Study setting

Ghana is a West African nation of 29 million people who are mostly concentrated in the southern and coastal regions. Ghana is bordered by Ivory Coast to the west, Burkina Faso to the north, Togo to the east, and the Atlantic Ocean to the south [14]. Our study was conducted in Upper East Region (UER) with 13 administrative districts, a predominantly agricultural area of approximately one million inhabitants, bordered by Burkina Faso to the north and Togo to the east [15–17]. The UER is one of the five northern regions of Ghana. The total land area is about 8,842 square kilometres (2.7% of the total land area of Ghana) and has savannah woodland. However, human activities with the ecology are leading to near semi-arid conditions. The climate is characterized by one rainy season from May/June to September/October. It also has low humidity and high temperatures. There are a number of tourist attractions in the UER such as the Paga Crocodile Pond and the Bolgatanga Museum [17]. The distance from Accra to UER is about 782 km while 822 km to Paga border (the Ghana–Burkina Faso border) and 860 km to Pusiga border (Ghana–Togo border). There is one main standard road from Accra through Paga to Burkina Faso, while the road leading from Bolgatanga to Togo is in a poor state and partly untarred. Often, it takes about 12–15 hours to travel by road from Accra to either border. The UER is far from the national capital with a higher likelihood of epidemics importation due to its borders.

Ghana’s health care system is organized in a three-tier system (district/sub-district/community, regional, and national levels) and the main implementing agency is the Ghana Health Service. The smallest unit of the health system is the Community-based Health Planning and Services (CHPS) [15, 18]. Each district is served by a hospital, health centers, private and mission clinics, and CHPS Compounds. The Regional Health Directorate is responsible for the overall health service planning, organization, monitoring, supervision, evaluation, and provision of technical support to districts/municipalities [17].

Ebola case definition during the outbreak in West Africa

During the peak of the Ebola outbreak in West Africa, the case definition was revised to reflect the prevailing situation in Ghana and distributed to all levels of the health system in the country. Alert Case: any ill person with onset of fever who is not responding to treatment for common causes of fever in the area; any ill person with at least one of the following signs: bleeding, bloody diarrhea, blood in the urine; or any sudden death. Suspected case: any person alive or dead, suffering or having suffered from a sudden onset of high fever and having had contact with a suspected, probable, or confirmed Ebola case; or any person with sudden onset of high fever and at least three of the following symptoms: headache, anorexia (loss of appetite), lethargy, aching muscles or joints, breathing difficulties, vomiting, diarrhea, stomach aches, difficulty swallowing, and hiccups. A suspected case is also any person with inexplicable bleeding or any sudden inexplicable death [19]. Probable case: any deceased suspected case (where it has not been possible to collect specimens for laboratory confirmation) having an epidemiological link with a confirmed Ebola case or any suspected case evaluated by a clinician. Confirmed case: any suspected or probable case with a positive laboratory result, laboratory-confirmed cases must test positive by reverse transcriptase-polymerase chain reaction (RT-PCR) or by detection of IgM antibodies directed against Ebola.

Travelers health declaration on entry in Ghana

At every entry point into Ghana, travelers disembark from their flights/vessels/vehicles and they provide a completed health declaration form. The first section of the form collects personal information, contact address in Ghana, and potential symptoms of EVD. The personal information section includes name, age, sex, nationality, country of departure, country visited on this trip, passport number, flight/vessel/vehicle, and seat number. The second section deals with the contact address, location, street name, and phone number of the traveler in Ghana. The third section includes questions such as: have you had close contact with patients or suspects suffering from Ebola/Viral Haemorrhagic Fever in the past 21 days? Have you had close contact with domestic animals or wildlife in the past 21 days? Do you have any of the under-listed signs and symptoms: a) fever, b) cough, c) headache, d) vomiting, e) bodily weakness, f) diarrhea, g) sore throat and h) bleeding from any part of the body.

The port health officer then collates the number of travelers screened and detected with fever or febrile illness each day. The information is then recorded in the weekly epidemiological/surveillance report by the port health unit and submitted to the district disease control officer. The recorded information is submitted to the district health administration using weekly reporting forms. All suspected cases are referred for further investigation and confirmation. The district's rapid response team then conduct further investigation, including obtaining specimens for confirmation at Noguchi Memorial Institute for Medical Research which is the only laboratory with the capacity for confirming suspected EVD cases in Ghana.

