have lower costs, given its generally lower intensity of care. Therefore, we ran a specific analysis that included a group of patients with diagnosis-related group (DRG) 089 (simple pneumonia) to examine just the final 24 hours before discharge. We found that the variable costs for these final 24 hours were 35 percent lower than the average across all days (data not shown), so we used this lower, more conservative estimate to better match true cost savings in reducing LOS for patients with CAP.

On average, a 0.8-day reduction in LOS therefore produced $456 in direct per-patient savings. Assuming that a hospital is reimbursed $8,000 for a given pneumonia patient (based on DRG 089) and earns 10 percent margins at baseline, this $456 savings increases per-patient profit margin by nearly 60 percent.

Factors that Influence Results

Three provider-specific factors affect the actual amount of value that can be captured through this type of QI effort. First, a given hospital's case mix and types of patients admitted will affect the level of savings from any change initiative. Second, the hospital's baseline performance clearly influences how much improvement is possible. Third, the payment system and contract structure a provider has with a payer determines whether savings from reduced LOS accrue directly or indirectly to the hospital. In prospective payment systems (e.g., Medicare), the hospital is at risk for incremental costs; thus, any savings accrue directly to it. But when a hospital is reimbursed on a per diem basis, LOS reductions can only lower costs indirectly, through negotiating tactics to achieve price concessions (e.g., share-the-value savings). Without such concessions, the savings accrue to the payer; indeed, LOS reductions could be value-destroying for the provider (HFMA 2005). In our study, however, the vast majority of CAP patients with long LOSs were Medicare patients, which suggests that reducing hospitalization duration presents an important economic opportunity.

An additional limitation of our calculations must be noted: we did not quantify several other potentially important drivers of economic value, such as the impact of improved quality metrics on pay-for-performance revenue or negotiating leverage with payers. Furthermore, we did not include the revenues gained from refilling liberated beds in capacity-constrained hospitals or the impact marketing campaigns could have on nonconstrained hospitals. Had we included these other drivers in our calculations, the impact on margins would likely have been higher.

CAPTURING VALUE FROM QI

We identified two quality metrics that are associated with both reduced LOS for patients with CAP and favorable hospital economics. Although our results are promising, we recommend that providers replicate our analyses to confirm the clinical and economic value of these metrics at their institutions. Providers can also use our analyses to assess other quality metrics that are likely to correlate with economic value. These analyses require time and expertise, however: two things that may be lacking at the front line.

If QI efforts are to succeed, providers must invest in training programs for their frontline QI staff; of particular importance is training in analytic and organizational tools (Sullivan, Bretschneider, and McCausland 2003; Woodard 2005). Furthermore, senior executives must lead the way if they want QI efforts to have sustained clinical and financial impact (Barron et al. 2005). Finally, broad awareness of, and commitment to, the quality agenda among all constituents (physicians, nurses, and ancillary staff) is crucial for successful QI efforts (HFMA 2004).

Unfortunately, getting buy-in from physicians for change initiatives is notoriously difficult (Audet et al. 2005). One method that has proved effective is to report accurate quality metric profiles on a physician-by-physician basis (Ballard 2005; HFMA 2004). (The physicians' identities are disguised in the public report, but each of them knows his or her own results.) Presenting these reports to physicians requires not only a firm grounding in clinical data but also the confidence to challenge existing perspectives.

Improving frontline capabilities is the best way to ensure the success of change initiatives. However, some hospitals have another option available to them: system-level clinical quality organizations. These organizations generally have more resources and better access to the databases that can help define the optimal set of quality metrics for each hospital. Based on the value at stake, we believe a stronger role for and significant upfront investment in these organizations are justified.

CONCLUSION

Many providers can perform better on their dual mission--to deliver high-quality patient care while ensuring economic stability and strength--if they focus their QI efforts on a rigorously derived set of metrics that takes into account both clinical care and financial results. To achieve this goal, a provider must focus its QI program on identifying opportunities and driving improvement efforts.

We did not analyze the incremental cost required to successfully transform QI efforts, because they vary substantially from provider to provider. (They are heavily dependent on each provider's baseline skill mix and infrastructure.) It seems reasonable, however, to assume that for many providers, expenses may rise initially, but in many cases these costs will soon be outweighed by the financial and clinical benefits of the quality improvement. Further research to fully understand the impact of QI efforts on provider economics would be valuable. In particular, analyses are needed to quantify the size and financial impact of volume shifts driven by patients, physicians, payers, and employers in response to quality initiatives and to assess the impact of QI efforts on providers' pricing leverage in negotiations with payers.

