Disaster management: findings from a systematic review

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0965-3562.htm Disaster management: findings from a systematic review Disaster management: findings Emanuele Lettieri, Cristina Masella and Giovanni Radaelli Department of Management, Economics and Industrial Engineering, Politecnico di Milano, Milan, Italy 117 Abstract Purpose – The paper aims to discuss a systematic review of the literature about disaster management within the period 1980-2006. Design/methodology/approach – The research protocol is based on the methodology that is commonly used in healthcare for analysing the literature and provides a state-of-art medical discipline. The paper presents both a descriptive analysis and a thematic analysis in order to provide a state-of-art of international literature. The research protocol is provided in order to make transparent the review process. Findings – The descriptive analysis highlights the peculiarities of the literature in terms of attention paid during the years, country of provenience and clusters of content of the selected papers. The thematic analysis deepens the content of the papers formalising the state of art. Research limitations/implications – The review considered only academic journals and peer-reviewed published papers, excluding working papers and books. Practical implications – Through both the analyses the authors argue for scholars in disaster management specific streams for further research and for providing practitioners with a state of art of disaster management discipline. Originality/value – The paper is original and is aimed at translating to the disaster management discipline the methodology of the systematic review commonly used in healthcare disciplines. Keywords Disasters, Risk management, Defence sector Paper type Literature review 1. Introduction The literature about disaster management is becoming wide and more and more spread on different journals. Even a cursory review of literature would identify that scholars of disaster management claim different theoretical foundations and argue different theoretical frameworks. The definition of disaster management is also fuzzy and disagreed on. For the purposes of the present literature review, disaster management has been defined as the body of policy and administrative decisions, the operational activities, the actors and technologies that pertain to the various stages of a disaster at all levels. The present paper aims to contribute to the existing knowledge on disaster management by reviewing its literature, through a systematic review methodology that has been used in the medical sciences over the last 15 years (Tranfield et al., 2002). This research does not focus on a specific hazard or issue related to disaster management, but aims to define the state of the art of the discipline. In this view, the paper will present both the descriptive analysis and the thematic analysis. The structure of the paper is as follows: first, the literature review methodology is presented; second, the descriptive findings are described; third, a state of the art about disaster management is argued through content analysis. Finally, conclusions about Disaster Prevention and Management Vol. 18 No. 2, 2009 pp. 117-136 q Emerald Group Publishing Limited 0965-3562 DOI 10.1108/09653560910953207 DPM 18,2 118 the findings, limitations found during the research, and issues for further research are presented. 2. Methodology A systematic review differs radically from the traditional narrative literature reviews. Narrative reviews often lack rigour, are susceptible to researchers’ bias (Mulrow, 1994) and can lack a means for making sense of what the collection of studies reviewed is saying. Researchers in medical sciences would claim that narrative reviews have a modest level of evidence and they are not a genuine piece of investigatory science. On the contrary a systematic review is a research methodology characterised as being a pragmatic, transparent, and reproducible manner of analysing existing literature (Cook et al., 1997; Cooper, 1998). In fact it clearly specifies how the researcher carried on the review, what type of records (documents, papers, books, etc.) he/she reviewed, and where those records were found. This allows other researchers to fully replicate the review. The main goal of a systematic review is to make sense of a mass of often contradictory evidence in order to help both academics and practitioners to improve their decision-making and practice, narrowing the knowing-doing gap that exists between research and practice. A systematic review (Tranfield et al., 2002) involves five stages: (1) planning the review; (2) identifying and evaluating studies; (3) extracting and synthesising data; (4) reporting descriptive and thematic findings; and (5) utilising the findings in order to inform research and practice. Stages (1), (2) and (3) are briefly presented in this paragraph; the descriptive and thematic findings obtained in stage (4) and the content of stage (5) are discussed in the following sections. Stage (1), planning of the review, includes different phases: (1) the constitution of a panel of experts that will inform the process and assess the findings; (2) the mapping of the field of investigation in order to identify the bodies of literature relevant to the topic; and (3) the formalisation of a review protocol that will enable other researchers to replicate the review. For the present Systematic Review, in phase (1) faculty members of Politecnico di Milano who are involved in the PROMETEO project[1] and practitioners from civil defence and civil protection were selected as review panel. Regarding phase (2), the map of the disaster management field was defined with the help of the review panel and the following areas of investigation were identified: (1) the theoretical framework used in disaster management; (2) phases of the general process of disaster management; (3) actors involved and responsibilities within disaster management; and (4) technology and information as resources for disaster management. This map of the field has had a critical role during the review process, since it has facilitated the identification of those studies that were out of or within the scope of the topic being investigated. The review protocol developed in stage (3) detailed how the literature review should be conducted by the research team. The protocol should include the different sources that are going to be used for identifying relevant studies, the intended search strategy, the specific criteria for including and excluding studies, the criteria for assessing the quality of the studies selected, and any other information that would allow someone else to reproduce the