Study design

This was an observational study that used quantitative and qualitative data. The quantitative data was from epidemiological weekly reports of travelers entering Ghana in the UER while qualitative data was from 12 key informant interviews (7 port health officers and 5 district directors of Ghana Health Service) in the five out of 13 districts in the UER that have designated border entry points to query them on screening of travelers at the entry points. Note that in the five districts with designated entry ports, there are a total of seven points of entry: Paga border (Kassena-Nankana west district), Mognore and Missiga borders (Bawku Municipal), Zebilla border (Bawku west district), Kulungugu and Pulmakom borders (Pusiga district) and Namoo border (Bongo district).

Eligibility of key informants

Eligibility for participation included: a) working with port health unit of the district health system in one of the five districts with a designated entry port in the UER; b) working as director of health service in one of the five districts with a designated entry port in the UER; c) involvement in EVD screening and d) completion of written consent. In each district with a border entry, the director of district health service was asked and agreed to participate in the study (total of 5 district directors). In addition, at each border entry, the responsible port health officer was also asked and agreed to participate in the study (total of 7 port health officers).

Data collection

Quantitative data were collected from the paper-based weekly epidemiological reports collected at each border entry point (October 2014 –March 2015) and entered into a Microsoft Excel (Redmond, WA) spreadsheet. October 2014 reflects the beginning of screening at the Paga border of the Kassena-Nankana West district before the other four districts with border crossings in the UER began screening for EVD in 2015. Data recorded included the number of individuals screened, the country of origin (beginning in 2015), and the number of suspected cases of EVD.

Qualitative data were collected from 12 key informants using face-to-face interviews. The main issues addressed in the semi-structured interview guide were: a) perceived objectives of EVD screening; and b) problems associated with screening, identifying, and recording of suspected EVD. Examples of the questions included: “What is the objective of EVD screening? What are the problems associated with EVD screening?” The fieldwork was conducted from 25^th May - 6^th June 2015.

Data analysis

Data analysis was based on the number of travelers screened for EVD and disaggregated by the three most affected countries with EVD. Frequencies of screened travelers were produced from the weekly reports, as well as the number of suspected cases of EVD. The qualitative aspect captured the perspective of the informants on the EVD screening at border entry. we produced transcripts from each interview and these were read multiple times to identify specific themes and categories using an inductive approach [20]. These were data preparation, cleaning, a close reading of texts, creation of themes or categories accordingly. All the transcripts were read by the lead author while subsample transcripts were read by the co-authors. The repeated reading by all authors addressed issues of credibility, transferability, dependability and confirmability as well as coding consistency of the findings of the study. After the reading and discussion, a coding framework was developed and transcripts were coded by the lead author. This approach allows findings to emerge from the frequent, dominant or significant themes inherent in the data [20, 21]. The purpose of the approach adopted in the qualitative analysis include to develop a brief and summary format from extensive raw data; to identify clear links between the study objective and summary findings through transparent and justifiable standard [20, 21]. Finally, the Attride-Stirling thematic network analysis framework technique was used to break up the themes into basic themes, organizing themes, and global themes of the findings. This approach was employed because of its frequent use in health and social science-related studies [20–22].

Ethical statement

Ethical approval for the study was obtained from the Navrongo Health Research Centre Institutional Review Board (NHRCIRB155), Ghana. We also obtained permission from the Regional Health Directorate of the Upper East Region to conduct the study. All participants were informed about the nature and processes involved in the study, the research objectives, and the confidentiality of the data. The participation of the subjects was completely voluntary in nature, and signature/thumbprint consent was obtained from each participant.

Quantitative data

Tables 1 and 2 show weekly epidemiological data on screened travelers at the points of entry. A total of 41,633 travelers were screened between October 2014 and March 2015. Only one suspected case of EVD was identified through border screening. The person was considered a suspected case because of high temperature (>38°C), headache, breathing difficulties, coughing with blood, and vomiting. The suspected case was a traveler from Togo screened at the Namoo border (Bongo district) point of entry, yet serum samples were negative after further investigation at the Noguchi Memorial Institute for Medical Research in Accra, Ghana.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Epidemiological weekly reports on screened travelers for EVD in Upper East Region, Ghana, 2014.

https://doi.org/10.1371/journal.pone.0245039.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Epidemiological weekly reports on screened travelers for EVD in Upper East Region, Ghana, 2015.

https://doi.org/10.1371/journal.pone.0245039.t002

Screening in the UER started at the Paga point of entry in the Kassena-Nankana West district in October 2014, resulting in 10,905 individuals screened in 2014. Data were not initially recorded on the country of origin. In 2015, 30,633 travelers were screened from the seven points of entry. Fifty-two travelers originated from one of the countries affected by EVD: 40 were Guineans, 10 were Sierra Leoneans, 1 was a Liberian, and 1 was a Malian. Their contact and personal information were reported to the national level per EVD reporting protocols.