For more information on the concepts in this article, please contact Dr. Grote at [e-mail].

RELATED ARTICLE: Practitioner application.

Louis M. Cocorullo, FACHE, vice president/chief financial officer, Martin Memorial Health Systems, Stuart, Florida

This article is a valuable addition to the literature for hospital administrators and clinical quality staff who are concerned with earning a return on investment for their clinical quality initiatives.

As a practicing hospital system CFO, I am part of the decision to invest hundreds of thousands, if not millions, of the organization's dollars to improve our clinical quality and patient safety outcomes. Much of this investment delivers a clear return in terms of improved clinical outcomes, reduced error rates, and increased patient satisfaction. However, many of us have long appreciated the fact that the clinical improvements that result from this investment also have important financial implications. The financial impact of these clinical benefits is typically not considered when we establish our quality priorities and when we follow those set by the Centers for Medicare and Medicaid Services (CMS).

This study, to my knowledge, is the first to provide a useable framework for evaluating the return on these quality investments. It should motivate hospital administrators not only to critically look at the quality measures they are tracking but also to ensure that they are directly involved in setting the agenda for quality improvement initiatives. In my healthcare system, a reduction of a day or more in length of stay for pneumonia patients (who are predominantly covered by Medicare) could generate financial savings--something that would clearly be on my radar screen, and I suspect that many other hospital administrators would think the same way.

The authors focus on just a few examples of quality measures that correlate with financial value, but their work outlines an approach that could be expanded to deal with other financially oriented questions about quality improvement. For example, what other quality measures we use correlate with a reduction in complications, readmissions, or in-hospital costs? While the authors' analytic approach is rigorous, even a simplified approach could have great merit. I can imagine working with the quality staff to identify interventions expected to correlate with reduced cost, testing those hypotheses against retrospective hospital data, and using those findings to shape our focus on quality improvement.

The article indicates the need for an evolving relationship among quality, finance, and operations. Quality leaders understand the impact of their initiatives on patient outcomes; financial leaders can make a similar assessment of quality initiatives to evaluate the initiatives' financial implications. Operational leaders must communicate to frontline staff the importance of quality initiatives and must ensure that staff members (especially the quality staff) are trained appropriately and rewarded for their performance on achieving quality goals as well as financial outcomes.

I do not expect this study to shut down many aspects of my organization's current quality measurement and reporting methods, because measuring and reporting CMS metrics, regardless of their direct financial value, will continue to be important. Nonetheless, this article is very useful in that it points out that clinical quality has a broader role in achieving financial objectives. With the emergence of more pay-for-performance programs, center-of-excellence designations by payers, and volume shifting by payers on the basis of quality and cost, the insight that hospitals should reorganize to connect quality to operations and finance is a timely one.

The approach in this article is intriguing and useable for practitioners in the field. It has the potential to substantially improve a hospital's financial performance.

References

Alexander, S. 2005. "Early Results from the Premier-CMS Hospital Quality Incentive Demonstration Program." Presented at the Quality Colloquium, Cambridge, Massachusetts, August 23.

Audet, A-M. J., M. M. Doty, J. Shamasdin, and S. C. Schoenbaum. 2005. "Measure, Learn, and Improve: Physicians' Involvement in Quality Improvement." Health Affairs 24 (3): 843-53.

Ballard, R. 2005. "Strategic Supply Cost Management: Physician Preference Without Deference." Healthcare Financial Management 59 (4): 78-84.

Barron, W. M., C. Kresk, D. Weber, and J. Cerese. 2005. "Critical Success Factors for Performance Improvement Programs." Journal on Quality and Patient Safety 31 (4): 220-26.

Bradley, E. H., E. S. Holmboe, J. A. Mattera, S. A. Roumanis, M. J. Radford, and H. M. Krumholz. 2003. "The Roles of Senior Management in Quality Improvement Efforts: What Are the Key Components?" Journal of Healthcare Management 48 (1): 15-28.

Centers for Medicare & Medicaid Services (CMS). 2005. "Overview of Specifications of Measures Displayed on Hospital Compare." [Online information; retrieved 1/10/06.] new.cms.hhs.gov/ HospitalQualityInits/downloads/Hospital OverviewOfSpecs200512.pdf.

Dedier, J., D. E. Singer, Y. Chang, M. Moore, and S. J. Atlas. 2001. "Processes of Care, Illness Severity, and Outcomes in the Management of Community-Acquired Pneumonia at Academic Hospitals." Archives of Internal Medicine 161 (17): 2099-104.

Healthcare Financial Management Association (HFMA). 2004. "Key Strategies for Sustained Performance Improvement." Healthcare Financial Management 58 (11, suppl): 1-12.

