Int. J. Epidemiol.

Advance Access published November 20, 2015

International Journal of Epidemiology, 2015, 1–15

doi: 10.1093/ije/dyv307
Original article

Original article

Risk factors for transmission of Ebola or

Marburg virus disease: a systematic review and
meta-analysis
Julii Brainard,1 Lee Hooper,1 Katherine Pond,2 Kelly Edmunds3 and
Paul R. Hunter1*

Downloaded from http://ije.oxfordjournals.org/ by guest on December 9, 2015

1
Norwich Medical School, University of East Anglia, Norwich, UK, 2Robens Centre for Public and
Environmental Health, University of Surrey, Guildford, UK and 3School of Biological Sciences,
University of East Anglia, Norwich, UK
*Corresponding author. Norwich Medical School, University of East Anglia, Norwich NR4 7TJ, UK.
E-mail: Paul.Hunter@uea.ac.uk
Accepted 23 October 2015

Abstract
Background: The Ebola virus disease outbreak that started in Western Africa in 2013 was
unprecedented because it spread within densely populated urban environments and
affected many thousands of people. As a result, previous advice and guidelines need to
be critically reviewed, especially with regard to transmission risks in different contexts.
Methods: Scientific and grey literature were searched for articles about any African filo-
virus. Articles were screened for information about transmission (prevalence or odds
ratios especially). Data were extracted from eligible articles and summarized narratively
with partial meta-analysis. Study quality was also evaluated.
Results: A total of 31 reports were selected from 6552 found in the initial search. Eight
papers gave numerical odds for contracting filovirus illness; 23 further articles provided
supporting anecdotal observations about how transmission probably occurred for indi-
viduals. Many forms of contact (conversation, sharing a meal, sharing a bed, direct or
indirect touching) were unlikely to result in disease transmission during incubation or
early illness. Among household contacts who reported directly touching a case, the
attack rate was 32% [95% confidence interval (CI) 26–38%]. Risk of disease transmission
between household members without direct contact was low (1%; 95% CI 0–5%). Caring
for a case in the community, especially until death, and participation in traditional funeral
rites were strongly associated with acquiring disease, probably due to a high degree of
direct physical contact with case or cadaver.
Conclusions: Transmission of filovirus is unlikely except through close contact, espe-
cially during the most severe stages of acute illness. More data are needed about the
context, intimacy and timing of contact required to raise the odds of disease transmis-
sion. Risk factors specific to urban settings may need to be determined.

C The Author 2015. Published by Oxford University Press on behalf of the International Epidemiological Association
V 1
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/),
which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact
journals.permissions@oup.com
2 International Journal of Epidemiology, 2015, Vol. 0, No. 0

Key words: Ebola virus disease, Marburg virus, risk factors, bodily fluids, systematic review

Key Messages
• Human-to-human transmission of filoviruses usually requires direct contact with a symptomatic individual.
• Transmission through indirect contact has been reported, but appears to be uncommon.
• There is a need for more primary epidemiological research in urban communities.
• During outbreaks, provision of appropriate care through designated specialist health facilities reduces transmission

rates.

Introduction Concerns have been raised that the shift in the 2013–15
The 2013–15 epidemic of Ebola virus disease (EVD) in epidemic towards infected patients being managed in

Downloaded from http://ije.oxfordjournals.org/ by guest on December 9, 2015

Western Africa is by far the largest and most widespread urban communities exacerbated disease transmission.7
outbreak of this disease to date and case numbers far Consequently there is a greater need to better understand
exceed the total from all previous EVD emergences. It is the mechanisms and risk factors behind intra-community
the first outbreak of the Zaire species of Ebola in this disease transmission. Only then can appropriate commu-
region and the first in urban high population density set- nity control measures be implemented. Although there are
tings where sustained transmission has occurred. In previ- some previous reviews on Ebola viruses and their epidemi-
ous outbreaks the main focus of attention was on ology,8,9 systematic evaluation of evidence on community
nosocomial transmission of the disease and on risks associ- human-to-human transmission risks has been limited.
ated with funeral practices. However, the occurrence of With the development of a highly effective vaccine against
cases in high population density urban environments raised Ebola,10 such rigorous evaluation will be essential to de-
concern about alternative transmission pathways. The size signing optimal strategies for immunization campaigns.
of the 2013–15 epidemic was often unmatched by suffi- In this systematic review we searched for all published
cient clinical capacity, which resulted in community-based evidence which identified and/or quantified the risk factors
care rather than hospitalization of cases.1 for community acquisition of filovirus infection. We also
Ebola virus is part of the Filoviridae family which also included papers where the authors expressed an opinion as
includes Marburg viruses. Both Ebola and Marburg dis- to how patients acquired infection, even if the evidence to
eases are generally understood to be zoonotic infections support this suggestion would not usually meet standards
whose primary hosts are thought to be bats.2–4 Once the of acceptable epidemiological evidence.
virus crosses from wildlife into humans, subsequent per-
son-to-person spread propagates the outbreak until it is
Methods
brought under control. Given the experiences of previous
human filovirus infections, the primary focus of interest Medline and Scopus were searched from inception through
has been in nosocomial spread and spread associated with 13 August 2015 using the search string filovir*.af. OR
funeral practices.5,6 The 2013–15 epidemic differs from ebola.af. OR ebolavir*.af OR Marburg-virus.af (af means
previous outbreaks not only in number of people afflicted ‘all fields’ including all text words and relevant indexing)
and geographical spread, but also in its setting. Many of without restrictions for date or language. Twelve sources
the reported cases have been among people living in high of grey literature were searched (see Appendix A1, avail-
density and impoverished urban environments. Indeed, able as Supplementary data at IJE online) and screened by
because of the lack of adequate health care facilities, many a single investigator. Included papers were also checked for
people remained in their home community during the en- further studies.
tire course of their illness, receiving care from family mem-
bers and neighbours. Hence, the scale of the Western
Africa outbreak resulted in many Ebola treatment centres Inclusion criteria
being built within or close to these newly-affected urban The preferred study for inclusion in this review provided
communities. data that enabled us to assess the odds of filoviral infection
International Journal of Epidemiology, 2015, Vol. 0, No. 0 3