Qualitative data

Satisfaction with EVD screening

The majority (5/7) of the port health officers reported that they were not satisfied with the screening of travelers at the points of entry for suspected EVD cases, given the challenges described above. The key informants declared that the EVD screening at the border entry points is unsatisfactory due to inadequate isolation rooms, logistics (supplies and PPEs), to few trained staff, lack of capacity to confirm suspected EVD cases, irregular or no feedback on suspected cases to the peripheral levels, inaccurate personal information of travelers, porous borders and language barriers. The combined effect of these limitations renders the border screening for EVD ineffective and unreliable for detection and prevention of disease spread in the country.

Knowledge about EVD border screening

Overall, the study demonstrated that EVD screening is carried out at different entry points in the country. These include seaports, airports, and inland border entry [11]. The purpose and importance of EVD screening are known among frontline health workers. The knowledge of EVD includes routes of transmission and behavioral implications. The screening of travelers using a thermometer is able to detect travelers presenting with fever. However, it has the potential to miss about 20% of suspected cases [13]. Despite the knowledge of Ebola among informants, we concluded that EVD screening is only fairly effective in the study area. This is due to challenges such as inadequate personal protective equipment, insufficient space, and isolation rooms, too few trained staff and delays at border entry points, lack of cooperation, language barriers and porous borders. These pose a threat to the effectiveness of EVD screening at the border entry points. This coupled with the weak health system in the country has the potential to affect detection and prevention of disease spread. This conclusion is similar to previous studies that reported that screening interventions for Ebola and SARS were found to be ineffective at early detection and prevention of disease spread [2, 25].

Isolation of suspected cases among travelers

Generally, the study found that isolation is being implemented at some border entry points while efforts are made to establish more centres in the country. However, the 48 hours holding period for two diagnostic tests to be conducted before national and international travelers are not implemented [11]. For effective EVD screening, each border entry point is required to have an isolation centre or treatment centre for holding suspected EVD cases among travelers. However, not all districts or border entry points had isolation centres for suspected EVD cases. This leads to inconsistencies in the implementation of the isolation approach as recommended by the World Health Organization at the different border entry points in the country. In Ghana, a previous study found that regions and districts did not have EVD treatment or holding centres, though isolation or holding centres are critical in EVD preparedness and screening [1, 11]. Another study has reported the existence and utilization of EVD isolation centres and increasing case find of EVD cases [26]. Timely isolation of suspected EVD cases is vital for early detection and prevention of disease spread. In the study setting, the existing EVD screening lacks the key building blocks for effectiveness.

Documentation and EVD screening information

EVD screening uses the recommended health declaration form to collect information from travelers. This form is similar to the traveler public health declaration form recommended for countries by WHO [11]. The frontline health workers use the integrated disease surveillance and response (IDSR) system guidelines to compile and report the number of travelers and suspected cases weekly to the next level of the health system in the country [27]. The existence of health declaration forms for collecting information from travelers and schedule epidemiological reporting is important for data compilation, reporting and feedback. This is an important aspect of EVD screening in Ghana. This study also found that the health system is unable to effectively conduct contact tracing and monitoring due to inadequate health declaration forms. Some of the respondents use their own resources to make photocopies of forms for data capture and reporting. The inability to effectively implement the WHO guidelines coupled with the weak health system makes EVD screening at the border entry points ineffective [11].

Challenges for EVD screening

Key informants indicated a variety of challenges to effective screening, yet no known cases of EVD were identified in Ghana. Considering that Ghana was classified as the country with the highest risk of detecting suspected EVD cases [3, 4], an effective screening system was both appropriate and timely. Yet this study illustrated profound challenges associated with screening and disease surveillance. About eight (8) major areas were derived from the findings to describe challenges during EVD screening. These include; inadequate personal protective equipment and supplies, insufficient space or isolation rooms, porous borders, too few trained staff, delays at border entry points, lack of capacity to confirm suspected EVD cases locally, lack of cooperation by some travelers, and language barriers. Findings from this study raise questions about screening effectiveness at the points of entry, potentially leading to ineffectiveness in early detection and prevention of disease spread in Ghana.