review. For the purpose of the present Systematic Review the Social Science Citation Index (SSCI) electronic database was the main source for identifying studies on disaster management and only published peer-reviewed papers were used as documents, excluding books and un-published papers or reports. The SSCI database allowed above all the identification of a list of journals dealing with disaster management. The full papers were downloaded from CILEA and EBSCO databases or directly from the web site of the journal. For the selection of relevant studies, the research team used two different sets of criteria: one set of inclusion and exclusion criteria (Table I); and another set of quality assessment criteria (based on methodology or research design, contribution to theory or implication for practise, generalisability of findings). The information collected from each selected study was: . Reference details, like as author(s), journal name, year, etc. . Methodology, distinguishing between theoretical or empirical. . Country, which refers either to the country where the case study was developed in the case of empirical study or to the country where the author is affiliated in the case of theoretical study. . Perspective, which could be any of the areas identified during the mapping the field exercise. Disaster management: findings 119 Once the search process has been carried out and the relevant studies have been selected, two separate analyses of the information obtained can be produced, namely: a descriptive and a thematic analysis. The descriptive analysis helps to clarify the main characteristics of the field that is under investigation (methodologies used, classification of countries and evolution of the key words, etc.). The thematic analysis consists on synthesising the main outcomes extracted from the literature and its main purpose is to inform future research and practice. Inclusion criteria Exclusion criteria iii. Research related to the process of disaster management without preference for the hazard investigated iii. Research within the scope set by the map of the field iii. Theoretical and empirical research iv. Academic and practitioner research iii. Research related py organisational behaviour or network management iii. Research which is out of the scope set by the map of the field (e.g. stress management and sense making) iii. Research that did not pass the quality assessment iv. Pre-1980 Table I. Research protocol for including or excluding the found papers DPM 18,2 120 3. Descriptive analysis The descriptive analysis aims at providing an overview of the characteristics of the field. The papers from the systematic review are collected in Table II together with their peculiarities. In detail the papers are recorded by: . The year of publication within the period 1980-2006. . The country of origin according to the affiliation of the first author. . The focus of the paper according to four clusters: theoretical framework; phases of disaster management process; actors involved into the process; and technology and information. . The phases of the disaster management process that were investigated within the paper. In the following each issue is briefly discussed. Figure 1 shows the number of papers that were included according to the inclusion criteria within the period 1980-2006. The evolution clarifies that disaster management is a relatively recent issue and that the large part of studies were published within the period 1997-2005. The figure related to year 2006 is partial because the systematic review terminated on June 2006. The highest number of publications was in the year 2003 and after that year the number decreases quickly. Probably authors were focusing to other issues about civil defence. It would be relevant to clarify this argument reviewing the table of content of the more relevant journals about disaster management after 2003. Figure 2 shows the provenience of the selected papers according to the country of affiliation of the first author. This is relevant in order to understand how heterogeneous are the cultural backgrounds of the studies and the authors. The results are peculiar and nevertheless not surprising; they allow the gathering of at least two conclusions. First, the large part of studies come from the USA and the European Union (EU). Less research comes from Asia despite the tremendous attention that countries such as Japan and India pay to disaster management because of the high frequency and magnitude of earthquake and tsunami. Second, the authors could prefer to share the main results of their investigations with a national audience preferring national journals instead of international ones. This could be relevant in terms of knowledge sharing and cross-fertilisation about disaster management across the different countries. This argument is strengthened by the lack of papers that discuss knowledge management or performance management. Figure 3 shows the classification of the included papers according to the content. The classification is based on four different clusters of content: theoretical framework; phases of disaster management process; actors involved into the process; and technology and information. The findings clarify that the large proportion of papers deal with the phases of the disaster management process, formalising different sequence (please, see the next thematic analysis). Relevant attention was paid also to the discussion about which are the actors who are involved in the disaster management network and what role they should play. One-third of the selected papers deal with issues that are related to the information management process and the information and communication technologies that should be adopted to improve actual performance. Focus Author 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Phases Country Year Inclusion Exclusion Theory Phases Actors Hensgen Heino Buisson Ody Ray Elsubbaugh Bertrand Quarantelli Hite Liu Hwacha Pearce Wybo Ritchie McEntire Alexander Kunruether Beroggi Tobin Aspinall Faulkner Wilson Dai Mansor Choularton Batho Heath Trim Menoni Simpson Shaluf USA FIN FRA UK USA UK FRA USA USA CINA CAN CAN FRA AUS USA UK USA OLA USA UK AUS USA CINA MAL USA UK UK UK ITA USA MAL 2003 1996 1995 1995 2001 2004 2002 1997 2003 1997 2005 2003 1998 2003 2002 2003 2003 1995 2002 2003 2000 2001 2001 2004 2001 1999 1998 2004 1996 2002 2003 £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ Tech. and info. Strategy Mitigation Preparedness Response Signalling Recovery Learning £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ (continued) Disaster management: findings 121 Table II. Papers analysed for the systematic review of the literature Author 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 Phases Country Year