The importance of community resilience and public health in natural disaster management

(1)

The importance of community resilience and public health

in natural disaster management

The example of Palu (Indonesia)

By Laura Vroni van Tinteren

August, 2018

Supervisors NOHA: Dr. B. J.W. Pennink –

Rijks Universiteit Groningen, the Netherlands RDI: Dr Saut Sagala – Resilience Development Initiative, Bandung, Indonesia

This thesis is submitted for obtaining the Joint Master’s Degree in International Humanitarian Action. By submitting the thesis, the author certifies that the text is from her own hand, does not include the work of someone else unless clearly indicated, and

(2)

2

Acknowledgements

The writing of this thesis marks the end of the Erasmus Mundus master in International Humanitarian Action. During the two years, I moved from an interesting and meaningful first year in Europe, at the Rijks Universiteit Groningen (Netherlands) and Deusto Universidad in Bilbao (Spain), to South East Asia for a semester at Universitas Gadjah Mada, Yogjakarta (Indonesia) and a final research internship at the Resilience Development Initiative in Bandung (Indonesia). While travelling different countries and learning much about the ins and outs of humanitarian assistance, I have been fortunate to meet many interesting fellow students and inspiring professors from various backgrounds. In particular, I would like to express my gratitude for three professors that have added their valuable time and critical notes to this final work; Dr. BartJan Pennink (Groningen University), Dr. Saut Sagala (Institute of Technology Bandung & senior researcher at Resilience Development Initiative (RDI) and Professor Yodi Mahendradhata (Universitas Gadjah Mada).

(3)

3

Indonesia is extremely prone to a variety of natural disasters. Fortunately, both in Indonesia and globally, the importance of locally available soft capital in natural disaster management is increasingly recognized. A term in the respective domain that quickly gained popularity is community resilience. In theory, healthy and resilient communities are associated with lower risks of disasters, are more adaptable and recover faster. Yet, the variation in understanding of what healthy and resilient communities consists of, have led to difficulties in operationalization. Analysis of a recent natural disaster that affected Palu in terms of (health) impacts and (health) response, aimed at identifying characteristics of resilience at a community level. The study is using a qualitative case study methodology, based on a semi-systematic literature review and enriched with interviews from Palu emergency responders and other relevant experts.

Among others, this research highlights the need to prioritise health and social capita at a community level in Indonesian disaster management. Local communities and their potential in disaster management are often disregarded in emergency situations both in Indonesia and elsewhere. Yet, the wealth of local knowledge and close networks in Indonesian communities allow for flexible and rapid responses, reducing the adverse impacts of disasters. Health has been identified as a prerequisite for community resilience. Integrating a public health lens allows for conceptualising resilience based on a systems wide thinking approach that looks further than the narrow and sometimes vague descriptions provided in the literature thus far. Moreover, this thesis stresses that no single organisation (governmental or non-governmental) can solely make communities resilient. Rather, community strengths have to be supported by various levels and (health) vulnerabilities addressed.

(6)

6

Abbreviations

AAR After Action Review

ADRC Asian Disaster Reduction Centre

BNPB Badan Nasional Penanggulangan Bencana (in English: the National Agency for Disaster Management)

CBDM Community Based Disaster Management DRR Disaster Risk Reduction

MoH Ministry of Health

NGO Non-Governmental Organisation

PASIDOMO Palu, Sigi, Donggala, Moutong District (names of the affected areas) PMI Pulang Merah Indonesia (Indonesian Red Cross)

PTSD Post-traumatic Stress Disorder RDI Resilience Development Initiative

SCDRR Safer Communities through Disaster Risk Reduction UGM Universitas Gadjah Mada

UN United Nations

UN OCHA United Nations Office for the Coordination of Humanitarian Affairs WHO World Health Organisation

(7)

7

1 Overview and Rationale

1.1 Background

The magnitude of humanitarian need as a result of natural disasters is greater today than ever recorded before (WHO, 2015). Natural disasters, combined with population growth, unplanned urbanization, and a fast changing climate, currently pose the biggest threats to human safety and sustainable development (Baxter, 2019; Djalante, Garschagen, Thomalla & Shaw, 2017; IFRC & UNDP, 2013; OCHA & CDA, 2016). A report by UN OCHA reflecting on Extreme Weather Events (2018 Year in Review) concluded that approximately 61,7 million people were affected by natural disasters in 2018. Noteworthy, although the terms natural hazards and natural disasters are often used interchangeably, their meanings differ significantly. Whereas natural hazards refer to sudden extreme weather and climate events that may pose a considerable risk, natural disasters only occur when affected communities are not capable to prevent, respond or adapt to the natural hazard.

This can be illustrated by the following equation; Risk of Disaster = (Hazard X Exposure X Vulnerability) / (Capacity). Natural disasters typically lead to public health emergencies, due to a disruption of the functioning of a community and exceeding local capacity to respond (Keim, 2008; Khan et al., 2018). In general, natural disasters have broad human, material, economic or environmental impacts and may ultimately lead to an excess in morbidity and mortality (Cui, Han & Wang, 2018, "Terminology - UNISDR", 2017). Interestingly, within the aforementioned equation, most components can be influenced. This thesis focuses specifically on the role of community resilience and public health as influencing factors and aims to emphasize the importance in disaster (risk) management.

Natural disaster management has traditionally focused its attention on, and monetarily invested in, ‘hard capital’ (physical critical infrastructure such as roads, facilities, equipment). Although important, local knowledge, health and other capita within communities are equally essential in reducing disaster risk and rapid recovery. Recognition of this ‘soft capital’ (e.g peoples traditions and behaviour, connectedness and health) has only recently gained more appreciation (Plough et al., 2013).

(8)

8

1.2 Community resilience

Particularly the concept of community resilience, referring to the capability of communities to adapt and recover in a timely manner after emergency situations, has become more important in the academic discourse and national guidelines, public policies and programmes in natural disaster management (Abramson et al., 2014; Baxter, 2019; Kurniawan, Ascholani, Irawan, Nurdin & Wermasubun, 2019; Lovell, Bahadur, Tanner & Morsi, 2015; Patel, Rogers, Amlot & Rubin, 2017; WHO Regional Office, n.d.; Wulff, Donato & Lurie, 2015). An increasing body of literature has tried to define community resilience, but a significant gap remains between theory and operationalization (Imperiale & Vanclay, 2016; Parsons et al., 2016; Zamboni, 2017).

