A Review into the Acute Emergency Response of Operational Partnerships During the Rohingya Refugee Crisis in Cox’s Bazar


 Background: The unprecedented influx of Rohingya refugees into Cox’s Bazar, Bangladesh, in 2017 led to a humanitarian emergency requiring large numbers of humanitarian workers to be deployed to the region. The World Health Organization (WHO) contributed to this effort through several well-established deployment mechanisms: the Global Outbreak Alert and Response Network (GOARN) and the Standby Partnerships (SBP). The aim of the study was to capture the views and experiences of those humanitarian workers deployed by WHO through operational partnerships between December 2017 and February 2019.Methods: A mixed methods design was used. A desktop review was conducted to describe the demographics of the humanitarian workers deployed to Cox’s Bazar and the work that was undertaken. Interviews were conducted with a subset of the respondents to elicit their views relating to their roles and contributions to the humanitarian response, challenges during their deployment and how the process could be improved. Thematic analysis was used to identify key themes.Results: We identified sixty-five deployments during the study period. Respondents’ previous experience ranged between 3 and 28 years (mean 9.7 years). The duration of deployment ranged from 8 to 278 days (mean 67 days) and there was a higher representation of workers from Western Pacific and European regions. Forty-one interviews were conducted with people who experienced all aspects of the deployment process. Key themes elicited from interviews related to staffing, the deployment process, the office environment and capacity building. Various issues raised have since been addressed, including the establishment of a sub-office structure, introduction of online training prior to deployment, and a staff wellbeing committee. Conclusions: This study identified successes and areas for improvement for deployments during emergencies. The themes and subthemes elicited can be used to inform policy and practice changes, as well as the development of performance indicators. Common findings between this study and previous literature indicate the pivotal role of staff deployments through partnership agreements during health emergency response operations and a need for continuous improvements of processes to ensure maximum effectiveness.


Background
In 2017, over 700,000 Rohingya refugees crossed from Myanmar into Bangladesh joining refugees from previous waves of displacement (1)(2)(3) to form the largest refugee camp in the world, currently hosting over 850,000 refugees, and straining the resources of one of Bangladesh's poorest districts (4).
The Rohingya refugees faced considerable public health challenges, including severe and acute (mental) trauma, housing in overcrowded camps, poor water sanitation and hygiene (WASH) conditions, poor nutritional status (5), low vaccination coverage (6) and others.
The number of refugees rapidly overwhelmed the limited available capacity of local communities and health systems, and additional health service capacity was required. The Government of Bangladesh requested international assistance, and the World Health Organization (WHO) took on the leadership of the Health Sector Coordination in support of the Government on 1st October 2017. In an emergency of this scale, it was necessary for WHO to collaborate with operational partners in line with the Regional Framework on operational partnerships for emergency response to achieve a timely, coordinated, effective and e cient emergency response (7). Standby Partnerships (SBPs) and the Global Outbreak Alert and Response Network (GOARN) contributed signi cantly to the sta ng of WHO operations in Cox's Bazar.
Standby Partnerships are bilateral agreements between organizations and UN agencies, with partners providing support to UN agencies responding to humanitarian emergencies through the secondment of gratis personnel. Each Standby Partner maintains its own roster of humanitarian experts (8). These personnel have been pre-identi ed and can be rapidly mobilized.
GOARN is a global partnership which can deploy personnel with technical and operational skillsets to support public health emergency response (9). The GOARN secretariat and operational support team are based in WHO Headquarters in Geneva, Switzerland, and at several WHO regional o ces. Partners include networks of academic institutions, laboratories and regional technical networks, medical and surveillance initiatives, United Nations agencies, international humanitarian non-governmental organizations and national public health institutions.

Previous research into deployments
Humanitarian emergency response globally involves the deployment of large numbers of staff, often at signi cant nancial cost. Research is needed to guide organisations to undertake these deployments e ciently, while ensuring the thorough vetting and wellbeing of deployed staff.
Limited literature has been published on staff providing humanitarian assistance (10). Where it is available, the majority has been concentrated on the mental wellbeing of humanitarian workers, or focuses speci cally on recruitment or the nature of the work. In addition to the operational constraints of research in humanitarian settings, workers deployed are often part of a transient workforce for which follow up can be challenging (11). To date, there has been little written on the process of their deployment, or the technical and operational aspects of the deployment process in the emergency response.
This article aims to describe the characteristics of deployees through operational partnerships mechanisms to the emergency response in Cox's Bazar, and to describe their deployment experiences. Results of this study can inform future changes in policy and practice related to deployments in humanitarian emergencies.

