Decomposition-based multiobjective evolutionary algorithm for community detection in dynamic social networks.

Community structure is one of the most important properties in social networks. In dynamic networks, there are two conflicting criteria that need to be considered. One is the snapshot quality, which evaluates the quality of the community partitions at the current time step. The other is the temporal cost, which evaluates the difference between communities at different time steps. In this paper, we propose a decomposition-based multiobjective community detection algorithm to simultaneously optimize these two objectives to reveal community structure and its evolution in dynamic networks. It employs the framework of multiobjective evolutionary algorithm based on decomposition to simultaneously optimize the modularity and normalized mutual information, which quantitatively measure the quality of the community partitions and temporal cost, respectively. A local search strategy dealing with the problem-specific knowledge is incorporated to improve the effectiveness of the new algorithm. Experiments on computer-generated and real-world networks demonstrate that the proposed algorithm can not only find community structure and capture community evolution more accurately, but also be steadier than the two compared algorithms.

Measuring racial segregation in health system networks using the dissimilarity index.

Racial disparities in the end-of-life treatment of patients are a well observed fact of the U.S. healthcare system. Less is known about how the physicians treating patients at the end-of-life influence the care received. Social networks have been widely used to study interactions with the healthcare system using physician patient-sharing networks. In this paper, we propose an extension of the dissimilarity index (DI), classically used to study geographic racial segregation, to study differences in patient care patterns in the healthcare system. Using the proposed measure, we quantify the unevenness of referrals (sharing) by physicians in a given region by their patients' race and how this relates to the treatments they receive at the end-of-life in a cohort of Medicare fee-for-service patients with Alzheimer's disease and related dementias. We apply the measure nationwide to physician patient-sharing networks, and in a sub-study comparing four regions with similar racial distribution, Washington, DC, Greenville, NC, Columbus, GA, and Meridian, MS. We show that among regions with similar racial distribution, a large dissimilarity index in a region (Washington, DC DI = 0.86 vs. Meridian, MS DI = 0.55), which corresponds to more distinct referral networks for black and white patients by the same physician, is correlated with black patients with Alzheimer's disease and related dementias receiving more aggressive care at the end-of-life (including ICU and ventilator use), and less aggressive quality care (early hospice care).

A methodology for reliability analysis in health networks.

A reliability model for a health care domain based on requirement analysis at the early stage of design of regional health network (RHN) is introduced. RHNs are considered as systems supporting the services provided by health units, hospitals, and the regional authority. Reliability assessment in health care domain constitutes a field-of-quality assessment for RHN. A novel approach for predicting system reliability in the early stage of designing RHN systems is presented in this paper. The uppermost scope is to identify the critical processes of an RHN system prior to its implementation. In the methodology, Unified Modeling Language activity diagrams are used to identify megaprocesses at regional level and the customer behavior model graph (CBMG) to describe the states transitions of the processes. CBMG is annotated with: 1) the reliability of each component state and 2) the transition probabilities between states within the scope of the life cycle of the process. A stochastic reliability model (Markov model) is applied to predict the reliability of the business process as well as to identify the critical states and compare them with other processes to reveal the most critical ones. The ultimate benefit of the applied methodology is the design of more reliable components in an RHN system. The innovation of the approach of reliability modeling lies with the analysis of severity classes of failures and the application of stochastic modeling using discrete-time Markov chain in RHNs.

Networks: better language required.

While networks have proliferated in literature and in our health system, our day-to-day language has not kept up in sophistication. This commentary builds on the work presented by Huerta, Casebeer and VanderPlaat to further explore the language of networks. An expansion of our "network literacy" needs to be reflected in a broader vocabulary for describing particular networks and identifying patterns of relationship that are not appropriately labelled a network. Dimensions along which network managers often understand and place their networks are reported, and the implications of various network images are considered. The distinction between the image of a fishing net and that of a spider's web explores the difference between networks as system substrates and as centres. A moratorium on the term network is called for, to ensure an expanded vocabulary is applied to emerging new relationship patterns between or independent of organizations.

Using networks to enhance health services delivery: perspectives, paradoxes and propositions.

