Background
Ebola Viral Disease (EVD) outbreaks have occurred with increasing frequency in the last 15 years and the disease constitutes one of the biggest public health problems in Africa [
1]. Many countries have reported EVD in the period between 1976 and 2018. Most notably is the Ebola outbreak in West Africa between 2014 and 2016 that affected mainly Liberia, Sierra Leone and Guinea [
2]. Many frontline health care workers (HCWs) were affected due to lack of preparedness, poor and weak health care systems. In 2018, Democratic Republic of Congo (DRC) reported an EVD outbreak that appears to be still active [
3]. Both the West African and DRC Ebola virus disease (EVD) outbreaks led countries globally to step up preparedness efforts. The World Health Organization (WHO) provided technical assistance to 14 priority countries to strengthen public health activities [
1]. Furthermore, half (7/14) of the priority countries had achieved heightened level of preparedness according to WHO assessment criteria [
1]. This WHO assessment checklist was developed for the EVD outbreak in West Africa in 2014 to look at factors such as proximity to highly affected countries, health systems developed and transport. In the wake of the DRC EVD outbreak of 2018, the World Health Organization Regional Office for Africa (WHO - AFRO) worked with nine neighboring countries including Uganda, to assess readiness and preparedness for EVD [
4]. Uganda has since been tasked to scale up preparedness particularly border districts near North Kivu that experience frequent incursions of people from Eastern DRC (including Ituri province) into Uganda.
The first Ebola outbreak in northern Uganda of 2000 and 2001 is believed to have caught the country unawares with a naïve and inexperienced staff in response to contain the EVD outbreak. There were 425 cases with a case fatality rate of 53% [
5]. In subsequent years, Uganda was in a position to quickly contain EVD outbreaks and minimize fatalities [
5,
6] by putting in place incident command system, institute burial teams and multisectoral teams to manage the outbreak. This showed that the epidemic preparedness, response planning and training of multidisciplinary teams improved the country’s preparedness, alertness and response capabilities in controlling Ebola [
7]. Other well managed outbreaks were Marburg disease around Rubirizi, Kamwenge and Kampala districts [
8‐
10]. Both EVD and Marburg disease are priority zoonotic diseases in Uganda [
11]. However, the low CFR for Marburg disease and the Bundibugyo Ebola virus disease outbreaks may have been due to their low virulence compared to the Zaire Ebola strain.
Yahaya et al. [
4] reports that Uganda had made some progress on EVD in terms of coordination, having rapid response teams, Infections Prevention and Control (IPC), care management, dignified burials, and strengthening laboratory, epidemiological surveillance, risk communication and contact tracing. In addition, Uganda did have some level capacity building on EVD at the points of entry including having budgets and all logistics needed. However, details of preparedness for districts neighboring DRC are scarce. The country developed a Health Sector Strategic Plan (HSSIP III) for 2011–2015 that emphasized capacity building for response, early detection, prevention and control of newly emerging and endemic zoonotic diseases such as Ebola [
12]. The level of preparedness for a disease outbreak determines the impact it has on the health care system and the individual health care workers. However, health care workers are known to lack knowledge of Ebola and other zoonotic diseases [
13]. For instance, previous studies in Uganda found poorly prepared health care workers and non-adherence to the Universal Infection Prevention and Control precautions helped facilitate the rapid spread of Ebola during an outbreak. Starting in 2013, the United States Centers for Disease Control and Prevention (CDC) has supported the Ministry of Health in Uganda to establish an Emergency Operations Center (EOC) [
14,
15]. Under the EOC, the Uganda Ministry of Health has followed the WHO Viral Hemorrhagic Fevers management guidelines and standard operating procedures (SOPs) as a means of ensuring preparedness for EVD and training health care workers [
1].
Districts such as Kasese and Rubirizi, that are near the eastern border of the Ebola prone belt of DRC, are presumed to be high risk of EVD. The communities in these districts are poor and underserved and, as a result, still depend on hunting and consumption of game meat as source of livelihood. This increases their likelihood of interaction with wildlife and, therefore, the risk of contact with EVD reservoirs directly or indirectly [
16,
17]. Uganda has porous borders whereby communities from neighboring countries travel easily to Uganda. The level of preparedness by districts near the border is not well known especially in terms of HCW knowledge of EVD, presence of infrastructure, logistics, rapid response teams, burial teams and simulations exercises to prepare health workers for an EBV outbreak. We set out to assess preparedness of the health care system and identify appropriate preparedness measures for Ebola outbreak response in Kasese and Rubirizi districts, Western Uganda.