Methods
A mixed methods approach was used: rstly, a desk review of deployments utilizing quantitative data was performed, followed by a qualitative study of interviews carried out with personnel involved in the response by operational partnerships in Cox's Bazar.

Study design
The desk review was a cross-sectional study from which descriptive statistics were derived to characterize the demographics and input of people deployed by GOARN and SBPs in Cox's Bazar.

Study population
The study population comprised of people deployed to Cox's Bazar through either GOARN or a Standby Partner as part of the emergency response to the Rohingya refugee crisis between December 2017 and February 2019.

Data collection
Data were extracted from a database held centrally at WHO Headquarters in Geneva, which included demographic details of deployees, dates of deployment, deployee role (job title) and incident management system function (wider technical area of working), releasing entity (operational partner) and grade. Previous experience and nationality were extracted from CVs submitted to WHO. Descriptions of contributions were extracted from technical reports and performance evaluation reports submitted by the workers at the end of their deployment.

Study population
Interviewees were selected persons with signi cant operational experience related to the deployment of operational partnership personnel. This included WHO staff from all levels of the organization (Headquarters, Regional O ce, Country O ce and Cox's Bazar Sub-O ce), staff from operational partners who contributed to the deployment process, and deployees themselves.

Study design
A semi structured interview with a set of core questions was used, with exibility to probe interviewees' answers further if needed. Interview questions used are included in Supplementary Materials.

Data collection
The interviews were conducted by an external consultant, unknown to the majority of interviewees. The setting of the interview varied: ideally interviews were conducted face to face, but where this was not practical, Skype or telephone interviews were used. All interviewees were assured that their responses would be anonymous and they could withdraw from the process at any point.

Sampling technique
Initially a purposive and strati ed sampling frame was used, identifying twenty people representing each technical area or role of interest. This includes at least one representative from each area listed in Table 1 and their a liation or location is listed in Table 2. As interviews progressed, a snowball sampling technique developed, with interviewees suggesting others to interview. Interviews were conducted until suggested interviewees had responded and saturation point was reached.

Analysis
Thematic analysis of the interviews was undertaken by the interviewer. The process of analysis followed the seven steps outlined by Braun and Clarke: transcription, reading and familiarization, coding, searching for themes, reviewing themes, de ning and naming themes, and nalizing the analysis (12). Discussions were held by members of the authorship team (CB, AP, AB) to clarify and triangulate themes and gain a fuller understanding of the interviewees' discourse.

Ethical approval
As this study is a service evaluation and no patients were involved, there was no requirement for ethical approval. Involvement in interviews was optional with no remuneration, and participants were informed of the aims of the study.

Desk review
Deployments of personnel There was a total of 65 deployments of personnel from operational partners to the WHO Cox's Bazar subo ce between December 2017 and February 2019: 43 deployments through GOARN and 22 through SBPs. Four people were deployed twice, two from GOARN and two from SBPs, giving a total of 61 individuals who were deployed during that period.
The majority of deployees ful lled roles within the health expertise and operations function of the Incident Management System (IMS), as outlined in the WHO Emergency Response Framework (13), with the largest number of deployees ful lling epidemiologist or surveillance o cer roles. Speci c job roles are included in Table 1.

Level of experience and education
Deployees' relevant occupational experiences were not always directly linked to humanitarian or the public health sector, with some deployees having relevant experience from the private sector. The number of years of experience ranged from three to 28 (mean 9.7). This is shown in Figure 1. All deployees except one provided educational achievements on their CV, with all being in possession of at least a Bachelor's degree. Eight were medically quali ed and ten had PhDs.

Duration of deployment
The length of deployment ranged from eight days to 278 days, shown in Figure 2. The mean length of deployment was 67 days, with GOARN deployments tending to be shorter (mean 40 days, range 8-91) and deployments through SBPs lasting longer (mean 119 days, range 23-278). The longest deployments were between March and August 2018. This is also re ected when looking at the number of deployees in the Cox's Bazar o ce at any given time ( Figure 3). This peaked at 18 deployees in May and July 2018, then reduced gradually over time.