Inclusion Exclusion Theory Phases Actors Kaiser Kuo Khan Drabek Guarnieri Mendonca Stephenson Cohen Streeter Sutphen Phillips Kreps Fitzgerald Demichela Jaber Alexander Perry Bakir Dixit Haque Joninetz Drabek Henderson Kletz Neal USA CINA IND USA FRA USA USA USA USA USA USA USA USA ITA FRA ITA USA TUR NEP CAN USA USA USA UK USA 2003 1997 1999 2001 1995 2000 1997 2003 1991 1998 1993 1983 1996 2003 2001 2005 2003 2004 2002 1999 2004 1999 2004 1997 1997 £ £ £ £ £ £ £ £ £ £ £ £ £ Strategy Mitigation Preparedness Response Signalling Recovery Learning £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ Tech. and info. £ £ £ £ £ £ DPM 18,2 122 Table II. Focus £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ Disaster management: findings 123 Figure 1. Spread of the papers included into the review within the period 1980-2006 Figure 2. Classification of the included papers according to the country of affiliation of the first author Figure 3. Classification of the included papers according to the main content DPM 18,2 124 Fewer papers cope with the design of a theoretical framework for the disaster management discipline. Figure 4 shows which phases of the disaster management process – that is, according to Figure 3 the most frequent content of selected papers – are more investigated. The results show that the phases related to strategy and response got more attention than the others. It is relevant to be precise that the phase of learning received modest attention. This result should be connected to the lack of specific studies about performance management and knowledge management. The risk is that the network of actors that is responsible for disaster management does not learn from the past practice and does not improve time after time. Surely further research on these issues should be recommended. 4. Thematic analysis The thematic analysis is as follows: review of the theoretical framework; phases of the process of disaster management; actors involved and responsibilities; and technology and information as resources for disaster management. The previous structure meets the map of the field came from the panel of experts. 4.1. Theoretical framework The systematic review points out that the main contributions to disaster management theory agree a common theoretical framework that is independent by the specific hazard. This can be considered as a recent achievement in disaster management theory. The hazard-related framework is simplistic and often neglects the theoretical deployment relevant interactions between different typologies of hazards. Moreover the focus on technical issues neglects key contributions from the organisational and psychological fields. These limitations lead to difficulties in making these frameworks actionable. The new orientation can be synthesised in the term of all-hazard approach, holistic approach and continuous process. These are the relevant keywords that contributors highlight. Figure 4. Classification of the included papers according to the phases of disaster management An all-hazards approach is quite self-explanatory: several unlinked and single hazard-based strategies, are substituted by a single strategy which collects in a single glance all the hazards and they mutual relationships. This modus operandi improves risk analysis, decision making for both prevention and response to disasters (Alexander, 2005; Hwacha, 2005; Jonientz-Trisler et al., 2005; McEntire et al., 2002; Wilson and Oyola-Yemaiel, 2001; Hite, 2003). An holistic approach is characterised by two complementary perspectives. On the one hand it claims that input variables for decision making should be as many as possible, and involving both technical and managerial issues. On the other hand it claims the relevance of building a comprehensive picture of reality through continuous observation. This picture would test and improve the decisional variables and their linkages. This approach revaluates the contributions from sociology and psychology fields (McEntire et al., 2002). The concept of continuous process is close to the holistic approach. It is based on the idea – widely accepted – that disasters will happen seamlessly. Any organisation should work to reduce the risks – i.e. the product of magnitude and probability to occur – even if it is aware that a probability close to zero does not mean that the disaster will not happen. Consequently disaster management should be an endless process that clearly cannot stop. This idea brought to a contraposition in the literature between two concepts: resilience and resistance. These concepts are often referred to a different role that a community should play for disaster management. Resilience focuses on after-crisis activities and claims for a community educated to be as flexible as possible in order to be able to cope with the wider variety of hazards and disasters. By contrast resistance focuses on before-crisis activities and claims for a community educated to prevent risks reducing of the magnitude or the probability to occur. The recent literature (primarily McEntire et al., 2002) argues a comprehensive vulnerability management approach that grounds on and synthesises the two concepts. 4.2. Phases The process of disaster management can be deployed into phases that are characterised by peculiar goals and resources. A phase can be deployed into specific functions while each function can be deployed into activities. The literature review has been focused only to a higher level of aggregation (i.e. the level of phase) because functions and activities are hazard- and context-related and the degree of generalisation would have be modest. In order to fully discuss phases it is relevant to be precise about which are the main temporal stages that host the phases. The literature argues three different temporal (and logical) stages: pre-crisis, crisis and post-crisis. In detail pre-crisis is the period that goes before the occurrence of a disaster, crisis is the aftermath of the disaster and post-crisis is the period between the fading of crisis and the return to a normal condition (Hensgen et al., 2003; Bertrand and Lajtha, 2002). Stages have a logical relationship with disasters. In fact during the pre-crisis organisations apply for future disasters; during the crisis they apply for present disasters; and during the post-crisis they apply for past (just terminated) disasters. The literature agrees a reference model for the process of disaster management based on the following phases: mitigation and preparedness (mainly in pre-crisis); response (in crisis); and recovery (in post-crisis) (Hwacha, 2005; Mansor et al., 2004; Disaster management: findings 125 DPM 18,2 126 McEntire et al., 2002; Hensgen et al., 2003; Faulkner, 2001; Henderson, 2004; Shaluf et al., 