While community resilience, in theory, has the potential to shift the way of thinking in the field of disaster management and sustainable development, it is also at risk of solely being a popular buzzword (Keating et al., 2016). One way of trying to avoid the latter is by applying concepts and theoretical frameworks to real life situations. This thesis aims to investigate and highlight valuable lessons on community resilience and public health level by looking at the rather unique combination of natural hazards that affected areas of mainly Palu but also surrounding areas Donggala, Sigi and the Moutong district in Central Sulawesi (altogether called PASIDOMO) on the 28th_{of September. By analysing the}

impacts of the disaster on a community level, community strengths, well-being and the disaster response by various stakeholders, this thesis aims to show the importance of social capital in disaster management. The basis consists of interviews with various stakeholders employed in healthcare, professionals dealing with emergency situations and community resilience, supplemented with public health literature studies.

The need to further include and integrate community resilience and health in disaster management is on a global level emphasized by resolutions of the World Health Assembly and the World Health Organisation, national strategies, the International Strategy for Disaster Risk Reduction, the agenda for Humanitarian Action and is perhaps most prominently emphasized in the Sendai Framework for Disaster Risk Reduction (SFDRR) (Baxter, 2019; Chandra et al., 2011; Cui, Han & Wang, 2018; World Health Assembly, 2011; Wulff, Donato & Lurie, 2015; Plough et al., 2013). These international commitments emphasize that countries have the responsibility to ensure population health, safety and well-being and the importance of resilience and self-reliance in decreasing vulnerabilities and more effective disaster response (Baxter, 2019). However,

(9)

9

this broad interest has not yet sufficiently led to real empowerment of locally affected communities in their own preparedness or recovery in Indonesia. Indonesia is, for several reasons (discussed below), an interesting country to study. In Indonesia, the importance of public health in the field of community resilience is still a relative new concept. Both in Indonesia and internationally, a more profound understanding on the role of communities in disaster management and the dynamics of how health, people’s traditions, behaviour and best practices may support community resilience are warranted (Lovell, Bahadur, Tanner & Morsi, 2015).

1.3 The Indonesian Context; A rapidly developing Vulnerable Country

Worldwide, the Asia-Pacific region was the most disaster-prone region in 2018 (OCHA, 2019). Indonesia in particular, located on the so-called ring of fire, experienced a series (approximately 2000) of natural hazards that killed over 4000 people and left an estimated number of three million people displaced only in 2018 (Emmet, 2018). It is worldwide the country with the highest total death toll due to natural disasters (Djalante, Garschagen, Thomalla & Shaw, 2017; UNISDR & CRED, 2019). Due to its high frequency of experiencing natural disasters, Indonesia is internationally referred to as the ‘disaster laboratory’ or the ‘supermarket for disasters’ (Djalante, Garschagen, Thomalla & Shaw, 2017; Viverita, Kusumastuti, Husodo, Suardi & Danarsari, 2014). Indonesian people have a wealth of experiences and strategies on how to respond to disasters, yet every disaster is context specific and provides new valuable lessons to be used as preparation for the next disaster in terms of response and preparedness (Imperiale & Vanclay, 2016).

Besides, the government in Indonesia has also become increasingly interested in enhancing community resilience and supporting community-based initiatives (Djalante, Garschagen, Thomalla & Shaw, 2017; Pauwelussen, 2016; The World Bank, 2012). This interest was emphasized by the launching of several community resilience projects and guidelines. For instance, the BNPB launched an initiative on Community-based Disaster Management (CBDM) to enhance resilience at local community levels (Guarnacci, 2016). However, different reports and academic sources argue that based on current regulations, DRR and community resilience are still being dealt with on a sectoral basis via top-down approaches (Ali, Arsyad, Kamaluddin, Busthanul & Dirpan, 2018; Wimbardana & Sagala, 2018).

(10)

10

This is worrisome as many areas in Indonesia are presented with multi-hazards, which are only expected to increase in frequency and intensity. Indonesia is very susceptible to the effects of climate change (Keim, 2011). Strategies focusing on community resilience and public health are valuable in disaster management as they are inherently cross-disciplinary and located at the local level (Chandra et al., 2011; Wulff, Donato & Lurie, 2015). In cases such as Palu, when time to respond takes long, and health emergencies worsen fast, community resilience can provide the means to a fast recovery and decrease of adverse health effects.

1.4 Thesis Structure

This thesis is organized in six chapters. This first chapter stresses the importance and objectives of this research, its scope, and introduces the key terms and Indonesian context. Chapter two elaborates on the methodological approach used to collect and analyse the data. Following, chapter three includes the background literature review and the theoretical framework. The theoretical framework highlights different perspectives to community resilience, public health and DRR and discusses overlapping areas. A thorough understanding of the different terms facilitates practical application and scholarly experimentation between the different concepts. Results in chapter four, describe the case study and main findings from the literature review and interviews. Next, an analysis based on the literature review and insights from key informants is provided. This covers overarching associations between community resilience and public health. In the discussion, chapter five, main findings are presented and discussed in a wider context. The discussion also covers the methodological limitations. Since this research is of explorative nature, a number of points and questions that arose while conducting this research is presented in a food for thought section. Lastly, the conclusion, chapter six, highlights the importance of further integration of health and community resilience in disaster management in Indonesia and briefly states the main findings.

1.5 The researcher’s motivation and topic development

Before eventually deciding on the topic of community and health resilience, many other topics had been considered, started and disregarded. Two years of studying international humanitarian assistance let to the understanding that the field has many drawbacks, learns

(11)

11

and improves at a slow pace and contains a high risk of doing more harm than good. In essence, humanitarian assistance is internationally recognized to be based on the following core principles; humanity, neutrality, impartiality and independency.

However, in practice, humanitarian assistance is strongly depended on and geared by politics and donor money. Trying to be as politically neutral as possible provides access to areas and people that are difficult to access, but it also allows other organisations to misuse humanitarian assistance, which sometimes leads to benefitting the ‘wrong’ organisations or sides. Although humanitarian assistance should solely consist of short term emergency aid, in many instances, this has turned into a long term emergency aid dependency, also known as protracted crisis. Providing mere emergency services for a long time does not relieve the suffering of people, but sustains it (Iserson, 2014).

