Self-Adaptation in Dynamic Disaster Environments: Response to the 2006 Avian Influenza Crisis in Turkey

Abstract

This paper examines the response to the 2006 avian influenza crisis in Turkey. Using complex adaptive systems as the theoretical framework, the paper discusses the extent to which the Turkish disaster management system showed self-adaptation during the crisis. Self-adaptation requires organizational flexibility that facilitates sufficient information flow through technical and cultural infrastructures. This study uses qualitative methods to analyze the data. The research findings indicate that during the crisis, Turkish disaster management was faced with critical difficulties related to organizational, technical and cultural capacities that undermined its capacity to adapt to changing conditions. The system was able to manage these difficulties in seven to ten days; however, Turkey’s contemporary disaster response services still require a transformation to effectively respond to any influenza pandemic.

Keywords

Disaster management self-adaptation avian influenza public administration public policy organizational learning complex adaptive systems

Introduction

Avian influenza is an important source of risk for local communities. When avian influenza appears in the form of a pandemic, people and livestock are faced with a significant threat that spreads easily to different parts of the world. Past influenza pandemics killed 40 to 50 million people in 1918, two million people in 1957, and one million people in 1968 (World Health Organization, 2005). In the last decade, avian influenza has once again captured international attention due to outbreaks that killed humans as well as livestock. One type of avian influenza, the H5N1 virus, first appeared in China in 1996 and has the potential to be a pandemic disease.

Turkey is one of the risk-prone countries that faced an H5N1 outbreak in late 2005. The virus was first identified in a flock of domesticated turkeys infected by migrating birds in Manyas. By 2006, it had turned into a life threatening health crisis, spreading to 254 different local communities, large and small, in 54 provinces. This is a case study that focuses on the avian influenza outbreak of 2005 in Turkey. The research problem of this study is to identify the factors that contribute to or inhibit self-adapted intergovernmental response operations. As will be discussed in the theoretical section below, the organizational, technical, and cultural capacity of a disaster management system is closely related to its self-organization concept. Based on a theoretical framework that draws on the complex adaptive systems model, this study analyzes the organizational, technical, and cultural capacities of the Turkish disaster management system during the H5N1 outbreak. The research proposes a self-adaptive disaster response model in which many organizations with various cultures from different jurisdictions and economic sectors, as well as cities, can adapt to the dynamic conditions of health-related crises.

Theoretical framework

The tension between differentiation and integration is a critical problem that organization theory has recognized since the industrial revolution (Jaffee, 2001). While differentiation refers to an increasing demand for specialization and departmentalization, integration refers to the need for coordination and collaboration among different units and specialists. The tension is closely related to the lateral and vertical organization of tasks, people, units, and organizations, as well as the search for a good fit between differentiation and integration. Bureaucratic organization provides a model of coordination represented by the machine metaphor. It offers a rational design for tasks and organizational activities based on previously determined goals. According to Weber, a bureaucratic organization achieves speed, precision, reliability, efficiency, and objectivity through the fixed division of tasks, formal rules, hierarchical command, and control type management (Morgan, 1997: Weber, 1996).

Information flows follow hierarchical lines. Bureaucrats operating at the same level under the control of top management can share information if they can overcome departmental barriers and defensive routines. An important assumption of such a system is that it operates under stable or slowly changing environmental conditions. Therefore, the system can effectively manage routine tasks and programmed decisions. In this sense, a bureaucratic organization is a linearly-designed organizational model for integrating different tasks, specialized personnel, and units.

Drawing on the literature, it can be concluded that control and command oriented bureaucratic organizations face critical difficulties in adapting to new conditions in complex and dynamic environments (Axelrod and Cohen, 1999; Jones et al., 2013; Lewin and Sanger, 1994; Marion, 1999; Osborne and Gabler, 1992; Senge, 1994; Uri, 1995).

Moreover, the linear approach becomes ineffective when dealing with complex public policy problems in which many local, national, and international public, nonprofit, and private organizations must be integrated for collaborative work towards a common goal. This is especially characteristic of disaster management, which typically involves hundreds of organizations from different public jurisdictions, cities, and economic sectors to save lives in local communities during response and recovery operations. These organizations can operate with different missions, mandates, accountability criteria, and agendas (Bjorn and Olsen, 2012). Therefore, as a field of public policy, disaster management involves a high degree of uncertainty and complex relationships that require a complex organizational model based on improved organizational, technical, and cultural capacities.

The balance between differentiation and integration requires an increased capacity for speed, flexibility, and adaptation in today’s complex, uncertain, and changing disaster environments, especially at the local level (Jones et al., 2013). Complex adaptive systems theory better represents the complexity of disaster environments (Comfort, 1996). Based on complexity theory, the complex adaptive systems model can help multiple organizations from multiple jurisdictions achieve more effective coordination and collaboration, as well as adapting to changing conditions. Moreover, sufficient organizational capacity also requires contingency planning supported by sufficient staff to be able to respond to changing conditions (Lalonde, 2010). An important characteristic of complex adaptive systems is their sensitive dependence on initial technical, organizational, and cultural conditions prior to disasters.

A complex adaptive model requires a dynamic organization to satisfy the balance between structure and flexibility provided by sufficient organizational capacity. Organizational flexibility is critical for inter-organizational adaptation and learning, along with the capacity for information search, exchange, dissemination, and retrieval (Comfort, 1999). Lateral coordination and teamwork gain critical importance over vertical coordination; the inhibition of local initiative by central authorities, rigid rules and procedures, and a control and command oriented management style are all rejected in the complex model (Morcol, 2001). Organizational actors focus on adapting to changing conditions of disaster environments through reorganizing their subsystems and coordinating with other subsystems. As general norms for the operation of system-guided actor behavior, timely and valid information about local conditions assists organizations in adapting to each other and to changing environmental conditions.