transmission between humans according to particular Reporting Items for Systematic Reviews and Meta-
characteristics, behaviours or contacts (data could be pre- Analyses; a PRISMA checklist was included with
sented as odds ratios, risk ratios or raw numbers). In add- submission).16
ition, we also included papers where the authors expressed In order to assess the risk of household transmission, we
an opinion about how infection was acquired, although identified papers that presented incidence rates in house-
not based on analytical epidemiology (i.e. anecdotal obser- hold contacts (where possible) with and without direct
vations). Eligible filoviral infections were Marburg virus, contact with another case. A random-effects meta-analysis
TM
Ravn virus, Zaire, Sudan, Taı̈ Forest and Bundibugyo spe- of proportions was conducted using Stats Direct and het-
cies of Ebola. Species of filovirus not present in Africa or erogeneity checked visually. Separate meta-analyses were
not known to be dangerous to humans were excluded. The conducted of incidence rates in household contacts with
filovirus disease outbreak had to be laboratory confirmed and without a history of direct contact.
(using RT-PCR, NAAT or Vero culture tests), as antibodies
or inflammatory factors in body fluids were deemed inad-
equate by themselves to verify the outbreak, because they Results
are widespread in the regional population including in Of 6552 mostly unique articles found in Medline or
many people with no relevant clinical history.9,11–15 Data Scopus, 114 were immediately excluded for being obvious
were included from mixed patient groups where all or conference abstracts, protocols, news reports, commenta-

Downloaded from http://ije.oxfordjournals.org/ by guest on December 9, 2015

some had laboratory confirmation of disease cause, all had ries or editorials (Figure 1). A further 2001 items lacked an
a compatible clinical history and survivors were confirmed abstract and,on brief review, most of these appeared to be
as having filovirus antibodies. short commentaries, news summaries and possible confer-
Titles and abstracts were screened for inclusion by a ence presentations. Of the 2001 items, only nine were
single reviewer and verified by a second researcher. screened directly because their titles contained keywords
Conference presentations, protocols, news reports, com- most relevant to our research questions. Thus, 4560 scien-
mentaries and editorials were excluded. Where titles were tific articles were duplicate screened on title and abstract,
not accompanied by abstracts they were only assessed in of which 52 were not excluded. The grey literature search
full text if they included the word(s) risk or transmission. (see Supplementary Appendix A1) yielded two inclu-
Full-text articles were assessed for inclusion independently sions.17,18 One additional article with potential primary
in duplicate. Decision differences at all stages were data19 was identified in the discussion text of selected art-
resolved by discussion. Where several articles reported on icles; 55 articles were thus chosen for full-text review. Full
the same primary data, the articles were grouped to ensure text was unavailable for one article20 and, after full-text
data were not duplicated within the review. review, 23 articles did not meet eligibility criteria. Four art-
Data were extracted into tables and verified by a se- icles11,17,19,21 reported at least partly duplicated informa-
cond researcher. Extracted data included bibliographic tion on two patient groups (a list of outbreaks in the
details, viral species, date and place of outbreak, risk or selected articles to check for duplicated data is in
exposure factor(s) identified, assay methods and calcu- Supplementary Appendix A3, available at IJE online). The
lated odds, hazard or risk ratios or relevant raw data. final number of articles included in the final review was 31
Unadjusted and adjusted data (where available) were ex- covering 29 distinct patient groups, for which data extrac-
tracted and unadjusted odds ratios calculated from raw tion was undertaken. Study quality and validity assessment
data. Where anecdotal opinions of acquisition were pre- for all selected studies are shown in Supplementary
sented, these were also extracted into a separate list. Our Appendix A2.
study validity assessment was based on attributes most
likely to undermine the utility of the studies, including
delay in investigating the cause of disease transmission Characteristics of included studies
(for studies identified at low risk of bias this was  3 Most data (Table 1) came from retrospectively adminis-
months from contact), study aim (to quantify human-to- tered interviews with survivors or their close contacts or
human transmission or not), whether there was a stand- from clinical notes, using standardized questionnaires, and
ard methodology for data collection and whether the risk usually collected less than 3 months after illness.
factors assessed were pre-specified. A positive answer Community disease transmission occurred from 1967
suggested low risk of bias for all of these. Validity was as- to 2015 in 10, primarily African, countries (Angola,
sessed by a single researcher (see Appendix A2, available Democratic Republic of Congo (DRC)/Zaire, Guinea,
as Supplementary data at IJE online). The review is re- Liberia, Republic of Congo, Sierra Leone, South
ported in accordance with PRISMA guidelines (Preferred Africa, Sudan, Uganda, West Germany). Quantitative
4 International Journal of Epidemiology, 2015, Vol. 0, No. 0

Downloaded from http://ije.oxfordjournals.org/ by guest on December 9, 2015

Figure 1. Study selection procedure.

data were available for BUDV, EBOV, MARV and SUDV confirmed filovirus infection as the cause of disease in
species, but accounts for how people contracted disease each outbreak; but most studies included some
were overwhelmingly anecdotal for EBOV. Laboratory cases identified from clinical history and antibody presence
methods [culture or polymerase chain reaction (PCR) tests] only.
International Journal of Epidemiology, 2015, Vol. 0, No. 0 5

Table 1. Included study characteristics, table ordered by filovirus speciesand chronological date of relevant outbreaks

Species Outbreak date, location, authors Type of information relevant to this review

BUDV Aug–Dec 2007, Bundibugyo Uganda38 Delayed recognition; unconfirmed, risks of attending childbirth
Aug–Dec 2007, Bundibugyo Uganda27 Numerical risk ratio data, various attributes, OR
MARV 1967, Germany and Yugoslavia17,21 Documents transmission of disease from sexual contact
Feb–Mar 1975, Johannesburg South Africa85 Likely transmission moment ¼ handling wet paper tissues from
bereaved incubator
Mar–Jul 2005, Uige, Angola26 OR data
SUDV 31 Jul–6. Nov 1979, Nzara, Yambio, Sudan25 34 patients, concentration in blood, one OR þ anecdotal, during &
after illness
Aug 2000–Jan 2001, Gulu, Uganda22 PPRs, fomites suggested, many factors
Aug 2000–Jan 2001, Gulu, Uganda40 Children under 18 survive better, close contact risk
EBOV 1 Sep–24 Oct 1976, Bumba, Yambuku, Zaire11,19 ORs (also in Breman et al.) Non-intimate contact risk, touching dry
skin, sexual partners, attending childbirth or a funeral, intimate fu-
neral tasks, needle sharing, bedbugs, rats?
1976–77, Sud-Ubangi subregion, Zaire (Tandala)12 1981–85 surveillance report: direct contact implicated, asymptomatic,
antibody prevalence