2003). These phases are considered canonical in the debate on disaster management. It is relevant to point out that the phases are mutually inclusive and multidimensional, because they are strongly interconnected (Neal, 1997). Mitigation consists of the efforts/actions aimed to minimise the degree of risk, to prevent disasters and to reduce the vulnerability of both the ecosystem and social system (i.e. the community) (Menoni and Pergalani, 1996; Bakir, 2004; Dai et al., 2002; Mansor et al., 2004). The last two authors specify how the main functions should be hazard assessment, vulnerability and risk reduction. On the contrary, preparedness involves actions to prepare responders and common people to post-disaster activities (Simpson, 2002). Response consists of actions to manage and control the various effects of disaster (also the ripple effects) and minimise human and property losses. The main functions (Kreps, 1983; Perry and Lindell, 2003) are evacuation (Tobin and Whiteford, 2002), sheltering (Phillips, 1993), medical care, search and rescue, property protection, and damage control. Recovery consists of those actions that bring the disrupted area back to an often improved normal condition. After the analysis of some contributions, it is possible to argue the logical existence of three more phases: strategy, learning and signalling. Strategy came from business process literature that claims that any process requires a strategy to address long-term goals and outputs. Learning came from literature on continuous improvement and quality management, which claims that any organisation should discuss processes and performance to find the gaps between the as-is and the should-be configuration and solve them (Choularton, 2001). Signalling emerges from the four “canonical” phases and from the contributions by Hensgen et al. (2003) and Drabek (1999, 2001). They highlight that: . warning is present in each of the four “canonical” phases; and . warning cannot be classified eeither as a function, or as an activity. The intensity and the content of signalling change from phase to phase having a climax during the response. The main functions of signalling are: reception of alert and investigation; and supply and delivery of warning messages. Signalling requires an endless involvement in order to define which warnings are reliable and which organisations should take care of them. If organisations are fed by wrong alerts, they could act when they are not required to (wasting time and resources) or they could not act promptly when they are required (fostering deaths among people and property damages). Similarly, if a proper signal is sent to a wrong organisation the results could be distressing (Hensgen et al., 2003). Alternately, there are many examples of how an effective warning system could prevent disasters or anticipate the response before the crisis (Batho et al., 1999). Many authors focused their investigation towards a limited group of phases rather than on the whole process of disaster management. For instance, mitigation and response are the phases much studied. The ratio is linkages to the concept of resilience and resistance. Contributions on mitigation focus on theoretical models and technologies generally applicable to a specific hazard (Mansor et al., 2004; Dai et al., 2002; Guarnieri and Wybo, 1995). On the other hand, contributions on response focus on past disasters, suggesting an improvement of performance through the use of new technologies. The other phases have been modestly investigated. For instance speculation on preparedness is poor and limited to a few assumptions on community involvement. It lacks insights on how to bring theory and training into action. Also recovery has been neglected despite a few contributions. Kunreuther and Linnerooth-Bayer (2003) investigated after-crisis financing. Bertrand and Lajtha (2002) proposed innovative ways to improve disrupted areas. Strategy has rare, but relevant, contributions by Alexander (2003, 2005), Perry and Lindell (2003), Fitzgerald (1996), and Bakir (2004). Learning is mostly neglected and surely further studies should be carried out in order to apply the contributions on learning to the disaster management field. 4.3. Actors The systematic review highlights the involvement of different actors within the process of disaster management. The involvement may be direct or indirect. Actors have a direct involvement when they are responsible for a phase/function/activity within the process of disaster management. By contrast actors have an indirect involvement when they are affected or affect the process with their behaviours and/or actions without being fully aware of their impact. An example of indirect involvement is when the neighbours tried to help damaged people after a landslide on an inhabited area without the required training. Their efforts could be actually counterproductive. The literature offers numerous different points of view about actors and their involvement without delivering reference models or taxonomies. For the purpose of this systematic review the authors argued four clusters reviewing the different roles that actors play within the disaster management. Each actor plays at least one of the four following roles: agents, researchers, population and media. Population and media are clearly two indirect roles. The contributions to the literature highlight an increasing interest to these actors and to the linkages between their behaviours and performance of disaster management. Population and media have been recognised as influential actors because of their presence within the process. They represent two different forces that, when properly exploited, could address tremendous advantages; otherwise they could be serious sources of risk and causes of under target performance. In the following the two actors will be analysed separately. Population. Population has been recognised to not be a passive actor to be evacuated, healed and sheltered. It has one active task at least: evaluation. This task is decisive to the success of disaster management. Organisations should get people’s trust, otherwise success could be compromised. Furthermore a severe academic debate is ongoing about the role that population could play within both strategy and pre-crisis (Pearce, 2003; Haque, 1999; Dixit, 2003; McEntire et al., 2002). Public roundtables (Hite, 2003; Hwacha, 2005), public preparedness and mitigation are becoming frequent keywords in the literature. The concept of public participation, indeed, can be seen as one of the strongest examples of holistic approach. Some authors have also recognised the influence of quasi-official actors (Perry and Lindell, 2003). They are common people who act as volunteers during response activities such