Methods
Study area and setting
A cross-sectional study was carried out in health centers in Kasese and Rubirizi districts, in Western Uganda [
18] from August to November 2016. Kasese district shares borders with the DRC with two administrative counties of Bukonzo and Busongora. About 26.5% of households are located within 5 km of the nearest Public Health Facility [
19]. Uganda has guidelines for designating, establishing and upgrading different health care units ranging from Health Centre (HC) II to Hospital [
20]. For instance, by virtue of their levels, the HC IV are by policy supposed to be equipped with logistics and supplies ready to identify, detect index cases and if possible, respond to Ebola and other deadly epidemic prone disease outbreaks. Kasese district is served by three hospitals, three HC IVs, 44 HC Level III, and 57 HC II [
21]. The district has a total land area of 2724 km
2 of which 885 km
2 is reserved for Queen Elizabeth National Park and 652 km
2 is reserved for Rwenzori Mountains National Park. The population density of Kasese District in 2014 was estimated to be 236 people per km
2 [
19]. On the other hand, Rubirizi district with a population density of 118 people per km
2 [
19] shares a border with Kasese and Bushenyi districts with two administrative counties of Bunyaruguru and Katerera counties. About 50% of Queen Elizabeth National Park lies in Rubirizi District. About 28% of the households are located within 5 km of the nearest Public Health Facility [
19]. Rubirizi does not have a hospital whether Public or Private. However, it has one HC level IV, three HC Level III and 12 HC Level II facilities [
21].
Study design and variables
Study units composed of all the hospital and HC IV level facilities, and randomly selected HC level III facilities were established in both Kasese and Rubirizi districts [
18]. Health Care Workers (HCWs), defined as any personnel working within a health unit such as medical doctors, medical clinical officers, nurses, laboratory technicians, midwives and nursing assistants, were recruited into the study through random selection within each health care facility. Other respondents recruited were district health management team members, in-charge of hospitals and health unit administrators. A total of 22 HC, including four hospitals, were assessed.
The sample size was estimated using Kish formular [
22]. Assuming a 95% confidence interval and if the proportion of health care workers with knowledge on contact with body fluids of symptomatic person is 58.9% [
23] and allowing alpha of 5% and non-response rate of 5%, approximately 391 health care workers were taken to be the number to be interviewed. However, only 187 HCWs were at the health facilities. These were interviewed using a questionnaire developed by the authors (
S1). All hospitals and HC IVs in both districts were purposively selected because, by virtue of their levels, they are supposed to be equipped with logistics and supplies ready to respond to Ebola and other deadly epidemic prone disease outbreak. Fifty percent of the HC IIIs in each district were randomly selected using a ballot paper system, whereby a list of HC IIIs was obtained and a ballot paper with their names was written. The names were put on separate pieces of paper, thrown into a bag, shook and then randomly selected, one at a time, until the desired number was reached. From the selected health facilities, the names of health care workers were obtained from either the District Health Officers or the respective health facility administrators. Then the number of respondents was selected using simple random sampling. This was done using the computer-generated method whereby a list of names was generated into a column in a Microsoft Office (MS) Excel spreadsheet. Using the function = RAND the list of names and the random number were sorted by the random numbers.
In order to assess the level of knowledge on containment of EVD, names of HCW from the selected health facilities were obtained from either the District Health Office or the respective health facility administration. The names were then confirmed by fellow HCWs at facility level. Then the number of HCWs was selected using simple random sampling and proportionate to the number of HCWs in a given district.
Independent and dependent variables
There were two categories of independent variables at 1) health facility level, and 2) individual HCW level. The health facility level variables considered were structural design of a health facility, district physical structural scale, and the catchment area /distance from facility to the vulnerable community. Regarding the structural design of the health facility, the district physical structural scale [
24] was used to calculate the distances and spaces used for ascertaining safety precautions. A district health map was used to determine the distance or catchment area from each health care facility to the vulnerable community. The individual HCW variables included the respondent’s age, sex, type of employment, and job designation of the HCWs. The level of knowledge by a HCW was used as an dependent variable.
A WHO consolidated preparedness checklist developed for the West Africa Ebola disease outbreak of 2014 [
1] was used to develop an adapted checklist for health facilities to measure the presence of: a VHF incident management center, a high-level isolation unit, clinical notification systems, a triage area spacious enough to ensure isolation of a patient in a holding area, and sufficient space to enable maintaining a meter distance among patients and between patients and HCWs. Data was also collected on the availability of protocols such as Infection Prevention and Control (IPC) guidelines, EVD management guidelines, burial and disposal guidelines, EVD treatment unit SOPs, policy guidelines/standard operating procedures, presence of surveillance systems, rapid response teams, Ebola and other deadly epidemic prone disease focal persons, burial teams, and table-top simulations and referrals [
1]. Furthermore, the health facility logistics and supplies were measured to determine the capacity of a health facility based on the availability of the following: a health facility integrated disease surveillance and response check list, a standard case definition book for the 16 notifiable diseases, report forms for the 16 notifiable disease displayed, a budget, funds, lab and medical equipment, personal protective equipment (PPE), disinfectants & detergents, triple packaging kits, running water with soap and transportation of specimens. The knowledge of HCWs was measured based on the number of health care workers trained in Ebola and other deadly epidemic prone disease outbreak containment, the use of PPEs, infection control and practice, barrier nursing, quarantining, and triaging and isolation. The individual’s inclusion criteria were health care workers who have consented to the study, are working at health care facility of Health centre III and above and have worked in the district for the last 6 months.
