Data & Analytics

A distributed coordination platform for home care: analysis, framework and prototype

Description
A distributed coordination platform for home care: analysis, framework and prototype
Published
of 17
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  International Journal of Medical Informatics (2005)  74 , 809—825 A distributed coordination platform for homecare: analysis, framework and prototype Nathalie Bricon-Souf  a , ∗ , Franc¸oise Anceaux b , Nadia Bennani b ,Eric Dufresne a , Ludivine Watbled a a CERIM, Facult´e de M´edecine, Universit´e de Lille 2, 1 Place de Verdun, 59 045 Lille, France b CNRS, LAMIH, UVHC, Le Mont Houy, 59313 Valenciennes, Cedex 9, France Received 10 January 2005; accepted 22 March 2005 KEYWORDS Medical informatics;Home care;Coordination;Communication Summary  Good cooperation between health care (HC) professionals, patient, andfamily is indispensable during homecare as mentioned in reports and analyses fromdifferent countries. In a French National project named coordination for the qualityof care (COQUAS), we aimed to address the problem of improving such cooperationwith current tools and techniques. We hypothesized that, as in some other domains,a better integration of use and users in informatics systems could improve the use-fulness of the cooperative tool.The first part of this paper is devoted to the cognitive analysis of the home-care process and highlights the requirements which should be met according to thisanalysis. We describe some specific features of asynchronous cooperation and somecommunication issues in the cooperation of HC workers. We then detail the analy-sis of the homecare process: methodology, description of the processes, cognitiveactivity analyses, and of the requirements which flow from this analysis.The second part of this paper proposes a framework and then describes a modu-lar system prototype, designed to take into account these requirements, includingaspects of both cooperation and interoperability. It uses a meta-description of actions and information derived from a cognitive study to build dynamically theinterface settings; it respects the current trend in distributed architecture and usesXML communication of messages, manages complex coordination with a workflowand allows mobile work.The last part of the paper presents the evaluation which has been done with theimplemented prototype, with actual homecare users.© 2005 Elsevier Ireland Ltd. All rights reserved. ∗ Corresponding author. Tel.: +33 320 62 68 29; fax: +33 320 52 10 22. E-mail address:  nsouf@univ-lille2.fr (N. Bricon-Souf).1386-5056/$ — see front matter © 2005 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/j.ijmedinf.2005.03.020  810 N. Bricon-Souf et al. 1. Introduction 1.1. Home care requires cooperation by thehealth care workers Many reports underline the increase in the numberof elderly people and the importance of home careservices in the follow-up of chronic diseases or end-of-life situations. For example, in a study done inCanada [1], Arundel states, ‘‘ Home care programsare an increasingly important component of thecontinuum of health care services ’’ (p. 14), while astudy using Australian life table models reaches thefollowing conclusion on the Australian population(p. 6): ‘‘  A woman at age 65 faces a probability ashigh as 0.46 of using an aged care home for perma-nent care before her death compared with 0.28 for a man at the same age ’’[2]. At the same time, most countries encounter major problems with healthcarecostsandratiosofHealthCare(HC)profession-als. Thus, improving home care as an alternativeto classical hospitalization appears as a one of themain challenges for the 21st century.Most research associated with home careaddresses improvements in tele-surveillance, dis-tant follow-up, and home care maintenance forelderly people or disabled people, using new tech-nologies to increase the communication betweenpatient and HC providers. Helping an individual tobecome a real participant in his own care plan is acurrent challenge in home care. Information Tech-nologies (IT) such as wireless networks or mobiletools allow us to imagine new communication sys-tems between patient and HC providers throughwearable captors [3], easy-to-use interfaces [4,5] and the so-called ‘‘intelligent home’’ [6,7]. Otherprojects like the European project, Topcare [8], areworking on a home care platform for home monitor-ing and treatment.It is also well-recognised that a good coordi-nation of healthcare professionals is an essentialelement in quality health care, particularly whendifferent structures have to organize themselves totreat a patient in a cooperative way. Homecare isno exception: ‘‘ In most western countries, thereare certain dilemmas to overcome within the homehealth care organizations  [ . . . ].  Patients often havedifferent care providers at the same time, whichimplies a need for improved co-operation betweencaregivers within the different home health careorganizations, and also a need for changed rolesand responsibilities .’’ [9] (p. 130). An Americanworkshop on Home Care Technologies for the 21stCenturyreportsthat‘‘ Insufficientknowledge,coor-dination, and communication across medical per-sonnel, organization and agencies ’’ appears in thetop 15 highest rated gap in knowledge base [10]. ACanadian study on the improvement of home careshows that co-ordination is one of the key implica-tions for the decision makers (p. 5): ‘‘ Co-ordinationof services to meet the care plan for home carerecipients is necessary for continuity of care ’’[11].Improvingthecommunicationofinformationisthenimportant to support this coordination: the abovecited Ontario report [11] gives the following advice (p. 5) ‘‘ Better communication among stakehold-ers is required. Communication—–especially dur-ing transition periods—–is inadequate between casemanagers and service providers, resulting in poor continuity of care ’’. In the Canadian report [1]entitled ‘‘An Analysis of Blockage to the Effec-tive Transfers of Clients from Acute Care to homecare’’, three types of factors are determined asmajor points to improve those transfers: (i) for-mal systems, divided into legitimization of therelationship between acute care and home care,access to compatible and/or common informationsystems, and flexible use of resources, (ii) relation-ships and informal networks, (iii) system capacity.It is also underlined (p. 16) that: ‘‘ Many other spe-cific types of barriers were identified in the lit-erature. Perhaps the most important in this final grouping was the barriers created by lack of infor-mation and the challenge of incompatible informa-tion. Despite information systems and technologi-cal advances, basic patient data barriers continueto exist between hospitals and home care ’’. 1.2. The cooperation of the health careworkers within home care has somespecificity Home care represents a particular and difficult con-text of cooperation. Home care takes place on acomplex system organized in four main sectors:COMMUNITY: politicians, patients, hospitals, boardof directors, CONTROL: managers of hospitals, of home care,  . . . , CURE: acute hospitals, physicians,community of physicians, and CARE: nurses, otherprofessionals,  . . . , and a strong collaboration isneededbetween,butalsowithin,eachofthesesec-tors [1]. As shown in Fig. 1, a complex collective home care case involves a number of heteroge-neous participants. It relies on the constitution of a network including the person who asks for thehome care (the patient, his family, the hospital orthe general practitioner), the persons involved inthe logistic implementation (coordinator in chargeof the evaluation of material and human needs,pharmacy) or in the financial aspect of home care(health insurance), the HC team (nurses, practi-tioner, physiotherapist, and so on), and the team  A distributed coordination platform for home care 811 Fig. 1  Home care system organization. involved in the patient’s well-being (family, home-help, and so on) [12].Generally, home care is managed with less infor-mationexchangesthanhospitalcare.Inthecontextof a hospital, even if sometimes difficult, coordi-nation between healthcare workers is helped byfrequent meetings (clinical rounds, corridor discus-sions  . . . ) and by a large number of exchanged docu-ments (medical records, laboratory results, letters, . . . ). In the home care setting, coordination andcooperation are even more complex, albeit lessfrequent and with a lesser quantity of exchangedinformation.As shown in [13], the team is made of diverse people who know each other but who rarely meetin common. The family doctor, the nurse, the phys-iotherapist act when they want and when they can,and interactions between them are poor. Moreover,in hospital services the organization is based upon adivision according to different type of pathologies.In home care, such a division does not exist: thesame persons manage all types of them.Coordination is more often organized by a homecare organization and introduce a new ‘‘role’’:the care coordinator. In France home care is oftensupervised by home care organizations which haveelaborated some tools to support their activities.Some of these tools are oriented to support coop-eration between the care coordinator and the dif-ferent structures (hospital, laboratory, and so on)or between the different health care professionalsproviding the care. However, in a significant pro-portion, home care may be organized by very smallgroups: for example a GP (or a group of GPs) oran independent nurse (or a group of nurses). Inthese cases, the lack of tools supporting activityand cooperation can slow down the initiation or themaintenance of home care.Thus, home care introduces the need for newcooperative activities. Such cooperation is, how-ever, asynchronous; the team members cannotdirectly communicate during task realization or inan informal way during a meeting or a coffee break[14]. Transmission of information between the dif-ferent workers is also elementary, and medicalinformation is frequently minimal. On the oppositeside, the role of the family is enhanced and some-times essential to gather the information from thedifferent parties, as also underlined in [1]: ‘‘ it hasbeen estimated that families and informal care- givers provide approximately 80% of all home caresupports ’’ (p. 8). 