Moreover, international humanitarian assistance is known to mainstream its services to provide quick and ready-to-go programs and services in different contexts. However, local contexts usually differ so much that these ready-to-go programs and services often mismatch local customs and contexts. Besides, international humanitarian aid is likely to disrupt local markets and economies, making it harder for local people to recover. Most notably, it is often ignored by foreign aid helpers and programs, how much knowledge and ability locally affected people possess. Beneficiaries are often viewed as helpless and unable to recognize what is best for them. An interesting article by Elayah (2016) concludes that many of the ineffectiveness within the field of humanitarian assistance is related to ‘a vicious cycle of corruption, weak policies and fragile institutions in recipient countries, and objectives within the donor countries and organizations themselves, on the other’ (Elayah, 2016).

Fortunately, although slowly, the humanitarian landscape is changing (OCHA & CDA, 2016). More and more focus is geared to the exact needs of affected people and they are increasingly involved in the decision making of the kind of services that are provided and needed. Research and results distribution play an important role in this change. Although relatively new, enhancing community resilience provides a bottom up and sustainable approach to disaster management and little harm is known to result from it as of yet. After consulting with supervisors, other relevant local experts, and scoping the available literature, the additional aspect of analysing a case study and looking at the association between public health and community resilience, was decided upon.

(12)

12

1.6 Research Objectives & Research Questions

A body of literature is emerging on the importance of community resilience. However, to date, few studies have focused on the added value of Public Health for community resilience in relation to natural disaster management. Especially studying factors underlying healthy and resilient communities in an Indonesian context may be important, given the high incidence of hazards. This research is conducted at an Indonesian research institute, named RDI, in Bandung. Responding to the importance and need to further integrate resilience in various disaster management disciplines, they produce working papers based on results from field research and topic specialists. These are used for academic purposes and ultimately, advocacy. RDI had a particular interest in learning more about how public health fits in a resilience paradigm. Moreover, since the Tsunami in Palu had only recently happened and resulted in certain public health challenges, it provided an interesting case-study for evaluation and lessons to learn.

Responding to the above mentioned gap in the literature and the interest from RDI, the goals of this thesis are threefold. First, to analyse the impacts of the recent natural disaster in Palu at a community level. Second, to explore resilient characteristics of Indonesian communities. And third, to evaluate the role of communities in Indonesian disaster management. Ultimately, findings from the interviews were scrutinized with the literature and more general associations on community resilience and public health are provided which might hold true in various contexts.

The process of developing this thesis has meant to be cross-disciplinary, with various inputs from academia, government and non-governmental organisations as well as people from the community in Palu. Due to the explorative character of this paper, generating hypothesis, providing recommendations and proposing future areas of research has also been valued as important. Results will hopefully contribute to a greater effort of promoting and enhancing healthy and resilient communities in Indonesia.

The following research questions have been established in accordance with the objectives;

1. What is the impact of natural disasters, taking Palu as case example, on communities and their well-being?

a) What happened in Palu and what was the response by various actors? b) What are the social and health impacts of the natural disaster in Palu on

(13)

13

2. What are important characteristics of Indonesian community resilience and health?

a) What role does health play in disaster resilience?

3. Which role do local communities play in Indonesian disaster management?

4. How can health and community resilience be further integrated and enhanced in DRR activities?

5.

What are the associations between public health and community resilience?

1.7 Scope

The scope of this research was not predefined in very strict terms. For a better understanding, literature on community resilience from case studies around the world has been examined, while emphasising on the situation of Indonesia. Clearly, resilience has many context dependent aspects. It is based on local factors such as available resources, system capacity, and vulnerabilities. Community resilience here, is further considered from a public health point of view, using articles that look at community resilience and population health simultaneously, and speaking to interviewees associated with the health sector.

Especially an analysis of vulnerable and strong components and impacts during the disaster in Palu and and a review of performance after the event were studied. Baseline information on community characteristics or preparedness was limited. The thesis focuses on the potential of health and community resilience in terms disaster risk reduction on a community and system level, as opposed to community resilience from an individual or global level resilience. It is thought that the community level captures more of the interactions between the environment, social factors, vulnerabilities, relevant stakeholders, culture and their impact on livelihoods and wellbeing of people (Mochizuki, Keating, Liu, Hochrainer-Stigler & Mechler, 2017). Definitions for a ‘community’, ‘health’ and community resilience are mentioned in the literature background, although their boundaries remain ambiguous and demand further investigation. Lastly, this research focused solely on natural disasters and natural disaster management as opposed to technical and man-made disasters.

(14)

14

2 Methodology

This chapter describes the overall research design, e.g. the various methods used to collect data, and how the data is analysed. The research was carried out between March and August 2019.

2.1 Research design

Overall, this study is of qualitative nature and employs an explorative and descriptive manner to answer the research questions. A single case study (Palu, a natural disaster affected city) was used to examine community resilience and community health in context. Baxter & Jack (2008) refer to such approach as a qualitative case study methodology. This approach particularly fits studies that aim to understand the ‘how and why’ of phenomena, when there are no static boundaries between the phenomena and the context, when it is impossible to influence the people which are part of the observation, and lastly, when the aim is to explore contextual conditions which are thought to be relevant for the phenomena of focus (Baxter & Jack, 2008; Savenye & Robinson, 1996). When these conditions match the study focus and goals, benefits of this methodological approach include its utility as means of in-depth learning and the studying of complex phenomena within their context (Baxter & Jack, 2008). This methodological design is particularly used by health science researchers who are interested in testing a certain theory, hypothesis or evaluate ongoing programs or situations, provide recommendations, develop interventions or generate hypothesis.

The research started in March, interviews were conducted in April, May & June, after which the results were analysed in June & July. The thesis was finalised in August. Conducting the literature search was a continuous process between March and August and supported the overall research. Most of the data collection and writing happened in Bandung, at the Resilience Development Initiative, a global think tank. Founded in 2013, RDI’s research focus is on different forms of resilience, natural disaster management, sustainability, finance, security, renewable energy, urban planning and many other rather ‘progressive’ themes for Indonesia. Doing research at RDI also meant having access to field related contacts, networking, and various other data sources in possession of RDI. Besides, living in Indonesia, the same country as the case study, provided a good insight

(15)

15

in the context of the study area and characteristics of different communities in Indonesia. At RDI I was also able to attend interesting guest lectures on relevant themes, and meet with many of my interviewees. Further details about the different data collection methods are described in paragraph 2.2. Paragraph 2.3 focuses on the strategies used for data analysis. Methodological limitations and challenges are described in the Discussion (chapter 5).