--.2005. "The Case for Quality: Effective Clinical and Financial Case Management." Healthcare Financial Management 59 (6, suppl): 1-8.

Island Peer Review Organization (IPRO). 2006. "Compare Hospitals." [Online tool; retrieved 1/10/06.] compare.ipro.org/ index/hosp_compare.

Joint Commission. 2002. "Overview of Community Acquired Pneumonia Order Set." [Online information; retrieved 1/10/06.] www.jointcommission.org/Performance Measurement/PerformanceMeasurement/ Pneumonia+Core+Measure+Set.htm.

--. 2006. "Quality Check." [Online information; retrieved 1/10/06.] www.jcaho .org/quality+check.

Leatherman, S., D. Berwick, D. Iles, L. S. Lewin, E Davidoff, T. Nolan, and M. Bisognano. 2003. "The Business Case for Quality: Case Studies and an Analysis." Health Affairs 22 (2): 17-30.

Meurer, S. J., M. A. Counte, D. M. Rubio, and B. Arrington. 2004. "Measuring a Hospital's Ability to Improve." American Journal of Medical Quality 19 (5): 214-22.

Mundy, L. M., T. L. Leet, K. Darst, M. A. Schnitzler, and W. C. Dunagan. 2003. "Early Mobilization of Patients Hospitalized with Community-Acquired Pneumonia." Chest 124 (3): 883-89.

Office on Women's Health (OWH). 2006. "National Centers of Excellence in Women's Health." [Online information; retrieved 1/10/06.] www.4woman.gov/ COE/.

O'Hare, P. 2005. "Pay for Performance: Will Your Hospital Be Ready?" Healthcare Financial Management 59 (5): 46-49.

Ramirez, J. A., S. Vargas, G. W. Ritter, M. E. Brier, A. Wright, S. Smith, D. Newman, 1. Burke, M. Mushtaq, and A. Huang. 1999. "Early Switch from Intravenous to Oral Antibiotics and Early Hospital Discharge." Archives of Internal Medicine 159 (20): 2449-54.

Ramsdell, J., G. L. Narsavage, J. B. Fink, American College of Chest Physicians' Home Care Network Group. 2005. "Management of Community-Acquired Pneumonia in the Home: An American College of Chest Physicians Clinical Position Statement." Chest 127 (5): 1752-63.

Rhew, D. C., D. Hackner, L. Henderson, A. G. Ellrodt, and S. R. Weingarten. 1998. "The Clinical Benefit of In-Hospital Observation in 'Low-Risk' Pneumonia Patients After Conversion from Parenteral to Oral Antimicrobial Therapy." Chest 113 (1): 142-46.

Safavi, K. 2006. "Patient-Centered Pay for Performance: Are We Missing the Target?" Journal of Healthcare Management 51 (4): 215-18.

Saint, S., T. P. Hofer, J. S. Rose, S. R. Kaufman, and L. E McMahon. 2003. "Use of Critical Pathways to Improve Efficiency: A Cautionary Tale." American Journal of Managed Care 9 (11): 758-65.

Shulkin, D. J. 2000. "Why Quality Improvement Efforts in Health Care Fail and What Can Be Done About It." Commentary. American Journal of Medical Quality 15 (2): 49-53.

Stone, R. A., M. K. Mot, J. R. Lave, L. J. Hough, and M. I. Fine. 2005. "Implementation of an Independent Management and Discharge Strategy for Patients with Community-Acquired Pneumonia." American Journal of Managed Care 11 (8): 491-99.

Sullivan, J., J. Bretschneider, and M. P. McCausland. 2003. "Designing a Leadership Program for Nurse Managers. An Evidence-Driven Approach." Journal of Nursing Administration 33 (10): 544-49.

United Health Care (UHC). 2006. "Hospital Comparison Took." [Online information; retrieved 1/10/06.] www.geoaccess.com/ uhc/po/Default.asp.

U.S. Department of Health and Human Services (HHS). 2006. "Hospital Compare: A Quality Took For Adults, Including People With Medicare." [Online information; retrieved 1/10/07.] www.hospitalcompare.hhs.gov.

U.S. News & World Report. 2005. "Best Hospitals 2005." [Online article; retrieved 1/10/06.] www.usnews.com/usnews/ health/best-hospitals/tophosp.htm.

Woodard, T. D. 2005. "Addressing Variation in Hospital Quality: Is Six Sigma the Answer?" Journal of Healthcare Management 50 (4): 226-36.