Interviews of personnel
Forty-one interviews were conducted with people involved in the deployments of personnel to Cox's Bazar. Their a liation is shown in Table 2.

Key themes
Four key themes were elicited by thematic analysis of the interviews: "sta ng", "deployment process", "o ce" and "capacity building". Within these four key themes, there are between two and four master themes, and several subthemes. These are shown in Table 3, with example quotes where interviewees suggested areas for development for each subtheme. Table 3 Themes and subthemes identi ed from interviews with example quotes "It is better to give speci c technical functions to deployees, rather than lead positions or management responsibility, as these roles should be given to people who are present for a longer period of time"

Continuity and transition
Duration of deployment "There should be a minimum deployment length of three months" "The same time is spent on recruitment regardless of how long they stay for, but the deliverables are very different" "Those who contributed the most were those who stayed for more than 3 months" "GOARN deployees are usually only for six weeks, it would be good if this could be extended" "An alternative to the PER could be a structured conversation" "There is a need for an internal record that is distinct from the PER" Roster "A roster should be formed of individuals who have been to CXB and performed well, who we could ask to return" Performance review "There should be a performance review process and an early evaluation" "More feedback to deployees and deploying organizations is needed" "An early assessment should be undertaken after two weeks"

Deployment process Preparation and arrival
Preparation "Before initiating the deployment, you should receive letters, documents, and in good time… you need some internal documents and information" "Deployees are usually informed about procedures for payment, leave etc by their deploying organization before leaving, so usually this is straightforward" "Personnel deployed through GOARN have been briefed by GOARN, and there is a GOARN focal point" SBP/GOARN deployment "Deployments from GOARN/SBP were a lifesaver when we needed one" "Deployments need to be more timely and reliable" "GOARN personnel deployed were highly trained and deployed quickly" Organizational structure "Administrative questions at times must pass through 3-4 layers: eld o ce administration, country-o ce, regional o ce and at times headquarters/global service center… which introduces a delay" Training "Training needs to be provided on the systems required" "Different SBPs may offer different training, but all will receive a degree of mandatory training on operations, nance, security" "In every position there are some particular training needs" Orientation/brie ng "It would be good to have a brie ng for deployed personnel in a more structured way" "The response from Western institutions is disproportionate" Regional focus "It can be useful to use regional [operational partners] due to culture and regional solidarity" "We should have a regionally focussed GOARN mechanism for this WHO region. This way the experience can be used in the region and we build local capacity" "If people are from the region or have experience in the region, they nd it easier and blend in well" Key theme 1: sta ng Sta ng was an issue highlighted by all interviewees. All those who were interviewed felt that staff deployed through partnership agreements played a pivotal role in the emergency response.
The need for a clear team structure and reporting lines was emphasised, including communication of management decisions. Some interviewees believed the vertical structure of the Incident Management Structure (IMS) affected communication, at times preventing information from being shared.
Some interviewees recommended that team lead positions should be assigned to persons with considerable WHO experience and who would remain in the sub-o ce for longer, allowing for institutional memory and more effective functioning of the IMS. This also linked with continuity, with shorter deployments viewed as having some valuable contributions but at times being disruptive. However, it was recognised that the intensive workload, especially in early stages of the response, may not be sustainable for longer periods. Many interviewees called for a longer-term sta ng plan as soon as it was clear that the emergency would be prolonged.
Clear terms of reference were raised by the majority of interviewees. Some deployees suggested for these to include a degree of exibility and to be nalized on arrival with their supervisor to ensure that they are clear on the role requirements. Handover and debrie ng were also seen as important elements, and that these processes should be formalised, ideally with both the inbound and outbound deployments overlapping on site.
Many persons interviewed expressed that the standardized performance evaluation report (PER) was not seen as sensitive enough to act as an adequate evaluation tool and had limited use in distinguishing successful deployments. Linked to this, an internal roster was seen by some as a way to positively distinguish deployees who had worked well in Cox's Bazar. Many suggested an early performance review, within the rst two weeks of a deployment, in order to identify any potential problems and correct course as soon as possible.

