Networkcentric Healthcare and the Entry Point into the Network

Introduction

The concept of e-health gains rapid and widespread international acceptance as the most practical means of reducing burgeoning healthcare costs, improving healthcare delivery, and reducing medical errors. However, due to profit-maximizing forces controlling healthcare, the majority of e-based systems are characterized by non-existent or marginal compatibility leading to platform-centricity that is, a large number of individual information platforms incapable of integrated, collaborative functions. While such systems provide excellent service within limited range healthcare operations (such as hospital groups, insurance companies, or local healthcare delivery services), chaos exists at the level of nationwide or international activities. As a result, despite intense efforts, introduction of e-health doctrine has minimal impact on reduction of healthcare costs. Based on their previous work, the authors present the doctrine of network-centric health care operations that assures unimpeded flow and dissemination of fully compatible, high quality, and operation-relevant healthcare information and knowledge within the Worldwide Healthcare Information Grid (WHIG). In similarity to network-centric concepts developed and used by the armed forces of several nations, practical implementation of WHIG, consisting of interconnected entry portals, nodes, and telecommunication infrastructure, will result in enhanced administrative efficiency, better resource allocation, higher responsiveness to healthcare crises, and—most importantly—improved delivery of healthcare services worldwide.

BACKGROUND: CURRENT ISSUES OF E-HEALTHCARE

Major shifts in political and economical structure of the world that took place in the 20th century were instrumental in focusing global attention on healthcare and its importance in maintaining stability and growth of nations. At the same time, the cost and complexities of national and global healthcare operations became increasingly apparent (World Health Organization Report, 2000, 2004). In order to be efficient, healthcare providers and administrators became progressively more dependent on a broad range of information and knowledge that spans the spectrum stretching from purely clinical facts to the characteristics of local economies, politics, or geography. Consequent to the elevating demand for knowledge is the flood of a wide variety of uncoordinated data and information that emerges from multiple and equally uncoordinated sources (von Lubitz & Wickramasinghe, 2005b, 2005c). It has been hoped that vigorous use of IC2T (Information/Computer/ Communications Technology) will, in similarity to some forms of business operations, obviate the growing chaos of global healthcare. While IC2T changed many aspects of medicine, the explosive growth of worldwide healthcare costs indicates that a mere introduction of advanced technology does not solve the problem (Fernandez, 2002: von Lubitz & Wickramasinghe, 2005). The quest for financial rewards provided by the lucrative healthcare markets of the Western world led to a plethora of dissonant healthcare platforms (e.g., electronic health records) that operate well within circumscribed (regional) networks but fail to provide a unified national or international service (Banjeri, 2004; Olutimayin, 2002; Onen, 2004). There is a striking lack of standards that would permit seamless interaction or even fusion of nonhealthcare (e.g., economy or local politics) and healthcare knowledge creation and management resources. The “inward” concentration of the Western societies on their own issues causes progressive growth of technology barriers between the West and the less developed countries, while the essentially philanthropic efforts to address massive healthcare problems of the latter continues to concentrate on “pretechnological” and often strikingly inefficient approaches (Banjeri, 2004; Olutimayin, 2002).

Thus, despite the massive amount of information that is available to healthcare providers and administrators, despite availability of technologies that, theoretically at least, should act as facilitators and disseminators, the practical side of access to, and the use and administration of healthcare are characterized by increasing disparity, cost, and burgeoning chaos (Larson, 2004). Solutions to many of these acute and disturbing problems may be found in the recent approach chosen by the defence establishments of many countries to the information needs of the battlefield and to the modern, highly dynamic combat operations (von Lubitz & Wickramasinghe, 2005a).

Doctrine of network-centric healthcare operations

Our previous publications (von Lubitz & Wick-ramasinghe, 2005a, 2005b, 2005c) discussed the general principles and applicability of the military network-centric operations concept and its adaptation to modern worldwide healthcare activities. Network-centric healthcare operations are physically facilitated by the World Healthcare Information Grid (WHIG)—a multidimensional communications network connecting primary information collecting sources (sensors) with information processing, manipulating, and disseminating nodes. The nodes also serve as knowledge gathering, transforming, generating, and disseminating centres (Figure 1).