2.2 Data Collection

Another advantage of the selected methodological research design, is the possibility of using various data sources, adding to the overall internal validity (via triangulation) and understanding of the ‘whole picture’. For this particular study, two main data collection methods have been exhausted; a semi systematic literature review and in depth interviews with key informants. The literature review provided a solid understanding of the key terms in the context of Indonesia and resulted in the theoretical framework (paragraph 3.2) which further guided the data collection and analysis. The interviews, on the other hand, provided meaningful insights on the context of Palu and other Indonesian case studies and examples. Key informants were also generous in sharing excel sheets and PowerPoints covering the situation in Palu (numbers on the amount of NGO’s present, the type of diseases that occurred and their frequency) as well as situational and organisational reports, which are not readily available online.

The literature review and interviews were further substantiated with the following secondary data sources: local and national disaster plans, archive records (of RDI), lectures, news reports, and discussions with relevant experts or community members. The different pieces of information in different forms are seen as vital in the researchers understanding in the overall phenomena of focus, in this case, the association between community health and resilience and their contribution to disaster management in Palu (Savenye & Robinson, 1996).

2.2.1 Literature review

To understand the dynamics between the concept of community resilience, health, and disaster management, this study has made use of a semi systematic literature review, as opposed to solely narrative or systematic reviews (Ferrari, 2015). Systematic analysis are

(16)

16

generally used to cover large amounts of material (or, meta-analysis) within a precisely defined scope, and extensively present what is currently known about a subject (Ferrari, 2015). The review is based on scientific articles and official policy documents, systematically retrieved from PubMed, Science direct, the online library of Groningen University (SMARTsearch) and complemented by Google Scholar. Additional information was gathered from governmental websites (such as the website of BNPB and Indonesian MoH), as well as from the WHO website, Reliefweb and the OCHA website. The selection of relevant scientific articles and other documents was conducted using different combinations of controlled keywords (MeSH-terms).

These MeSH terms were classified in two groups. General terms included; ‘community resilience’ ‘public health’ ‘disaster’ ‘resilience characteristics’ and ‘natural disaster’. Whereas the second set of keywords reflected the aim of finding resilient characteristics within Indonesian communities ‘social capital’. ‘Gotong Royong’ (mutual community assistance), ‘traditional knowledge’, ‘community engagement’, ‘community collaboration’, ‘social volunteers’, ‘local wisdom’, ‘risk communication’, ‘community initiatives’, ‘koperasi’ (cooperation), and ‘masyawarah’ (communal consensual decision making). The geographical labels Palu and Indonesia were included to obtain articles related to the pre-set geographical area of research.

Articles were retrieved making combinations (‘’and’’) of the key terms and after an initial screening of the abstracts, results and discussion paragraphs. Additionally, reference lists of relevant papers were screened for keywords and relevance to identify additional useful literature (“snowballing” technique). Results of the literature review are used throughout this paper and have in particular yielded the selected conceptual models in the theoretical framework. These show the intersections of different systems related to community resilience (O'Sullivan, Kuziemsky, Toal-Sullivan & Corneil, 2013), the building blocks for healthy and resilient communities (Chandra et al., 2011) and lastly, a practical toolkit on how humanitarian/ developmental agencies could enhance health and resilience at the community level (McCaul & Mitsidou, 2016).

Criteria for considering studies for inclusion and exclusion

Preference was given to public health papers. This either meant papers having the key words ‘public health’ ‘health’ AND ‘community resilience’ in their title, or papers on

(17)

17

community resilience from Public Health journals, evidently including a public health perspective. Furthermore, all selected articles addressed community resilience and disaster management in terms of natural disasters. Although perhaps a similar methodology could be used for man-made or technical disasters (Zamboni, 2017), these papers were excluded after inspection. Solely articles in either English, Dutch or Bahasa Indonesia were selected for review. Although Bahasa Indonesia is not a widely spoken language, reports in Bahasa Indonesia provided in-depth information about the local context, perspectives and different organisational structures. Nevertheless, help was needed with translation which is associated with a translation bias (further elaborated on in the discussion). Moreover, only studies published after the year 2000 are included in the literature analysis. This was to prevent including outdated information. For books there the timeframe was less strict, also books before 2000 have been used. Geographically, preference was given to studies that described the Indonesian context. Articles not related to humans were also excluded.

2.2.2 Interviews

As mentioned previously, in-depth interviews were useful to gather an understanding of context specific challenges and insights. Specifically, the interviews had two main objectives. Firstly, to understand from people from different relevant backgrounds how they viewed and incorporated community resilience in their line of work. Second, to understand what happened during the disaster in Palu at a community and on a health level and how that affected resilience. Given the limited time frame, interviewees were mostly contacted via the extensive Network of RDI and by means of the snowballing technique and some personal contacts.

Disaster risk reduction and enhancing community (health) resilience are cross cutting and complex issues which cannot be addressed by single entities alone. Thus, interviews were held with different people possessing a wealth of experience and knowledge on either disaster management, community resilience, emergency health response, urban planning or other relevant themes. Interviewees consisted of officials/ representatives of the Ministry of Health, a local Community Disaster Management NGO, the National Agency for Spatial Planning, UN agencies (UNICEF and the FAO), the medical Agency of Universitas Gadjah Mada, the Dutch News Reporter of Southeast Asia as well as local NGO practitioners. A total of 13 semi structured interviews/ discussion were held with

(18)

18

key informants. With permission, most skype and email discussions were recorded, later digitally transcribed and saved. Although this process was handled with care, two recordings were of too little quality to later be transcribed and one interview was thought to be of invaluable quality for this paper. Table 2 (in chapter 4) illustrates the ten key informants whose data has been used for this paper.

Of these ten, all but one interviewee were Indonesian citizens, with a good understanding of how disaster management is organized in Indonesia. Interviews were conducted in the cities of Jakarta, Bandung and Jogjakarta, but interviews were also held via skype with relevant stakeholders elsewhere. The researcher was also able to consult and interview professors from the faculty of Medicine and Health of Universitas Gadjah Madah. The Medical School of UGM also hosts the Disaster Health Management Division. The contact person from this Division has been active in the recovery of health centres after the Tsunami in September and provided many preliminary reports and data on the health response and locally active health NGO’s and health facilities.

