Health systems delivery is a global term used to define the structures and processes by which health care is provided to individuals and populations. The term generally refers to the collective availability of services rather than to an individual organization by itself, although larger organizations such as academic health science centers may use the term to reflect the extent of their capacity. The features that distinguish health systems from other connected services are their level of differentiation, their extent of centralization, and their degree of integration (Bazzoli, Shortell, Dubbs, Chan, &Kralovec, 1999).Health systems have a single owner and some type of decision-making oversight group, whereas health networks (which also provide an array of services) are more loosely linked and each participating organization maintains its original ownership (Bazzoli et al.).Probably the most significant influence on health systems delivery has been the introduction of managed care, which places restrictions on access and consumption of services and has prompted delivery systems to identify ways to provide a variety of services more efficiently and at a lower cost (Cook, J. A., Ingersoll, & Spitzer, 1999).
Investigations of managed care delivery systems have identified five characteristics common to the service delivery processes associated with these systems—the use of population-based strategies for cost containment, a focus on wellness rather than illness care, the increased influence of consumers on services offered and selected, the interdependence of professionals involved in care, and the use of delivery systems reengineering to improve services (Ingersoll, Spitzer, &Cook, 1999). Comprehensive studies of these processes are limited, with even less information available concerning the impact of systems delivery models on care delivery outcome. Subsumed within the broader classification of health services research, studies pertaining to health systems delivery focus on which collection of services are most effective and efficient for achieving maximum care de-livery outcome. Few large-scale investigations of delivery systems are available, how-ever, with most studies examining specific types of organizations (e.g., hospitals, long-term care, hospice care). Although these studies are not focused on health delivery systems, perse, they provide useful information about the structures and processes that may achieve favorable outcomes in individual or linked organizations.
The organizational characteristics investigated in prior research were examined recently in a comprehensive analysis of the ways in which care delivery systems influence patient safety. The state of the science was reviewed by an Institute of Medicine (IOM)expert panel, which identified four environmental factors that consistently contribute to the quality of care delivered and the patient outcomes seen (Page, 2003). The IOM expert panel described these systems characteristics as sources of threats and labeled them management, workforce, work processes, and organizational culture. Using the literature available, they proposed several safeguards for addressing these systems components and improving patient safety. Included in the recommendations were: developing governing boards that focus on safety, introducing evidence-based management of organizational structures and processes, assuring high levels of leadership ability, providing sufficient staffing, promoting ongoing learning and decision support at the point of care, encouraging interdisciplinary collaboration, creating work designs that promote safety, and achieving an organizational culture that continuously addresses patient safety.
Multisite studies supporting the IOM expert panel’s recommendations have been drawn from acute care, long-term care, and home care settings. Among the studies providing data to support these recommendations were several multisite investigations conducted by nurse researchers—four of which are summarized here. The studies included in this description were selected because of the variables they investigated, their inclusion of institutions from a variety of locations, and their potential application to health delivery systems regardless of size or type of services delivered.
The most commonly measured structure variable in health delivery systems research is nurse resources, with several large-scale investigations exploring the impact of nurse staffing mix and nursing care hours on employee, organizational, and patient outcome.
Three studies of hospitals drawn from across the U.S. have demonstrated consistent evidence of the beneficial effect of registered nurse (RN) care hours on length of stay (Kovner & Gergen, 1998; Needleman, Buerhaus, Mattke, Stewart, & Zelevinsky, 2002), mortality (Aiken, Clarke, Sloane, Sochalski, & Silber, 2002), and adverse events (Aiken, Clarke, Sloane, et al.; Kovner & Green; Needleman et al., 2002).
Comparable findings were seen in a multisite investigation of the best and worst performing nursing homes, although the investigators of this study defined nurse resource variables as an indicator of care delivery process, rather than organizational structure (Anderson, R. A., Hsieh, & Su, 1998). Although a case can be built for defining nurse resources as both a structure and a process variable, simply identifying skill mix or percent of RN hours does little to clarify what actually occurs during the delivery (or process) of care by one type of provider (RN)versus another (licensed practical nurse [LPN] or patient technician). Nonetheless, in this study the greater the number of RNs, the better the patient outcome. Structure variables associated with type of ownership, size of nursing home, and percentage of private-pay residents were not associated with any of the outcomes measured. Few studies have explored the impact of processes of care on care delivery outcome, primarily because this variable is difficult and costly to assess. Care delivery processes evolve over time and change in response to work-group makeup, leader vision, standards used to guide care delivery, and types of patients served.
Monitoring what transpires during the interactions that take place among care providers and between care providers and patients and families requires an understanding of group relationships, individual motivation and need, and the ways in which work gets done. Consequently, studies of organizational processes are inherently complex and difficult to carry out. As a result, employee perceptions are often used as proxy indicators for work-group or leadership behaviors and the processes they use to get work done. For example, nurses are commonly surveyed about their perceptions of ideal and actual nurse leaders or work groups. Favor-able perceptions of both are frequently related to nurse satisfaction and retention.
What processes nurse leaders and work groups use to produce these favorable or unfavorable perceptions are less clear, with most reports describing general categories of behaviors (e.g., inclusiveness, cohesiveness) to denote the characteristics of ideal leaders and group members. How they go about creating an inclusive and cohesive process is unknown. One area of increasing interest in health systems delivery research is the impact of health care teams and team functioning on care delivery outcome. This interest is generated by evidence linking poor interpersonal interactions among team members with health care errors (Ingersoll & Schmitt, 2003). Because the number and makeup of teams varies significantly even within a single institution, measuring the effect of team performance on care delivery outcome is troublesome.
Moreover, the structure of the team, including its hierarchical nature, its placement within the organizational system, and its mission and purpose all contribute to its potential for effectiveness and ultimate impact on care. Consequently, measuring one team’s performance will not necessarily help with understanding what processes result in favorable care delivery outcomes. Ideally, a variety of teams should be monitored to identify differences in the ways the team members work together to achieve a good effect.
Studies of comprehensive health delivery systems are in their infancy, with limited information available from comprehensive multisite investigations of health care organizations. Additional research is needed that focuses on both the structures and the processes that promote favorable outcomes for employees, patients, and organizations. A combination of approaches will be required to achieve this goal, with qualitative methods used for understanding care delivery processes and expectations of providers and quantitative methods for examining causal relationships between organizational structures, processes, and outcomes seen.