A complex adaptive system must also have sufficient technical capacity parallel to organizational flexibility (Comfort, 2002). The need for information is one of the most critical aspects of effective disaster response (Celik and Corbacioglu, 2010; Yang, 2010). Because disaster management can be described as decision making subject to uncertainty, sufficient information channels to support the collection, analysis, and dissemination of information are needed (Bjorn and Olsen, 2012). Therefore, the disaster management system must invest in sufficient technical capacity for timely information flow among public, nonprofit, and private organizations operating at local, national, and international levels (Celik and Corbacioglu, 2012). Disasters require investments in advanced information technology in addition to traditional communication instruments. Such technology provides response support through the enhancement of situational awareness, with a focus on hazards, destruction, and resource allocation along with contributions to mitigation, preparedness, and recovery (Hamp et al., 2014). The 1999 Marmara earthquake revealed that “symmetry breaking” crises can disrupt telecommunication and the Internet through damage to and overloading of telephone lines or lost power (Celik and Corbacioglu, 2010). Sometimes cell phones may not operate in rural areas or will operate only sporadically. Therefore, satellite-based communication and information systems, GPS, GIS, and interactive databases are needed to ensure continuous and effective information flows before and after disasters. Technical capacity extends beyond information technology to include materials, tools, vehicles, medication, special protective clothing, proper construction practices, and infrastructure such as roads, water, and sewage systems that play critical roles during response and recovery operations.

Cultural capacity is the third type of capacity required by a self-adapting disaster response. A change and adaptation-oriented organizational culture (Schein, 1997) is a fundamental feature of a self-adapting disaster management system (Comfort, 1999). Inter-organizational coordination can only be achieved through organizational flexibility and technical infrastructure supported by a change-oriented organizational culture (Corbacioglu and Kapucu, 2006). A self-adapting disaster management system must be open to information sharing and interaction among organizations from different jurisdictions. The system must develop its cultural capacity for facilitating collaboration among organizations with different cultures and work methods. Moreover, a change-oriented organization culture encourages lower-level bureaucrats to take initiative (Comfort, 1999; Hosseini et al., 2013). Organizations that learn from and adapt to changing conditions can update governing values through valid information by achieving double-loop learning in disaster environments (Argyris, 1993; Maor, 2010). This happens through substantially modifying existing roles in response to a disruptive and non-routine event through adaptive behavior at the organizational and inter-organizational levels (Lalonde, 2010). Such a need requires timely and valid information and a flexible organization structure, both of which are critical for decision making.

The theoretical framework of this research involves the following assumptions regarding inter-organizational coordination and self-adaptation in response to infectious diseases such as avian influenza (H5N1):

As interactions among organizations increase through increased organizational capacity, they share more information and resources.

As the information flow rises, the capacity of managers for innovative decision making increases.

A sufficient technical capacity for continuous information flow among organizations and jurisdictions depends on a sufficient technical infrastructure.

As organizations search for, share, and disseminate more information and relocate more resources through increased cultural capacities, their ability to adapt to the environment increases.

Increasing adaptation capacity facilitates a timely inter-organizational cooperation that minimizes the vulnerability and losses of local communities.

Methodology

The methodology of this study follows from the problem statement and literature review. This study is an exploratory small-n case study analysis (Yin, 1993, 1994) that seeks to identify the factors that can contribute to the development of a self-adaptive emergency health services system as a part of a disaster management system. The research focuses on factors that inhibited and facilitated a timely and coordinated inter-organizational response to during the 2006 avian influenza (H5N1) outbreak in Turkey. The study analyzes the Turkish disaster system in three categories developed and suggested by Comfort (1999): organizational, technical, and cultural. The following data sources are used:

37 semi-structured interviews with important public and nonprofit managers from the local, national, and international levels;

official reports;

related laws and regulations;

analyses by researchers belonging to national and international professional organizations;

daily news reports from two national newspapers, Cumhuriyet and Sabah, for 21 days following the avian influenza crisis.

The research used purposive sampling for the selection of the case and the communities affected. The avian influenza crisis of 2006 was the only health-related disaster that affected Turkey in recent years, and it helped us see the capacity of the Turkish disaster management system to respond a wide-spread health-related crisis in Turkey and access the units of observations. The study used purposive sampling in selecting among the 254 communities through which the virus spread, and ultimately the Dogubeyazit district and the provinces of Agri, Erzurum, and Van were selected for the sample. Although avian influenza was detected in many different regions, Eastern Turkey, especially the provinces of Agri and Van, were at the center of the crisis, and the Dogubeyazit district of Agri experienced the crisis at the highest level. All human losses (four individuals) were from Dogubeyazit. Considering the total number of infected human cases, 6 of 21 occurred in Dogubeyazit. Van province, as Agri’s neighbor, experienced both human and poultry cases and served as a medical center. Erzurum province experienced poultry cases, served patients, and became the regional center for crisis response sites in Van and Agri.

The study used stratified sampling to choose the organizations for interviews. Stratified sampling is recommended for heterogeneous groups comprising the population (Kerlinger and Lee, 2000). Stratified sampling increased the validity of the results and helped save time and money. The interviews were conducted with two groups of organizations: public organizations that were officially responsible or played critical roles in response and recovery operations, and nonprofit organizations and universities that were actively involved in operations or field research.

The sampled public organizations included five types: (1) central organizations at the national level; (2) provincial organizations; (3) district organizations; (4) municipalities; and (5) international organizations. When selecting among public organizations, three criteria were used:

legal responsibility;

participation in decision making processes during response operations;

frequency of interaction with other organizations involved.

The criterion for the selection of nonprofit organizations was active involvement in response operations. The criteria for the universities included operating at the disaster site and active involvement in the avian influenza crisis.

The units of analysis for this research were organizations that were involved in the response operations while the units of observation were managers and researchers who were involved with the crisis.

The research used qualitative methods to explore the factors that influence inter-organization coordination and adaptation. Analyses also involved descriptive statistics.

Context of the case

Avian influenza (H5N1 virus) was first identified on October 8, 2005, in a flock of domesticated turkeys, most likely infected by migrating birds in Manyas (World Health Organization, 2006). The Ministry of Food, Agriculture, and Livestock (FAL) attempted to control the disease by culling the poultry birds in quarantined areas.

