The impact of the COVID-19 pandemic on healthcare access and utilisation in South Sudan: a cross-sectional mixed methods study

Due to high levels of institutional and social fragility as well as ongoing conflict in South Sudan, the World Bank Group has classified the country as a fragile state since the country gained independence in 2011 [23, 24]. Chronic conflict has left the health system underdeveloped and an estimated 70% of health services are provided by Non-Governmental Organisations (NGOs) and faith-based organisations [25]. South Sudan’s poor health outcomes, such as the world’s highest maternal mortality rate [26], are ascribed to low access to healthcare services, particularly for women, new-borns and children [23]. Furthermore, service coverage varies considerably at the county and state level [27].

The Health Pooled Fund (HPF) is a multi-donor funding mechanism that supports the Ministry of Health in providing basic primary healthcare services in eight of South Sudan’s ten states. HPF was launched in 2012 and was in its third phase (HPF3) in 2022. Funding comes from several donors¹ and is delivered through HPF to eleven implementing NGOs. At the time of this study, the programme supported a total of 794 health facilities at different levels of care including 25 hospitals, 192 Primary Health Care Centres, and 577 Primary Health Care Units. HPF also supports the government-led Boma Health Initiative (BHI) which aims to increase demand, access and awareness of health services by strengthening community health systems.

For both the qualitative and quantitative arms of the study, we purposefully selected three states as study sites taking into account several selection criteria, such as accessibility, security, the availability of both urban and rural areas and the demographic characteristics of the population. Within each state, we purposively selected safe counties to achieve a diversity of the same selection criteria used at the state level. Central Equatoria, specifically Juba County, was chosen to represent an urban setting, whereas Western Equatoria and Warrap were selected to reflect a rural setting and rural-pastoralist setting, respectively. Figure 1 shows the different data collection sites.

The interview and FGD topic guides as well as the household survey questionnaire were based on the healthcare access and utilisation framework of Levesque et al. [29] including aspects of both demand- and supply-side characteristics of access and utilisation.

Data was collected between August 2020 and June 2021 by the Forcier Consulting field team. In total, 10 KIIs, 63 semi-structured interviews, 27 FGDs and 1,223 household surveys were conducted. Interviews and FGDs were conducted by three trained researchers (one per state) who were selected based on their qualitative research experience, their ability to speak the state’s predominant language as well as English, and their knowledge of the state context. Translations of concepts in the topic guides from English to the language used in each state were agreed upon during trainings. Audio files of completed interviews and FGDs were downloaded to password-protected computers and transcribed. Audio files were directly translated into English, capturing as much of the original content as possible. Transcripts were reviewed for accuracy and completeness. The household survey was administered by trained enumerators. One survey was completed per household, to be answered by the heads of household or main care giver, answering on behalf of all household members. Forcier Consulting programmed and conducted the survey in Open Data Kit (ODK) on the SurveyCTO platform.

To contextualise findings from the household survey, FGDs and interviews, a purposive document review was conducted from February to June 2021. Documents were collected through scientific and general search engines. Searches were limited to documents published in the English language and peer-reviewed scientific literature as well as grey literature. The latter included mainly newspaper articles, as well as several policy documents such as South Sudan’s National COVID-19 Response Plan [30]. Abstracts and conclusions were read and, if information relevant to understanding South Sudan’s COVID-19 response was found, the document was included in the study. From reviewing these documents, we developed a narrative of South Sudan’s COVID-19 response. This narrative focuses on the time period up to the end of qualitative and quantitative data collection in July 2021 to contextualise the qualitative and survey responses describing this period of time. All epidemiological data included in this study is also limited to this time range.

Following a qualitative content analysis approach, all interview and FGD transcripts were read repeatedly, and coded deductively based on the healthcare access and utilisation framework by Levesque et al. [29], after which sub-codes were created inductively. This led to a coding tree, which guided data analysis using Atlas.ti v8.4.25. Household survey data was analysed using Stata v15.1 SE. Descriptive statistics were calculated to understand the demographic and socioeconomic composition of the study population, as well as on variables of interest related to health seeking behaviour and COVID-19. These included variables at both the household and individual levels. Descriptive statistics were initially disaggregated by gender of the respondent and head of household, but the limited differences in responses by gender means the results presented in this paper are disaggregated by state only. We made a distinction between three distinct types of COVID-19 effects: direct health effects (infection with the virus), indirect health effects (effects of COVID-19 on communities’ ability to access and utilise care), and indirect non-health effects (e.g., school closures). Survey weights were applied when calculating frequencies to account for the sampling design.

To validate the findings, a research validation workshop was held with researchers involved in the HPF3 project. The workshop aimed to synthesise the study findings, by discussing main themes emerging from data analysis and sharing ideas on the most pressing issues with regards to healthcare access and utilisation in South Sudan during the COVID-19 pandemic. The validation workshop also allowed the researchers to contextualise and better interpret the findings in light of recent developments in the country.