Downloaded from http://ije.oxfordjournals.org/ by guest on December 9, 2015

Jan–Jul 1995, Kikwit DRC24 PPRs, not recognized until May 1995; households of 27 cases inter-
viewed 17 May–3 June about risk factors (no risk after 1 May);
stage of illness relevance
Apr–May 1995, Mosango DRC28 Related to Kikwit outbreak, 23 only in Mosango; forms of dangerous
contact
Jan-Jul 1995, Kikwit DRC23 Matched ORs
1994–96, Gabon29 Occupation and economic activity
2002–03, Rep. of Congo35 Cases linked to direct contact between people following primary con-
tacts with wildlife
2005, Etoumbi DRC41 Gender factors, funerals, cremation controversy
May–Nov 2007, Occidental Kasaı, DRC37 Suggests via sweat, dead animals
2014, Sierra Leone39 Transmission after caring for ill patient, organizing funeral, caring for
infant or attending during childbirth
2014, Sierra Leone patient taken to Germany36 Believed transmission in office or lavatory; high levels of virus de-
tected in sweat
2014, Sierra Leone34 Funerals, health care workerss affected; people who left clinic but had
EVD after all
2014–15, Sierra Leone32,45 Touching bodies at funerals, contact with or caring for patients,
touching cadavers
2014, Conakry, Guinea33 Reproduction numbers, chains of transmission
2014–15 Guinea18 and Liberia30,42–44 Care in community, funerals and cremation. Also, assistance into taxi
2015 Liberia31 Following sexual contact

Seven papers gave different forms of numerical odds or consequently have more confidence in these results than
risk ratios for developing disease.22–28 Two further art- others in our discussion.
icles11,19 gave data that enabled us to calculate unadjusted A further 23 articles are included in the review although
odds ratios for one outbreak. The available odds, risk and they provide only anecdotal or contact-tracing observa-
prevalence ratio data are summarized in Table 2. The ana- tions on how disease transmission probably occurred in
lysis of Dowell et al24 is unique because it broke down risk confirmed cases.12,17,18,21,29–45 Only one paper covering
of transmission by stage of illness at exposure (incubation the 2013–15 epidemic had a primary aim to quantify as-
period, early or late illness). Other sources do not have pects of human-to-human transmission,33 which publica-
detailed linkage to disease onset, but still can be used to tion calculated R0 numbers rather than identifying risk
support observations about overall trends of evidence. ratios for individual risk factors. Details of the information
Most of the numbers in Table 2 are crude odds ratios (not provided by this anecdotal group of articles are in
adjusted for confounding variables). Data from two Appendix A4 (available as Supplementary data at IJE
papers22,24 are mostly adjusted for co-variates, and we online).
6 International Journal of Epidemiology, 2015, Vol. 0, No. 0

Table 2. Numerical odds, risk or prevalence ratios for filovirus disease transmission

Risk factor Details Unadjusted effect size Adjusted effect size

(95% CI) (95% CI)

Demographics and personal attributes

Age Being > 18 years24 PRR* 6.8 PRR* 3.6 (1.3-10.1)a
Being > 30 years old22 PPR 1.38 (0.64-2.97)
Being 30 years old26 OR 1.32 (0.60-2.92)
Being 34 years old29 OR 0.83 (0.35-1.95)
Being 41–60 years old27 OR 2.0 (0.8–4.9) Not reportedb
Being 40 years old26 OR 0.99 (0.37-2.68)
Sex Being female27 OR 0.63 (0.28–1.43) Not reportedb
Being female22 PPR 1.54 (0.7-3.6)
Being female24 PRR* 2.1 PRR* 1.0 (0.5-2.1)a
Being female26 OR 2.46 (1.03 – 5.90)
Occupation Working in forest23 MOR 1.3 (0.4-6.0)
Fishing23 MOR 3.0 (0.04-235)
Fisherman29 OR 3.12 (0.59-16.41)

Downloaded from http://ije.oxfordjournals.org/ by guest on December 9, 2015

Health care worker23 MOR 9.0 (1.6-91.2)
Health care worker26 OR 1.52 (0.41-5.64)
Student26 OR 0.81 (0.34-1.94)
Housewife26 OR 1.23 (0.50-3.04)
Housewife29 OR 0.87 (0.24-3.09)
Farmer29 OR 1.27 (0.15 -10.81)
Trader29 OR 0.77 (0.22 -2.75)
Gold-panner29 OR 1.33 (0.56-3.17)
Setting Urban or suburban (versus rural)26 OR 0.82 (0.23-2.89)
Recent travel To areas with known cases27 OR 1.4 (0.5–3.8) Not reportedb
Outside own local area23 MOR 3.0 (0.2-41.4)
Recurring non-intimate contact
Commerce-related Frequenting markets23 MOR 1.1 (0.3-4.5)
Conversation with case During incubation period24 PRR* 1.5 PRR* 0.7 (0.2-3.0)a
During early illness24 PRR* 3.3 PRR* 0.7 (0.3-2.0)a
During late illness24 PRR* 10.6 PRR* 3.9 (1.2-12.2)a
Washing clothes of a case (Point of disease onset unclear)22 PPR 1.68 (0.78-3.60) PPR 1.02 (0.47-2.2)d
Indirect contact with case Household or similar contact without OR 6.88 (1.35-35.1)
direct physical touching26
Sharing same hut Without sharing bed/sleeping mat22 PPR 2.16 (0.90-5.19) PPR 2.34 (1.13-4.8)d
Entered same room but no physical contact25 OR 0.06 (0.00-1.06)
Slept in same room11 OR 1.65 (0.95-2.85)
Visiting cases In hospital or their own home, before or after OR 8.7 (3.0–26.3) Not reportedb
diagnosis27
Visit to ill (with fever and bleeding) friend MOR 10.6 (3.8-36.3)
(in own home)23
Recurring intimate contact
Shared a meal During early illness24 PRR* 2.5 PRR* 1.2 (0.5-2.7)a
During late illness24 PRR* 7.0 PRR* 2.2 (1.2-4.0)a
With index patient22 PPR 1.94 (0.89-4.22) PPR 1.69 (1.0-2.8)d
Sharing a bed or sleeping During incubation24 PRR* 2.9 PRR* 1.4 (0.8-2.4)a
mat During early illness24 PRR* 3.8 PRR* 1.3 (0.7-2.5)a
During late illness24 PRR* 7.4 PRR* 2.2 (1.2-4.2)a
Point of disease onset unclear22 PPR 2.78 (1.15-6.70) PPR 2.93 (1.2-7.4)d
Direct physical contact – During incubation period24 PRR* 2.9 PRR* 0.8 (0.4-1.8)a
touching During early illness24 PRR* 12.5
During late illness24 PRR* 12.5

continued
International Journal of Epidemiology, 2015, Vol. 0, No. 0 7

Table 2. Continued
Risk factor Details Unadjusted effect size Adjusted effect size
(95% CI) (95% CI)

With person who had fever or bleeding, at work or MOR 24.0 (3.2-1065)
in the market23
Contact with body or body fluids of a suspected OR 11.0 (2.6-46.1)
case26
Touched case11 OR 1.45 (0.73-2.87)
Touching during illness22 PPR 3.53 (0.52-24.11) PPR 1.56 (0.2-13.0)c
Touching but no nursing care25 OR 0.40 (0.11-1.45)
Contact with body fluids Contact with body fluids22 PPR 5.30 (2.14-13.14) PPR 4.61 (1.7-12.3)c
Direct contact with individuals potentially infected OR 12.0 (3.6-39.6)
with MHF or their bodily fluids or direct contact
during funeral26
Body fluid contact in early illness24 PRR* 6.1
Body fluid contact in late illness24 PRR* 5.9
Likely sexual contact Being spouse of index case24 PRR* 3.8 PRR* 1.3 (0.7-2.5)a
Caring for patient Nursing a patient25 OR 8.9 (3.1-25.4)

Downloaded from http://ije.oxfordjournals.org/ by guest on December 9, 2015

Cared for case11 OR 0.99 (0.56-1.76)
Early care at home, not until death22 PPR 6 (1.3-27.1) P for trend for these
At hospital until death22 PPR 8.57 (1.9-37.7) 3 < 0.001
In home until death22 PPR 13.33 (3.2-55.6)
Aided patient in childbirth11 OR 2.46 (1.02-5.92)
Funeral-related activities
Viewed body Without touching24 PRR* 4.8 PRR* 1.6 (0.5-4.9)a
Attended Special (pre-funeral) rituals23 MOR 0.8 (0.2-3.2)
Funeral itself23 MOR 3.0 (1.2-7.6)
Funeral itself11 OR 0.86 (0.41-1.79)
Communal meal As part of funeral event22 PPR 2.84 (1.35-5.98) PPR 1.5 (0.98-2.28)d
Touched body Before or during funeral22 PPR 1.95 (0.91-4.17) PPR 1.84 (0.95-3.55)c
Before or during ceremony24 PRR* 4.9 PRR* 2.1 (1.1-4.2)a
Ritual handwashing22 PPR 2.25 (1.08-4.72) PPR 1.16 (0.54-2.49)d
Washing and dressing body27 OR 7.4 (2.9–19.3) OR 3.83 (1.78-8.23)b
Direct contact with corpse, its body fluids or soiled OR 38.5 (4.2-352.1)
items26
Prepared for burial23 MOR 13.1 (1.4-631)
Prepared cadaver OR 1.07 (0.63-1.82)
Previous use of health services (nocosomial indicators)
Taking regular medication Kikwit outbreak 199523 MOR 2.0 (0.5-9.8)
Admitted previously to hos- Before outbreak was recognized, Kikwit outbreak, MOR 9.9 (3.1-41.0)
pital for something else 199523
Received injection Before outbreak was recognized, Kikwit, 199523 MOR 30.0 (4.3-1302)
Admission or visit to For any reason27 OR 8.7 (3.0-26.3) Not reportedb
hospital
Number of types of direct No direct contact22 PPR 1.0 P for trend for these
contact (touching ill pa- One type of contact22 PPR 0.18 (0.01-2.45) 4 < 0.001
tient, touching dead Two types of contact22 PPR 1.94 (0.30-12.44)
body, touching body Three types of contact22 PPR 4.00 (0.64-25.02)
fluid)

PRR*, prevalence rate ratio; PPR, prevalence proportion ratio; MOR, matched odds ratio; OR, odds ratio. Bold text indicates a 95% confidence interval that
is entirely above 1.0. Otherwise, where figures are missing, figures were not provided or not possible to calculate.
a
Adjusted for direct physical contact during illness and contacts with body fluids.
b
Dropped from multivariate logistic regression by authors due to lack of significance at P < 0.05.
c
Using multivariate log-binomial regression models, factors included touching patient during illness, touching dead body and contact with patient fluids.
d
Using multivariate log-binomial regression models, factors included shared meals, washed clothes, slept in same hut or mat, ritual handwashing during funeral
and communal meal during funeral, and also controlled for intensity of contacts (two or more indirect contacts versus less).
8 International Journal of Epidemiology, 2015, Vol. 0, No. 0

Demographic variables infection.24 Washing clothes of a case yielded adj. PPR

Outbreak reports often state40,46,47 that, compared with 1.02 (95% CI 0.47–2.2).22 Disease transmission to people
other age groups, children were less frequently infected who had regular but non-intimate contact with cases
and more frequently survived; but this is not consistently (household or workplace) had quite variable association
confirmed by statistical analysis on EBOV.24,48 There is with disease transmission (OR range from 0.06, 95% CI 0-
mixed evidence (minimum OR 0.83, 95% CI 0.35–1.95, to 1.06,25 to OR 6.88, 95% CI 1.35–35.126).
maximum prevalence rate ratio 6.8, no 95% CI Disease contraction from casual contact with sweat has
stated),22,24,27 as to whether increased age among adults is been suggested,37 although most studies fail to detect virus
a risk factor for contracting filovirus illness. Gender- in sweat.36,51–55 Shared office or lavatory facilities was
related risk ratios ranged from OR 0.63 (95% CI 0.28– implicated by Kreuels et al36 but was not the only possible
1.43) to OR 2.46 (95% CI 1.03–5.90).22,26,27,47 Recent opportunity for the relevant case. Sexual partners of con-
travel yielded OR 1.4 (95% CI 0.5–3.8) or matched odds valescents are potentially at chronic but low risk of getting
ratio 3.0 (95% CI 0.2–41.4).23,27 Although high preva- disease,11,56 due to persistence of detectable virus for many
lence of Marburg virus antibodies has been reported weeks in semen and vaginal excretions57 of convalescents.
among forest workers49 and miners,29,50 these occupations Filovirus transmission from sexual contact with a convales-
(or being a housewife) were not associated with raised risk cent has been documented only twice since 1967.17,31
Being a spouse of a victim24 was not shown to be risky as

Downloaded from http://ije.oxfordjournals.org/ by guest on December 9, 2015

of contracting disease (ORs range from 0.87, 95% CI
0.24–3.09, to 3.12, 95% CI 0.59–16.41).29 it had a 95% CI for prevalence rate ratio between 0.7 and
2.5 when adjusted for other factors.
During later stages of illness (when bodily fluids with
Intensity of contact high viral loads are mostly likely to be shed), even rela-
Within Table 2, contact was put in order of (approxi- tively non-intimate contact posed some remaining risks
mately) increasing intimacy: non-intimate or intimate with even after adjustment for direct contact (e.g. conversa-
live cases, or with cadavers. It is notable that even within tion, PRR 3.9, 95% CI 1.2–12.2).24 During all active dis-
the same categories, there was often substantial heterogen- ease stages, sharing a meal had PPR 1.69 (95% CI 1.0–
eity in reported risk ratios, making it difficult to provide 2.8) and sharing a bed had PPR 2.93 (95% CI 1.2–7.4).22
single estimates of risk. This difficulty is compounded by During late disease, sharing a meal had adjusted PPR 2.2
the fact that many studies did not adequately control for (95% CI 1.2–4.0) and sharing a bed had adjusted PRR
confounding. In the discussion below, stated risk ratios are 2.2 (95% CI 1.2–4.2).24 Touching feverish people had
unadjusted for any confounding factors except where MOR 24 (95% CI 3.2–1065)23 and touching bodily fluids
noted. had OR 11 (95% CI 2.6–46.1).26 That disease often fol-
Only one article24 provided data for prevalence of risk lows directly touching body fluids is corroborated by
ratios according to stage of infection (incubation, early or much anecdotal evidence.18,21–23,26,28,44) Visiting known
late illness; late illness was defined as post-hospitalization). patients in their homes raised risk of acquiring disease
The adjusted relative risks associated with sharing a bed (MOR 10.6, 95% CI 3.8–36.3).23 Attending the birth of
with a case during the incubation period was PRR 1.4 a child to an ill patient conferred OR 2.46 (95% CI 1.02–
(95% CI 0.8–2.4) and during early illness PRR 1.3 (95% 5.92)11,19 for contracting disease, and transmission from
CI 0.7–2.5), but bed-sharing in later illness yielded a much attendant to mother-in-labour has also been implicated.39
higher risk ratio (PRR 2.2, 95% CI 1.2–4.2).24 Such Other forms of direct physical contact with ill patients
chronological data were missing from other reports, which were identified as being associated with a high risk of
only gave combined risks from contact at all stages of incu- transmission.12,35,38,39 However, the greatest risks are
bation or illness. associated with caring for an actively ill patient (min-
Faye et al.33 observed that 72% of transmission was be- imum OR 0.99, 95% CI 0.56–1.76, and maximum OR
tween family members. These and other data strongly sug- 8.9, 95% CI 3.1–25.4)22,25 or visiting patients receiving
gest that non-intimate contact such as frequenting markets care (OR 8.7, 95% CI 3.0–26.3 and MOR 10.6, 95% CI
(MOR 1.1, 95% CI 0.3–4.5),23 conversation with a case in 3.8–36.3).23,27 Caring for patients at home until death
early illness [adjusted (adj.) PRR 0.7, 0.3–2.0]24 or sharing carries very high risk of disease transmission (OR 13.33,
a home but not sleeping space11 had low risk of transmis- 95% CI 3.2–55.6).22,30
sion before and during early illness. Even touching dry skin The quantitative and adjusted data presented in Table 2
(adj. PRR, 0.8 95% CI 0.4–1.8), sharing meals (adj. PRR suggest that in themselves neither viewing a body(adj.
1.2, 0.5–2.7) or a bed (adj. PRR 1.3, 0.7–2.5) during early PRR 1.6, 95% CI 0.5–4.9)24 nor attending a funeral
disease stages lacked a significant risk of contracting acute (MOR ranges from 0.8, 95% CI 0.2–3.2, to 3.0, 95% CI
International Journal of Epidemiology, 2015, Vol. 0, No. 0 9

Downloaded from http://ije.oxfordjournals.org/ by guest on December 9, 2015

Figure 2. Attack rates without direct contact between household members.

Figure 3. Attack rates after direct contact between household members.

1.2–7.6)11,19,23,33 pose elevated risks, but rather that it is consistently associated with increased risk (range of adj.
the nature of traditional funeral-associated activities that PPR 1.16, 95% CI 0.54-2.49, to adj. PRR 2.1, 95% CI
increases risk. One anecdotal report18 found that among 1.1–4.2).22,24 Sharing a communal meal during the funeral
cases that arose after a specific (traditional) funeral cere- was found to pose an elevated risk (adj. PPR 1.5, 95%
mony, 21% of cases had directly touched the cadaver and CI ¼ 0.98–2.28),22 perhaps due to crowded conditions.
the other 79% of linked cases had physical contact with The intimate tasks (washing and dressing) associated with
those who touched the body either during the service or preparing a body for funeral and burial tend to confer a
afterwards. However, briefly touching the cadaver is not very high risk of disease transmission, although again data
10 International Journal of Epidemiology, 2015, Vol. 0, No. 0

are inconsistent (range from OR 1.07, 95% CI 0.63–1.82, person; or (iii) when preparing the recently deceased for
to MOR 13.1, 95% CI 1.4–631).11,19,23,26,27,30,41 burial. These findings are not surprising, and our review
has strengthened the evidence base for them. Importantly,
we provide a more nuanced understanding of the risks, es-
Contact with health services and other possible pecially around risks associated with indirect contact. For
risks example, we have found no evidence of risk associated
Elevated transmission risk following contact with health with casual contact with asymptomatic individuals outside
services appears in Table 2 and is also recounted in the home. Even between household contacts who did not
Supplementary Appendix A4, particularly for the 1995 have direct physical contact, the risk of disease transmis-
Kikwit outbreak in the DRC.23 Transmission rates associ- sion is relatively minor (1%, CI ¼ 0-5%), although not
ated with health service contact were high in these out- zero. We have also confirmed that there is negligible if any
breaks, partly because the disease was recognized belatedly risk of contracting disease during the incubation period
with subsequent delayed implementation of control meas- and only low risk in the first week of symptomatic illness.
ures.33,58 Unsafe needle-sharing practices as part of a vac- Our review also confirms the high risk of transmission
cination programme were also specifically blamed for associated with funerals, and that disease transmission
rapid disease spread during the 1976 emergence of appears to most often follow after touching the body of a
EBOV.11,19 Visiting a hospital or caring for filovirus cases case.

Downloaded from http://ije.oxfordjournals.org/ by guest on December 9, 2015

in hospital raised transmission risks dramatically in most Visiting or caring for filovirus cases in hospital raised
outbreaks (OR 8.7, 95% CI 3–26.3, and MOR 9.9, 95% transmission risks dramatically across most outbreaks.22–
CI 3.1–41.0).22–25,27 25,27
This is probably due to high viral loads associated
with severe disease and insufficient protection measures. It
has been suggested33 that earlier hospitalization and longer
Meta-analysis hospital stays (provided sufficient isolation and protection
Three papers gave sufficient data to calculate incidence procedures are followed) could significantly shorten the
rates in household contacts with and without a history of duration of filovirus outbreaks.
direct contact.22,24,25 The proportion meta-analyses Although this review concerned person-to-person trans-
(Figures 2 and 3) drew on a total of 254 direct contacts mission in the community, it was not always possible to
and 135 indirect contacts. Among those household con- distinguish primary from secondary cases. We excluded
tacts reporting direct contact with a case, the attack rate reports where disease clearly resulted from wildlife con-
was 32% (95% CI 26–38%) without important heterogen- tacts, but some risk ratios were undoubtedly calculated for
eity. In household contacts with no history of direct con- mixed groups of primary cases (disease acquired from
tact, the attack rate was 1% (95% CI 0–5%). Only one wildlife or wildlife environments) and secondary cases
confirmed case with no direct contact in any of the three (from humans). The ecological niche that each filovirus
studies was reported.22 This person ‘slept wrapped up in a occupies would certainly impact on some risk factors for
blanket left by his brother’, recently deceased from EVD. It primary infection59 (such as occupation). However, the
was possible that this was a case of transmission from dir- ratio of secondary to primary cases in most outbreaks is so
ect contact with body fluids. Breman et al.19 also reported large9 that ecological factors responsible for primary dis-
data sufficient to calculate attack rates in household con- ease acquisition are unlikely to have a major influence on
tacts, but without distinction between direct and indirect calculated risk ratios.
contact. Instead, they reported that the attack rate was Why adulthood would increase the risk of illness is un-
27% between spouses, brothers and sisters or between par- certain, although the risk of illness does not appear to
ents and children. In all other relatives the attack rate was depend on lower total viral load, and contradicts intuition
only 8.0%. that small children would tend to have greater risks due to
more physical contact with their carers.60 The association
with adulthood may be primarily explained by the fact
Discussion that adults tend to be the carers.25 Funeral-related events
We present the first systematic review that addresses the are another well-recognized opportunity for infection
behavioural and other risk factors associated with filovirus transmission.18,32–34,39,41,61 Viable virus has been isolated
(Ebola and Marburg disease) infection within the commu- from animal tissues or fluids in the laboratory as late as 7
nity. The key findings of this review are that infection risk days post-mortem.62 However, not all of the studies found
is primarily associated with three behaviours: (i) close con- such an association between attending funerals and disease
tact in the later stages of infection; (ii) caring for a sick risk. Furthermore, even for those attending funerals where
International Journal of Epidemiology, 2015, Vol. 0, No. 0 11

transmission occurs, only people who engage in certain casual contact), which we discuss at length in this paper, is
behaviours are particularly at risk. It is possible for digni- not the same as saying that such contact poses low risk
fied funerals to be held without high risk to those attend- overall to human health after disease has developed. The
ing.63 Unfortunately, efforts to persuade local populations total risk to human health results from risk of transmission
to change funeral traditions during outbreaks, and in combined with likely consequences of disease, and filovirus
particular to allow cremation, often meet cultural diseases are usually very dangerous.
resistance.35,41,44,64 We excluded articles which identified cases solely from
Of particular relevance to policy is the strong evidence symptomology or antibody counts13,15,49,50,60,73,74,76–78
that risk to family members is high in those caring for their The symptoms of Ebola and Marburg virus can mimic
relatives up to death, and that this risk is much higher in other diseases endemic to sub-Saharan Africa,9 and filo-
those caring for people within the home (unadj. PPR virus antibodies from asymptomatic infection are in
13.33, 95% CI 3.2–55.6).22 Such an observation strength- 2–15% of the regional population not otherwise known to
ens the need to ensure that sufficient spaces are available in have ever had relevant illness.9,11–14,49,72,73 Concerns have
health care facilities so that no one suspected of filovirus been raised about problems of high rates of false-positives
infection has to be cared for in their own home until death. from Ebola virus antibody tests,79,80 which is why, ideally,
Our review focused on community human-to-human we wanted to only report results from patients who had at
infection, and excluded transmissions to health care work- least one positive laboratory test for filovirus infection. In

Downloaded from http://ije.oxfordjournals.org/ by guest on December 9, 2015

ers within clinical environments actively treating filo- reality, too few articles met that strict criterion. Some
virus,25,28,34,65,66 accidents in research laboratories67 or reports also did not clearly state whether interviews, con-
wildlife contact. We omitted cases in health care or labora- tact tracing and risk ratios were calculated only with
tory workers because they tend to lead to few subsequent laboratory-confirmed cases. Therefore, our rule was to
cases33 (with notable exceptions particularly when EVD include risk ratio and anecdotal data in articles where at
has not yet been identified68–70). The risk of transmission least some of the patients among the identified cases in the
in laboratory or clinical environments can clearly be disease cohort were laboratory confirmed, with others
greatly minimized with stringent control measures33,66,71 identified by clinical and contact history. We are also not
that are unlikely to be replicable in the community. in a position to comment on the impacts of possible virus
Wildlife contacts are extremely important in filovirus epi- mutations on transmission risks; there is mixed evi-
demiology because outbreak starts are nearly always dence13,81,82 about whether the large 2013–15 epidemic
linked to contact with animals, but calculating the risks of may have caused the virus to become more or less infec-
contracting disease is very difficult due to lack of data on tious or deadly to humans. Although it is common prac-
wildlife contacts in Africa that do not result in filovirus dis- tice in epidemiological reviews, grouping Marburg and
ease. Wildlife contacts may be at least partly responsible Ebola viruses together may not be ideal. Better risk assess-
for widespread filovirus antibody seroprevalence after ments may result as more detailed information emerges
asymptomatic illness.11–14,49,50,72–74 Genetic analysis about the transmission risk factors for each individual
strongly suggests that the 2013–15 outbreak was driven by filovirus. Nevertheless, there was no evidence within the
human-to-human transmission rather than new imports studies we found that suggested that the risk factors for
after the initial emergence.75 human-to-human transmission differ between the two
genera.
It is also unfortunate that we do not have specific odds
Limitations or risk ratio data for the 2013–15 outbreak. Most of the
We were able to produce only limited pooled estimates of information emerging from this epidemic is still anecdotal.
risk of transmission because the included papers were in- Undoubtedly, useful risk prevalence or odds ratio figures
consistent in defining risk factors or which measures of will emerge from the large Western Africa outbreak.
association were used; plus, many of the earlier studies pre- However, this may not be until after vaccination pro-
sented only unadjusted estimates. The partial exceptions to grammes have commenced these inoculation programmes
this are the data on disease transmission risk to household need information as soon as possible about the best strat-
contacts with, and without, direct contact. Consequently egy for containing the current and likely future outbreaks.
we are unable to specify with sufficient degree of certainty
how risky specific behaviours may be. Nevertheless, the
general findings of low risk for contracting disease from Implications
relatively casual contact is reassuring. It is also important The 2013–15 epidemic was unprecedented due to the large
to emphasise that low risk of contracting disease (e.g. from number of cases in densely populated urban settings.
12 International Journal of Epidemiology, 2015, Vol. 0, No. 0

Only one of our selected papers33 discussed factors which direct physical contact. Once an outbreak has been identi-
may be especially relevant in this but not previous out- fied, care for patients in well-equipped health care facilities
breaks, including (but not limited to): population density cuts transmission rates. There is wide variation in the con-
in home area, level of education, income or affluence lev- fidence intervals and magnitude, in many suggested risk
els, urban occupation group, type of home or household factors even when adjusted for confounders, suggesting
construction materials, religious practices, funeral prac- that understanding of community filovirus transmission
tices and proximity to medical care. Better future disease could be greatly improved.
containment in urban areas may depend on identification
of risk factors specific to urban Africa. Reducing possible
risks may also mean long-term cultural changes (e.g. trad- Supplementary Data
itional funeral practices) that are difficult for local popula-
Supplementary data are available at IJE online.
tions to accept.
The implications of our review are that when vaccin-
ation is not widely available, the primary control measure
for filovirus infections should remain the early identifica- Funding
tion of cases, contact tracing and subsequent quarantine This research was funded by the National Institute for Health
and care of cases in suitably equipped treatment centres. Research (NIHR) Health Protection Research Unit in Emergency

Downloaded from http://ije.oxfordjournals.org/ by guest on December 9, 2015

Preparedness and Response in partnership with Public Health
When vaccination is possible, then priority should be given
England (PHE). The views expressed are those of the authors and
to individuals engaged in activities that involve high-risk not necessarily those of the NHS, the NIHR, the Department of
direct contact with confirmed or suspected cases (or bodily Health or PHE.
waste): e.g. those providing physical care within the com-
munity and treatment centres. A second-level priority ring
Acknowledgements
for vaccination would for those who have other forms of
Many thanks to MSF staff and filovirus experts who fielded our
direct contact with known or suspected cases. Such a ring
emails. We are also grateful for the diligent reading skills of two
approach to disease containment is already in use, in the anonymous reviewers.
2013–15 EBOV outbreak.83
Conflict of interest: None declared.

Conclusions References
We have shown that risk of acquisition of filovirus infec- 1. Kucharski A, Camacho A, Checchi F, Waldman R, Grais R,
Cabrol J. Evaluation of the Benefits and Risks of Introducing
tions primarily follows from only close personal contact
Ebola Community Care Centers. Washington, DC: Himmelfarb
and generally only in later stages of illness. By making this
Health Sciences Library, The George Washington University,
statement, in no way do we deny that filovirus infections 2014.
are dangerous. The EVD transmission paradox (colloqui- 2. World Health Organization. Ebola Virus Disease: Fact Sheet
ally summarized as ‘Hard to catch, Easy to die from’) has No. 103. Geneva: WHO, 2015 15.
been discussed previously84 but never summarized with as 3. Olival KJ, Hayman DT. Filoviruses in bats: current knowledge
much quantitative and documentary evidence as we pro- and future directions. Viruses 2014;6:1759–88.
vide. Caring for patients until death is particularly risky, 4. World Health Organization. Marburg fever outbreak leads sci-
entists to suspected disease reservoir. Bull World Health Organ
especially within domestic settings. Among people experi-
2007;85:659–732.
encing only indirect contact, even when living in the same
5. MacNeil A, Rollin PE. Ebola and Marburg hemorrhagic fevers:
house, the risk of contracting disease is actually quite low. neglected tropical diseases? PLoS Negl Trop Dis 2012;6:e1546.
There is little evidence that more distant contact or that 6. Feldmann H, Geisbert TW. Ebola haemorrhagic fever. Lancet
contact with people incubating the disease poses any risks. 2011;377:849–62.
More studies are needed that correlate context, timing and 7. World Health Organization. The Ebola Outbreak in Liberia is
intimacy of contact with days after disease onset and exter- Over. WHO Statement. Geneva: WHO, 2015.
nal symptoms or severity of illness. There is evidence that 8. Van Kerkhove MD, Bento AI, Mills HL, Ferguson NM,
Donnelly CA. A review of epidemiological parameters from
transmission from non-intimate contact is low during early
Ebola outbreaks to inform early public health decision-making.
illness, but there is no simple indicator for the transition to Sci Data 2015;2; doi: 10.1038/sdata.2015.19.
late illness when disease transmission is highly likely from 9. Kuhn J, Calisher CH. Filoviruses: a Compendium of 40 Years of
any contact without adequate protective measures. Meta- Epidemiological, Clinical, and Laboratory Studies. New York,
analysis showed that transmission is very unlikely without NY: Springer Science & Business Media; 2008.
International Journal of Epidemiology, 2015, Vol. 0, No. 0 13