The analysis of two Turkish newspapers, Cumhuriyet and Sabah, indicated that 482 organizations participated in response operations. Although international organizations such as the World Health Organization (WHO) and the United Nations Food and Agriculture organization made important contributions, national organizations from several Turkish jurisdictions dominated the response operations. While public organizations dominated the response operations, small numbers of private and nonprofit organizations were also involved.

It was necessary to address the avian influenza crisis as a process. The first phase of the crisis included the appearance of animal cases and was dominated by the efforts of FAL¹ to bring the disease under control. The period between October 8, 2005, when the H5N1 virus first appeared in Manyas, and January 1, 2006, the date that the first patient from Dogubeyazit died, is the first phase. The second phase of the crisis began on January 1, 2006, the point at which the disease became a health crisis, and extended to the time when outbreaks of the disease were extinguished by the end of March 2006 (Ministry of Food, Agriculture, and Livestock, 2006). In the second phase of the crisis, the Ministry of Health and the Ministry of Food, Agriculture, and Livestock carried out parallel and intensive work, especially at the local level.

Organizational capacity

The organizational capacity of Turkish disaster management is described in this section under the subtitles of organization and coordination, emergency planning, availability of professional and educated human resources, and open communication channels.

Organization and coordination

According to the Turkish crisis management structure of the time, the Prime Ministry Crisis Management Center (PMCM) starts operations at the central and local levels if the situation is called a crisis by the Prime Minister. Different public organizations, such as the National Security Council, Turkey Emergency Management General Directorate of the Prime Ministry, Economy Ministry and Health Ministry take charge for the coordination of crisis management in different type of crises. For example, the Health Ministry is responsible for outbreaks of infectious disease. Interestingly, the PMCM structure does not give the responsibility of coordination to FAL or specifically relate it to any type of crisis. During a crisis, local search and rescue committees become district/provincial crisis management centers.

Turkey did not officially use the crisis management structure because the avian influenza outbreak was not declared a crisis by the Prime Minister; however, although the PMCM structure was not officially activated, the disaster management system’s operation at the central and local levels was based on PMCM principals (interview with Turkish General Directorate of Disaster Affairs).

Although the Health Ministry was responsible for coordinating any health-related crisis according to the PMCM regulation, the H5N1 outbreak started as an animal disease concerning what was then the Ministry of Food, Agriculture, and Livestock. As a result, FAL focused its efforts on controlling the disease without any close coordination with other public organizations and emerged as the main actor between the identification of the H5N1 virus in Manyas on October 8, 2005, and the first loss of human life on January 1, 2006. During this period of time, the local rescue and search committees were not sufficiently activated and medical surveillance efforts for the virus were not satisfactory. The identification of human cases preceded that of animal cases in many places (Ceyhan, 2006; World Health Organization, 2006). As a result, systematic culling efforts of poultry in Dogubeyazit district were not started until after the loss of a human life (interview with Agri Food, Agriculture, and Livestock Directorate).

After the scientific confirmation of H5N1 infection in humans on January 4, 2006, the Health Ministry and FAL set up crisis desks and informed the WHO of the developments, at which point the Health Ministry began working together with WHO (World Health Organization, 2006). Beside the crisis desks, on January 14, 2006, following a meeting of the Health Ministry, FAL and the Environment and Forestry Ministry, the Avian Influenza National Coordination Center was established (Sabah Daily, January 15, 2006).

Despite the Health Ministry’s transparent management of the crisis and its close cooperation with WHO (World Health Organization, 2006), critical problems occurred during the initial outbreak:

It is not true to say everything was going well in the Health Ministry…Two right things done by the Ministry cannot cover up negative issues. Then, what happened? Did not we have information that the avian influenza would spread to our country [Turkey]? We did. But our plan was not ready. It was at the first stage. If we were prepared very well, we would not confuse for the first 10 days…Then we put things in order. But that time period was a nightmare (interview with the Health Ministry General Directorate of Primary Health Care Services, Department of Infectious Diseases).

As shown in Table 1, when interviewees were asked about the extent of timely intergovernmental action, out of 37 respondents, 19.4% stated that the extent of coordination was very good, while 27.8% stated that it was good. The remainder of the informants took a more pessimistic view: while 27.8% of respondents thought there was a timely intergovernmental response to some extent, 25% felt that the intergovernmental response was not timely to any extent.

Table 1.

Intergovernmental coordination during avian influenza response operations.

To what extent were organizations operating at the central, provincial, and local levels coordinated to ensure timely intergovernmental action after the crisis?
	Number	Percentage (%)
To a great extent	7	19.4
To a good extent	10	27.8
To some extent	10	27.8
Not to any extent	9	25.0
Not at all	0	0.0
N/A	1	0.0
Total response	37	100

Source: Semi-structured interviews conducted between January 6, 2008, and May 30, 2008.

The findings from the interviews and content analysis indicate that the coordination of the crisis at the provincial and district levels varied. The interviews indicated that in comparison to Agri province, Erzurum and Van provinces were quicker to organize to take steps to control the outbreak.

Then [the coordination of crisis] differed from one province to another. Let’s name it; Agri was insufficient. But Van was sufficient to a good extent. The coordination capability of the governorate was the determinant…The governorate in Agri did not even realize what was going on. A crisis management center could neither be set up nor overcome [the crisis] until the crisis ended. Then things found their way and returned to routine (interview with the Health Ministry General Directorate of Primary Health Care Services, Department of Infectious Diseases).

Erzurum emerged as a regional coordination center. As it responded to animal cases in its Horasan district, Erzurum also sent specialists to neighbor provinces. An important factor distinguishing Erzurum from its neighbors was the existence of a 24-hour active provincial crisis management center that worked as the information center (interview with Erzurum Civil Defense Department). The presence of Palandoken Mountain, a tourism center known as a climbing, hiking, and skiing destination, was an important motivation for establishing an active crisis management center supported by some information technology to encourage timely responses to any accident or crisis before the avian influenza outbreak:

The event got a timely response (in Erzurum). Received information was checked by the related unit and was acted upon accordingly. The biggest advantage of Erzurum was the existence of the crisis management center. Our crisis management center can easily reach all units and give orders. [It] had the capacity to do what was required in time…We had the telephone numbers of all units…The crisis management center could be easily reached by us or the citizens (interview with Erzurum Crisis Management Center).