In similarity to the already proved attributes of network-centric military operations (Cebrowski & Garstka, 1998) of which, at the simplest level, the command centre of a joint naval task force is the simplest example and the execution of Operation Iraqi Freedom probably the most complex one, healthcare activities are characterized by multidirectional and unrestricted flow of multispectral data (von Lubitz & Wickramasinghe, 2005b, 2005c). All data, information, and node generated knowledge are characterized by fully compatible formats and standards that allow automated meshing, manipulation, and reconfiguration. Essentially, network-centric healthcare operations are based on the principles of high order network computing, where the WHIG serves as a rapid distribution system, and the nodes as the sophisticated processing centres that function not only as data/information/knowledge generating elements but also as DSS/ESS platforms providing high level, query-sensitive networkwide outputs. The nodes are also capable of extracting and analyzing data and information from healthcare-relevant sensors and electronic data sources (e.g., financial, political, military, geological, law enforcement, infrastructure level, etc.) and mesh these with the relevant biomedical elements. Incorporation of external information in healthcare operations provides readily available, rich, and necessary background that has, typically, a highly significant bearing on the success of activities that are either planned or conducted within the strict healthcare domain. The complications resulting either from the failure to include elements external to the essential healthcare activities or consequent to the exclusion caused by incompatible resource platforms have been amply demonstrated by major difficulties encountered during relief operations following tsunami-mediated destruction in December 2004.

Figure 1. Schematic diagram of a WHIG segment

Schematic diagram of a WHIG segment

 

Sensors feed raw data/information into the network through network-distributed portals. Likewise, data, information, and knowledge queries enter through portals as well. The latter provide entry level security screening and sorting/ routing. Subsequent manipulation, classification, and transformation into information/pertinent knowledge is executed by interconnected nodes. Whenever required, each node can access information/knowledge existing within non-WHIG networks and databases and compare/merge the contents with the contents existing within the WHIG. While portals are associated with the nodes, implementation of ASP philosophy allows reaching the portal from anywhere within the WHIG.

Figure 2. Integrated entry portal/node

Integrated entry portal/node

In addition to functioning as data/information/ knowledge generating/manipulating/disseminating centres, the nodes also serve as the network points of entry (entry portals, Figure 2). However, contrary to the classical Web portal, where the client determines the information gathering path (O’Brien, 2004), the WHIG portal provides automated query classification, direction, and integration functions. Its operations are fuzzy logic-based, and the principal function of the WHIG portal is that of a “sorting/distribution station” which distributes the original query throughout the entire WHIG and collects and weighs the relevant outputs generated by multinodal analysis of the available resources. As the final step, the portal assigns the relevance level of the cumulative output, and provides automated pathways toward its further refinement. The WHIG portal operates thus not only as an entry point but also as either redirection station or WHIG exit site. Some of the functions of the WHIG portal are exemplified by the response to a hypothetical NGO query requiring decision support on the conduct of healthcare activities within the scope of a humanitarian relief operation in a costal region of “State X.” The query will be automatically distributed within the network and the response will (equally automatically) provide multifaceted analysis of the essential medical needs of the affected population (e.g., most threatening diseases, the type and quantity of the required vaccines, need for other pharmaceuticals, tenting, water supplies, etc.). However, the response will also provide information on the local infrastructure and its nature and quality (e.g., air/sea port off loading/storage capacity, availability of beaches as the off-loading sites, capacity of local healthcare human and physical resources, quality and distribution density of roads/railways/means of transport, etc.), whether as an adverse factor, political stability/law enforcement efficiency within the region as a factor influencing distribution of aid, or movement of support teams. Clearly, even within such a simplified example, the range and complexity of factors that may significantly (and adversely) affect only one of many critical elements within a major relief operation is strikingly large. Correspondingly, the need for germane information/knowledge is equally substantial. Yet, due to the prevailing platform-centricity, despite the existence of such information, its dispersal within several, largely incompatible, systems makes it essentially inaccessible. Moreover, its retrieval demands clear awareness of the need followed by human-based/human guided search and extraction. Consequently, in situations of stress or in environments that pose acute demand for a wide range of simultaneous responses, the potential for major errors of omission and commission increases dramatically. A classical chain of such errors can be seen, for example, in the response to the events immediately preceding the destruction of World Trade Center in September 2001 (National Commission on Terrorist Attacks on the United States, 2004).