as search and rescue, evacuation, etc. They could be a relevant source of risk, since they are not led by training but by improvisation, which, even if heartfelt, can compromise the whole process. Media. Recent literature has investigated the role that media plays within disaster management and above all the main responsibilities of after-disaster failures. Examples such as Mt St Helens (1980) and East Kootenay (1985) highlight failures due Disaster management: findings 127 DPM 18,2 128 to wrong relationships between organisations and media (Ritchie, 2004). Media can be helpful when it properly informs the population and supports the various direct actors (Ritchie, 2004; Faulkner, 2001; Ray, 2001). Media is a surely a critical bridge between the population and the organisation (Quarantelli, 1997). McLuhan (1964) defines the media as a cold actor because it does not offer much detailed data and requires public involvement to integrate missing information. Drabek (1999) claims that people show an incredible creativity in interpreting information related to disasters: in this view other authors suggest the need of appropriate filters between organisations and media (i.e. spokespeople; information sorting, etc.). On the other hand agents and research are considered as direct actors. Perry and Lindell (2003) call these actors as “official” because they work for disaster management and they are responsible for performance. In this view they have been the natural target of academic research during recent years. Despite the fact that main attention has been paid to agents, investigation about research has been rare (Hwacha, 2005; Hite, 2003; McEntire et al., 2002). Focusing on agents two different research efforts can be identified: classifications of actors; and insights on their peculiar features. The literature shows mainly two classifications of agents. On one hand Wybo and Kowalski (1998) argues a functional classification that they use for both actors and technologies. These classification has been developed specifically for command centres. Agents are classified against four functions: (1) perception (data collecting and processing); (2) analysis (decision making); (3) communication (inside organisations); and (4) information (communication outside organisations). This classification is coherent with the incident command system organisational architecture. On the other hand, authors such as Kreps (1983) and Perry (1991) argue a sectorial classification that defines the various sectors involved in disaster management. In detail they identify: . public sector (the most relevant and responsible actor); . private industry (including all economic sectors); and . volunteer agencies (i.e. Red Cross and NGOs). The large body of literature focuses on Government as the main actor, the one with the ultimate responsibility for disaster management, and neglects both private industry and volunteers. Recent studies uncovered the relevance of the other agents: for instance Fitzgerald (1996) and Cohen (2003) claimed the need to involve private industry also within the strategy in order to boost performance. Government can be deeply analysed focusing on the various agencies and ministries involved following the responsibility chain. Or more fruitfully it can be investigated on the principle of subsidiarity (Henderson, 2004; Hwacha, 2005; Wilson and Oyola-Yemaiel, 2001). Three different government levels can be recognised. The first is local, which is the closest to disasters and the one with better knowledge of the area. Therefore, this should intervene first. If it does not succeed in resolving the crisis (e.g. for lack of means or competences) the responsibility must be taken by regional/provincial authorities and, at the end, by national authorities (i.e. the president/king and defence minister), which provide all available means. It is possible to identify another, considering international authorities. They may not take the responsibility, but they could offer additional means and competences. A well-known example is the aftermath of the tsunami where a massive amount of international aid helped the local authorities. The literature is also rich with insights about peculiar features and trends in official organisations. The contributions are quite heterogeneous but it is possible to trace an harmonised picture. The main features are: . Coordination between all organisations and inside each organisation through team working along all the phases, from strategy to recovery and not only during response (Quarantelli, 1997; Ritchie, 2004; Kreps, 1983; Faulkner, 2001; Perry and Lindell, 2003; Bertrand and Lajtha, 2002; Hite, 2003; Hwacha, 2005; Elsubbaugh et al., 2004). A wrong coordination would lead to conflicts, to resource and time wasting and, ultimately, to human and property losses. This feature is linked to efficient labour division and delegation and had lead to important technological implementations such as RDBMS, the internet and GIS. . Professionalisation – disaster management cannot be an extemporaneous profession but it is an actual job. Professionalisation has lead to a relevant workforce rationalisation and it is currently leading to its demilitarisation (Sutphen and Waugh, 1998; Wilson and Oyola-Yemaiel, 2001, Trim, 2004). . Effectiveness in resource allocation and mobilisation – “Good disaster management does not involve the mobilisation per se of personnel and resources – that will happen anyway [. . .] Effective means that a desired and intended result has been produced; this definition differs from that of efficiency which requires that the results be obtained in the best possible way” (Quarantelli, 1997, p. 43). . Redundancy of actors and technologies so that, if one stops working for any reason, it can been rapidly substituted by someone else. Streeter (1991) suggests the use of “active” redundancy instead of “stand-by” one, since the latter would cause useless inefficiencies. . Team working both inter- and intra-organisational – The holistic approach suggests that team working should better be accomplished with the employment of as many competences as possible. Incident command system structures show a distinctive example of team working. . Trust between employees and their superiors that goes beyond hierarchy reliance. Since they are employed in a stressful and potentially harmful environment and there is little chance for a participatory approach, trust is the best way to minimise the chance that responders would disobey to superiors’ orders and compromise operations outcomes (see various contributions in Trim (2004)). . Ramp-up capability that means the possibility of rapidly increasing both the individual and the organisational outcomes during a stressful and heavily demanding