Measuring Outcomes


Introduction

Outcome measurement can be defined various ways. One definition is “a measure of the quality of medical care, the standard against which the end result of an intervention is assessed.” Another definition is “the determination and evaluation of the results of an activity, plan, process or program and the comparison with the intended or projected results.” Although these definitions differ somewhat, both measure some level of care and both assess care against some standard, which then leads to the implementation of changes if the health intervention is found to be suboptimal. Outcome measures may be used by an individual physician to convey expected results to a patient or to evaluate his or her own results. Outcome measures also can be used to assess systems of care. Almost always, outcomes are measured to ensure that the results of care are optimal, or if not, to trigger strategies that will improve the outcome of interest.

Measuring outcomes dates back several centuries. During the Crimean War (1854), Florence Nightingale measured morbidity and mortality rates and related them to the quality of the nursing care given to wounded soldiers. She found that adequate care reduced the mortality rate by 32%. Early in the 20th century, Dr. Emory Groves suggested that guidelines for postoperative follow-up could reduce adverse outcomes. At The Massachusetts General Hospital in 1914, Dr. E.A. Codman prophetically recommended that the result of patient care should be used as an indicator of the quality of that care. Unfortunately, his progressive views met with resistance, and he was fired. Some 50 years later, Donabedian was the first person to refer to the three aspects of measuring quality in health care: structure, process, and outcome. He said that these three aspects were complementary and equally important.

During the past 20 years, greater emphasis has been placed on measuring and knowing outcomes, at either a system level or an individual level. The concept of value-based care, with value being defined by quality and cost, has come to affect both reimbursement and reputation, which makes measurement of outcomes an integral part of modern-day health care. This chapter is intended to introduce the field of outcome measurement; the concepts involved will be defined in general terms, and then some measures used to evaluate care given to patients with colorectal disease who are undergoing colorectal surgical procedures will be discussed.

Types of Outcome Measures

Structural Measures

Structural measures are aimed at human and physical resources used in the delivery of health care. These measures reflect the setting or system in which the care is delivered. Examples of structural measures include the presence of an adequately resourced specialized unit for care of patients with special needs (such as a closed intensive care unit); the volume of procedures performed (such as pancreatectomy); and the number of faculty with advanced training (such as colorectal surgical certification). Thus structural measures used to assess the quality of colorectal surgery at an institution might include the presence of a fully equipped laparoscopic operating suite, access to advanced laparoscopic and endoscopic equipment, and the inclusion of an enterostomal therapist and board-certified colorectal surgeons on staff. When structural measures are used as indicators of quality, the assumption is made that having these structures in place leads to improved care. Thus, they are surrogate or proxy measures for quality. In some situations, relatively poor evidence may exist to support their validity. Even if evidence exists, at best, one can conclude that these structures are associated with improved care, not that they actually improve care. When structural measures are used, they tend to focus on relatively few variables. On the other hand, structural indicators often can be obtained easily and inexpensively from administrative data. They usually can reliably identify differences among groups of providers but not between individual hospitals or surgeons. One of the major limitations of these measures is that they do not report outcomes and usually are not actionable. In other words, suboptimal care is not identified, and thus strategies to improve the process cannot be implemented.

Process Measures

Process measures assess the quality of procedures and activities involved in delivering health care. Process measures are often used to assess compliance with guideline recommendations. For instance, rather than assessing the rate of surgical site infections, the appropriate use of antibiotics, intraoperative warming, and skin preparation are often measured and reported. Although these outcome measures are more clinically relevant than structural measures, they still are proxy measures. An assumption is made that by being compliant with these processes, the clinical outcome of concern will be high. However, this assumption should only be made if strong evidence supports it. If not, these measures may lack face validity and fail to predict clinical outcomes. Process measures are often used in quality initiatives because they are actionable. For instance, if an institution was trying to decrease its surgical site infection rate, measuring compliance with antibiotic prophylaxis, intraoperative warming, and use of skin preparation might be helpful in identifying which area requires attention to improve outcomes. Compared with structural indicators, data on process measures are more expensive to collect. Historically, process measures have been used more frequently in medicine to assess quality, whereas surgeons have tended to measure clinical outcomes to assess quality of care. That approach likely is changing because systemwide assessment of quality is now being performed more often.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here