Given the relatively short time after the natural disaster in Palu, interviews with community members in Palu were deemed unethical. This means that analysis of the impact of the Palu Tsunami among community members from their point of view are unfortunately limited. Moreover, given the timeframe and period of data collection colliding with Indonesia’s Ramadan month of fasting and the many public holidays prior, it was unfortunately not possible to meet with all originally planned experts and visit Palu.

Interviews were held in English and in the form of structured discussions, with several questions pre-determined but room and time to focus on specific information depending on the role/ function of the interviewee. Pre-determined questions were inspired by the ‘Analysis of the resilience of communities to disasters, Toolkit’ (McCaultwente & Mitsidou, 2016) and further adapted in consultation with the local supervisor in Bandung (see Anex 1 for interviews). The interview was in almost all cases started by asking what community resilience meant for the interviewee. Other general questions focused on challenges, gaps and proposed community solutions or initiatives. Via these questions information on community resilience, characteristics of the Palu community and the role of different stakeholders were tried to be extracted.

(19)

19 2.3 Data analysis

Two important characteristics are associated with qualitative research. Firstly, humans construct their own truth, based on norms and values in which they believe. As such, truth and reality are relative and subjective to researcher’s interpretation and sense making (Baxter & Jack, 2008). To combat the issue of subjectivity, triangulation is used to assure the validity of the research (Carter, Bryant-Lukosius, DiCenso, Blythe & Neville, 2014). Four main types of triangulation include; data triangulation (using various data sources, investigator triangulation (using different evaluators), theory triangulation (using more theories on the same phenomena) and methodological triangulation (using different styles of data analysis). This paper has employed data, methods and theory triangulation to verify to what extent perspectives and literature were all in line with each other (e.g the interviews with facts from situational reports, and Indonesian case studies of community resilience compared with case studies internationally). Importantly, in reference to this paper, triangulation was not mainly used to cross-validate but rather to capture different dimensions of community resilience and health in an Indonesian context. The second important characteristic is that the researcher becomes one of the research tools, by interacting directly with the subjects of the study (Savenye & Robinson, 1996). The closer the relationship with key informants, the more information can be extracted, providing a better understanding of what happened.

Since the overall study design was of explorative nature, data collection and analysis happened mostly simultaneously. As for data synthesis and presentation, a narrative analysis is chosen to cover the interviews. A narrative analysis is used to analyse stories and highlight important results (Ferrari, 2015). In reference to this paper, a narrative approach, using quotes, was considered useful to present the vast extent of experiences and examples of community resilience of Indonesian communities, shared by the key informants.

Besides general methods for data analysis, as described by Yin (2009), the following six data analysis techniques belong to a qualitative case study methodology; 1. pattern matching 2. linking data to propositions, 3. explanation building, 4. time-series analysis 5. logistic models and lastly, 6. cross case synthesis (Yin, 2009). While these techniques have not been used very consistently in this thesis, a mixed methods approach has been exhausted to attend all the evidence gathered and in line with methodological triangulation. Firstly, linking of data to propositions, in this case the interviews and

(20)

20

Indonesian reports to the theoretical framework and conceptual models. Secondly, the cross case synthesis, in which multiple case studies on community resilience in different contexts were studied, especially aiming to extract the lessons learned and challenges. Thirdly, explanation building, relevant to explanatory case studies, eventually aims to develop ideas for further studies or interventions or in case of this paper. Explanation building may also lead to generating hypotheses. This is also one of the aims of a narrative analysis and was one of the objectives of this thesis.

3 Background Literature review

3.1 Literature Review

Although multidisciplinary in nature, within the frame of this thesis, community resilience (the role of local communities) and their well-being are discussed in an Indonesian context in relation to disaster management, disaster risk reduction (DRR) and humanitarian assistance. Disaster management is referred to in its broadest meaning, encompassing the strategies Indonesia uses to manage disasters in terms of preparedness, response and recovery ("Disaster Management UNISDR Terminology", 2019). Further associations between these concepts will be discussed in the theoretical background (section 3.2) and discussion section of this thesis.

3.1.1 Disaster Risk Reduction

As defined by the UNISDR, DRR concerns applying policies, strategies and practices to prevent and reduce existing disaster risks (UNISDR, 2017). On an international level, the Sendai Framework for Disaster Risk Reduction (2015-2030) is currently the most prominent DRR guideline (Pascapurnama et al., 2018). Noteworthy, following up the ‘Hyogo Framework for Action (2005-2015)’ and ‘Building the resilience of Nations and communities to Disasters’, the SDFF underlines the importance of DRR being more people centred. Explicit measures to do so include, building resilience to disasters, while explicitly focusing on people and their health and livelihoods, active identification and reducing of risks, and establishing community centres for the promotion of public awareness (Zwijnenburg, 2016). It was endorsed by the UN General Assembly in 2015 and signed by 18 UN member states, among which, Indonesia.

(21)

21

3.1.2 Community resilience and DRR

The concept of resilience firstly emerged around 1970 and finds its roots in socio-ecological systems theory (Imperiale & Vanclay, 2016). It referred to the ability of different ecological systems to cope, adapt and even thrive when exposed to external stresses and shocks (e.g natural hazards). Over time, a wide variety of disciplines (e.g psychology, engineering and management) have adopted the term and adapted it to the characteristics of their respective sector or discipline (Lovell, Bahadur, Tanner & Morsi, 2015). However, due to its increased prominence in different fields and wide variety of (sector specific) definitions, the term has become confusing, and critics refer to it as a mere ‘umbrella term’ (Abramson et al., 2014; Chandra et al., 2011; Wulff, Donato & Lurie, 2015; Zautra, Murray & Hall, 2010).

What constitutes of a community in community resilience?

A study exploring the indicators of community resilience observed different definitions for community based on individual and geographic levels, suggesting no agreement of what a community exactly entails (Zamboni, 2017). Likewise, when Pauwelussen (2016), attempted to study community resilience in coastal Indonesia, she mentioned having to reconsider her original understanding of a village as a local community. Exploring where and how communities were constructed became one of the main research objectives. In her particular setting, Pauwelussen found that communities were more related to the type of fishing and trade practices than geographical boundaries (Pauwelussen, 2016). To date, having no universally accepted meaning or distinguishable boundaries of a ‘community’ that are agreed upon in the literature, poses one of the main challenges in the ‘resilience discipline’ (Sharifi, 2016). Obviously, lacking a clear scope may especially become an issue during the operationalization and practical application of enhancing community resilience.

