Methodological issues of retrospective surveys for measuring mortality of highly clustered diseases: case study of the 2014-16 Ebola outbreak in Bo District, Sierra Leone.

BACKGROUND: There is a lack of empirical data on design effects (DEFF) for mortality rate for highly clustered data such as with Ebola virus disease (EVD), along with a lack of documentation of methodological limitations and operational utility of mortality estimated from cluster-sampled studies when the DEFF is high. OBJECTIVES: The objectives of this paper are to report EVD mortality rate and DEFF estimates, and discuss the methodological limitations of cluster surveys when data are highly clustered such as during an EVD outbreak. METHODS: We analysed the outputs of two independent population-based surveys conducted at the end of the 2014-2016 EVD outbreak in Bo District, Sierra Leone, in urban and rural areas. In each area, 35 clusters of 14 households were selected with probability proportional to population size. We collected information on morbidity, mortality and changes in household composition during the recall period (May 2014 to April 2015). Rates were calculated for all-cause, all-age, under-5 and EVD-specific mortality, respectively, by areas and overall. Crude and adjusted mortality rates were estimated using Poisson regression, accounting for the surveys sample weights and the clustered design. RESULTS: Overall 980 households and 6,522 individuals participated in both surveys. A total of 64 deaths were reported, of which 20 were attributed to EVD. The crude and EVD-specific mortality rates were 0.35/10,000 person-days (95%CI: 0.23-0.52) and 0.12/10,000 person-days (95%CI: 0.05-0.32), respectively. The DEFF for EVD mortality was 5.53, and for non-EVD mortality, it was 1.53. DEFF for EVD-specific mortality was 6.18 in the rural area and 0.58 in the urban area. DEFF for non-EVD-specific mortality was 1.87 in the rural area and 0.44 in the urban area. CONCLUSION: Our findings demonstrate a high degree of clustering; this contributed to imprecise mortality estimates, which have limited utility when assessing the impact of disease. We provide DEFF estimates that can inform future cluster surveys and discuss design improvements to mitigate the limitations of surveys for highly clustered data.

Main findings: For humanitarian organizations it is imperative to document the methodological limitations of cluster surveys and discuss the utility.Added knowledge: This paper adds new knowledge on cluster surveys for highly clustered data such us in Ebola virus disease.Global health impact of policy and action: We provided empirical estimates and discuss design improvements to inform future study.

Health conditions of migrants, refugees and asylum seekers on search and rescue vessels on the central Mediterranean Sea, 2016-2019: a retrospective analysis.

OBJECTIVES: This study will contribute to the systematic epidemiological description of morbidities among migrants, refugees and asylum seekers when crossing the Mediterranean Sea. SETTING: Since 2015, Médecins sans Frontières (MSF) has conducted search and rescue activities on the Mediterranean Sea to save lives, provide medical services, to witness and to speak out. PARTICIPANTS: Between November 2016 and December 2019, MSF rescued 22 966 migrants, refugees and asylum seekers. PRIMARY AND SECONDARY OUTCOME MEASURES: We conducted retrospective data analysis of data collected between January 2016 and December 2019 as part of routine monitoring of the MSF's healthcare services for migrants, refugees and asylum seekers on two search and rescue vessels. RESULTS: MSF conducted 12 438 outpatient consultations and 853 sexual and reproductive health consultations (24.9% of female population, 853/3420) and documented 287 consultations for sexual and gender-based violence (SGBV). The most frequently diagnosed health conditions among children aged 5 years or older and adults were skin conditions (30.6%, 5475/17 869), motion sickness (28.6%, 5116/17 869), headache (15.4%, 2 748/17 869) and acute injuries (5.7%, 1013/17 869). Of acute injuries, 44.7% were non-violence-related injuries (453/1013), 30.1% were fuel burns (297/1013) and 25.4% were violence-related injuries (257/1013). CONCLUSION: The limited testing and diagnostics capacity of the outpatient department, space limitations, stigma and the generally short length of stay of migrants, refugees and asylum seekers on the ships have likely led to an underestimation of morbidities, including mental health conditions and SGBV. The main diagnoses on board were directly related to journey on land and sea and stay in Libya. We conclude that this population may be relatively young and healthy but displays significant journey-related illnesses and includes migrants, refugees and asylum seekers who have suffered significant violence during their transit and need urgent access to essential services and protection in a place of safety on land.