When interviewees were asked about the factors that impacted inter-organizational coordination and collaboration, 37 respondents (who were allowed to select multiple responses²) identified the following factors:

quality of personnel (21%);

realist contingency planning (15.4%);

capacity of response units (15.4%);

new organizational design that helps coordination and collaboration (13.3%);

decentralization of administration (11.9%);

mitigation policies for reducing disaster risk (10.5%);

timely information (8.4%);

change-oriented organization culture (4.2%).

Emergency planning

The Health Ministry initiated a national-level meeting to develop a national pandemic influenza plan with FAL, universities, and the firms that import influenza medicine in July 2005 after an international meeting that brought the WHO, the European Commission, the European Center for Disease Presentation, and 52 countries together to prepare national influenza pandemic plans in March 2005 (Beyazıt, 2005). The plan was shared with the public in October 2005 (Buzgan and Uzan, 2006). While the plan positively influenced the central organizations’ decision making and coordination processes, it could not be localized to districts and provinces affected by the time of the crisis (World Health Organization, 2006). According to the WHO, the problems with preparation, combined with bad weather conditions and material insufficiencies, negatively affected the response operations at the provincial and district levels in the first several days (World Health Organization, 2006).

Table 2 shows the extent to which emergency plans and procedures were exercisable.

Table 2.

Execution of emergency plan during avian influenza response operations.

To what extent were emergency plans and procedures implementable?
	Number	Percentage (%)
To a great extent	0	0.0
To a good extent	5	13.9
To some extent	18	50.0
Not to any extent	12	33.3
Not at all	1	2.8
N/A	1	0.0
Total response	37	100.0

Source: Semi-structured interviews conducted between January 6, 2008, and May 30, 2008.

Of 37 respondents, 13.9% said that the plans and procedures were implemented to a good extent, 50% said they were implemented to some extent, and 33.3% said they were not implemented to any extent. One person, 2.8% of respondents, said that the plans and procedures could not be exercised at all.

The interviews indicated that the local authorities in Van, Dogubeyazit, Agri, and Erzurum did not have any specific plan of action regarding avian influenza (interviews with Van Health Directorate, Dogubeyazit State Hospital, Van State Hospital, and Erzurum Health Directorate). A respondent from Van Province explained the process as follows:

We had not faced such a crisis [before]. Therefore, we did not have any written bird flu emergency action plan. Rapidly, by using initiative, we prepared it. In a week, the [Health] Directorate and us, had an emergency management plan indicating who would do what.

Professional and educated human resources

Having professional and educated human resources in disaster response operations is crucial to minimize the risks to the safety of lives and property in local communities. In this regard, although the directorates of Health and Food, Agriculture, and Livestock play major roles, other public, private, and nonprofit organizations can also provide substantial contributions during the crisis.

When asked about the degree to which professional, educated human resources were available to be utilized during the avian influenza disaster response operations, 3% of respondents claimed that they had access to professional and educated human resources to a very good extent, whereas 8,3% noted they had access to a good extent, 41,7% to some extent, and 47.2% said they did not have access to any extent. Of the respondents that answered, 2.8% that they had no professional and educated human resources at all.

There was a particular need for support personnel and veterinarians to be employed in culling poultry during the first days of crisis (interviews with Van, Erzurum, Agri, and Dogubeyazit Food, Agriculture, and Livestock Directorates). A shortage of doctors and support personnel was deeply felt in Dogubeyazit and Agri as worried citizens overcrowded the hospitals when they learned of human and animal cases (interviews with the Health Ministry General Directorate of Primary Health Care Services, Agri Women and Children Hospital, and Dogubeyazit State Hospital). However, the shortage of personnel in these provinces was largely brought under control in the first 10 days of the crisis. Personnel assigned from the Health Ministry, FAL, and international organizations, as well as from neighboring provinces, made up for the personnel deficiency in the stricken communities (interview with the Health Ministry General Directorate of Primary Health Care Services).

Open communication channels

There is a close relationship between inter-organizational collaboration and the adequacy of information search, exchange, and dissemination. Sufficient information sharing during a crisis requires the development of open communication channels for facilitating information flows among organizations before a crisis.

When asked about the degree to which there were open communication channels among disaster response organizations prior to the crisis, 30.6% of 37 responses said there were communications to a good extent, 44.4% said to some extent, 22.2% said not to any extent, and 2.8% of respondents said that they had no open communication channels at all.

When we evaluate the interview data, communication between organizations varied at the district and province levels. In this respect, the province of Van shows a deficiency in information sharing (interviews with Van Health Directorate and Van State Hospital). In contrast, before the crisis, a positive relationship developed between the personnel of Ataturk University, Aziziye Education and Research Hospital, and the Directorates of Erzurum Health and Food, Agriculture, and Livestock, which provided significant support during the emergency operations (interviews with Ataturk University, Aziziye Education and Research Hospital, Erzurum Health Directorate, and Erzurum Numune Hospital).

The interview data also show that no special efforts or preparedness for information exchange and sharing were enacted, other than personal communication and formal correspondence between bureaucrats in Agri and Dogubeyazit prior to the crisis (interviews with Agri Crisis Management, Agri Health Directorate, and Agri Food, Agriculture, and Livestock Directorate). It is difficult to conclude that this personal communication, described as a “sympathetic relationship,” was sufficient to create open communication channels through which to exchange and share information during the avian influenza crisis. The following anecdote from a respondent provides some insight regarding the situation in the field:

The ones who came from outside did not know the job well. They also learned the job here…We did not have any communication before. The Ministry sent them. There was a case in Manyas, Balikesir…sent us expert veterinaries. They came and sat in a room…[They said] “We will inform you (about the disease).” [We said], “We do not need information today. We need someone who will jump on and catch the bird” (interview with Dogubeyazit Food, Agriculture, and Livestock Directorate).

