Overview

Eight years ago, the Institute of Medicine's (IOM) report To Err is Human revealed that between 44,000 and 98,000 Americans die each year as a result of medical errors.^[1] Medication errors alone are estimated to cause more than 7000 deaths annually,^[1,2] compared with 6000 annual deaths from workplace injuries.^[1] Although much of the discussion of medical errors has focused on the hospital setting, errors can occur in other settings, such as physician offices, outpatient surgical centers, nursing homes, and retail pharmacies.^[1]

Total costs of medical errors resulting in injury are estimated to be between $17 billion and $29 billion, with healthcare costs comprising over 50%.^[1] In 2003, Medicare paid hospitals an additional $300 million per year, or 0.3% of annual Medicare hospital spending, for 5 types of adverse events. These extra payments covered less than one third of the extra costs that hospitals incurred in treating these adverse events.^[4] Healthcare professionals also pay with a loss of morale and may view the medical liability system as an impediment to learning from errors.^[1]

Aims for Healthcare Improvement

In the report Crossing the Quality Chasm,^[3] IOM identified 6 aims for improvement in healthcare:

1. Safe: avoiding injuries to patients from care that is intended to help them.
   
   
2. Effective: providing services based on scientific knowledge to all who could benefit, and refraining from providing services to those unlikely to benefit (avoiding underuse and overuse).
   
   
3. Patient-centered: providing care that is respectful of and responsive to individual patient preferences, needs, and values and ensuring that patient values guide clinical decisions.
   
   
4. Timely: reducing waits and sometimes harmful delays for both those who receive and give care.
   
   
5. Efficient: avoiding waste, such as waste of equipment, supplies, ideas, and energy.
   
   
6. Equitable: providing care that does not differ in quality because of personal characteristics such as gender, ethnicity, geographic location, and socioeconomic status.

Efforts Toward Error Reduction

In the United States, healthcare lags behind other industries with respect to attention to ensuring safety. Aviation has focused on building safe systems since World War II. Although the risk of dying as a result of a medical error far exceeds that of dying in an airline crash, much more public attention has focused on improving safety in the airline industry than in healthcare. Some believe that public concern about airline safety, in response to the impact of media stories, has played an important role in the improvement of safety in the airline industry.^[1]

Healthcare has begun to follow companies such as General Electric and Motorola in embracing the six sigma philosophy to reduce errors. Six sigma seeks to reduce variation in processes that lead to defects.^[5] It relies on statistical analysis of data and strong problem-solving techniques. Sigma is the Greek letter of the alphabet used to describe variability, or standard deviation. By achieving six sigma, the failure rate is minimized to 3.4 defects (errors) per million opportunities, which translates to a 99.9996% success rate.^[6] Like six sigma, continuous improvement emphasizes scientific methods to seek continual improvement through suggestions of workplace teams. The Deming PDCA cycle (plan, do, check, act) underlies continuous improvement.^[5]

Quality in healthcare organizations may be evaluated using Donabedian's trilogy: structure (how care is organized, such as ICU staffing with intensivists); process (what is done by caregivers, such as the percentage of patients with diabetes who have their glycated hemoglobin measured); and outcomes (the results achieved, such as mortality rates following coronary artery bypass graft).^[7]

Evaluating Improvement

Process measures can be based on scientific evidence, can be relatively easily measured and compared, and are within physicians' control.^[8] Most publicly reported quality indicators are process measures because of the methodologic challenges in measuring outcomes, including the need for case-mix adjustment.^[7] However, measuring process of care has limitations. Only a fraction of care is evidence-based, and clinical care often relies on physician judgments that do not easily translate into the development of indicators.^[8] The absence of strong scientific evidence linking processes to meaningful outcomes has limited the usefulness of process measures for studying patient safety.^[7]

Outcome measures are clearly valued by patients and appear to be more closely aligned with the goal of clinical care. However, they are heavily confounded by patient mix, (ie, one physician's patients may be sicker and therefore have increased mortality rates). Even poor practitioners may have mortality rates that are not statistically worse than those of excellent practitioners if the sample sizes are small.^[8]

Hospitals that perform well on process measures have better outcomes, although different studies have found that the strength of this association varies. One study found a robust relationship, while another found that only 6% of the variation in mortality was explained by variations in the process measures examined.^[8]

Although healthcare organizations may use administrative databases to obtain quality-of-care information, the demands for clinically relevant data are increasing and may dictate the need for an electronic medical record.^[9] The Joint Commission on Accreditation of Healthcare Organizations' standard set for pneumonia measures performance with respect to oxygenation assessment, pneumococcal vaccination, blood cultures, smoking-cessation counseling, and mean time to initiation of antibiotics. The first 4 of these are discrete measures that reflect either the presence or absence of an item.