Health of migrants, refugees and asylum seekers in detention in Tripoli, Libya, 2018-2019: Retrospective analysis of routine medical programme data.

Libya is a major transit and destination country for international migration. UN agencies estimates 571,464 migrants, refugees and asylum seekers in Libya in 2021; among these, 3,934 people are held in detention. We aimed to describe morbidities and water, hygiene, and sanitation (WHS) conditions in detention in Tripoli, Libya. We conducted a retrospective analysis of data collected between July 2018 and December 2019, as part of routine monitoring within an Médecins Sans Frontières (MSF) project providing healthcare and WHS support for migrants, refugees and asylum seekers in some of the official detention centres (DC) in Tripoli. MSF had access to 1,630 detainees in eight different DCs on average per month. Only one DC was accessible to MSF every single month. The size of wall openings permitting cell ventilation failed to meet minimum standards in all DCs. Minimum standards for floor space, availability of water, toilets and showers were frequently not met. The most frequent diseases were acute respiratory tract infections (26.9%; 6,775/25,135), musculoskeletal diseases (24.1%; 6,058/25,135), skin diseases (14.1%; 3,538/25,135) and heartburn and reflux (10.0%; 2,502/25,135). Additionally, MSF recorded 190 cases of violence-induced wounds and 55 cases of sexual and gender-based violence. During an exhaustive nutrition screening in one DC, linear regression showed a reduction in mid-upper arm circumference (MUAC) of 2.5mm per month in detention (95%-CI 1.3-3.7, p<0.001). Detention of men, women and children continues to take place in Tripoli. Living conditions failed to meet minimum requirements. Health problems diagnosed at MSF consultations reflect the living conditions and consist largely of diseases related to overcrowding, lack of water and ventilation, and poor diet. Furthermore, every month that people stay in detention increases their risk of malnutrition. The documented living conditions and health problems call for an end of detention and better protection of migrants, refugees and asylum seekers in Libya.

Predictors of virological failure among people living with HIV receiving first line antiretroviral treatment in Myanmar: retrospective cohort analysis.

BACKGROUND: Progress toward the global target for 95% virological suppression among those on antiretroviral treatment (ART) is still suboptimal. We describe the viral load (VL) cascade, the incidence of virological failure and associated risk factors among people living with HIV receiving first-line ART in an HIV cohort in Myanmar treated by the Médecins Sans Frontières in collaboration with the Ministry of Health and Sports Myanmar. METHODS: We conducted a retrospective cohort study, including adult patients with at least one HIV viral load test result and having received of at least 6 months' standard first-line ART. The incidence rate of virological failure (HIV viral load ≥ 1000 copies/mL) was calculated. Multivariable Cox's regression was performed to identify risk factors for virological failure. RESULTS: We included 25,260 patients with a median age of 33.1 years (interquartile range, IQR 28.0-39.1) and a median observation time of 5.4 years (IQR 3.7-7.9). Virological failure was documented in 3,579 (14.2%) participants, resulting in an overall incidence rate for failure of 2.5 per 100 person-years of follow-up. Among those who had a follow-up viral load result, 1,258 (57.1%) had confirmed virological failure, of which 836 (66.5%) were switched to second-line treatment. An increased hazard for failure was associated with age ≤ 19 years (adjusted hazard ratio, aHR 1.51; 95% confidence intervals, CI 1.20-1.89; p < 0.001), baseline tuberculosis (aHR 1.39; 95% CI 1.14-1.49; p < 0.001), a history of low-level viremia (aHR 1.60; 95% CI 1.42-1.81; p < 0.001), or a history of loss-to-follow-up (aHR 1.24; 95% CI 1.41-1.52; p = 0.041) and being on the same regimen (aHR 1.37; 95% CI 1.07-1.76; p < 0.001). Cumulative appointment delay was not significantly associated with failure after controlling for covariates. CONCLUSIONS: VL monitoring is an important tool to improve programme outcomes, however limited coverage of VL testing and acting on test results hampers its full potential. In our cohort children and adolescents, PLHIV with history of loss-to-follow-up or those with low-viremia are at the highest risk of virological failure and might require more frequent virological monitoring than is currently recommended.

