DEFINITIONS

Key terms related to outcomes and their measurement and management include outcomes, indicators, outcomes management, outcomes research, nursing outcomes research, and risk adjustment. Simply put, an outcome is the result or results obtained from the efforts to accomplish a goal (Huber & Oermann, 1998). When most nurses consider outcomes, they think of the consequences of a health care intervention or treatment. The term outcomes has also been defined as the conditions in patients and others that health care delivery aims to achieve (Peters, 1995). Donabedian (1985) described outcomes as changes in the actual or potential health status of individuals, groups, or communities.

Indicators are “valid and reliable measures related to performance” (Oermann & Huber, 1999, p. 41). They are the specific tools used to make quality visible to stakeholders in health care. Outcomes are measured (quantified) by observing or describing indicators. Recently, the Agency for Healthcare Research and Quality (AHRQ, n. d.) developed a set of three broad categories of desirable attributes of a quality indicator: (1) importance; (2) scientific soundness, including clinical logic and measurement properties; and (3) feasibility.

Because quality is so important yet can be so elusive to define, a variety of accrediting and regulating bodies and a number of trade and professional associations, as well as health care quality assessment organizations (sometimes in alliances), have developed standardized health care performance indicator data sets to measure outcomes. For example, the AHRQ has developed a series of Patient Safety Indicators (PSIs) that include eighteen preventable adverse events and complications that patients may experience in their contacts with the health care system. A recent study explored using the AHRQ’s PSIs to identify nursing-specific opportunities to improve care (Zrelak et al., 2012).

There are growing efforts to develop indicator sets involving nursing-specific or nursing-sensitive outcomes in an attempt to quantify nursing’s contribution to quality and safety (Doran et al., 2011; Loan et al., 2011; McGillis-Hall, 2002; Naranjo-Gil, 2009). For example, the American Nurses Association (ANA) developed the National Database of Nursing Quality Indicators (NDNQI) based on their Nursing Quality Indicators initiative (ANA, 1996; NDNQI, n.d.). According to the ANA, outcome measures or indicators measure how nursing care affects clients (e.g., urinary tract infection incidence 72 hours after admission as an indicator of nosocomial infections potentially related to nursing care).

Indicators are used to measure all three of Donabedian’s (1985) aspects of quality: structure, process, and outcomes. Donabedian’s framework is useful to understand the relationship between outcomes and the structure and processes that have produced them. This suggests that nurse managers and leaders should attend to structure and process factors as precursors to patient outcomes (Donabedian, 2005).

Outcomes management, as originally described by Ellwood (1988), is a process used to assist managers and others make rational patient care–oriented decisions based on what is known about the effect of those choices on patient outcomes. To understand outcomes, the entire care process needs to be carefully examined, and variations must be analyzed. Outcomes management is defined as “a multidisciplinary process designed to provide quality health care, decrease fragmentation, enhance outcomes, and constrain costs. The core idea of outcomes management is the use of care process activities to improve outcomes” (Huber & Oermann, 1998, p. 4).

Outcomes research is a field (or subfield) in health services research that examines the extent to which services achieve the goals of health care. What makes outcomes research distinct from other bodies of research that examine end points in patients (i.e., much clinically-oriented research) is that outcomes researchers seek to tease out the effects of patient-level care and systems-level environments from the background demographic, psychosocial, and clinical characteristics of patients as influences on end points. The purpose is to understand which patients or clients fare well and which do not in relation to treatments selected and/or the organizational context of care delivery (Kane, 2006; Mitchell et al., 1998). An example of a provider characteristic that might be investigated as a predictor of patient outcomes might be the professional background of providers (e.g., physicians versus advanced practice nurses; RNs versus LPNs/LVNs).

Nursing outcomes research is a subspecialty within the larger field of health outcomes research that focuses on determining the effect of different contexts and conditions, related specifically to nurses and nursing care, on the health status of patients. Nursing outcomes researchers are interested in the structures or management strategies for nursing care delivery, as well as the mix of health care workers best equipped to care for them. Other types of outcomes research are intended to determine the types of patients who benefit most from certain nursing interventions.