There is a growing need to better understand and address the consequences of an increasing reliance on networks used to enhance health services delivery. Networks seem to have emerged as the definitive solution for tackling complex healthcare problems together that we have not been able to adequately address separately. Emphasizing the collective and the collaborative, networks are assumed to address healthcare issues in ways that are superior to previous service-delivery models. While this assumption would appear to be sound theoretically, we have little empirical information available to actually understand what networks are, what they do and whether they achieve their stated goals--truly making a difference in the delivery of care and the maintenance of health. With a diversity of networks within Canada focused on health services delivery, this paper offers a multi-dimensional framework for conceptualizing how these complex inter-organizational relationships generate both challenges and opportunities. We identify six paradoxes that the networks create when used to enhance the delivery of health services and posit several propositions concerning the evaluative work that needs to be done to enhance our understanding of and confidence in this inter-organizational form. Unless these paradoxes are adequately recognized and addressed, the value and costs associated with developing and using networks in healthcare contexts will remain unclear at best. Given the broad interest in and use of networks proliferating in health-related arenas, it is time to amass the evidence and than align the perspectives. Are networks here to stay in healthcare because they make a difference or because we got tired of talking about the need for greater collaboration and so gave it a new name and frame? At the very least, it will be important to build on what we have already learned through research into collaboration in healthcare and related fields, and even more critical to be mindful of the pitfalls and possibilities of using networks as the solution of choice as we move forward.

Health services delivery networks: what do we know and where should we be headed?

Networks of collaborating organizations have become critical mechanisms for the effective delivery of healthcare and related human services. Despite their importance, there is much about health networks that is not understood. The article by Huerta, Casebeer and VanderPlaat is an effort to discuss the importance of health services delivery networks and to point out ways in which such networks might best be studied. Their article offers a number of useful and interesting ideas for both practice and research. Many of these ideas are not, however, well organized, integrated or fully developed. This commentary provides a critique of their work, while offering some of our own suggestions about how the study of health delivery networks might be advanced.

A taxonomy of health networks and systems: bringing order out of chaos.

OBJECTIVE: To use existing theory and data for empirical development of a taxonomy that identifies clusters of organizations sharing common strategic/structural features. DATA SOURCES: Data from the 1994 and 1995 American Hospital Association Annual Surveys, which provide extensive data on hospital involvement in hospital-led health networks and systems. STUDY DESIGN: Theories of organization behavior and industrial organization economics were used to identify three strategic/structural dimensions: differentiation, which refers to the number of different products/services along a healthcare continuum; integration, which refers to mechanisms used to achieve unity of effort across organizational components; and centralization, which relates to the extent to which activities take place at centralized versus dispersed locations. These dimensions were applied to three components of the health service/product continuum: hospital services, physician arrangements, and provider-based insurance activities. DATA EXTRACTION METHODS: We identified 295 health systems and 274 health networks across the United States in 1994, and 297 health systems and 306 health networks in 1995 using AHA data. Empirical measures aggregated individual hospital data to the health network and system level. PRINCIPAL FINDINGS: We identified a reliable, internally valid, and stable four-cluster solution for health networks and a five-cluster solution for health systems. We found that differentiation and centralization were particularly important in distinguishing unique clusters of organizations. High differentiation typically occurred with low centralization, which suggests that a broader scope of activity is more difficult to centrally coordinate. Integration was also important, but we found that health networks and systems typically engaged in both ownership-based and contractual-based integration or they were not integrated at all. CONCLUSIONS: Overall, we were able to classify approximately 70 percent of hospital-led health networks and 90 percent of hospital-led health systems into well-defined organizational clusters. Given the widespread perception that organizational change in healthcare has been chaotic, our research suggests that important and meaningful similarities exist across many evolving organizations. The resulting taxonomy provides a new lexicon for researchers, policymakers, and healthcare executives for characterizing key strategic and structural features of evolving organizations. The taxonomy also provides a framework for future inquiry about the relationships between organizational strategy, structure, and performance, and for assessing policy issues, such as Medicare Provider Sponsored Organizations, antitrust, and insurance regulation.

Physician equity alliances: attractive alternatives to PHOs.

Physician equity alliances are becoming attractive alternatives to PHOs as integrative models for partnering with physicians, securing managed care contracts and increasing revenue. Unlike many PHOs, these alliances provide mechanisms for asset integration and long-term relationships along with utilization management, sophisticated information systems, access to capital and opportunities for physicians to integrate clinically. There are six major types of physician equity alliances: majority physician-owned, clinic without walls, health system joint venture, publicly held physician practice management company, specialty network, and venture capital. The type of alliance that a physician group practice ultimately develops depends on vision, values, method of capitalization, initial organizer of the alliance, level of involvement of physicians in business issues, corporate structure desired, and characteristics of the managed care market in which the alliance will operate.