Resilience to maintain quality of intrapartum care in war torn Yemen: a retrospective pre-post study evaluating effects of changing birth volumes in a congested frontline hospital.

BACKGROUND: Fragile and conflict-affected states contribute with more than 60% of the global burden of maternal mortality. There is an alarming need for research exploring maternal health service access and quality and adaptive responses during armed conflict. Taiz Houbane Maternal and Child Health Hospital in Yemen was established during the war as such adaptive response. However, as number of births vastly exceeded the facility's pre-dimensioned capacity, a policy was implemented to restrict admissions. We here assess the restriction's effects on the quality of intrapartum care and birth outcomes. METHODS: A retrospective before and after study was conducted of all women giving birth in a high-volume month pre-restriction (August 2017; n = 1034) and a low-volume month post-restriction (November 2017; n = 436). Birth outcomes were assessed for all births (mode of birth, stillbirths, intra-facility neonatal deaths, and Apgar score < 7). Quality of intrapartum care was assessed by a criterion-based audit of all caesarean sections (n = 108 and n = 82) and of 250 randomly selected vaginal births in each month. RESULTS: Background characteristics of women were comparable between the months. Rates of labour inductions and caesarean sections increased significantly in the low-volume month (14% vs. 22% (relative risk (RR) 0.62, 95% confidence interval (CI) 0.45-0.87) and 11% vs. 19% (RR 0.55, 95% CI 0.42-0.71)). No other care or birth outcome indicators were significantly different. Structural and human resources remained constant throughout, despite differences in patient volume. CONCLUSIONS: Assumptions regarding quality of care in periods of high demand may be misguiding - resilience to maintain quality of care was strong. We recommend health actors to closely monitor changes in quality of care when implementing resource changes; to enable safe care during birth for as many women as possible.

Mortality and health survey, Walikale, Democratic Republic of the Congo, 2017: an example of the use of survey data for humanitarian program planning.

BACKGROUND: During humanitarian crises, health information systems are often lacking and surveys are a valuable tool to assess the health needs of affected populations. In 2013, a mortality and health survey undertaken by Médecins Sans Frontières (MSF) in the conflict affected Walikale territory of North Kivu, Democratic Republic of the Congo (DRC), indicated mortality rates exceeding humanitarian crisis thresholds and a high burden of mortality and morbidity due to malaria. In late 2017, after a period of relative stability, MSF reassessed the health status of the population through a second survey to guide ongoing operations. METHODS: A two-stage cluster survey, selecting villages using probability proportional to size and households using random walk procedures, was conducted. Household members were interviewed on morbidity and mortality, healthcare use, vaccination status, and bednet availability. RESULTS: The sample included 5711 persons in 794 households. The crude mortality rate (CMR) and under-five mortality rate (U5MR) were 0.98 per 10,000 persons/day (95% confidence interval (CI) 0.78-1.2) and 1.3 per 10,000 persons/day (95% CI): 0.82-2.0), respectively. The most frequently reported causes of death were fever/malaria (31%), diarrhoea (15%) and respiratory infections (8%). In 89% of households at least one person was reported as falling ill in the previous 2 weeks, and 58% sought healthcare. Cost was the main barrier amongst 58% of those who did not seek healthcare. Coverage of measles-containing-vaccine was 62% in under-fives. Sufficient bednet coverage (1 bednet/2 people) was reported from 17% of households. CONCLUSION: The second survey illustrates that although mortality is now just below crisis thresholds, the area still experiences excess mortality and has substantial health needs. The study results have supported the further expansion of integrated community case management to improve access to care for malaria, diarrhoea and respiratory infections. Such surveys are important to orient operations to the health needs of the population being served and also highlight the ongoing vulnerability of populations after humanitarian crises.