periods, such as during disasters (Ray, 2001; Cohen, 2003). Disaster management: findings 129 DPM 18,2 130 4.4. Technology and information The systematic review informed the authors of what technologies are or should be utilised within the disaster management process, to which extend and how often they have been investigated by academics. The same analysis has been carried on regarding data and information that is utilised and delivered through the disaster management process. The perspective was managerial and not technical: in this view there was a modest interest to specific algorithms or technical tools, because they are often related to specific hazards or countries, but the focus was to the role that technology plays within the process, understanding which phases are supported. The literature argues that data and/or information support mainly three functions: (1) understanding of hazards and disasters; (2) decision making; and (3) signalling and communication. Wybo and Kowalski (1998) classify the technologies that support these functions respectively as perception, analysis, communication, and information agents. These agents should be interconnected because they are complementary constituents of the disaster management process. They should be able to operate and communicate in a unique interoperable system in order to address a performance improvement. The technologies that constitute the present information systems that have largely investigated are: . geographical information system (GIS). . relational data base management system (RDBMS); . analysis techniques; and . real-time communication systems. Less investigated technologies are: . satellites; . ground sensors; and . specific decision support system (DSS). The GIS has been largely studied during the last years in order to design more performing configurations or to extend the running implementations to other environments (Guarnieri and Wybo, 1995; Hite, 2003; Kaiser et al., 2003; Jaber et al., 2001; Mansor et al., 2004; Dai et al., 2002; Kaiser et al., 2003; Heino and Kakko, 1998). Generally, the attempts to design the aforementioned system begin from a GIS analysis because it is the most comprehensive technology and it provides a relevant support to collection, assessment and communication of the available information. It offers also a visual and user-friendly representation of hazards, environmental evolution and supports the linkages of various descriptive variables. Nowadays, the variety of applications of current bibliography, altogether show how visual representation and real-time communication are considered common requirements in the disaster management field (Mendonca et al., 2000; Buisson and Cligniez, 1995; Beroggi et al., 1995; Ody, 1995). Relational databases offer a decisive contribution to collection and communication of information. Moreover they offer the tremendous opportunity to link the local databases of any organisation involved in the process according to the incident command system architecture for information management (Heath, 1998). The possibility to get a unique collection of data available of all the organisations involved into the disaster management process could improve both effectiveness and efficiency through a redesign coordination. Stephenson and Anderson (1997) pointed the Gopher protocol as a decisive step towards real-time communication. An other research stream is dedicated to the techniques of analysis that support the organisations to identify the causes for a disaster, to understand the context, to build proper scenarios and to make appropriate decisions. There are various techniques; the most investigated are surely: . Event and Fault Tree Analysis (Demichela et al., 2004; Kuo et al., 1997); . HazOp (Khan and Abbasi, 2000; Kletz, 1997); . Markov, Petri and Bayesian networks (Aspinall et al., 2003; Liu and Chiou, 1997). It is interesting to observe how a considerable part of this material comes from industrial risk analysis and can be stretched to disaster management without significant upsets. While the large part of authors concentrate on improving a specific technique through different algorithms, other authors focus on different topics for disaster management. For instance Kuo et al. (1997) offers an interesting attempt to integrate some of these techniques, while Aspinall et al. (2003) shows an implementation of Bayesian networks for a specific hazard (volcano) assessment. 5. Conclusions The systematic review of the literature has revealed the peculiarities and the gaps in the knowledge on disaster management and number of subject areas that could be researched further. In particular the combination of a descriptive and thematic analysis informed the “state of the art” and suggested some implications for both the scholars and the practitioners who operate in the field of disaster management. For instance the descriptive analysis has suggested some conclusions: . The number of papers related to the disaster management process has decreased in the last years and contributions to the literature are focusing to other issue although several gaps still remain in the understanding of disaster management theory. . The large part of contributions comes from the USA and Canada, while the contributions from the EU and Asia are still modest: a shortcoming could be the polarisation of the general theoretical background and a limited enrichment of the theory. . The contributions to the literature focus on the identification of the various phases of the disaster management process, while papers on organisational issues, such as network management and meta-organisations (e.g. the incident command system framework) are uncommon. . Within the general framework of disaster management the phases more investigated are the ones related to the strategy, the mitigation, the response and the recovery, while only a few papers dealt with other more specific phases such as learning. A serious risk for practitioners in disaster management is to miss the opportunity of build adequate knowledge management systems and improve performance time after time. Disaster management: findings 131 DPM 18,2 132 The descriptive analysis remarked on main areas for further research: performance management and knowledge management. Despite the fact that the incident command system framework introduced those as areas of relevance, there is a scarcity of studies that cope with the formalisation of a performance management system and key performance indicators. Without measuring the performance in term of effectiveness (timeliness and quality) and efficiency (cost) it is difficult to understand how value for money is the current practice of disaster management and to promote improvements. The same could be argued about the lack of