In the context of this paper, a community might be best referred to as ‘a group of people living in the same geographical area who share common interests and dynamic socio-economic interactions and engage in collective action’ (Sharifi, 2016). A master dissertation titled Community empowerment as a way of improving public health in Indonesia, concluded that especially trying to define communities in urban areas and crowded cities (in Indonesia) is complex (van Loosen, 2019).

(22)

22

A definition for community resilience in a Disaster Risk Reduction Context

In relation to DRR, a commonly used definition for community resilience comes from Twigg (2009) (Anwar, 2016; IFRC, 2012). He describes a resilient community as ‘a community that has the capacity to anticipate, minimize and absorb potential stresses or destructive forces through adaptation or resistance’ (Twigg, 2009). Twigg further divides the determinants for a resilient community in the following five thematic areas: Governance, risk assessment, knowledge & education, risk management & vulnerability reduction and lastly, disaster preparedness & response. However, Twigg’s framework is critiqued for its empirical nature, while being limited in theoretical and practical explanation (Mochizuki, Keating, Liu, Hochrainer-Stigler & Mechler, 2017).

Examples of community resilience mainly surface during and after disasters (Abramson et al., 2014; Imperiale & Vanclay, 2016). It is in these times that ‘normal’ systems and procedures are disrupted and resilience can best be assessed and evaluated (Wulff, Donato & Lurie, 2015). This is why the Red Cross in 2012 started conducting multiple community resilience projects with at-risk communities with the aim of operationalizing the concept of community resilience (IFRC, 2012; IFRC, 2018). During these projects, the following six characteristics of a resilient community surfaced; a resilient community is knowledgeable, organised, healthy, capable of assessing context specific risks, assertive in solving problems, and finally, a resilient community is socially connected and receives the necessary support from a wider system (Keating et al., 2016; Madrigano, Chandra, Costigan & Acosta, 2017).

Social support systems are found in neighbourhoods, family networks, self-help groups and mutual interest groups. The wider system support refers to higher level support from governmental and non-governmental organisations (NGO’s) for instance. Since community resilience is multidisciplinary by nature, a variety of stakeholders are involved. In an Indonesian context these include, but are not limited to, engineers, faith based organisations, spatial planners, health care providers, educational committees, community representatives, social workers and a so called extension worker (who

In the context of this thesis, keeping in mind the fluidity of communities in Indonesia, the following definition of community resilience, by Abramson et al (2014), will be referred to ‘The enduring capacity of geographically, politically or affinity bound communities to define and account for their vulnerabilities to disaster and develop capabilities to prevent, withstand, or mitigate for a traumatic event’.

(23)

23

promotes the interests of communities at a governmental level). In an emergency context of scale, humanitarian aid workers, developmental organisations, government workers and cluster related other stakeholders also get involved.

Despite the variation in definitions for both resilience and a community, a few important assumptions lay the basis for understanding community resilience in terms of DRR. First, it is acknowledged that communities are the first and main ones to be affected by disasters while simultaneously possess most (local) knowledge on how to survive and recover from disasters (Lovell, Bahadur, Tanner & Morsi, 2015). Secondly, zero risk does not exist, not all threats can be avoided. This implies that mechanisms should be in place to ensure risks are kept at a minimum and communities are prepared as much as possible for possible shocks or hazards (Kruk, Myers, Varpilah & Dahn, 2015). Third, community resilience is no passive quality or inherited trait. Some ecological systems, communities or institutions are more often exposed to external shocks than others, or have less resources to their availability. However, resilience can in any situation be greatly enhanced by certain activities and having access to critical resources (Chandra et al., 2011). Noteworthy, enhancing resilience activities are context specific. It is important that communities are viewed and enabled to be actors of their own safety and well-being, both prior, during and after emergencies (Sharifi, 2016).

Overall, resilient communities experience a smaller disaster impact and recover faster, meaning a lower dependency on external aid and fewer financial and human losses. Moreover, resilient communities cooperate, protect and motivate each other while trying to improve structures and resources after disasters, to avoid future disastrous situations (Sharifi, 2016). This adds to sustainability and resonates well with the concept of Building Back Better, an important goal of humanitarian aid. Acknowledging that community resilience is vulnerable to changes in leadership, funding, culture and the intensity of natural hazards, this thesis regards community resilience as a dynamic process, as opposed to an outcome (Lovell, Bahadur, Tanner & Morsi, 2015; Zautra, Murray & Hall, 2010).

Globally, the interest in enhancing community resilience in policies and guidelines has increased rapidly (Davies et al., 2017). Governments worldwide translated resilience enhancing strategies to national plans and strategies. In Australia and the US for example, community resilience has become a vital aspect of national guidelines on disaster

(24)

24

management planning as well as in homeland security (Cui, Han & Wang, 2018). Despite community resilience being a frequently used term, the importance of public health as requirement of community resilience and important human and social capital aspect, has only fairly recent come to surface. In many circumstances, the importance of health and being healthy is overlooked, ignored or perhaps taken for granted (Morcelle & O'Connor, 2018). This thesis tries to emphasize the importance of integrating health in community resilience and thereby exploring knowledge or factors that could be practically used in Indonesian disaster risk reduction approaches.

3.1.3 A Public Health Lens in DRR

Equally multidimensional, the concept of (public) health was defined by the WHO in 1947 as ‘a state of complete physical, mental, and social well-being and not merely the absence of disease or infirmity’ (WHO SEARO, 2018). Although the definition received the necessary criticism for being unattainable, the definition is widely used. Interestingly, the definition encompasses the notion that health is not solely the responsibility of the individual but is influenced by external factors and a shared responsibility of different sectors. This could also be said for community resilience. Health underpins all aspects of resilience. All activities to prepare, adapt and recover involve health. Likewise, the outcomes of community resilience in times of emergency, is wellbeing and the lack of a health crisis.