Technical capacity

During the avian influenza crisis, the existing technical infrastructure was not faced with any crucial issues or damage that could prevent ordinary communications. Ordinary communications devices such as telephones, mobile phones, transmitters, and the Internet continued to provide standard performance. In the avian influenza scenario, which morphed into a medical crisis with the appearance of human cases, respondents were asked about the communications devices and methods they used in the initial three-day period. Respondents said that they were able to use regular telephones and mobile phones at first because the crisis did not damage communication channels. During the crisis, it was observed that satellite telephones were not used in the cities of Agri, Erzurum, and Van; however, the communications infrastructure in Erzurum was better prepared for crisis. Although it was not needed during the avian influenza crisis, a ready-to-use mobile center with the capacity for 100 subscriptions was kept in reserve to ensure uninterrupted communication at least between major actors in Erzurum. Furthermore, a wireless telephone that provided satellite-based contact with Mount Palandoken was in service to follow the risk of avalanche. Following the avian influenza crisis, the number of satellite telephones in Erzurum was increased to 10 (interview with Erzurum Crisis Management Center).

Field searches and managers’ responses indicated that some governmental institutions had access to geographic information systems, global positioning systems, satellite-based communication, and remote sensing systems at the local or central level. Despite the presence of those systems, however, there was no evidence of their use in crisis management. Only five respondents out of 37 mentioned that they had access to an information/data bank to share information with central- or provincial-level governmental organizations. One of these organizations, the Prime Ministry General Directorate of Emergency Management, later replaced by the Prime Ministry Disaster and Emergency Management Presidency, prepared software to enable the online exchange of information between central organizations; however, the software was not put into action during the avian influenza crisis (interview with Prime Ministry General Directorate of Emergency Management).

Because avian influenza did not become a pandemic, regular communication devices were sufficient during the crisis. A communications crisis similar to that experienced in the aftermath of the 1999 Marmara and Duzce earthquakes in Turkey could occur in the case of a national or international influenza pandemic, due to overloading of the telephone lines. In such a case, telephones or mobile phones, which were the essential communications technologies used during the 2006 crisis, would not be available when they were most necessary.

Interviewee institutions and organizations’ responses concerning the factors that provided continuous communication abilities and increased the existing communication and information capacity of the system are shown in Table 3.

Table 3.

Factors affecting information capacity.

What factors provide constant information flows among emergency organizations and increase existing information/communication capacity?
	Number	Percentage (%)
Planning for information/knowledge sharing within the institutions and organizations in advance	31	30.7
Utilization of advanced information/communications technologies	22	21.8
Training public officers in emergency communications	17	16.8
Sustaining the flow of information between organizations involved in crisis response	15	14.9
New organizational structure that allow the flow of information between organizations	11	10.9
Coordination in the field of information technology investment in governmental organizations	5	5.0
Total response	101	100.0

Source: Semi-structured interviews conducted between January 6, 2008, and May 30, 2008.

Respondents could select multiple answers. N=37.

Cultural capacity

The cultural capacity of the Turkish disaster management at the time of the avian influenza crisis is described in this section under two subheadings: learning from past experience, and learning and adaptation during the response operations. The second subheading comprises three subsections, namely information search, information sharing and distribution, and openness to change.

Learning from past experiences

According to Chris Argyris, organizational learning and change requires a change in the governing values of decision makers based on acquired information (Argyris, 1993). Avian influenza caused the loss of human lives around the world and has impacted different parts of the world on a daily basis since 2002. The risk was high for human beings, especially in countries where there was close contact between humans and poultry, which was precisely the case in Turkey’s most rural communities (interview with the Ministry of Food, Agriculture, and Livestock, General Protection and Control). The critical question to be asked here is to what extent local people and government officers were aware of the risk and whether action was taken to reduce it.

Following the identification of the H5N1 virus in Manyas on October 8, 2005, local authorities were unable to prepare themselves sufficiently until the virus killed a child on December 31, 2006 (World Health Organization, 2006). Moreover, ordinary citizens and some key personnel were not aware of the risk to which they were exposed (Kapucu et al., 2010).

Although the Health Ministry prepared posters and some written materials following the discovery of poultry cases in Manyas (Buzgan, 2006), it is difficult to conclude positively that these efforts were sufficient to increase awareness of local people (interview with Health Ministry General Directorate of Primary Health Services). This finding is also important in that it shows the need for, and importance of, community-based, local nonprofit organizations along with public organizations to increase awareness among communities (Hosseini et al., 2013).

Insufficient information regarding the disease among public personnel also negatively affected the risk evaluation. Despite the Health Ministry’s preparedness efforts at the national level, local health units did not have sufficient information or experience regarding the diagnosis and treatment of the virus (interviews with Agri Health Directorate, Van State Hospital, and Erzurum Health Directorate). Ataturk University Aziziye Education and Research Hospital was an exception, where doctors from the Infectious Diseases and Clinical Bacteriology department studied the issue and benefited from international experience in following the human cases to prepare themselves for patients.

When members of interviewee organizations were asked what factors are effective in learning from past experiences and increasing organizational adaptation to dynamic environmental circumstances, respondents ordered a series of factors from most important to least important:

qualified personnel (31.4%);

national disaster policy (24.5%);

change-oriented organizational culture and leadership (23.5%);

organizational structure that facilitates learning and adaptation (10.8%);

academic research (6.9%);

public pressure for change (2.9%).

Learning and adaptation during the response operations

Disasters require organizations to expedite their decision making processes based on valid and timely information. Moreover, an effective culture must enable them to act with other organizations guided by different working principles and methods from different jurisdictions. Therefore, issues regarding information search, exchange, and dissemination, as well as cultural openness to change, are critical to adapt to the changing conditions of dynamic environments.

Information search

The transparent dialog between the Health Ministry and the WHO that began before the crisis increased the information capacity of the Turkish disaster response system. The General Directorate of Primary Health Care Services had informed local health directorates about avian influenza, its current status in the world and the best protective measures, through a written notice in 2004. Moreover, the first version of the National Pandemic Plan had been publicized in October 2005 (Buzgan, 2006); however, it is difficult to say conclusively that local health organizations were knowledgeable about H5N1 virus, its symptoms, and treatment. The written notice tended to be filed rather than implemented to change the governing values of professionals regarding the risk of avian influenza. In Dogubeyazit, the epicenter of human cases of avian influenza, doctors were not aware of the efforts at the central level:

There had been written notices regarding the bird flu in 2005. I transferred the first patient on December 31 2005…[the Health Ministry’s] written notices were conveyed to us later…We were made to sign them hastily. In fact, we did not know about them. They appeared after the [avian influenza] cases (interview with Dogubeyazit State Hospital).