OUTCOMES MANAGEMENT

The process of managing outcomes includes the following five steps:

In managing outcomes, the information derived from measuring client outcomes is collected, trends are identified, variances are examined, and appropriate care needs are determined to improve care to an individual, group, or population. Goals of this process include quality improvement and risk reduction. Variance analysis is one outcomes management tool. A variance is a deviation from what is expected. For nurses, this may mean a departure from the anticipated clinical trajectory. Variances may be positive or negative but are most useful for trends analysis.

Outcomes research and measurement examines the effectiveness of nursing care in improving client outcomes. Outcomes data and information about factors or approaches that promote favorable outcomes can help nurses assist clients and their families in meeting health needs and care needs across the continuum of care. Reading outcomes research can also help nurses to locate and select interventions that are the most useful in accomplishing the desired improvement in the client’s health status. Identifying the most effective interventions can provide invaluable information for patient self-management (Oermann & Huber, 1997, 1999).

As in any area of clinical care or the management of health services, ideally practice is at least partially guided by research evidence. Although outcomes research has a great deal in common with other forms of research, it involves some special elements. In particular, outcomes researchers are especially concerned about understanding “real” differences between expected and observed outcomes and between outcomes on different units, in different institutions, or at different points in time.

Outcomes research can provide key data for managerial decision making to improve quality of care. Data derived from outcomes research can be used to answer the following types of questions:

Although the answers to each of these questions depend on individual and institutional contexts and economic considerations, data from outcomes research can be used to inform decision making.

INFLUENCES ON OUTCOMES

It is critical that all consumers of outcomes data, including managers, understand how to interpret measures. Clinicians and other workers attempt to foster positive patient outcomes and avoid negative ones. However, the specific treatments delivered to patients are only one factor influencing how well patients do. A model of factors influencing outcomes is useful as a guide for managers. Kane (2006) summarized the factors influencing outcomes and expressed this in the form of a mathematical “function” as follows:

Nurses and nurse leaders are obviously most interested in the effects of treatment on outcomes. For nurses, the focus is usually the process of nursing care on outcomes, but that often encompasses the actions of the entire multidisciplinary health team. However, correctly interpreting health outcomes data across settings or providers (whether in practice or in research) and attributing differences and outcomes to the right causes or sources requires attention to two major challenges. The first lies in ensuring that consistent definitions and data collection processes have been identified, and accurate measures of the phenomena of interest are used. This includes the outcomes, treatment, and any other risk factors thought to influence outcomes. The second challenge, shared with all research dealing with dependent variables influenced by many factors, is that of risk adjustment (Iezzoni, 2003). Risk adjustment involves accounting for patient factors, the intrinsic risks that a patient brings to the health care encounter in the form of clinical and/or demographic factors, before drawing conclusions about the meaning of different values for indicators. Comparisons of outcomes across settings or time periods are meaningful only when potentially important differences in the characteristics of patients involved are taken into account.

MEASUREMENT OF OUTCOMES

Jennings and colleagues (1999) have presented a framework classifying outcome indicators into three categories by stakeholder perspective: patient-focused, provider-focused, and organization-focused. Patient-focused outcomes can include such indicators as disease status, symptom experience, or pain. Other outcomes indicators incorporate a broader impact of disease and its management on clients’ lives. These perceptual outcomes include quality of life, functional status, health status, and patient satisfaction. There are also provider and organizational outcomes. Provider-focused outcomes include such phenomena as nurse burnout, turnover, and job satisfaction. Organization-focused outcomes may include patient or provider outcomes that are aggregated to the organizational level, such as rates of hospital-wide inpatient or 30-day mortality, errors, and other adverse events. Cost indicators are commonly measured at the organizational (hospital-wide) or unit level. A number of health system level outcomes are also receiving increased attention and include measures of successful movements of patients across settings (i.e., care transitions) (Naylor et al., 2011) and readmission rates (Epstein, 2009; van Walraven et al., 2011).