The protocol for the wider study into healthcare access was reviewed and approved by the ethics committee of both KIT and the Ministry of Health South Sudan. All research procedures were in accordance with the Declaration of Helsinki. Data collection adhered to KIT’s ethical standards with regards to neutrality, participation, informed consent, privacy and gender considerations. Respondents invited to the interviews, FGDs and household survey were informed about the intended (secondary) use of their data for dissemination and development purposes, and all participating individuals consented to this.

Interviews and surveys were conducted in a safe environment allowing for privacy and confidentiality. Only the research team had access to the data and identifiers were removed from transcripts. Prior to data collection, the research team was trained on research ethics to ensure guidance on ethical conduct was clearly understood and implemented. Confidentiality of the data collected was ensured by the means of password-protected storage.

In March 2020, before South Sudan confirmed its first COVID-19 case (on April 5, 2020), a High-Level Task Force, including members from various ministries and civil organisations, was established to coordinate and communicate the measures needed to limit viral transmission in South Sudan [31, 32]. The task force would be advised by the COVID-19 National Steering Committee, a larger body consisting of a variety of governmental and non-governmental health actors [33]. Even though the steering committee would be advisory in nature, it was reported that donors and international actors were able to push the COVID-19 response considerably towards their own priorities and targets. In some cases, this resulted in greater attention towards certain pillars of the response, namely the supply of inputs like the procurement of protective equipment, whereas those pillars of the response focusing on managing health services and longer-term health system development remained under-resourced.

“The national government should be the one making the decisions, but they don’t entirely make them. They’re made depending on what the partners kind of push towards, or what the partners are adhering towards.” – Health Systems Technical Manager, donor

Both the High-Level Task Force and the National Steering Committee were guided by the Country Preparedness and Response Plan (later: National COVID-19 Response Plan) [30]. Measures to curb the spread of COVID-19 were implemented in March 2020 [33], through a two-pronged approach, focusing on the reduction of social interactions as well as limiting cross-border transmission of COVID-19 [Table 2] [31, 33]. In response to South Sudan’s first confirmed case on April 5, 2020 [33], the national government reinforced earlier lockdown measures and urged the public to strictly adhere to the social distancing guidelines [34]. Implementation of COVID-19 response activities was sometimes observed to create frictions between the national government and international health actors, for instance on responsibilities and remit. The Ministry of Health and national government were deemed to be responsible for coordinating the COVID-19 response, yet participants from international health agencies felt a lack of ownership and responsibility from the national government for the response.

During the first two months of the COVID-19 outbreak in South Sudan, 1,317 official infections were recorded, mostly public officials in Juba, and 14 confirmed COVID-19 deaths [38]. Even though increasing numbers of confirmed COVID-19 cases were reported, the government decided to relax restriction measures on May 7^th, 2020; a decision which was criticised in local news for being ill-timed and dangerous [35]. Following these relaxation measures, the number of confirmed cases doubled in May, from 111 new confirmed COVID-19 cases per week on May 11 to 332 new confirmed cases on May 18, 2020. Nonetheless, the effects of the COVID-19 pandemic in South Sudan appeared to remain rather moderate, relative to other countries, with a maximum of 481 new confirmed cases per week by the end of May [Fig. 2] [3].

Despite an increase in testing capacity in the beginning of June 2020, testing remained confined to particular areas of the country. Even though Juba-centred testing capacity was rolled-out to the rest of the country, by distributing Gene Xpert machines to laboratories in some states, interview participants suggested that reagents and skilled laboratory staff were often lacking which resulted in the machines not being used to their full capacity. Indeed, an analysis of laboratory data by the Sudd Institute suggested this may have hidden the true magnitude of COVID-19 transmission in South Sudan [39]. Furthermore, the same analysis reported the vast majority of the samples (97.7%) tested at South Sudan’s National Public Health Laboratory between April 2 and May 20, 2020 came from Central and Eastern Equatoria states, and predominantly Juba (Central Equatoria), the nation’s capital, and Nimule (Eastern Equatoria), a border crossing between South Sudan and Uganda. Despite the potentially hidden magnitude of COVID-19 transmission in South Sudan, participants from international health agencies felt that national leadership sometimes disproportionately focused on COVID-19, represented by the size of leadership bodies in charge of the response, and in contrast to the limited number of public health officials at the national and sub-national levels.

“At the national level of course, we [the national steering committee] have everybody from the donors, to the implementing partners to the fund managers, to just technical people, and then all kinds of random organisations. So, the national steering committee has I think [a] membership of over 200 at any given point” – Health Systems Technical Manager, donor

The combination of limited surveillance of COVID-19 transmission, due to weak testing capacity, and human resourcing—i.e., public health officials who would normally be in charge of overseeing the overall coordination of the health sector, redirecting their efforts towards the COVID-19 response—has had severe implications for the COVID-19 response.