1.3. Objective In France, implementation of home care has beenrecognized as an interesting alternative to classicalhospitalization, and cooperation in home care as achallenging field of research. Our research projecthas been funded in a national project namedCOQUAS (French acronym for ‘‘Coordination forthe Quality of Care’’) which involves 15 differentpartners (research laboratories in Health Informat-ics, Computer Science and Psychology; Hospitals;the Social Insurance agency; commercial com-panies, and associations devoted to home care).The different research laboratories have extensiveexperience in the health care domain, and more-over, have a specific approach which consists in theassociation of cognitive ergonomics and computerscience. Indeed, health care activities are bothcognitively very complex, and confidential, ren-dering them difficult to observe, to describe andto analyze. Moreover, users themselves are oftennot aware of the procedures, inferences, reasoningand knowledge which underline their activities.When they describe their ‘‘users’ requirements’’,they generally focus on security, exhaustivenessof data, information quality and structure. Theygenerally consider than their activities and thecontext in which they take place are known by thedesigners. Application specifications mention, atbest, use scenarios and ergonomic requests, butrarely activity descriptions and even less cognitivedescriptions. It seems necessary to better integrateuse and users in such informatics projects in health  812 N. Bricon-Souf et al.care in the same way as has been done in other‘‘risky’’domains(transportation,powerplants . . . ).Nowadays the computerization of the healthcare systems is increasing. It is often done throughheterogeneous applications, leading to numeroustechnical problems such as difficulties with secu-rity and interoperability. In addition to the actualintegration of users, we wished to develop a systemable to adapt to such constraints. One of our indus-trial partners has already commercialized an inter-mediation platform (Rithme [15]) dedicated to sup- port information exchanges between health careprofessionals (hospital/GP, laboratories/GP, etc.),which addresses some of these problems (security,notification of medical events to medical profes-sionals for example). Thus, we decided to focuson the following main requirements of the system:interoperability and adaptation of the cognitiveactivity knowledge.We intend to define new services to help thehome care professionals to perform their daily tasksand to share their data, according to the context of care, their role, and their needs. The three mainpurposes of our research are then the following:(i) understand the work situation, and the collec-tive work organization, (ii) propose technical archi-tectures able to improve the cooperation betweenhealth care workers, (iii) innovate with new coor-dination tools.The next sections will present first, the homecare cooperation analysis, and then the home carecooperation system. Finally, we report on the firstsystem evaluation, and conclude. 2. Home care cooperation analysis 2.1. Brief state of the art 2.1.1. Characteristics of asynchronous anddistributed cooperation One of the main characteristics of numerousdynamic situations is their collective dimensionwhich implies different types of cooperative activ-ities: synchronous or asynchronous, direct or medi-ated. According to Schmidt [16,17], ‘‘  peopleengage in cooperative work when they are mutu-ally dependent in their work and therefore arerequired to cooperate in order to get the workdone ’’. This dependence is not only ‘‘ the interde- pendence that arises from simply having to sharethe same resources ’’. It implies the articulationand the coordination of the respective activities of the different operators from logical and organiza-tional points of view [18]. For this author, there is a continuum between two extreme points: onone side, collective activity (operators have a com-mon and shared goal) [19,20] and on the other side, co-activity (operators have no common goal) [21].On this continuum, it is possible to range a largerclass of cooperative and distributed activities inwhich there is a high level common goal but distinctindividual goals. For Schmidt [17], the cooperative work is, in essence, distributed. The action plans,elaborated before performing the activity, are notsimply executed during this process; each opera-tor can encounter events that may not have beenanticipated and can be managed as a local situa-tion (opaque to the others). This distributed natureis reinforced when the different operators have dif-ferent specialties, as is the case in home care activ-ities. In this framework, according to Rogalski 18]and Schmidt [17], we consider cooperative work asinvolving a set of operators having individual goalsand interests and ‘‘[ . . . ]  in principle, distributed in the sense that decision making agents are semi-autonomous in their work in terms of