Critical pathway effectiveness: assessing the impact of patient, hospital care, and pathway characteristics using qualitative comparative analysis

Health Services Research, April, 2005 by Sydney M. Dy, Pushkal Garg, Dorothy Nyberg, Patricia B. Dawson, Peter J. Pronovost, Laura Morlock, Haya Rubin, Albert W. Wu

Critical pathways are being implemented in many U.S. hospitals, primarily to improve the efficiency of hospital care while maintaining or improving quality (Pearson et al. 2001). Critical pathways are structured multidisciplinary care plans that detail essential steps in the care of patients with a specific clinical problem (Campbell et al. 1998). They display goals for patients and provide the ideal sequence and timing of staff actions to achieve those goals efficiently (Pearson, Goulart-Fisher, and Lee 1995). Largely because of pressure from payers to reduce length of stay, pathways have been developed and implemented for a large variety of procedures and illnesses where there appears to be excess variation in length of stay (Campbell et al. 1998).

However, several systematic reviews of the many uncontrolled and few, mostly small, randomized, controlled trials have not supported this widespread use. These reviews conclude that evidence for reducing length of stay and costs is weak because of the preponderance of retrospective studies that use historical controls and do not adjust for risk or secular trends (Conner et al. 2000; Kim et al. 2003; Kwan and Sandercock 2003). Moreover, the use of pathways may have risks. Although most studies do not measure patient-centered outcomes, the review in patients with stroke found that patient satisfaction and quality of life were significantly lower in patients on pathways (Kwan and Sandercock 2003). These reviews also found that results may differ substantially for different types of illnesses and that there was wide variation in, and often insufficient description of, pathway interventions.

Critical pathways require substantial resources to develop, implement, and maintain, and are only one of many potential tools to improve the quality and efficiency of complex medical care. Critical pathway programs often tend to be comprehensive rather than focused, and may target the most common reasons for admission rather than patients for whom the potential for quality improvement is greatest (Darer, Pronovost, and Bass 2002). In particular, the process of pathway development, which includes forming an interdisciplinary team, reviewing data, identifying quality concerns, building consensus, and educating staff on standards, may impact hospital care independent of actual pathway use (Holtzman, Bjerke, and Kane 1998). Efficient investing in critical pathway programs would be facilitated by knowledge of why certain pathways in certain situations succeed while others do not, as well as by understanding the relative importance of the pathway development and implementation processes.

Based on Donabedian's quality assessment model of structure, process, and outcomes (Donabedian 1988), the effectiveness of critical pathways could be affected by characteristics of hospital care, pathways, and patients, and could result from both the process of pathway development and the actual use of the pathway document. We asked the following research question: what patient, hospital care, and critical pathway characteristics are qualitatively associated with pathway effectiveness for reducing length of stay for surgical procedures? We hypothesized that four factors might be associated with effective as opposed to ineffective pathways: (1) length of stay was not already decreasing at the time of pathway implementation; (2) the pathway was the first implemented on a service; (3) patients eligible for effective pathways had lower severity of illness than those eligible for ineffective pathways; and (4) pathway documentation was used more frequently as part of clinical care for effective as compared with ineffective pathways.

METHODS

Study Design and Setting

We conducted a qualitative, retrospective cohort study of a group of critical pathways using qualitative comparative analysis (QCA) (Ragin 1999). QCA is a method in comparative case-oriented research (Ragin 1987) for studying a small-to-moderate number of cases in which a specific outcome has occurred, compared with those where it has not. Using principles of both in-depth case-oriented (qualitative) and variable-oriented (quantitative) research, this method helps to identify cross-case commonalities and differences to construct a general explanation of how an outcome occurs. QCA is based on Boolean algebra and on examining the minimum combination of variables that may result in either the presence or the absence of the outcome. Based on a conceptual model, the method identifies different logical combinations of variables, using AND or OR expressions, that might be necessary and/or sufficient to produce the outcome. In addition, QCA uses the concept of "prime implicants" to test hypotheses about the factors that are associated with the outcome and to eliminate variables that are subsets of these key factors.

We studied critical pathways in the Department of Surgery in a large tertiary care center, The Johns Hopkins Hospital. Pathways were implemented between 1990 and 1996, and covered all major surgeries performed at the hospital. We included all pathways implemented in the study period that applied to a procedure primarily responsible for an admission and had more than 20 patients per year potentially eligible for the pathway. Twenty-six of the 48 surgical critical pathways (54 percent) were eligible for inclusion. Details of some of the pathways (Dardik et al. 1997; Zehr et al. 1998; Pitt et al. 1999; Berenholtz et al. 2001) are described elsewhere.