It is particularly noteworthy that some of these challenges come from outside the screening system itself: some travelers conceal their personal information or falsify their contact information. The likely reasons for which travelers conceal their personal information (e.g. telephone numbers) include avoiding contact tracing when a suspected case is confirmed, to avoid quarantine or deportation to the country of origin, and lack of appropriate travel documents (e.g. fake passport). This is against the WHO recommendation for EVD screening where probable and suspected cases are to be isolated and their travel restricted to prevent the spread of the disease [11]. A previous study also reported of mistrust as an issue that can lead to concealment of information from health authorities [28]. This has the potential to affect the ability of the health system to trace such individuals and their contacts when it becomes necessary. During future outbreaks, policies that implement stiff penalties for falsification may be warranted.

Existence of porous borders is a common phenomenon in the study setting and Africa as a whole. The existence of porous borders has been observed in Uganda [28]. The reasons for porous borders include economic, cultural, and social. Some individuals reside in one country while conducting their economic (market trading, agriculture and small businesses) activities in neighboring communities across the borders. There is also evidence of marriages across or along borders of neighboring countries. These undermine formal border entry points but rather favor the informal border entry points. It also affects effective implementation of the WHO guidelines for EVD screening. Accessing the formal border entry points result in delays, increase travel time and additional financial cost. Thus, travelers prefer to use informal border crossing points. These should be considered when epidemic control measures are being developed and implemented in rural settings. This overall affects the effectiveness of EVD screening at the border entry points for suspected cases.

This study supports earlier research that found that the health systems in the most affected countries were unable to identify suspected cases or confirm them in a timely manner [2, 8]. It was repeatedly reported by the informants that weekly screened data were reported irregularly due to inadequate reporting forms. To address this problem, some port health officers traveled to the districts to make photocopies using their own funds. This particular challenge of inadequate reporting forms has been reported in previous studies [27, 29]. This is one example in which integrated disease surveillance could improve data capture: if forms included multiple diseases it is plausible there might be a higher priority placed on ensuring sufficient numbers of forms were available. Our data suggest that feedback to port health officers on the EVD screening data they provide is rarely done, which is similar throughout much of sub-Saharan Africa [27, 30, 31]. Yet an integrated system could help prioritize feedback to the districts to ensure that district officers are both providing and acting upon timely data.

The findings also showed inadequate trained staff on EVD screening and related activities. This leads to delays at the border entry points during screening. Though, efforts are being made to train staff at the regional and district levels. This finding is similar to studies that found that many frontline health workers did not receive training on EVD in the country [1, 28]. The too few trained staff is a major contributory factor to the risk of exposure among health workers and other travelers. When health workers are overburdened with work, the tendency for them to make mistakes due to poor judgment increases. Not only will potential suspected EVD cases be missed, but travelers will be delayed resulting in poor cooperation and use of informal entry points.

Limitations

The study has several limitations. First, as the investigation was limited to districts with border entry points in the Upper East Region, the findings are not representative of the entire health system or border screening system in Ghana. Second, key informants were asked to discuss the issues associated with EVD screening, which was a key component of their job responsibilities. Thus, it could be perceived that we were asking informants to tell us how well they were doing their jobs, which has the potential to lead to social desirability bias and downplaying of challenges. However, respondents were forthcoming with challenges and we do not believe social desirability significantly biased our findings.

Despite the limitations, this study has several noteworthy strengths. First, the paper addresses a very important aspect of EVD screening that has not been well examined in the research literature: the difference between formally recommended protocols and what actually happens in practice. While the protocols for border screening are of obvious importance, it is often the on-the-ground logistics that can ultimately determine whether they are successful. The findings from this study suggest that further work is necessary to bridge the gap between screening protocols and on-the-ground logistics. Second, this study illustrates the enormous number of travelers screened to identify a single potential case of EVD. While the lack of a confirmed case of EVD is ultimately a positive outcome of the screening process, it is important for governments and national and international health organizations to recognize the enormous resources required to patrol a national border to implement effective, comprehensive screening.

These findings have important implications. First, the need for adequate training, supplies and space is paramount if border screening is to be effective, each requiring sufficient human and financial resources dedicated to the effort. Second, screening guidelines should be tailored to meet local context dynamics and geographical needs for effective implementation and outcomes. Although, we do not know where and when the next Ebola outbreak affecting West Africa will occur, the spread of COVID-19 has provided additional lessons in managing the spread of infectious diseases. There are various in predictions and projections of between countries and continents. These findings and lessons learned are important for the overall disease surveillance strengthening in Ghana, West Africa and beyond.