Data, information, or queries from WHIG enter through the portal where they are subjected to security/standards/protocol screening then transfer to the manipulation site (DM). The latter provides detailed sorting and redirection via intra and extra nets, and/or Internet/Web to other locations within the node, for example, patient records, information storage sites, analysis and knowledge generating sites, and so forth (unidirectional arrows). All sites within the node are capable of multidirectional communication (not indicated for the sake of clarity). Their output is transmitted to the knowledge manipulation and generation site which, in turn, generates final output stored within the node and also disseminated throughout the network (Out). If needed, the node can distribute additional WHIG-wide queries. Replies are collected, manipulated at the KM level, and incorporated into the final node output. Although neither the portal nor individual functional aspects of the node need be collocated, their operations are conducted as a single, self-contained unit; that is, none of the constituting elements can participate individually in the functions of another node. Self-containment of each node adds to its security and reduces the risk of inadvertent networkwide dissemination of integrity-compromising factors (e.g., viruses, spurious data, etc.).

Future trends: operational theory of network-centric activities

The operational philosophy of network-centric healthcare operations is based on the principles of Boyd’s (OODA) Loop (Boyd, 1987; von Lubitz & Wickramasinghe, 2005a, 2005b, 2005c) that defines the nature and the sequence of interactions with dynamic, rapidly changing environments characterized by a high degree of structural and event complexity. Accordingly to Boyd, each complex action can be subdivided into a series of consecutive cycles, loops, with the preceding cycle strongly influencing the initial stages of the following. Each revolution (cycle) of the Loop comprises four stages: observation, orientation, determination, and action. During the observation stage, all inputs describing the action environment are collected and organized into coherent entities. At the orientation stage, the organized data are converted into meaningful information that provides as complete image of the operational environment as possible based on the totality of the existing information. At this stage the weaknesses of the opposition are detected, and the centre of the future action determined. During the determination phase, the hypothesis, that is, the plan to respond to the pressure exercised by the operation environment, is formulated. The Hypothesis defines the plan of action, the required strength and nature of the response, its precise location, timing and duration, and so forth. During the Action phase, the Hypothesis is tested: the formulated plan is implemented and its results (and the consequent response of the action environment/opposition) set off the next revolution of the Loop—the new observation stage is initiated. Clearly, the nature of action determines the intervals between the stages.

Originally Boyd’s Loop had been created as a tool facilitating aerial combat, where each individual stage was extremely brief (milliseconds). Nonetheless, the principles of the Loop can be applied to virtually any rapidly evolving environment. Moreover, Boyd’s Loop helps to understand the critical role of the mistakes made during the initial data collection (e.g., selective or biased selection, rejection of non-conforming data as necessarily false, etc.) at the observation stage and their subsequent analysis (subjective analysis based on preconceived notions, influence of personal bias, inflexibility, etc.) at the orientation stage.

Errors made at these two stages influence the following two. Thus, at each subsequent cycle, error correction demands increasingly larger resources and removes them from where they should be otherwise committed—at the centre of action. Uncorrected errors compound at each new revolution of the Loop and exponentially increase the chance of failure. Probably the best example of Loop failure was the disastrous response of state and federal authorities to Hurricane Katrina in August 2005, while the response to Hurricane Wilma (its shortcomings notwithstanding) shows how application of Boyd’s Loop-based thinking can lead to positive outcomes in situations demanding flexible, ongoing, and dynamic response to the continuously but unpredictably changing operational environment.