'When Ebola enters a home, a family, a community': A qualitative study of population perspectives on Ebola control measures in rural and urban areas of Sierra Leone.

BACKGROUND: During the West Africa Ebola outbreak, cultural practices have been described as hindering response efforts. The acceptance of control measures improved during the outbreak, but little is known about how and why this occurred. We conducted a qualitative study in two administrative districts of Sierra Leone to understand Ebola survivor, community, and health worker perspectives on Ebola control measures. We aimed to gain an understanding of community interactions with the Ebola response to inform future intervention strategies. METHODOLOGY/PRINCIPAL FINDINGS: Participants (25 survivors, 24 community members, and 16 health workers) were recruited purposively. A flexible participatory method gathered data through field notes and in-depth, topic-led interviews. These were analysed thematically with NVivo10© by open coding, constant comparison, and the principles of grounded theory. The primary theme, 'when Ebola is real', centred on denial, knowledge, and acceptance. Ebola was denied until it was experienced or observed first-hand and thus health promotion was more effective if undertaken by those directly exposed to Ebola rather than by mass media communication. Factors that enabled acceptance and engagement with control measures included: access to good, proximate care and prevention activities; seeing that people can survive infection; and the co-option of trusted or influential local leadership, with bylaws implemented by community leaders being strongly respected. All participants noted that dignity, respect, and compassion were key components of effective control measures. CONCLUSIONS: Successful control approaches need strong community leadership, with the aim of achieving collective understanding between communities and health workers. Health promotion for communities at risk is best conducted through people who have had close interaction with or who have survived Ebola as opposed to reliance on broad mass communication strategies.

The factors affecting household transmission dynamics and community compliance with Ebola control measures: a mixed-methods study in a rural village in Sierra Leone.

BACKGROUND: Little is understood of Ebola virus disease (EVD) transmission dynamics and community compliance with control measures over time. Understanding these interactions is essential if interventions are to be effective in future outbreaks. We conducted a mixed-methods study to explore these factors in a rural village that experienced sustained EVD transmission in Kailahun District, Sierra Leone. METHODS: We reconstructed transmission dynamics using a cross-sectional survey conducted in April 2015, and cross-referenced our results with surveillance, burial, and Ebola Management Centre (EMC) data. Factors associated with EVD transmission were assessed with Cox proportional hazards regression. Following the survey, qualitative semi-structured interviews explored views of community informants and households. RESULTS: All households (n = 240; 1161 individuals) participated in the survey. 29 of 31 EVD probable/confirmed cases died (93·5% case fatality rate); six deaths (20·6%) had been missed by other surveillance systems. Transmission over five generations lasted 16 weeks. Although most households had ≤5 members there was a significant increase in risk of Ebola in households with > 5 members. Risk of EVD was also associated with older age. Cases were spatially clustered; all occurred in 15 households. EVD transmission was better understood when the community experience started to concord with public health messages being given. Perceptions of contact tracing changed from invading privacy and selling people to ensuring community safety. Burials in plastic bags, without female attendants or prayer, were perceived as dishonourable. Further reasons for low compliance were low EMC survival rates, family perceptions of a moral duty to provide care to relatives, poor communication with the EMC, and loss of livelihoods due to quarantine. Compliance with response measures increased only after the second generation, coinciding with the implementation of restrictive by-laws, return of the first survivor, reduced contact with dead bodies, and admission of patients to the EMC. CONCLUSIONS: Transmission occurred primarily in a few large households, with prolonged transmission and a high death toll. Return of a survivor to the village and more effective implementation of control strategies coincided with increased compliance to control measures, with few subsequent cases. We propose key recommendations for management of EVD outbreaks based on this experience.

Improving mapping for Ebola response through mobilising a local community with self-owned smartphones: Tonkolili District, Sierra Leone, January 2015.