In March 2021, South Sudan received 132,000 doses of the AstraZeneca vaccine through the COVAX Facility [40]. The vaccination campaign kicked off on April 6, 2021 and, over the following months, all frontline health workers were offered the vaccine, followed by people with known co-morbidities, including cardiovascular diseases and diabetes among others, and then people above 65 years of age [Fig. 3] [40, 41]. Following this cohort, vaccination progress slowed due to operational difficulties and low uptake. Consequently, South Sudan decided to return over half of its vaccines to the COVAX Facility as the doses would otherwise expire, and vaccination centres closed in mid-July 2021 [42].

To our knowledge, our study was the first to assess the impact of COVID-19 on healthcare access and utilisation in South Sudan. As far as we know, this is also the first comprehensive mixed methods study into healthcare access and health seeking behaviour in South Sudan. The inclusion of three states with different social, economic, cultural and political realities means the populations we drew our sample from represent a diversity of experiences in healthcare needs, access, utilisation and quality. Data collection was conducted by a team of locally-knowledgeable and competent researchers and enumerators, enabling interviews and surveys to be conducted in appropriate languages for each state, and transcripts were immediately translated to English for collation and analysis. This knowledge was invaluable for revising our research tools so questions were relevant, appropriate and could be readily understood by all data collection team members and research participants. We also implemented a thorough quality assurance strategy to detect and correct errors or logical inconsistencies in data, which heightened its quality. Nevertheless, while interviews and FGDs were conducted in five different languages across states (namely Juba Arabic, Arabic, Dinka, Bari and Azande), there were still instances where participants had to speak languages other than their mother tongue, due to the great linguistic variation in the country. This will have impeded their understanding of some questions and their prerogative to be understood. Another limitation that needs to be recognized is that this study focuses on the first 1.5 years of the COVID-19 pandemic only. Changes in COVID-19’s epidemiology and vaccine roll-out since July 2021 may result in some findings no longer be accurate at the time of publication [3]. Further, despite our efforts to limit the effects of social desirability bias, fears and stigmatisation regarding COVID-19 may be underreported. Last, we used data collected in three of South Sudan’s twelve states, meaning certain state-level dynamics may not be captured. Nonetheless, considering the three states were purposively chosen to represent three distinct South Sudanese settings, the results are believed to provide an accurate representation of the situation across much of South Sudan.

Given the evolving nature of the pandemic and its impact on health systems and society, it is important to consider what lessons can be drawn from this study for both future research and the health systems response to the pandemic. First, beyond focusing on indicators such as the supply of reagents and qualified laboratory staff, improving diagnostic capacity in an effective and sustainable manner also requires the alignment of donor targets and funding to ensure inputs can lead to testing readiness [53]. The Organisation for Economic Co-Operation and Development’s framework for Donor Co-Operation [54] could be particularly useful to achieve this. In terms of setting targets, moving away from the provision of inputs to effective testing capacity can be helped by setting output-, outcome- and impact-based targets, such as on the timely diagnosis and treatment of COVID-19. Additionally, surveillance and response strategies should be informed by operational research to areas of public health importance in order to ground them in local settings [53, 55].

Second, to supplement official COVID-19 surveillance, point-in-time questionnaires, patient interviews or other verbal reporting mechanisms may help improve surveillance of COVID-19 transmission in contexts where testing or surveillance is limited, particularly in countries where the national government is coping with significant resource and capacity constraints, such as in South Sudan [56, 57]. Assessing relatively rare cases of mortality which may carry stigma through survey instruments in a fragile or conflict-affected context, however carries limitations as well, often leading to underreporting, as found in relation to measurement of maternal mortality in Afghanistan [58]. More research is therefore needed into the way verbal reporting mechanisms can aid in assessing the true public health impact, including mortality, of current and future disease outbreaks [56].

Third, the insights from this study, based on self-reporting, are best combined with an analysis of actual healthcare utilisation trends in South Sudan. This can also help inform health planners and programmers where support is most needed. Since we found individuals perceived the effect of the pandemic to be mostly on non-health-related aspects of their lives, the support of basic needs, such as food, could be helpful in protecting South Sudanese people from further socioeconomic hardship, particularly those most vulnerable. Additionally, it would be valuable to consider communities’ prioritisation of the various health problems present in the South Sudanese context, outbreaks and chronic health challenges alike, particularly as this should inform the Ministry of Health’s own prioritisation of resources. In line with other literature on the perceived changes to healthcare access during shocks, we recommend to further explore how perceptions about COVID-19 actually change health seeking behaviours, as well as the effect of factors such as mobility restrictions and public information campaigns on these health perceptions and preferences.