Similar to community resilience, public health focuses on (health) vulnerabilities and strengths that are present at a community or public level (instead of medicine which focuses on an individual’s symptoms). Being uniquely situated at the community level, Keim (2008) argues that public health has the possibility to reduce community wide vulnerability towards natural disasters. Namely, public health as a discipline analysis the wider determinants that influence health from different levels in society. Interestingly, disasters are viewed both as determinants of public health as well as hazards that could cause a health crisis ("When disaster strikes - Indonesia at Melbourne", 2019).

Applying a public health perspective to any issue, means considering the wider system that influences individual and community health. A model widely cited and used to illustrate the wider determinants of health, comes from Dahlgren and Whitehead (1991) (Figure 1).

(25)

25

Figure 1 The Social Model of Health (Adapted from Dahlgren & Whitehead, 1991) As depicted in Figure 1, the public/community’s health is situated in the middle and is composed of non-modifiable factors such as age, sex and genetics. The different layers represent different aspects of the environment influencing individual and community health. The most outer layer illustrates general factors that shape the wider living conditions (policies, laws, justice system). Higher level decisions determine (access to) resources, services and infrastructure available to the individual/ community that are illustrated one layer closer. These factors again determine the type of communities people are part of. Education, parenting, beliefs and practices in our community largely shape the individual lifestyle choices and thus, the health status of the individual. In ideal situations, policies are put in place in the most outer layer that promote equity, empowerment and engagement. Or in other words, a set of policies, laws and human rights that enable supportive living and working environments which are accessible to all and enable individuals to make informed decisions that promote their own, community and environmental health (Murray, Aitsi-Selmi & Blanchard, 2015).

System thinking

Likewise, looking at community resilience or DRR from a public health point of view, it is understood that these concepts are no isolated islands, but rather a dynamic part of society which is largely depended on the wider system as well as culture and individual lifestyle factors and health (McCaul & Mitsidou, 2016). System wide thinking includes thinking about interconnections and relationships rather than individual dots or puzzle

(26)

26

parts. Exhausting a system wide thinking approach, is useful to understand the complexity paired with disasters and vulnerability. ‘’For health systems to be sustainable, thinking in terms of system-level resilience is becoming an urgency’’ (WHO Regional Office, n.d.). It can help anticipate for instance how disasters trigger livelihoods, what risk factors might be exacerbated by disasters and the identification of vulnerable groups. Dahlgren and Whitehead (1991) understood that good social relations and strong supportive networks are known to improve health and overall well-being both in normal and emergency situations. Being actively engaged in your community, socially connected to neighbours and able to take part in local decision making all contribute to good health and connected communities both in normal times as well as in emergency situations (WHO Regional Office, n.d.). In the health resilience discourse, this is referred to as the resilience dividend. Systems and communities must work well in normal contexts to be able to withstand in emergency situations (Kruk, Myers, Varpilah & Dahn, 2015).

If local capacity is not sufficient before and during exposure to natural hazards, livelihoods, infrastructure, health and community systems get disrupted. In a recent BMJ opinion paper, Moitinho de Almeida and Guha-Sapir (2019),emphasize that the human consequences following disasters are heavily depended on the vulnerability of the affected community as a whole, especially in developing countries. As their research concludes, human consequences to disasters are more determined by a lack of local community capacity then by the physical damage done by the natural disaster. This is also stressed by the WHO, which urges governments to invest in people’s health and community resilience in disaster management (WHO, 2015).

Resilience, with respect to health and disaster management, is best regarded as the process of preventive, health-promotive and curative action taken by the health sector (WHO Regional Office, n.d.). Health resiliency is referred to as the capacity to maintain in good health in the face of significant adversity (Sanders, Lim & Sohn, 2008). Often however, resilience in health is related to the resilience of health systems in the wake of emergencies, instead of people per se. The goal of effective public health systems and programs, is to provide conditions in which people are able to make healthy and informed decisions (Murray, Aitsi-Selmi & Blanchard, 2015).

(27)

27

3.2 Theoretical Framework

Similar to the wide variety of sector specific definitions on community resilience, there are numerous resilience frameworks and conceptual models described in the literature. Some illustrate the building blocks for healthy and resilient communities (Chandra et al., 2011; Twigg, 2012), while others focus on the position of community resilience within the disaster management cycle (IFRC, n.d.), on the social foundations of community resilience (Pfefferbaum, Van Horn & Pfefferbaum, 2015) or in relation to climate change adaptation (Watts, Campbell- Lendrum, Maiero, Montoya & Lao, n.d.; Keim, 2008). Besides, alongside the increased interest in enhancing community resilience in literature and policies, the need for measurable indicators, measurement tools and specific characteristics of resilient communities arose. Frameworks and models of Parsons et al., (2016) and Sharifi (2016) focus on measurable indicators, assessment and outcomes of community resilience.

Despite the large amount of frameworks and theories, community resilience remains weakly framed for several reasons (Baxter, 2019; Fabinyi, Evans & Foale 2014; Wulff, Donato & Lurie, 2015). Firstly, due to the ambiguous boundaries and definitions of communities and the resilience concept (Baxter, 2019). Secondly, although many strengths lie with the fact that resilience is specific, complex and dynamic, frameworks are often limited by being location and disaster specific (Abramson et al., 2014). Lastly, it is extremely challenging to capture diversity and specific needs of communities within a single theoretical framework or model (Baxter, 2019). Even though various models include essential aspects of different types of resilience, it becomes clear that no single discipline or model alone reflects all the aspects needed for communities to withstand, adapt and recover in a healthy and sustainable way (Wulff, Donato & Lurie, 2015).

While a common understanding and definitions of community and health resilience are of great importance for researchers, health policy leaders, government officials, emergency planners and local leaders, it should not delay efforts to practically enhance community resilience (Patel et al., 2017). There are certain individual characteristics that are to be included in any definition of community resilience. These include: local knowledge, health, community networks, relationships, communication, Governance & leadership, resources, economic investment, preparedness and mental outlook (Patel, Rogers, Amlot & Rubin, 2017). While other authors have mentioned similar

(28)

28

characteristics (Abramson et al, 2014; Sharifi, 2016; Morton & Lurie, 2013; Plough et al., 2013) they have often grouped them in the following four or five categories; 1. Human capital, 2. Economic Capital, 3. Political Capital, 4. Social Capital, 5. Physical Capital (e.g Infrastructure).