Clearly, to assure efficiency of action, the interval separating each individual stage of the Loop must be as short as possible, particularly when interacting with highly fluid, ultracomplex systems such as military or healthcare information. Here, the demand is not only on rapid, reliable sampling of the environment but also on a very high degree of automation at the level of multi-source data collection, analysis, manipulation, and classification into larger information/germane knowledge entities.

Contrary to the prevalent platform-centric operations, network-centricity allows vast increase in sampling speed, range, and data manipulation speed. Consequently, decision supporting outputs of the network are faster, more situation/operational environment-relevant and, most importantly, allow robustly elevated rate of stimulus-response cycle (operations “inside the Loop”). Moreover, by increasing reaction relevance and speed, network-centric operations facilitate goal-oriented manipulation of the operational environment and also increase both the level (accuracy) and predictive range of responses to environment induced pressures. Military benefits of such operations have been frequently demonstrated. However, the acceptance of Boyd’s (OODA) Loop principles in the civilian world (e.g., global financial/banking operations, lean manufacturing, just-in-time supply chains, etc.) led to demonstrable gains in efficiency and productivity as well.

conclusion

The preceding description is, of necessity, vastly simplified. Yet, the existence and highly efficient use of the network-centric approach to military operations has already resulted in the significant enhancement of the C3I (Command, Control, Communications, and Intelligence) concept (Alberts, Garstka & Stein, 2000; Department of Defense, 2001). The most palpable consequences of network-centricity in warfare are increased efficiency in the use of available resources, application of resources appropriate to the operational environment, reduction of casualties, and transformation of conflict whose face changes rapidly from aggression by overwhelming force to prevention and de-escalation. Similar principles can be applied to healthcare operations, particularly in view of the already existing major technological components of the WHIG. However, in order to implement network-centricity in healthcare, a major conceptual transformation is required.

Presently, the ruling healthcare doctrine is that of e-health, which while supporting implementation of IC2T, promotes development of individual, largely noncollaborative (particularly in the global sense) systems. While there is no doubt that the existence of such systems (for example, electronic patient records) facilitates many aspects of healthcare delivery and administration, their effect is predominantly regional. On a larger scale (national, international) most of these platforms function in isolation and major (predominantly through human interaction) effort is needed in order to extract relevant information and convert it into pertinent knowledge.

Transition to the network-centric doctrine of healthcare will greatly facilitate interoperability of multiple electronic healthcare platforms and enhance their usefulness in the broadest sense of global health. There is also no doubt that, similar to other domains in which a network-centric approach has been successfully implemented, the consequence of the proposed doctrine will be improvement of access, better delivery, increased efficiency in the use of resources, accompanied by the concomitant reduction of presently staggering expenditure.

key terms

E-Health: The application of technology, primarily Internet based technology, to facilitate in the delivery of healthcare.

Germane Knowledge: The relevant and critical knowledge, or contextualized information, required to enhance a particular decision.

Information Symmetry: The gap between the available information between two entities.

Network-Centric: In contrast to a platform-centric approach, a network-centric approach is made up of interconnecting technology grids that enable and facilitate the seamless transfer of data, information and knowledge.

OODA Loop: A framework developed by John Boyd that facilitates rapid decision making in dynamic, rapidly changing environments characterized by a high degree of structural and event complexity. Each complex action can be subdivided into a series of consecutive cycles, while each revolution (cycle) of the Loop comprises of four stages: Observation, Orientation, Determination, and Action.

Platform-Centric: Based on and exploiting the exclusive properties of an employed system or specific technology platform. Useful on a small scale but does not enable seamless transferring of information and knowledge across platforms or systems.

World Healthcare Information Grid (WHIG): The technology backbone of network-centric healthcare operations, a network of interconnecting technology grids that together contain all the necessary information for effective and efficient healthcare delivery.

Next post:

Previous post: