Estimating the number of secondary Ebola cases resulting from an unsafe burial and risk factors for transmission during the West Africa Ebola epidemic

Abstract

Background: Safely burying Ebola infected individuals is acknowledged to be important for controlling Ebola epidemics and was a major component of the 2013-2016 West Africa Ebola response. Yet, in order to understand the impact of safe burial programs it is necessary to elucidate the role of unsafe burials in sustaining chains of Ebola transmission and how the risk posed by activities surrounding unsafe burials, including care provided at home prior to death, vary with human behavior and geography.

Methodology/principal findings: Interviews with next of kin and community members were carried out for unsafe burials in Sierra Leone, Liberia and Guinea, in six districts where the Red Cross was responsible for safe and dignified burials (SDB). Districts were randomly selected from a district-specific sampling frame comprised of villages and neighborhoods that had experienced cases of Ebola. An average of 2.58 secondary cases were potentially generated per unsafe burial and varied by district (range: 0-20). Contact before and after death was reported for 142 (46%) contacts. Caregivers of a primary case were 2.63 to 5.92 times more likely to become EVD infected compared to those with post-mortem contact only. Using these estimates, the Red Cross SDB program potentially averted between 1,411 and 10,452 secondary EVD cases, reducing the epidemic by 4.9% to 36.5%.

Conclusions/significance: SDB is a fundamental control measure that limits community transmission of Ebola; however, for those individuals having contact before and after death, it was impossible to ascertain the exposure that caused their infection. The number of infections prevented through SDB is significant, yet greater impact would be achieved by early hospitalization of the primary case during acute illness.

Fig 1. Estimates of risk factors among contacts of a primary case, in contributing to the chance of becoming a secondary case.

Circles show the estimate for the odds ratio based on Fisher’s exact test. Vertical lines enclose the 95% confidence interval surrounding the estimate. In calculating the odds ratio for caregivers in the baseline estimate (where caregivers are normally excluded), the inclusion criteria for the baseline dataset were amended to include anyone who was EVD positive, instead of the standard baseline criteria of EVD positives who were not caregivers, since the standard baseline criteria would lead to no cases who were caregivers, by definition.

Fig 2. Estimates and uncertainty in the number of secondary cases caused by an unsafe burial, pooled across districts, for baseline and ceiling estimates.

Thick horizontal lines show median estimate, boxes enclose the interquartile range in the estimate (between the 25^th and 75^th percentile) across primary cases and districts. Capped vertical lines encompass the total range of variability in the estimate.

Fig 3. Estimates and uncertainty in the number of secondary cases caused by an unsafe burial, by district.

Boxes enclose the interquartile range of each estimate (extending from the 25th percentile to the 75th percentile). The thick horizontal lines within each box correspond to the median. Capped vertical lines encompass the total range of variability in the estimate.

Fig 4. Reported cases in the WHO patient database and burials over time in Sierra Leone (left), Liberia (center) and Guinea (right).

Colored dots show reported cases from the WHO patient database. Dotted lines show the total number of recorded burials carried out. Solid lines show estimates of the number of burials of EVD casualties, reconstructed from laboratory testing of swab samples collected from community deaths.