Section 1. Overview and Background
Cost of Poor Quality or Waste in Integrated Delivery System Settings
On July 10, 2005, David Walker, head of the U.S. Government Accounting Office, testified before the Citizens' Health Care Working Group at its community hearing in Salt Lake City, Utah. As comptroller general of the United States, he noted that unfunded financial liability for the Federal Medicare program for citizens currently alive totals $29.9 trillion, or about two-thirds of the $48.5 trillion total net worth, including home equity, of all U.S. citizens. In comparison, unfunded liabilities for FICA (Social Security) total only $5.7 trillion. At present rates of health care cost increases, the Medicare Trust Fund will be exhausted by 2018. However, the assets of the Medicare Trust Fund are stored in the form of U.S. treasury bonds. Redeeming those bonds will require significant tax increases or redirection of existing tax funds from other parts of the U.S. Federal budget. By that measure, significant funding shortfalls for Medicare will appear by about 2011 or 2012.
Similar health care cost increases are burdening U.S. businesses, making them less competitive in international markets. As a result, an increasing number of companies are shifting health care costs to employees or dropping employment-related health benefits altogether. By any metric, continued growth in health care costs, well in excess of growth rates for the general economy, herald a looming cost crisis for health care. While the U.S. spends significantly more money per capita on health care than any other modern industrialized democracy, rates of increase in other countries are similar. Although starting from a lower base rate, many other countries regard health care cost increases in much the same light as does the United States.
While there is some question about the ability to contain health care costs by eliminating waste and inefficiency (e.g., Schwartz & Mendelson, 1994), documentation of waste/inefficiency in health care and the ability to reallocate these resources holds promise (Ovretveit, 2004b). According to Ovretveit (2004a, p. 7), "the absence of empirical studies and of strong evidence about costs and savings is surprising given the claims made about waste and the effectiveness of quality methods."
The primary purpose of this task was to collaboratively build a system that could be actively used to identify the costs of poor quality or waste in an effort to identify opportunities for improvement in hospital settings. The proposed identification strategy has been designed from a financial management perspective, recognizing that we are in the business of providing high-quality medical care. In doing so, our aim as an industry is to provide "all the right care, but only the right care" (James B).
The lead integrated delivery system (IDS) partners for this effort are Intermountain Healthcare and Providence Health System (PHS). We also solicited periodic input via in-person discussion throughout the task from our IDS partners—Baylor Health Care System, UNC Health Care, and UPMC Health System. Researchers at RTI International (RTI) supported and contributed to the investigative effort. A multidisciplinary team of clinical, quality improvement, finance, and management engineering research staff were engaged in this demand-driven, participatory research.
Although there are many definitions of inefficiency in health care, we chose to adopt a broad definition that has been used in quality of care assessment: ".a wasteful use of resources for no (or very little) benefit, or a failure to use resources on clearly beneficial activities. Inefficiency may arise because of apparently inappropriate, irrational, or misinformed decisions by individuals or organizations" (Severens, 2003, p. 366). The cost of such waste is typically framed in terms of people and resources. It is recognized that there are different kinds of costs—developmental/acquisition, planning, implementation, and maintenance—that can be further categorized as one time or recurring. However, no standard mechanism for capturing these costs and categorizing areas of waste/poor quality is available for health care. It has been argued that such guidance is necessary to make the business case for improvements and to prioritize such efforts across competing needs (Weeks & Bagian, 2003).
Clinical quality improvement (QI) provides a theory and a set of practical tools for measuring and managing health care delivery processes (Crosby, 1979; Donabedian, 1980; Deming, 1986). QI theory holds that all planned human work, including health care delivery, is accomplished through such processes. QI theory further posits that every such clinical work process produces parallel medical and cost outcomes, suggesting that medical outcomes are directly related to cost outcomes. Clinical QI describes three causal relationships that define this link. One of those relationships describes process changes that produce better medical outcomes, but only through higher resource investments. Reductions in the other two relationships—quality waste and inefficiency waste—generate lower costs by improving medical outcomes or while medical outcomes hold stable. We draw on this body of work as a core foundation to our thinking in this project (James, 1989).
Examples of poor quality and/or waste in health care are diverse and include clinician interruptions, duplicate or repeat testing/procedures, delays in care, inefficient use of clinician time, improper documentation/record keeping, iatrogenesis, and patient injuries. For example, a recent study funded by the Agency for Healthcare Research and Quality (AHRQ) (Zhan et al., 2003) demonstrated that medical injuries during hospitalization resulted in longer hospital stays, higher costs, and a higher risk of death. The study, "Excess Length of Stay, Charges, and Mortality Attributable to Medical Injuries During Hospitalization," was published in the October 8, 2003, issue of the Journal of the American Medical Association. Zhan et al. found that the impact of medical injuries varies substantially. Using data from the 2000 Nationwide Inpatient Sample, the study provides specific estimates for excess length of stay, charges, and risk of death for 18 of the 20 AHRQ Patient Safety Indicators. This study, and many like it, shows an association between cost and quality but does not pinpoint where or how this is occurring at the point of service—details required to address the problem.
An initial literature review conducted by research staff at Intermountain Healthcare resulted in the identification of actual markers that signify the cost of poor quality. More than 1,330 articles that showed cost improvements associated with clinical change were identified. This was not a systematic review of the literature. Even with our updated scan, we note that a wide net must be cast to identify such studies, which largely do not include terms such as "quality waste" or "inefficiency" as key words (Appendix A).
Estimates suggest that health care quality waste and inefficiency waste may account for more than 50 percent of all American health care expenditures. The Midwest Business Group on Health has estimated that 30 percent of health care costs are for waste, and the average cost of poor quality care per patient per year is $1,500. Growing pressure on health care costs have generated an interest in quality and inefficiency waste as means to control costs and possibly expand health care access while maintaining high health care quality given waste projections such as these. However, theory and estimates do not always translate into practice.
As a first step, we examined the peer-reviewed medical literature, seeking instances showing quality waste and inefficiency waste mechanisms at work within current American health care delivery. Our aim was to see whether such phenomena exist in practice as well as theory and to get some preliminary sense of their possible scale and, hence, potential. The findings from this preliminary search suggested that there is good evidence that quality waste and inefficiency are common in health care delivery and that attacking them can produce significant cost savings (at least as illustrated by individual projects where this occurred; for example, Ovretveit, 2004a, b).
Key leaders from our partner IDSs affirmed the need to address this problem. The literature and leading researchers have spoken in an abstract sense for years about making health care more efficient and eliminating waste without giving clear direction on how to identify such improvement opportunities. Quotes from senior executives in our partner IDSs illustrate the situation:
- "Patient injuries are expensive. Any class of patient safety is going to be part of this."
- "This idea of leveraging waste elimination creates capacity. Because of growth in potentially avoidable medical admissions, you are canceling elective surgeries."
- "We own a health plan that covers over 10 percent of our admissions. The health plan is more profitable if the hospital has less waste. More admissions and shorter length of stay with higher quality. Waste has a negative correlation with quality items."
With this in mind, we set out to construct an organizing framework for considering quality waste and inefficiency in health care that builds on existing constructs. The end point of this effort is intended to produce a toolset for eliminating/mitigating quality waste and/or inefficiency in health care and to provide a more refined set of waste/poor quality estimates to guide efforts to extract underutilized resources in hospitals.
Section 2 presents our aims, the context for this work, and our resultant organizing framework.
In Section 3, we illustrate how current approaches have addressed the problem of waste/poor quality and present specific examples of these estimates.