Key theme 2: deployment process
There were requests for more information prior to deployment, both role-speci c and general information about WHO systems and the WHO Emergencies programme. Staff from the Cox's Bazar sub-o ce commented that the timings of deployments would have the greatest added value if they coincide with greatest need and allow for handover from outgoing personnel. It was also felt that the process could be streamlined to reduce delays such as travel approvals and visas. Interviewees recognised that ideally everyone would be trained and familiar with WHO systems and procedures, but a minimum standard of deployees receiving training in relevant processes and IT systems would be bene cial. Alternatively, recruitment of more administrative support familiar with WHO systems could be considered. The importance of a thorough brie ng was noted, speci c to Cox's Bazar and including information on context, local culture and expectations.
More transparency was requested around recruitment of deployees, including selection of candidates. It was requested by those in the sub-o ce that they have a more active role in the recruitment process, and for a call prior to deployment between deployee and the receiving team to improve the preparation of both parties.

Key theme 3: o ce
The temporary nature of the Cox's Bazar sub-o ce premises, housed in hotel apartment blocks with several smaller rooms spread over different oors was raised. The layout was seen as contributing to fragmentation and detachment. It was raised that different o ce space might contribute to stronger coherence within and between teams.
Security restrictions and the cultural contextual challenges were raised, particularly by female deployees.
Although considered important for staff wellbeing and productivity by all interviewed, applicable policies on leave and rest and recuperation were not clear for different contractual modalities and deployment types, resulting in perceived barriers in accessing this entitlement.
Some deployees noted that opportunities for valuable research and documentation which could improve public health practice for similar settings existed in Cox's Bazar, together with some enabling factors. At the same time, it was noted that some of these were missed due to operational challenges of conducting research in an emergency, as well as lengthy and unclear approval processes, questions around ownership of data and authorships. It was felt that responsibility for coordination of research efforts should be assigned within the IMS structure to a staff member based in the o ce for medium term.
Many deployees reported challenges with IT equipment and access, with delays in allocations of o cial laptops and email accounts leading to temporary use of personal devices, emails and cloud accounts.
Concerns were raised around data storage, protection, and security. Establishing use of generic email addresses and function speci c accounts was recommended to promote continuity, particularly for highturnover roles.
Operational supervision and support for issues both inside and outside of work within the o ce were key issues, especially for less experienced deployees, although this was seen to improve with the introduction of a staff wellbeing committee later in the emergency response. The request for support also included a stronger relationship between the operational partners and WHO at eld level to permit more tailored identi cation of requirements, as well as to address concerns regarding deployees.

Key theme 4: capacity building
Interviewees recognised the importance of building local capacity. Suggestions for this included a roster for personnel with appropriate skillsets for different technical functions, particularly from within the region and increasing the number of local institutions and organizations with operational partnerships in place. This regional focus was seen as important to ensure that deployees have more familiarity with local customs and culture, and be better placed to quickly form relationships and build trust with the affected populations and local staff and administration.
Mentoring was mentioned several times in relation to the need for experienced staff, with suggestions for more experienced deployees to act as mentors to allow less experienced deployees to be deployed safely.
Similarly, sharing experiences was linked to building collaborations between WHO and other institutions, as well as between the institutions themselves.