procedure for knowledge management and knowledge sharing. These conclusions might not be surprising or unexpected; however, these notions about the field have never been tested before through such a rigorous review of the literature. Systematic reviews are not a panacea, they have proved their value in the medical context, but they still need to be experimented in the management field in order to improve them. This methodology can help researchers to create knowledge in a different way building on each other’s work and doing so to improve the knowledge in disaster management. A complete understanding of the state of the art of the literature could help practitioners in their decision-making processes and progressively inform research. Note 1. PROMETEO is a five-year long project carried on at Politecnico di Milano aimed at developing innovative methodologies to improve the performance of civil defence and civil protection. This is a multi-disciplinary project that involves Faculties of Engineering and Architecture. At present the PROMETEO project is at the end of the first year that has focused on a literature review of different issues and from different perspectives. References Alexander, D. (2003), “Towards the development of standards in emergency management training and education”, Disaster Prevention and Management, Vol. 12 No. 3, pp. 113-23. Alexander, D. (2005), “Towards the development of a standard in emergency planning”, Disaster Prevention and Management, Vol. 14 No. 2, pp. 158-75. Aspinall, W.P., Woo, G., Voight, B. and Baxter, P.J. (2003), “Evidence-based volcanology: application to eruption crises”, Journal of Volcanology: Application to Eruption Crises, Vol. 128 Nos 1-3, pp. 273-85. Bakir, P.G. (2004), “Proposal for a national mitigation strategy against earthquakes in Turkey”, Natural Hazards, Vol. 33 No. 3, pp. 405-25. Batho, S., Russell, L. and Williams, W. (1999), “Crisis management to controlled recovery: the emergency planning response to the bombing of Manchester city centre”, Disasters, Vol. 23 No. 3, pp. 217-33. Beroggi, G.E.G., Waisel, L. and Wallace, W.A. (1995), “Employing virtual reality to support decision making in emergency management”, Safety Science, Vol. 20 No. 11, pp. 79-88. Bertrand, R. and Lajtha, C. (2002), “A new approach to crisis management”, Journal of Contingencies and Crisis Management, Vol. 10 No. 4, pp. 181-91. Buisson, L. and Cligniez, V. (1995), “Development of a spatial knowledge base system for environment protection: the ARSEN project”, Safety Science, Vol. 20 No. 1, pp. 27-37. Choularton, R. (2001), “Complex learning: organizational learning form disasters”, Safety Science, Vol. 39 Nos 1-2, pp. 61-70. Cohen, M.J. (2003), “State-level emergency response to the September 11 incidents: the role of New Jersey’s Department of Environmental Protection”, Journal of Contingencies and Crisis Management, Vol. 11 No. 2, pp. 78-85. Cook, D.J., Mulrow, C.D. and Haynes, R.B. (1997), “Systematic reviews: synthesis of best evidence for clinical decisions”, Annals of Internal Medicine, Vol. 126 No. 5, March, pp. 376-80. Cooper, H. (1998), Synthesizing Research, 3rd ed., Sage Publications, Thousand Oaks, CA. Dai, F.C., Lee, C.F. and Ngai, Y.Y. (2002), “Landslide risk assessment and management: an overview”, Engineering Geology, Vol. 64 No. 1, pp. 65-87. Demichela, M., Piccinini, N., Ciarambino, I. and Contini, S. (2004), “How to avoid the generation of logic loops in the construction of fault trees”, Reliability Engineering and System Safety, Vol. 84 No. 2, pp. 197-207. Dixit, A. (2003), “Floods and vulnerability: need to rethink flood management”, Natural Hazards, Vol. 28 No. 11, pp. 155-79. Drabek, T.E. (1999), “Understanding disaster warning responses”, The Social Science Journal, Vol. 36 No. 3, pp. 515-23. Drabek, T.E. (2001), “Disaster warning and evacuation responses by private business employees”, Disasters, Vol. 25 No. 1, pp. 76-94. Elsubbaugh, S., Fildes, R. and Rose, M.B. (2004), “Preparation for crisis management: a proposed model and empirical evidence”, Journal of Contingencies and Crisis Management, Vol. 12 No. 3, pp. 112-27. Faulkner, B. (2001), “Towards a framework for tourism disaster management”, Tourism Management, Vol. 22 No. 2, pp. 135-47. Fitzgerald, M.E. (1996), “The emergency response plan: key to compliance with the emergency response provisions of the hazardous waste operations and emergency response standard (29 CFR 1910.210)”, Applied Occupational and Environmental Hygiene, Vol. 11 No. 9, pp. 1154-62. Guarnieri, F. and Wybo, J.L. (1995), “Spatial decision support and information management application to wildland fire prevention. The WILFRIED System”, Safety Science, Vol. 20 No. 1, pp. 3-12. Haque, C.E. (1999), “"Personal communication: GIS and new technology in flood mitigation”, mimeo. Heath, R. (1998), “Dealing with the complete crisis – the crisis management shell structure”, Safety Science, Vol. 30 No. 1, pp. 139-50. Heino, P. and Kakko, R. (1998), “Risk assessment modelling and visualisation”, Safety Science, Vol. 30 Nos 1-2, pp. 71-7. Henderson, L.J. (2004), “Emergency and disaster: pervasive risk and public bureaucracy in developing nations”, Public Organization Review: A Global Journal, Vol. 4 No. 2, pp. 103-19. Hensgen, T., Desouza, K.C. and Kraft, G.D. (2003), “Games, signal detection, and processing in the context of crisis management”, Journal of Contingencies and Crisis Management, Vol. 11 No. 2, pp. 67-77. Hite, M.C. (2003), The Emergency Manager of the Future, The National Academic Press, Washington, DC. Hwacha, V. (2005), “Canada’s experience in developing a national disaster mitigation strategy; a deliberative dialogue approach”, Mitigation and Adaptation Strategies for Global Change, Vol. 10 No. 3, pp. 507-23. Disaster management: findings 133 DPM 18,2 134 Jaber, A., Guarnieri, F. and Wybo, J.L. (2001), “Intelligent software agents for forest fire prevention and fighting”, Safety Science, Vol. 39 Nos 1-2, pp. 3-17. Jonientz-Trisler, C., Simmons, R.S., Yanagi, B.S., Crawford, G.L., Darienzo, M., Eisner, R.K., Petty, E. and Priest, G.R. (2005), “Planning for tsunami-resilient communities”, Natural Hazards, Vol. 35 No. 1, pp. 121-39. Kaiser, R., Spiegel, P.B., Henderson, A.H. and Gerber, M.L. (2003), “The application of geographic information systems and global positioning systems in humanitarian emergencies: lesson learned, programme implications and future research”, Disasters, Vol. 27 No. 2, pp. 127-40. Khan, F.I. and Abbasi, S.A. (2000), “Towards automation of HAZOP with a new tool EXPERTOP”, Environmental Modelling & Software, Vol. 15 No. 1, pp. 67-77. Kletz, T.A. (1997), “Hazop – past and future”, Reliability Engineering and System Safety, Vol. 55 No. 3, pp. 263-6. Kreps, G.A. (1983), “The organization of disaster