BACKGROUND: During the 2014-16 Ebola virus disease (EVD) outbreak, the Magburaka Ebola Management Centre (EMC) operated by Médecins Sans Frontières (MSF) in Tonkolili District, Sierra Leone, identified that available district maps lacked up-to-date village information to facilitate timely implementation of EVD control strategies. In January 2015, we undertook a survey in chiefdoms within the MSF EMC catchment area to collect mapping and village data. We explore the feasibility and cost to mobilise a local community for this survey, describe validation against existing mapping sources and use of the data to prioritise areas for interventions, and lessons learned. METHODS: We recruited local people with self-owned Android smartphones installed with open-source survey software (OpenDataKit (ODK)) and open-source navigation software (OpenStreetMap Automated Navigation Directions (OsmAnd)). Surveyors were paired with local motorbike drivers to travel to eligible villages. The collected mapping data were validated by checking for duplication and comparing the village names against a pre-existing village name and location list using a geographic distance and text string-matching algorithm. RESULTS: The survey teams gained sufficient familiarity with the ODK and OsmAnd software within 1-2 hours. Nine chiefdoms in Tonkolili District and three in Bombali District were surveyed within two weeks. Following de-duplication, the surveyors collected data from 891 villages with an estimated 127,021 households. The overall survey cost was €3,395; €3.80 per village surveyed. The MSF GIS team (MSF-OCG) created improved maps for the MSF Magburaka EMC team which were used to support surveillance, investigation of suspect EVD cases, hygiene-kit distribution and EVD survivor support. We shared the mapping data with OpenStreetMap, the local Ministry of Health and Sanitation and Sierra Leone District and National Ebola Response Centres. CONCLUSIONS: Involving local community and using accessible technology allowed rapid implementation, at moderate cost, of a survey to collect geographic and essential village information, and creation of updated maps. These methods could be used for future emergencies to facilitate response.

Ebola management centre proximity associated with reduced delays of healthcare of Ebola Virus Disease (EVD) patients, Tonkolili, Sierra Leone, 2014-15.

Between August-December 2014, Ebola Virus Disease (EVD) patients from Tonkolili District were referred for care to two Médecins Sans Frontières (MSF) Ebola Management Centres (EMCs) outside the district (distant EMCs). In December 2014, MSF opened an EMC in Tonkolili District (district EMC). We examined the effect of opening a district-based EMC on time to admission and number of suspect cases dead on arrival (DOA), and identified factors associated with fatality in EVD patients, residents in Tonkolili District. Residents of Tonkolili district who presented between 12 September 2014 and 23 February 2015 to the district EMC and the two distant EMCs were identified from EMC line-lists. EVD cases were confirmed by a positive Ebola PCR test. We calculated time to admission since the onset of symptoms, case-fatality and adjusted Risk Ratios (aRR) using Binomial regression. Of 249 confirmed Ebola cases, 206 (83%) were admitted to the distant EMCs and 43 (17%) to the district EMC. Of them 110 (45%) have died. Confirmed cases dead on arrival (n = 10) were observed only in the distant EMCs. The median time from symptom onset to admission was 6 days (IQR 4,8) in distant EMCs and 3 days (IQR 2,7) in the district EMC (p<0.001). Cases were 2.0 (95%CI 1.4-2.9) times more likely to have delayed admission (>3 days after symptom onset) in the distant compared with the district EMC, but were less likely (aRR = 0.8; 95%CI 0.6-1.0) to have a high viral load (cycle threshold ≤22). A fatal outcome was associated with a high viral load (aRR 2.6; 95%CI 1.8-3.6) and vomiting at first presentation (aRR 1.4; 95%CI 1.0-2.0). The opening of a district EMC was associated with earlier admission of cases to appropriate care facilities, an essential component of reducing EVD transmission. High viral load and vomiting at admission predicted fatality. Healthcare providers should consider the location of EMCs to ensure equitable access during Ebola outbreaks.

Successful Control of Ebola Virus Disease: Analysis of Service Based Data from Rural Sierra Leone.