Another respondent in Erzurum stated that they did not know “how to approach the case, how to get the sample when the avian influenza crisis started” (interview with Erzurum Province Health Directorate). The same situation prevailed in Van. Ataturk University’s Aziziye Education and Research Hospital quickly responded to the crisis by first learning about the disease in detail and then organizing conferences in two state hospitals in the first days following the crisis (interview with Ataturk University Medical School Hospital). Interestingly, all patients sent to Erzurum survived, whereas deaths occurred among the individuals sent to Van’s 100 Year University Medical School Hospital.

The other important subject closely related to information search is surveillance. These efforts can be described relative to two groups, birds and humans. The surveillance efforts targeting birds, both domesticated and wild, were insufficient. As a result, infections in birds were not revealed until after the H5N1 virus had infected humans (Ceyhan, 2006; World Health Organization, 2006).

Although Turkey participated in the International Influenza Surveillance Network and had begun monitoring seasonal influenza, these were general surveillances activities, and no systematic surveillance work targeting avian influenza appears to have been implemented prior to the crisis. In fact, following animal cases in Manyas, the Health Ministry had worked on the issue; however, the system was not developed or implemented until after the crisis (Buzgan, 2006). Comprehensive regulation regarding infectious disease surveillance and control were put into practice in May 2007 through a directive from the Health Ministry.

Information sharing and distributing

In addition to information search and acquisition, the use of reliable and accurate information in the decision making process helps coordinate organizational actors and adaptation to dynamic disaster environments. This requires focus on information sharing and distribution before disasters occur.

When examining how frequently the responsible crisis organizations exchange information on a regular basis, while 5.4% and 10.8% of respondents stated that they shared information or knowledge every day and every week, respectively, 24.3% responded that they exchange information or knowledge with others monthly. A total of 59.5% of respondents noted that they did not share information or knowledge on a regular basis.

Along with the progression of the avian influenza outbreak, some public organizations did not share sufficient information with other organizations or the public in order to avoid panic and negative consequences for the poultry industry and the national economy. It is understood that there was insufficient information sharing among public organizations due to the desire to avoid a public panic and harm to the poultry sector or the economy. The attitude of the FAL regarding information sharing was criticized by some interviewees:

[The] Ministry of Agriculture…knew about the H5N1 virus but covered it up. They hid it because it could affect the poultry business. Bornova Veterinary Institute, which is a reference institute, said that Ministry of Agriculture had identified the disease 20–30 days before it was publicly known. Even so they ignored it, hid it. They hoped to control it. But when it spread to human beings, there was nothing to do (interview with Health Ministry General Directorate of Primary Health Care Services, Department of Infectious Diseases).

The interviews showed that there was a tendency, especially among public organizations, not to share information with other organizations. When public, private, and nonprofit organizations are evaluated in terms of their willingness to share information, it appears that public agencies are more rigid in their interactions compared to private and nonprofit groups. When asked about the extent to which participants were able to gather sufficient information from other public and nonprofit organizations during the response operation, 8.1% of respondents said they were able to gather information to a very good extent, 32.4% said to a good extent, 37.8% said to some extent, and 21.6% said that they were not able to gather information to any extent.

The interviews suggest that there is a cultural resistance to information and knowledge sharing. One of the respondents described the issue as follows:

I mean every institution is somewhat inscrutable…if you ask any information, document about my work here, I would ask you 40 questions. What do you want? Why do you want it? I mean, there is no transparency. There is mistrust. I wonder what you/he will do with this document. Even if it is public institution that asks for the document (interview with Dogubeyazit Crisis Management Center).

A good example of the resistance to information sharing is found in the communication problems that occurred between Health Ministry organizations and the Van 100 Year University Medical Faculty Education and Research Hospital. According to the WHO (2006), the problems occurred due to doctors’ reluctance to share their data because they wanted to keep the data private with the goal of eventual academic publication. The Van 100 Year University Medical Faculty Education and Research Hospital claimed, however, that their doctors did not share information because it went against their working principles and endangered patient confidentiality.

Openness to change

A change-oriented organizational culture requires the elimination of barriers to open communication as well as adopting a decision making process based on valid, accurate, and timely information. The criteria of responses help us to understand the importance of information in decision making processes. Organizations learn from, as much as they use, valid and reliable information to find errors and anomalies during response operations.

Therefore, the interviewees were asked about the criteria for correcting errors and anomalies used during response operations. An analysis of interviewee responses in Table 4 shows that only 14% of respondents used valid and reliable information to correct errors and anomalies. The interviewee responses revealed that the organizations involved with the avian influenza crisis reflected the values of bureaucratic organization. The two criteria with highest frequency, “rules and regulations” and “the decision of the immediate manager,” are compatible with the nature of bureaucratic organizations.

Table 4.

Criteria used to correct anomalies during avian influenza response operations.

What were your criteria to correct errors and anomalies during the avian influenza response operations?
	Number	Percentage (%)
Existing rules and regulations	19	44.2
Valid and reliable information	6	14.0
Organizational strategy	3	7.0
There were no certain criteria	5	11.6
Decision of immediate manager	10	23.2
Total response	43	100.0

Source: Semi-structured interviews conducted between January 6, 2008, and May 30, 2008.

Respondents could select multiple answers. N=37.

Organizational learning cannot be achieved unless the acquired information is used to address problems. The bureaucratic organizational structure and culture enforces one-loop learning in Argyris’ (1993) sense by looking for solutions through pre-determined strategies, rules, and superordinate orders. In contrast, disaster environments require double-loop learning by which decision makers can revisit and update the governing values of the system through reliable and valid information. To achieve double-loop learning, however, decision makers should not be strictly bound by rules and regulations. Therefore, the extent to which organizations go beyond the status quo, take initiative, and make informed decisions is critical to evaluating openness to change.