Data management
The data collected from health facilities and participants was cleaned, confirmed using different data sources such as asking more than one respondent about a given facility and checking for presence of a particular data by reading through documents. Data was then edited before being entered into Epi-data software. For any missing data, we went back to the field to collect it and/or verified from the hard copies. During the initial interview, we indicated to the facility in-charge, that we would follow up by phone in case we had follow up questions. Any other missing data from analysis was indicated so in the tables. The minimum completeness of data was 91% at facility level. The data was then analyzed using Stata version 14. EVD Preparedness was operationalized as the mean score of 5 domains: (a) two of the domains relate to Infrastructure and Logistical capabilities, and (b) three of the domains, measure the level of knowledge of an individual health worker about Ebola disease etiology, control and prevention. Infrastructure capability was assessed on a 14-item binary scale, where zero represented that there was no adequate infrastructure. The key infrastructural components were measured on a 14-item scale and if this gave a 14 “Yes” response, then that would demonstrate full availability of all the infrastructure for EVD outbreak and response. A score of equal to 7 or more was taken to be adequate. Logistical capacity was operationalized on a 23 item binary scale where zero shows no logistical capabilities and then 23 “Yes” responses demonstrated the maximum logistical capacity that was utilized in EVD detection, response and eventual management in case of the outbreak. Knowledge of Ebola etiology was measured on a 11 item binary scale, zero representing no knowledge and 12 points indicating a very high level of Ebola etiology knowledge. Knowledge on preventing EVD was elaborated on a 19-item binary scale point scale ranging from 0 to 19 and to this effect, if a health worker scored zero, then this demonstrated no knowledge of Ebola prevention measures. A 11-item binary scale was used to measure the knowledge of health care workers on Ebola control measures, zero indicating no knowledge on control measures whereas 11 points indicated that the HCW was very knowledgeable about the control measures.
Aggregation of these indicators into the five domains of the Ebola preparedness composite indicator was done to get an overall preparedness picture. The indicators were transformed into five domains by linear aggregation through summation. All the Yes responses were scored 1 indicating availability of a capability in a given domain and scored zero where respondents indicated absences of capability in a given domain. Mean score for each domain was calculated by summing up all scores in that domain and averaging them. The cut off for the binary outcome was set at the 50th percentile. Thus, a health facility that scored less than the 50th percentile was categorized as not prepared. The scale was highly reliable with an alpha score of 85% and 79% for infrastructure and logistical preparedness respectively. The level of health care knowledge was assessed with a cutoff set at 50th percentile. Health workers who scored less than the 50th percentile were categorized as having low levels of knowledge. The scale scored a Cronbach’s alpha reliability coefficient of 0.83.
At the univariate level, background characteristics of participants, infrastructure, logistic and knowledge indicators were summarized using descriptive statistics. For Health facility infrastructure and logistic capabilities, confounding analysis at bivariate level was carried by disaggregating the level by district and type of health facilities to explain the variance in the levels of preparedness. For the HCWs knowledge, bivariate analysis was carried out using contingency table analysis with level of knowledge as the dependent variable and the social demographic characteristics as the independent variable. The independent covariates namely, district, gender, age, education level, job description and religion, were regressed on the level of knowledge using the equation of a straight line in multivariate logistic regression. The unadjusted and adjusted odds ratio was used to assess the level of association at 0.05 significance level. A mixed-effect multilevel and fixed-effect binary logistic regression models were compared and the fixed effect binary logistic that had the smallest AIC value was used. A cut-off p-value for selection at bivariate level was set at 0.20 and forward selection method was used to select variables to include in multivariable logistic regression. The Area under the ROC curve was not fitted since we had more than one independent variables that were categorical in nature.
Conclusions
Rubirizi and Kasese districts rated the same in terms of knowledge on EVD and preparedness in terms of logistics, such as laboratory equipment, budgets, PPEs and disinfectants. There was inadequate supply and preparation in terms of laboratory and medical equipment such as PPEs, triple packaging and special transportation mechanism. The knowledge of level of health care workers was slightly high as regards etiology while knowledge on EVD prevention and control was low. It is important that for a health system to be prepared all components such as knowledge of a disease, infrastructure and logistics should be in place. There is need to cascade preparedness and response efforts at global and national level to local community levels where disease outbreaks actually start from. District Rapid Response Teams should be constituted, trained, supported to hold regular meetings and conduct simulation drills. Absence of case definition books, burial teams and lack of dissemination of standard protocol like those on Infection Prevention and Control may put HCWs at risk of disease. The current study will provide a baseline of what is needed when it comes to preparing sub-national level health care systems, such as districts, in control and management of infectious diseases. Cross-border collaborations between Uganda and DRC is key to help in coordination of preparedness efforts as EVD spreads across borders during outbreaks.
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit
http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.