INTRODUCTION: The scale and geographical distribution of the current outbreak in West Africa raised doubts as to the effectiveness of established methods of control. Ebola Virus Disease (EVD) was first detected in Sierra Leone in May 2014 in Kailahun district. Despite high case numbers elsewhere in the country, transmission was eliminated in the district by December 2014. We describe interventions underpinning successful EVD control in Kailahun and implications for EVD control in other areas. METHODS: Internal service data and published reports from response agencies were analysed to describe the structure and type of response activities, EVD case numbers and epidemic characteristics. This included daily national situation reports and District-level data and reports of the Sierra Leone Ministry of Health and Sanitation, and Médecins Sans Frontières (MSF) patient data and internal epidemiological reports. We used EVD case definitions provided by the World Health Organisation over the course of the outbreak. Characteristics assessed included level of response activities and epidemiological features such as reported exposure (funeral-related or not), time interval between onset of illness and admission to the EVD Management Centre (EMC), work-related exposures (health worker or not) and mortality. We compared these characteristics between two time periods--June to July (the early period of response), and August to December (when coverage and quality of response had improved). A stochastic model was used to predict case numbers per generation with different numbers of beds and a varying percentage of community cases detected. RESULTS: There were 652 probable/confirmed EVD cases from June-December 2014 in Kailahun. An EMC providing patient care opened in June. By August 2014 an integrated detection, treatment, and prevention strategy was in place across the district catchment zone. From June-July to August-December 2014 surveillance and contact tracing staff increased from 1.0 to 8.8 per confirmed EVD case, EMC capacity increased from 32 to 100 beds, the number of burial teams doubled, and health promotion activities increased in coverage. These improvements in response were associated with the following changes between the same periods: the proportion of confirmed/probable cases admitted to the EMC increased from 35% to 83% (χ(2) p-value<0·001), the proportion of confirmed patients admitted to the EMC <3 days of symptom onset increased from 19% to 37% (χ(2) p-value <0·001), and reported funeral contact in those admitted decreased from 33% to 16% (χ(2) p-value <0·001). Mathematical modelling confirmed the importance of both patient management capacity and surveillance and contact tracing for EVD control. DISCUSSION: Our findings demonstrate that control of EVD can be achieved using established interventions based on identification and appropriate management of those who are at risk of and develop EVD, including in the context of ongoing transmission in surrounding regions. Key attributes in achieving control were sufficient patient care capacity (including admission to specialist facilities of suspect and probable cases for assessment), integrated with adequate staffing and resourcing of community-based case detection and prevention activities. The response structure and coverage targets we present are of value in informing effective control in current and future EVD outbreaks.

Electronic medical records in humanitarian emergencies - the development of an Ebola clinical information and patient management system.

By November 2015, the West Africa Ebola epidemic had caused 28598 infections and 11299 deaths in the three countries most affected. The outbreak required rapid innovation and adaptation. Médecins sans Frontières (MSF) scaled up its usual 20-30 bed Ebola management centres (EMCs) to 100-300 beds with over 300 workers in some settings. This brought challenges in patient and clinical data management resulting from the difficulties of working safely with high numbers of Ebola patients. We describe a project MSF established with software developers and the Google Social Impact Team to develop context-adapted tools to address the challenges of recording Ebola clinical information. We share the outcomes and key lessons learned in innovating rapidly under pressure in difficult environmental conditions. Information on adoption, maintenance, and data quality was gathered through review of project documentation, discussions with field staff and key project stakeholders, and analysis of tablet data. In March 2015, a full prototype was deployed in Magburaka EMC, Sierra Leone. Inpatient data were captured on 204 clinical interactions with 34 patients from 5 March until 10 April 2015. 85 record "pairs" for 32 patients with 26 data items (temperature and symptoms) per pair were analysed. The average agreement between sources was 85%, ranging from 69% to 95% for individual variables. The time taken to deliver the product was more than that anticipated by MSF (7 months versus 6 weeks). Deployment of the tablet coincided with a dramatic drop in patient numbers and thus had little impact on patient care. We have identified lessons specific to humanitarian-technology collaborative projects and propose a framework for emergency humanitarian innovation. Time and effort is required to bridge differences in organisational culture between the technology and humanitarian worlds. This investment is essential for establishing a shared vision on deliverables, urgency, and ownership of product.