As a result, respondents were asked what they did or would do during crisis response operations if/when they were faced with a situation in which existing rules and regulations were in conflict with the values and missions of their organizations. Of the respondents, 42.9% connoted that they would follow the organization’s values and missions and 40.5% said that they would inform their superiors of the situation and wait for their guidance, whereas 16.7% said that they would carry out existing rules and regulations. The interviewee responses again revealed the significant influence of bureaucratic culture by which the majority of respondents feel obligated to either wait for their superior’s directions or simply follow existing rules and regulations. One of the respondents explained the dilemma as follows:

Let me put it this way: complying with the existing rules and regulations is the principle. If you are bound by rules and regulations and are still being expected to act, the situation depends on your personal decision, initiative. Because you may get harmed at the end [for your action]…In normal times, if there is no crisis, rules and regulations are followed (interview with Agri Crisis Management Center).

Finally, the researchers inquired about the factors that could improve the cultural infrastructure of Turkish disaster management. Based on the experience of managers and specialists, interviewees were asked about the factors that could facilitate or inhibit learning from errors and anomalies and adaptation to the changing conditions of disaster environments. Respondents answered as follows:³

quality of personnel (31.4%);

existence of a national disaster policy (24.5%);

change-focused organizational culture and leadership (23.5%);

new organizational structures that facilitate learning and adaptation (10%);

academic research (6.9%);

public pressure for change (2.9%).

After evaluating the findings regarding the cultural infrastructure of Turkish disaster management, we argue that the response system and traditional organizational structures must be revisited. Therefore, creating a flexible organization structure and sufficient information channels will not be sufficient unless certain features of bureaucratic culture are also challenged.

Conclusion

Avian influenza, like many other types of influenza, exposes local communities to significant risk. Such infectious diseases become a great threat to human life if they develop into pandemics, as has occurred several times in the last century. Despite its problems, Turkish disaster management was able to control avian influenza and reduce the harm it inflicted on humans; however, this does not mean that response operations were effectively managed. Any influenza pandemic could have caused far higher losses in terms of human lives and the economy. This case study indicates the importance of the organizational, technical, and cultural conditions in place prior to an influenza outbreak in timely and effective response operations.

The research findings indicate that linearly-designed public policies and the bureaucratic organization of disaster systems cannot cope with the complexity of rapid change. Although the H5N1 virus was identified on October 8, 2005, the Turkish disaster management system took a step-by-step approach to an important health crisis that cost four human lives and more than 2 million domesticated birds (Arslan, 2007) by April 2006. This case shows that a flexible and integrated organizational structure designed using the principles of complex adaptive systems theory is needed for continuous information search, exchange, and decision making. The findings also indicate that preparedness for influenza at the central level does not produce effective results unless the local capacity for preparedness is increased through access to sufficient human resources, training, and open communication channels at the provincial and district levels. The difference in initial conditions is shown in terms of a 24-hour crisis management center that worked effectively and in close cooperation among various critical organizations at the local level. These factors made a positive difference for the operations in Erzurum following the outbreak.

Disaster response systems must be supported by sufficient technical infrastructure to facilitate information search, exchange, and dissemination. The avian influenza crisis in Turkey was not a symmetry-breaking event like the 1999 Marmara earthquake. As a result, contemporary communications infrastructure involving telephones, mobile phones, and two-way radios was helpful. If the H5N1 influenza outbreak had developed into a pandemic, however, the traditional telecommunication channels would have been overloaded. The avian influenza crisis showed that in addition to alternative communication technologies, investment in information technologies, data banks for effective virus surveillance, and resource management are also needed.

The study shows that the public’s awareness of risk is a critical aspect of disaster preparedness and response. During the avian influenza crisis, Turkish disaster management was not sufficiently effective in reducing the complacency of the people and some key public personnel. Moreover, FAL’s avoidance of sharing information regarding the outbreak in birds negatively affected the preparedness both of Turkish society as a whole and of local medical units. In conjunction with bureaucratic resistance to information sharing and cooperation with different organizations, this attitude decreased the capacity of the disaster management system to learn and respond to changing conditions. Moreover, research indicates that the bureaucratic organizational culture inhibits organizational learning through the ingrained hesitation of decision makers to use timely and valid information rather than the rules and regulations or supervisors’ orders that are relied upon during regular operations. Instead, a complex disaster management system for complex disaster environments requires decision makers to take initiative and collaborate with organizations from different economic sectors and policy arenas. The Turkish response to the avian influenza crisis reflects single-loop learning (Argyris, 1993), whereby all resources and personnel with the international support are deployed to the critical points. As a result, millions of domesticated birds were culled, and medical doctors were quickly trained in how to manage human cases. Ultimately, although the disease was controlled, the governing values of the system remain in need of change.

Footnotes

Acknowledgements

We would like to thank Dr Louise K Comfort for her time in reading and commenting on drafts of this paper.

An earlier version of this study was presented at the 7th Public Administration Forum of Turkey in 2009.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by a quick response grant from the Scientific and Technological Research Council of Turkey (TUBITAK) (Grant Number: 1002-7269/2009).

Notes

Author biography

Sitki Corbacioglu, PhD, is an Associate Professor in the Department of Political Science and Public Administration at the Eskisehir Osmangazi University, Turkey. He received his Master’s in Public Management at Carnegie Mellon University in 1998, and finished his PhD at the University of Pittsburgh in 2004. His research area involves disaster management, public policy and organizational learning.

Suleyman Celik, PhD, is an Associate Professor in the Department of Public Administration at the Gazi University, Turkey. He received his Master’s in Public Management at Carnegie Mellon University, and finished his PhD at the University of Pittsburgh in 2006. His research area involves disaster management, network analysis and public policy.

Ulvi Saran, PhD, is an Associate Professor (adjunct) at İstanbul Aydın University, Turkey. He received his Master’s in public administration at Ankara University in 1990, and finished his PhD at the same university in 2002. Mr. Saran worked as province governor, the deputy undersecretary of Health Ministry, and the undersecretary of Public Order and Security Agency of Turkey. His research area involves organization theory, public policy and public security.

References

Argyris

(1993) On Organizational Learning. Cambridge, MA: Blackwell.