Figure 2 Examples of capita contributing to community resilience (Adapted from Abramson et al., 2014) provides examples of these capita at an individual and community level. Overall community resilience depends on the resilience and available capita of each of these components. Since the article focused solely on how social resources promote adaptation and recovery after disasters, the physical category is not included. This would, at a community level, include aspects of the build environment, such as land using planning and engineering standards, and at an individual level include maintaining a healthy weight, exercising and healthy nutrition. Whereas the political and physical capita are more related to the performance of the availability of utilities and resources (e.g roads, infrastructure and number of hospitals), the human, social and economic capita are more influencing the overall performance and resilience (Viverita, Kusumastuti, Husodo, Suardi & Danarsari, 2014).

To date, several authors have focused specifically on the association of (public) health and community resilience, and refer to the overall discipline as community health resilience (NPRSB, 2014; O'Sullivan, Kuziemsky, Toal-Sullivan & Corneil, 2013; Watts,

Figure 2 Examples of capita contributing to community resilience (Adapted from

(29)

29

Campbell- Lendrum, Maiero, Montoya & Lao, n.d.; Wulff, Donato & Lurie, 2015). Although recognized as an upcoming and distinct concept, at this stage, community health resilience involves looking at the aspects of resilience and applying them to individual and community health (such as Figure 2 above), as well as behavioural health and society well-being (NPRSB, 2014).

Community resilience and public health have many overlapping areas. Both take place on the same societal level, are highly influenced by vulnerabilities (in systems at a community level and at an individual level) and are linked to sustainability (sustainable development goals), societal welfare and risk reduction (Sendai Framework) (Sagala, Yamin, Pratama & Rianawati, 2014). Moreover, both have (undiscovered) substantial economic benefits and are strongly linked to the development of a country and community well-being. A healthier community is more productive, knowledgeable and economically more efficient. Consequently, informed and knowledgeable communities are more aware of the health risks related to disasters, will likely focus more on prevention, are able to recognize which groups are vulnerable (in lesser health) and have more ability to help those groups during disasters. Vice versa, communities that are socially and economically more resilient are more likely to have health insurance, stable housing, and other facilities and abilities that make them able to sustain healthy behaviours (Wulff, Donato & Lurie, 2015; Plough et al., 2013).

3.2.1 Existing models looking at community resilience and public health.

Since no single framework well framed the goals and objectives of this paper, several models will be discussed. Models have been selected that show important intersections between health, community and disaster management. Derived from journals or papers focusing on public health and community resilience, these models do not contradict each other, but rather present different ways of looking at community resilience and well-being. At its core, community resilience and public health are underpinned by principles as equity and social support and aim to enhance well-being and core capacities of all people in the local community (especially those who are vulnerable), to be able to resist external shocks and recover in a timely fashion (Plough et al., 2013). Ideally, resilience is able to transform disastrous situations into new opportunities, or in humanitarian terms;

(30)

30

Figure 3 Framework for critical social infrastructure to promote population health and

resilience (adapted from O'Sullivan, Kuziemsky, Toal-Sullivan & Corneil, 2013)

‘building back better’ (Baxter, 2019). Besides good health, this also requires relevant sectors working together that are thus far still too segregated.

1. The complexity of different systems

The first model (Figure 3), by O’Sullivan et al (2013) presents the interconnectedness of different systems (soft and hard) at stake in enhancing community resilience. In particular, they propose local culture (referring to the community, situated at the top in the figure) being the connecting thread through the systems. The local culture is determined by four core themes (with a darker border in the figure); a dynamic context (e.g political situation, emerging hazards), connectedness, collaboration and lastly, situational awareness (resulting situation from the previous themes) which allows for adaptive responses (e.g flexible planning, open mindset, creative solutions) to uncertain events.

Interesting about this model, is the recognition and display of the complexity of different levels in society via which information and resources move and interact. Namely, those at the micro level that create a certain situation (via structures, protocols, information provided and people) versus the macro outputs at the community level (core themes). Important to note also is that there is a push and pull between the different levels. For instance, certain protocols and resources are made available in line with political priorities and funding allowance which influence the situation at the community level. Vice versa,

(31)

31

grass roots initiatives, protests and disasters at local community levels can pressure changes at higher levels. Ideally, resilient communities initiate what type of resources are needed to better prepare, adapt and recover from disasters, what type of information should be shared, via which channels and in what form (O'Sullivan, Kuziemsky, Toal-Sullivan & Corneil, 2013). The overall research method used to structure their research and form the framework presented in Figure 3 was the complexity theory (See discussion).

2. Building blocks of a healthy and resilient community

Unravelling the complexity and identifying building blocks to establish healthy and resilient communities, Chandra et al., (2011) has published extensively. From drawing on a literature review, the following core components (Figure 4 second column) of a healthy and resilient community were found; in physical and psychological good health, social and economic well of, communal self-organisation and reliance, effective risk communication, social connectedness between community members as well as organisational (government and NGO’s) (Chandra et al., 2011). Consequently, based on these outcomes, the following six building blocks (Figure 4 first column) were identified as required. Wellness and access, education, engagement, self-sufficiency, and partnership. These are again more or less similar to the human, economic, political, social and physical capita mentioned previously.

(32)

32

Figure 4 Building Blocks of Community Resilience (adapted from Chandra et al., 2011)

Figure 5 Practical toolkit to enhance community resilience (Adapted from McCaul &

Mitsidou, 2016)

This model is useful for policymakers and relevant organisations in identifying priorities for healthy and resilient communities both in normal and in disastrous times, consisting of factors that impact the community but on which the community itself has no or very limited influence (O’Sullivan, et al 2013). Models that show the situation from the side of the communities, which building blocks and strategies they can employ to become more resilient, are as of yet not widely available (Baxter, 2019).

3. From theory and building blocks to practice

From the previous two models it becomes clear that community resilience is highly depended on factors and policies in place in the wider system. It intersects with concepts and disciplines such as health, governance, leadership, vulnerability, risk, adaptation, recovery, culture, education but also development and sustainability (Chandra et al., 2011; Parsons et al., 2016; O'Sullivan., et al 2013). Keeping in mind the two previous models, the following model (Figure 5) depicts a practical toolkit on how to enhance community resilience and wellbeing from a practical perspective (McCaul & Mitsidou, 2016). Interestingly, this model is based on field experiences by GOAL, an international humanitarian organisation.

The importance of community resilience and public health in natural disaster management