Conclusions
Analysis of staff demographics showed an equal gender split, and an overrepresentation of staff recruited from Europe and Western Paci c regions. Only one deployee was from the South-East Asia region, despite interviewees stating that local deployees may have adjusted better to the environment and would have had more local context-speci c knowledge. Deployees generally brought considerable experience in humanitarian and public health work. In some situations, a compromise may be made between experience and offering opportunities to newer workers to develop capacities to ensure su cient sta ng numbers, including through early recruitment of local staff and subsequent capacity building.
There were different opinions on optimal duration of deployment. Some interviewees felt that six-week deployments were a su cient period of time to offer meaningful contributions within the high-pressure environment of the emergency response, while others felt that contributions made during six-week deployments often did not weigh up to the disruptive effects of staff turn-over that were experienced and advised against shorter term deployments. The need for early performance reviews of deployees was recommended to improve effectiveness. Other issues relating to team organisation and structure, and the need for clear and speci c terms of reference, had been identi ed by previous research (14)(15)(16). Likewise, requests for more information prior to deployment and a more streamlined deployment process have been found in prior studies (11,17), whilst requests for more information on the selection and deployment of staff from operational partnerships was speci c to this study.
Cultural challenges faced by deployees relating to the local context and the o ce environment were also not unique to this study. Working in a markedly different cultural context was often discussed in interviews and had been mentioned in the research carried out by Bjerneld et al. (11), and challenges relating to IT availability and access, data protection and continuity had been raised in other studies (18,19). It was important for deployees to be working in the same o ce area as their team but this was not always possible in the Cox's Bazar sub-o ce due to the o ce layout. Physical o ce space arrangements in future emergencies could consider the likely size and structure of the teams necessary, and be planned to promote teamwork. This can also assist with mentoring, which was discussed in several interviews. There was a positive view of more experienced deployees building capacity of junior colleagues to share their skills and knowledge while responding to the emergency.
This study demonstrates feasibility of organizational operational research in a protracted emergency. It has identi ed themes (sta ng, the deployment process, the o ce/base, and capacity building) which are seen as crucial to the quality and effectiveness of the deployment by both the people deployed and the receiving organisation. A number of these themes can lend themselves for policy and practice changes, as well as conversion into performance indicators for deployments for both the deploying and receiving organizations. Currently, only individual performance monitoring exists (in the form of performance reviews), but the development of wider monitoring and evaluation tools will help to assess the effectiveness of the deployments and improve their quality, for the individuals being deployed, their deploying organizations and the receiving organisation. The interviews conducted identi ed both strengths and areas for development in the themes identi ed, and it is hoped these can act as a trigger to conduct such monitoring and evaluation in future.
The common ndings between this study and ndings from literature indicate continued need to improve the process of deployment of humanitarian workers for effective response. This study provides the basis for further research into the operational and technical aspects of deployments during an emergency public health response. The aspects captured would constitute an important part of a comprehensive review of operations.
Since this exercise was conducted in 2019, various issues raised in this piece have been addressed, but are included here so other settings might learn from the Cox's Bazar experience and address any issues in a timely manner. Notably in 2018, a staff wellbeing committee was established, which has since worked to ag staff wellbeing needs within the Cox's Bazar o ce, to formulate advice to management and to help address wellbeing challenges where possible. Following a broad operational review in late 2018, the establishment of a sub-o ce structure in 2019, with delegations of authority to eld level, clear administrative standard operating procedures and recruitment of administrative support staff with appropriate skillsets have been observed to have streamlined administration considerably overall. In late 2019, a move to medium-term oriented o ce space addressed some of the limitations agged by deployees in the earlier o ce accommodation, which was perceived to limit team's functionality. A consultant was hired speci cally to coordinate and further research efforts conducted in the duty station in 2019, and GOARN has initiated an online interactive training to raise awareness among those eligible to be deployed on the realities and challenges of working with GOARN in the eld on an outbreak response mission. The development of a strategic plan for the health response, structured organizational workplans and mobilization of funding are some priorities to ensure sta ng can be transferred from operational partners to organizational staff as early as possible.

Limitations
The study does not cover internal deployments within WHO, which were largely from the South-East Asia region, particularly the WHO country o ce in Bangladesh. It also does not cover Government deployments, including from the national FETP programme.
The selection process for interviews was originally a strati ed sample but this transitioned to a snowball strategy which may have resulted in selection and reporting bias and resulting in the same issues being raised and interviews reaching saturation point earlier. The interviews, however, continued to elicit a range of both positive and negative comments on a range of topics. There are likely to also be some limitations with the transferability of ndings to other groups and organizations as some issues raised may be speci c to WHO policies and procedures.
What Is Added To The Literature Studies have recognised a dearth of organisational learning when examining the impact of emergency work on humanitarian workers (10). The protracted nature of the humanitarian crisis in Cox's Bazar has enabled this study to be designed and implemented whilst the emergency was ongoing. This allowed interviews to take place with key personnel, with many deployees still available for interview. The engagement of deployees in the evaluation of the response is something that should be done on a more routine basis.
This study has examined the experiences of eld workers deployed to a humanitarian emergency response and identi ed four main areas which contribute to the perceived quality and effectiveness of the deployments which may be considered to improve future deployments. The study has underscored the invaluable role of operational partnerships, early in emergency response, while medium-term oriented planning is being operationalized.

Declarations
Ethics approval and consent to participate The study was undertaken for service evaluation, therefore ethical approval was not required. Informed consent to participate in the interviews was obtained from participants, and they were assured that they were free to withdraw at any point, responses would be kept con dential, and that there would be no repercussions from their participation.

Consent for publication
There are no identi able details, images, or videos relating to an individual person. Those participating in interviews were informed that the study was conducted to produce a review of operational partnerships in Cox's Bazar, and that their responses would be used to inform the review.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.