Note

(1.) In this article, we use the term "quality improvement" to describe any type of program designed to enhance the care delivered to patients; it is not meant to characterize any particular type of program.

Kurt Grote, M.D., consultant, McKinsey & Company, Los Angeles; Edd Fleming, M.D., consultant, McKinsey & Company, Palo Alto, California; Edward Levine, M.D., consultant, McKinsey & Company, Palo Alto; Russell Richmond, M.D., chief medical officer, D2Hawkeye, Boston; Saumya Sutaria, M.D., consultant, McKinsey & Company, Palo Alto; Francine C. Wiest, M.D., Seattle, Washington; and Jennifer Daley, M.D., associate clinical professor of community health and family medicine, Dartmouth Medical School, Hanover, New Hampshire, and senior vice president of clinical quality, Tenet Healthcare, Dallas, Texas 

TABLE 1
How Timing Affects Who Benefits from Quality Improvement Efforts

                                    Stage of Care

                                    Acute Illness
                                    and               After
                  Preventive        Hospitalization   Hospitalization

Quality           Pneumococcal      * Conversion      * Pneumococcal
improvement                         from intravenous  vaccination
metric            vaccination       to oral anti-     * Provision
(examples)                          biotics in        of discharge
                                    patients with     instructions
                                    CAP*
                                    * Early ambulation
                                    in patients
                                    with CAP

Sources of        Reduced           * Reduced LOS **  Reduced
financial value                                       rehospitaliza-
through quality   hospitalization   * Reduced comp-   tion rate
improvement       rate              lication rate
(examples)                          * Pay-for-
                                    performance
                                    rewards

Primary           * Payer/employer  * Provider        * Payer/employer
beneficiaries of  * Patient         * Payer/employer  * Patient
financial                           * Patient
savings

* Community-acquired pneumonia; ** Length of stay

TABLE 2
Responses to Director of Clinical Quality Survey (n = 55)

Demographics

Experience as director of clinical quality              3.71 (0-15.0)
improvement (years, range)
Hospital type (percentage)
 Community hospital                                          91%
 Academic medical center                                     9%

Number of beds (percentage)
 [less than or equal to] 100                                 15%
 101-200                                                     35%
 201-300                                                     11%
 301-400                                                     28%
400                                                          11%

Quality-focused staff in hospital (number, range)     3.77 (0.2-10.0)
 Director of clinical quality improvement             0.77 (0.2-1.0)
 Patient safety officer                               0.44 (0-1.0)
 Risk manager                                         0.69 (0-3.0)
 Data coordinator                                     0.74 (0-3.0)
 Infection control manager                            1.13 (0-3.0)
Leadership Support
 Respondents who agree that "hospital leadership
 provides significant, frequent agenda-shaping
 guidance on quality improvement" (percentage)
  About which outcome measures are most important
  About which metrics should be pursued                     25.5%
  About how improvement in those metrics should             18.2%
  be driven                                                 25.5%

Process Mapping
Reported use of process mapping, when applicable,
to understand processes, diagnose bottlenecks, and
track improvement (percentage)
Consistently used                                           0.9%
Inconsistently used                                         60.0%
Never used                                                   9.1

Note: The directors surveyed were all employed by the 97-hospital
healthcare system described in the Data Analysis section.

TABLE 3
Estimated Average Cost Savings per Hospitalized Patient with CAP *,
Based on Two Quality Metrics
                                                    Savings
                                                          Earlier
                                                        Conversion
                                                        from Intrave-
                                           Earlier      nous to Oral
                                          Ambulation    Antibiotics
                                          ([dagger])     ([dagger])

A. Average daily variable hospital           $876           $876
cost for DRG 089 (simple pneumonia)
B. Average decrease in cost for              65%            65%
last day of stay
C. LOS ** reductions ([dagger])              0.2            0.8
Total (A x B x C)                            $114           $456

* Community-acquired pneumonia; ** Length of stay
([dagger]) Each one-day reduction in the time until patients could
ambulate was associated with a 0.2-day decrease in LOS; each one-day
reduction in the time until patients could be switched from
intravenous to oral antibiotics was associated with a 0.8-day
decrease in LOS.

FIGURE 1
Average Time Allocation by Hospital Quality Directors (Percentages *)

                                         Quality Improvement
                                              Process

Other quality improvement tasks                 10
Patient safety tasks                            11
Administrative tasks                            15
Joint Commission reporting compliance           17
Quality improvement                             46

Quality Improvement Process                Time Allocation

Selecting metrics                               11
Gathering data                                  29
Validating data                                 25
Distributing data                               14
Improving metric                                22

Note: Table made from pie graph.