Arslan

( 2007) Küresel Kamu Malı Olarak Sağlık Ve Kuş Gribinin Türkiye Ekonomisine Etkileri (Health in the Form of Global Public Good and Impacts of Avian Influenza on Turkish Economy). Eskisehir, Turkey: Anadolu University Press.

Axelrod

Cohen

(1999) Harnessing Complexity. New York: Free Press.

Beyazıt

(2005) Bir İnfluenza Pandemisine Hazır Mıyız? Refik Saydam Hıfzıssıhha Merkezi Aylık Epidemolji Raporu (Are we ready for an influenza pandemic? Refik Saydam Public Health Center Monthly Epidemic Report) 4(5), September–October.

Bjorn

Olsen

(2012) Knowledge creation and reliable decision-making in complex emergencies. Disasters 36(2): 212–232.

Buzgan

(2006) Avian influenza, Temel Sağlık Hizmetleri Genel Müdürlüğü Sunumu (Avian influenza, presentation of General Directorate of Primary Health Services). Available at: www.klimik.org.tr/etkinlikler/belgeler/20052006/ankara/kusgribi/turan_buzgan_23_ocak_2.pdf (accessed November 14, 2008).

Buzgan

Uzan

(2006) Avian influenza: Genel Değerlendirme ve Bu Konuda Yapılan Çalışmalar. Refik Saydam Hıfzıssıhha Merkezi Aylık Epidemolji Raporu (Avian influenza: general assessment and efforts undertaken. Refik Saydam Public Health Center Monthly Epidemic Report) 5(1), January–March.

Celik

Corbacioglu

(2010) Role of information in collective action in dynamic disaster environment. Disasters 34(1): 137–154.

Celik

Corbacioglu

(2012) Inter-organizational coordination: analysis of 2011 Van earthquake. In: Kapucu

Hawkins

Rivera

(eds) Disaster Resiliency: Interdisciplinary Perspectives. New York, NY: Routledge.

10.

Ceyhan

(2006) Avian influenza, yaşadığımız bir salgının ardından. Çocuk Sağlığı Ve Hastalıkları Dergisi (Avian influenza, following the experienced outbreak. Journal of Children Health) 49: 81–89.

11.

Comfort

(1996) Self -organization in disaster response: the Great Hanshin, Japan earthquake of January 17, 1995. Natural Hazards Center Quick Response Report, No. 78. Boulder, CO: University of Colorado.

12.

Comfort

(1999) Shared Risk: Complex Systems in Seismic Response. Amsterdam: Pergamon.

13.

Comfort

(2002) The Impact of Information Technology upon Disaster Mitigation and Management [Pre-publishment Copy]: University Of California Press.

14.

Corbacioglu

Kapucu

(2006) Organizational learning and self-adaptation in dynamic environments of disasters. Disasters 30(2): 73–102.

15.

Cumhuriyet daily newspaper, daily news reports between December 28, 2005, and January 17, 2006, Istanbul.

16.

Hamp

Reindl

Güthlin

(2014) Decision-making behavior during urban search and rescue: A case study of Germany. Disasters 38(1): 84–107.

17.

Hosseini

Hosseinioon

Pooyan

(2013) An investigation into the socioeconomic aspects of two major earthquakes in Iran. Disasters 37(3): 516–535.

18.

Jaffee

(2001) Organization Theory. Boston: McGraw-Hill.

19.

Jones

Aryal

Collins

(2013) Local-level governance of risk and resilience in Nepal. Disasters 37(3): 442–467.

20.

Kapucu

Corbacioglu

Gareyev

Saran

(2010) Managing public complacency and public preparedness in response to 2006 avian influenza crisis in Turkey. European Journal of Economic and Administrative Sciences 3(1): 41–60.

21.

Kerlinger

Lee

(2000) Foundations of Behavioral Research. Fort Worth, TX: Harcourt College Publishers.

22.

Lalonde

(2010) Organisational socialisation in a crisis context. Disasters 34(2): 360–379.

23.

Lewin

Sanger

(1994) Making Government Work. San Francisco: Jossey-Bass Publishers.

24.

Maor

(2010) Emergency preparedness consultants at the local government level: the Israeli experience. Disasters 34(4): 955–972.

25.

Marion

(1999) The Edge of Organization. Thousand Oaks, CA: Sage Publications.

26.

Ministry of Food, Agriculture, and Livestock (2006) Sunuşu Yapılan Hastalık Mihrakları, 4 Aralık Tarihli Faks Mesajı (Presented Disease Focal Points, December 4th Fax Message).

27.

Morcol

(2001) The practice of interdisciplinary: complex conditions and the potential of interdisciplinary theory. Issues in Integrative Studies 19:123–136.

28.

Morgan

(1997) Images of Organization. Thousand Oaks, CA: Sage Publications.

29.

Osborne

Gabler

(1992) Reinventing Government. Reading, MA: Addison Wesley.

30.

Sabah daily newspaper, daily news reports between December 28, 2005, and January 17, 2006, Istanbul.

31.

Schein

(1997) Organizational Culture and Leadership. Jossey-Bass: San Francisco.

32.

Senge

(1994) The Fifth Discipline. New York: Currency & Doubleday.

33.

Uri

(1995) Coping With Uncertainty. Westport: Praeger.

34.

Weber

(1996) Sosyoloji Yazıları (Sociology Writings Translator: Parla

Taha

). Istanbul: Hürriyet Vakfı Yayınları.

35.

World Health Organization (2005) Ten things you need to know about pandemic influenza. Available at: http://apps.who.int/csr/disease/influenza/pandemic10things/en/.

36.

World Health Organization (2006) Making preparation count: lessons from the avian influenza outbreak in Turkey. Available at: www.euro.who.int/document/e89139.pdf (accessed September 1, 2008).

37.

Yang

(2010) The 9/21 earthquake in Taiwan: a local government disaster rescue system. Disasters 34(1): 112–136.

38.

Yin

(1993) Applications of Case Study Research. Thousand Oaks, CA: Sage Publications.

39.

Yin

(1994) Case Study Research. Thousand Oaks, CA: Sage Publications.