response: core competences and processes”, International Journal of Mass Emergencies and Disasters, No. 13, pp. 439-65. Kunreuther, H.C. and Linnerooth-Bayer, J. (2003), “The financial management of catastrophic flood risks in emerging-economy countries”, Risk Analysis, Vol. 23 No. 3, pp. 627-39. Kuo, D.H., Hsu, D.S. and Chang, C.T. (1997), “A prototype for integrating automatic fault tree/event tree/HAZOP analysis”, Computers Chemical Engineering, Vol. 21, pp. S923-8. Liu, T.S. and Chiou, S.B. (1997), “The application of Petri nets to failure analysis”, Reliability Engineering and System Safety, Vol. 57 No. 2, pp. 129-42. McEntire, D.A., Fuller, C., Johnston, C.W. and Weber, R. (2002), “A comparison of disaster paradigms: the search for a holistic policy guide”, Public Administration Review, Vol. 62 No. 3, pp. 267-81. McLuhan, M. (1964), Gli strumenti del comunicare, Net, Milan, trans. by Capriolo, E.. Mansor, S., Abu Shariah, M., Billa, L., Setiawan, I. and Jabar, F. (2004), “Spatial technology for natural risk management”, Disaster Prevention and Management, Vol. 13 No. 5, pp. 364-73. Mendonca, D., Rush, R. and Wallace, W.A. (2000), “Timely knowledge elicitation from geographically separate mobile experts during emergency response”, Safety Science, Vol. 35 Nos 1-3, pp. 193-208. Menoni, S. and Pergalani, F. (1996), “An attempt to link risk assessment with land use planning: a recent experience in Italy”, Disaster Prevention and Management, Vol. 5 No. 1, pp. 6-21. Mulrow, C.D. (1994), “Systematic reviews: rationale for systematic reviews”, British Medical Journal, No. 309, pp. 597-9. Neal, D.M. (1997), “Reconsidering the phases of disaster”, International Journal of Mass Emergencies and Disasters, Vol. 15 No. 2, pp. 239-64. Ody, K. (1995), “Facilitating the ‘right ‘ decision in crisis: supporting the crisis decision maker through analysis of their needs”, Safety Science, Vol. 20 No. 11, pp. 125-33. Pearce, L. (2003), “Disaster management and community planning, and public participation: how to achieve sustainable hazard mitigation”, Natural Hazards, Vol. 28 Nos 2-3, pp. 211-28. Perry, R. (1991), “Managing disaster response operations”, in Drabek, T.E. and Hoetmer, G. (Eds), Emergency Management: Principles and Practice for Local Government, International City Management Association, Washington, DC, pp. 201-23. Perry, R.W. and Lindell, M.K. (2003), “Preparedness for emergency response: guidelines for the emergency planning process”, Disasters, Vol. 27 No. 4, pp. 336-50. Phillips, B.D. (1993), “Cultural diversity in disasters: sheltering, housing and long term recovery”, International Journal of Mass Emergencies and Disasters, Vol. 11 No. 1, pp. 99-110. Quarantelli, E.L. (1997), “Ten criteria for evaluating the management of community disasters”, Disasters, Vol. 21 No. 1, pp. 39-56. Ray, S. (2001), “A new flood rescue response model”, Advanced Rescue Technology Magazine, February/March. Ritchie, B.W. (2004), “Chaos, crises and disasters: a strategic approach to crisis management in the tourism industry”, Tourism Management, Vol. 25 No. 6, pp. 669-83. Shaluf, I., Ahmadun, F. and Mustapha, S. (2003), “Technological disaster’s criteria and models”, Disaster Prevention and Management, Vol. 12 No. 4, pp. 305-11. Simpson, D.M. (2002), “Earthquake drills and simulations in community-based training and preparedness programmes”, Disasters, Vol. 26 No. 1, pp. 55-69. Stephenson, S. and Anderson, P.S. (1997), “Disasters and the information technology revolution”, Disasters, Vol. 21 No. 4, pp. 305-34. Streeter, C.L. (1991), “Redundancy in social systems: implications for warning and evacuation planning”, International Journal of Mass Emergencies and Disasters, Vol. 9 No. 2, pp. 167-82. Sutphen, S. and Waugh, W.L. Jr (1998), “Organizational reform and technological innovation in emergency management”, International Journal of Mass Emergencies and Disasters, Vol. 16 No. 1, pp. 7-12. Tobin, G.A. and Whiteford, L.M. (2002), “Community resilience and volcano hazard: the eruption of Tungurahua and evacuation of the Faldas in Ecuador”, Disasters, Vol. 26 No. 1, pp. 28-48. Tranfield, D., Denyer, D. and Smart, P. (2002), “Developing an evidence-informed approach to management knowledge by means of systematic review”, working paper, Cranfield School of Management, Cranfield University, Cranfield. Trim, P.R.J. (2004), “An integrative approach to disaster management and planning”, Disaster Prevention and Management, Vol. 13 No. 3, pp. 218-25. Wilson, J. and Oyola-Yemaiel, A. (2001), “The evolution of emergency management and the advancement towards a profession in the United States and Florida”, Safety Science, Vol. 39 No. 1, pp. 117-31. Wybo, J.L. and Kowalski, K.M. (1998), “Command centers and emergency management support”, Safety Science, Vol. 30 No. 1, pp. 131-8. Further reading Mulrow, C.D. (1987), “The medical review article: state of the science”, Annual International Medicine, Vol. 106, pp. 485-8. About the authors Emanuele Lettieri, PhD is Researcher at the Department of Management, Economics and Industrial Engineering at Politecnico di Milano. His research activities are mainly related to performance management and management of technology within public administration with a focus on healthcare and civil defence. He is co-author of various publications in those research fields. He is Lecturer in Economics and Business Administration and he has been involved in various projects related to performance improvement and change management in healthcare organisations. He is also involved in the PROMETEO project aimed at innovating the civil Disaster management: findings 135 DPM 18,2 136 defence in Italy. Emanuele Lettieri is the corresponding author and can be contacted at: emanuele.lettieri@polimi.it Cristina Masella is Full Professor at the Department of Management, Economics and Industrial Engineering at Politecnico di Milano, where she teaches Management Accounting. She is also the Director of the Master in Public Administration Engineering at Politecnico di Milano. Her research activities before focused on innovation in management accounting and performance measurement, and are at present related to innovation to public administration, with a focus on healthcare and civil defence. She works on technology assessment, performance measurement and benchmarking. She chairs the postgraduate courses for healthcare practitioners (physicians, nurses and managers). Giovanni Radaelli, BA Eng. is a Student at the Master in Public Administration Engineering at Politecnico di Milano and he is working at the PROMETEO project aimed at innovating the civil defence in Italy. To purchase reprints of this article please e-mail: reprints@emeraldinsight.com Or visit our web site for further details: www.emeraldinsight.com/reprints