contingen-cies, criteria, methods, specialties, perspectives,heuristics , [ . . . ]’’ ([17] p.353).Our analysis of the cooperation underlying homecare activities relies on the cooperation cognitivearchitecture (cf. Fig. 2) proposed by Hoc [22,23]. This architecture, consistent with most of the lit-erature in the domain [24,25], was first elaborated by a multidisciplinary approach, combining the psy-chology and supervisory control points of view [25],in order to design human—machine cooperationtools.Itreliesontheconceptofinterference,stem-ming from planning literature [26]. This notion is also defined by Castelfranchi [27] who noted that‘‘ theeffectsoftheactionofoneagentarerelevant for the goals of another: i.e. they either favor theachievement or maintenance of some goals of theother’s (positive interference), or threaten someof them (negative interference) ’’ (p. 162).As shown in Fig. 2 this architecture also relieson the concept of ‘‘Common Frame of Refer-ence’’. Most of the work in the cooperation domainunderlines the necessity for the agents to have ashared knowledge, belief and representation struc-ture [19,29]. In this study, we used the definition of de Terssac and Chabaud [30] for which the CommonFrame of Reference (COFOR—–r´ef ´erentiel commun)is: ‘‘ the sharing of competencies to prepare and  perform an action; this sharing of competencies atthe same time complements each individual rep-resentation of the task, and enables the adjust-ment of each individual decision, considering theother’s knowledge ’’ (p.128—129, our translation).This notion is close to Clark’s concept [31] of com-mon ground, shared mental models [32], or shared situation awareness [33—35]. In Hoc’s architecture  A distributed coordination platform for home care 813 Fig. 2  Processing architecture of cooperation, from Hoc [22,23] and H´elie and Loiselet [28]. of cooperation [22,23], interference managementrequires the sharing of task representations, of partners’ activities and of the goal to be reached.Severaltypesofinterferencemayexistandtheyaremanaged at different levels of abstraction (Fig. 2):cooperation in action (goals and procedures man-agement activities, in the course of task execu-tion, in real time and in the short term); coopera-tion in planning (COFOR management—–elaborationor maintenance—–in the medium term); and metacooperation (high level of abstraction activi-ties producing useful support for other activitylevels). 2.1.2. Communication aspects Communication is often associated with coopera-tion. As emphasized by Clark [31], communicat-ing aims at understanding; cooperation betweenthe agents is necessary to reach this understand-ing. Here, as mentioned by Hoc [23], we will takecommunication as a means of cooperation. In thiscontext, communication is essential as: (i) a sup-port for the transmission of information necessaryto task execution, (ii) a support for team coordi-nation, and (iii) as means of COFOR management(elaboration and maintenance). Most of the situa-tions in which COFOR management was studied aresynchronous situations. In these situations, someelaboration process, such as explication [36,37] ornegotiation [13], may be implemented and make COFOR elaboration easier. Some studies [38] haveshown an improvement in cooperative work qualitywhen the team leader communicates to his part-ners a short summary about his task representation.This effect also appears in hospital healthcare [14]or during shift changeovers [39—41].Moreover, Schmidt [17] underlines the role of the observation of others’ actions (non verbal com-munication) which allows the actors to infer or toanticipate the behavior of the other one. For exam-ple, Loiselet and Hoc [42], in a study on coopera-tion between a pilot and a navigator in a fighteraircraft, have shown that the navigator can inferfrom the effects on the pilot’s behavior that thepilot is overloaded; in this case, the navigator willoften take over a task initially allocated to thepilot. Another work [43] has emphasized the factthat information about the operators’ activity rep-resents an important part of the COFOR content. Inan experiment studying cooperation between tworadar controllers, these authors found that mostof the communications were related to elaborationand maintenance of COFOR.Different studies show that communication isalso important in medical environments. •  First, the amount of communication is impor-tant as pointed out by Coiera [44]. Conversation is very important in medical care settings andeven if some communication seems to be disrup-tive, it is nevertheless fundamental to building acommon knowledge base and grounding for thepatient’s care. •  Second, the quality of communication is criticalto assuring quality of care. In the studies per-formed in Intensive Care Unit, it was that most of the breakdowns between workers were inducedby a lack of good communication [45,14].A major feature of home care is its asynchronouscharacter. In this context, the communicationbetween operators is rarely direct and they cannotobserve the others’ activity. But improving the
Search
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x