Quality-improvement measures may be reported as item-by-item measurement, with performance on each element of care reported separately as a percentage. The denominator is the total number of patients in the sample and the numerator is the total number of patients for whom the item was performed. In contrast to item-by-item measurement, all-or-none measurement raises the bar and is a more sensitive scale for assessing improvements. For example, with the pneumonia indicator, the denominator would be the number of patients eligible to receive at least 1 of the 4 discrete elements of care, and the numerator would be the number of patients who actually received all of the care for which the patient was eligible.^[10]

Performance Incentives

Medical practice in the United States has long been characterized by fee-for-service payment and a high degree of autonomy.^[11] Recently, pay-for-performance arrangements have become an increasingly popular way for payers to reward physicians and hospitals for adhering to evidence-based standards of care. More than half of US health maintenance organizations used pay-for-performance programs in their contracts with physicians and hospitals in 2005. Nearly 90% included programs for physicians and more than one third had programs for hospitals.^[12]

There has been skepticism surrounding pay-for-performance initiatives. A recent review suggests that small-scale bonus arrangements are insufficient to motivate significant changes by physicians or hospitals.^[13] High-quality healthcare requires a balance of risks, benefits, and patients' preferences, not necessarily rigid adherence to clinical guidelines. Some worry that pay-for-performance could cause more harm than good. If physicians are not convinced that risk adjustment is adequate, they could decide that the easiest way to realize high scores is to avoid sick or complicated patients. Systems serving disadvantaged patients could witness a decrease in revenues, which would undermine safety-net programs.^[11]

Despite the skepticism around pay-for-performance programs, the Premier, Inc. demonstration project with the Centers for Medicare and Medicaid Services (CMS) demonstrated that pay-for-performance resulted in better care and lower costs. The project involved almost 260 hospitals and showed that with financial incentives, outcomes improved in 5 clinical areas: acute myocardial infarction, heart failure, coronary artery bypass graft, pneumonia, and hip and knee replacement.^[14] Over a 2-year period, pay-for-performance was associated with improvements ranging from 2.6% to 4.1%.^[15]

Conclusion

In the 8 years since the IOM's landmark report To Err is Human was issued, the safety of the US healthcare system has received much attention. Frameworks for evaluating quality of care have been widely instituted. Pay-for-performance initiatives have been debated and implemented. Over the next 8 years, the effectiveness of pay-for-performance programs and their role in healthcare quality improvement will become more clearly delineated.

References

1. Institute of Medicine. To Err is Human: Building a Safer Health System. Kohn L, Corrigan J, Donaldson M, eds. Washington, DC: National Academies Press; 1999.
2. Hampton T. Health agencies update. JAMA. 2006;296:384.
3. Institute of Medicine. Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: National Academies Press, 2001.
4. Zhan C, Friedman B, Mosso A, Pronovost P. Medicare payment for selected adverse events: building the business case for investing in patient safety. Health Aff. 2006;25:1386-1393.
5. Chase R, Jacobs F, Aquilano N. Operations Management for Competitive Advantage. 11th ed. Burr Ridge, Ill: McGraw-Hill/Irwin; 2006.
6. Lanham B, Maxson-Cooper P. Is six sigma the answer for nursing to reduce medical errors and enhance patient safety? Nurs Econ. 2003;21:39-41. Available at: http://www.medscape.com/viewarticle/449692 Accessed August 17, 2007.
7. Pronovost P, Miller M, Wachter R. Tracking progress in patient safety: an elusive target. JAMA. 2006;296:696-699. Abstract
8. Jha A. Measuring hospital quality: what physicians do? how patients fare? or both? JAMA. 2006;296:95-97. Abstract
9. Baldwin K. Evaluating quality of primary care using the electronic medical record. J Healthc Qual. 2006;28:40-47. Abstract
10. Nolan T, Berwick D. All-or-none measurement raises the bar on performance. JAMA. 2006;295:1168-1170. Abstract
11. Fisher E. Paying for performance - risks and recommendations. N Engl J Med. 2006;355:1845-1847. Abstract
12. Rosenthal M, Landon B, Normand S, Frank R, Epstein A. Pay for performance in commercial HMOs. N Engl J Med. 2006;355:1895-1902. Abstract
13. Rosenthal M, Frank R. What is the empirical basis for paying for quality in health care? Med Care Res Rev. 2006;63:135-157.
14. Schoenbaum SC. The patient safety movement finally is saving lives and raising hopes. Medscape General Medicine. 2006;8(4):16 Available at: http://www.medscape.com/viewarticle/545749 Accessed August 17, 2007.
15. Lindenauer P, Remus D, Roman S, et al. Public reporting and pay for performance in hospital quality improvement. N Engl J Med. 2007;356:486-496. Abstract