10. Henao-Restrepo AM, Longini IM, Egger M et al. Efficacy and 27. Wamala JF, Lukwago L, Malimbo M et al. Ebola hemorrhagic
effectiveness of an rVSV-vectored vaccine expressing fever associated with novel virus strain, Uganda, 2007–2008.
Ebola surface glycoprotein: interim results from the Guinea Emerg Infect Dis 2010;16:1087–92.
ring vaccination cluster-randomised trial. Lancet 28. Ndambi R, Akamituna P, Bonnet M-J, Tukadila AM, Muyembe-
2015;386:857–66. Tamfum J-J, Colebunders R. Epidemiologic and clinical aspects
11. Burke J, Declerq R, Ghysebrechts G et al. Ebola hemorragic- of the Ebola virus epidemic in Mosango, Democratic Republic of
fever in Zaire, 1976. Report of an International Commission. the Congo, 1995. J Infect Dis 1999;179:S8–S10.
Bull World Health Organ 1978;56:271–93. 29. Bertherat E, Renaut A, Nabias R, Dubreuil G, Georges-Courbot
12. Jezek Z, Szczeniowski M, Muyembe-Tamfum J, McCormick J, M-C. Leptospirosis and Ebola virus infection in five gold-
Heymann D. Ebola between outbreaks: intensified Ebola hemor- panning villages in northeastern Gabon. Am J Trop Med Hyg
rhagic fever surveillance in the Democratic Republic of the 1999;60:610–15.
Congo, 1981–1985. J Infect Dis 1999;179:S60–S64. 30. Blackley DJ, Lindblade KA, Kateh F et al. Rapid intervention to
13. Leroy E, Baize S, Volchkov V et al. Human asymptomatic Ebola reduce Ebola transmission in a remote village – Gbarpolu
infection and strong inflammatory response. Lancet county, Liberia, 2014. MMWR Morb Mortal Wkly Rep
2000;355:2210–15. 2015;64:175–78.
14. Nkoghe D, Padilla C, Becquart P et al. Risk factors for Zaire ebo- 31. Christie A, Davies-Wayne GJ, Cordier-Lasalle T et al. Possible
lavirus–specific IgG in rural Gabonese populations. J Infect Dis sexual transmission of Ebola virus—Liberia, 2015. MMWR
2011;204:S768–S75. Morb Mortal Wkly Rep 2015;64:1–3.
15. Moyen N, Thirion L, Emmerich P et al. Risk Factors Associated 32. Dietz P, Jambai A, Paweska JT, Yoti Z, Ksaizek TG.

Downloaded from http://ije.oxfordjournals.org/ by guest on December 9, 2015

with Ebola and Marburg Viruses Seroprevalence in Blood Epidemiology and risk factors for Ebola virus infection in Sierra
Donors in the Republic of Congo. PLoS Negl Trop Dis Leone–May 23, 2014-January 31, 2015. Clin Infect Dis 2015,
2015;9:e0003833. Jul 15. pii: civ568. [Epub ahead of print.]
16. Liberati A, Altman DG, Tetzlaff J et al. The PRISMA statement 33. Faye O, Boëlle P-Y, Heleze E et al. Chains of transmission and
for reporting systematic reviews and meta-analyses of studies control of Ebola virus disease in Conakry, Guinea, in 2014: an
that evaluate health care interventions: explanation and elabor- observational study. Lancet Infect Dis 2015;3:320–26.
ation. Ann Intern Med 2009;151:W–65–W-94. 34. Fitzpatrick G, Vogt F, Moi Gbabai O et al. Describing readmis-
17. Martini G, Schmidt H. Spermatogene Übertragung des ‘Virus sions to an Ebola case management centre (CMC), Sierra Leone,
Marburg’. [Spermatogenesis Transmission of Marburg Virus]. 2014. Euro Surveill 2014;19.
Klin Wochenschr 1968;46:398–400. 35. Formenty P, Libama F, Epelboin A et al. [Outbreak of Ebola
18. Victory KR, Coronado F, Ifono SO, Soropogui T, Dahl BA. hemorrhagic fever in the Republic of the Congo, 2003: a new
Ebola Transmission Linked to a Single Traditional Funeral strategy?] Med Trop (Mars) 2002;63:291–95.
Ceremony—Kissidougou, Guinea, December, 2014–January 36. Kreuels B, Wichmann D, Emmerich P et al. A case of severe
2015. MMWR Morb Mortal Wkly Rep 2015;64:386–88. Ebola virus infection complicated by gram-negative septicemia.
19. Breman J, Piot P, Johnson K et al. The Epidemiology of Ebola N Engl J Med 2014;371:2394–401.
Hemorrhagic Fever in Zaire, 1976. Amsterdam: Elsevier/North 37. Leroy EM, Epelboin A, Mondonge V et al. Human Ebola out-
Holland, 1978. break resulting from direct exposure to fruit bats in Luebo,
20. Kunii O, Kita E, Shibuya K. Epidemics and related cultural fac- Democratic Republic of Congo, 2007. Vector Borne Zoonotic
tors for Ebola hemorrhagic fever in Gabon. [Nihon koshu eisei Dis 2009;9:723–28.
zasshi]. Jpn J Public Health 2001;48:853–59. 38. MacNeil A, Farnon EC, Morgan OW et al. Filovirus outbreak
21. Martini G. Marburg virus disease. Postgrad Med J detection and surveillance: lessons from Bundibugyo. J Infect
1973;49:542–46. Dis 2011;204:S761–S67.
22. Francesconi P, Yoti Z, Declich S et al. Ebola hemorrhagic fever 39. Moreau M, Spencer C, Gozalbes J et al. Lactating mothers
transmission and risk factors of contacts, Uganda. Emerg Infect infected with Ebola virus: EBOV RT-PCR of blood only may be
Dis 2003;9:1430. insufficient. Euro Surveill 2015;20:3.
23. Roels T, Bloom A, Buffington J et al. Ebola hemorrhagic fever, 40. Mupere E, Kaducu O, Yoti Z. Ebola haemorrhagic fever among
Kikwit, Democratic Republic of the Congo, 1995: risk factors hospitalised children and adolescents in nothern Uganda:
for patients without a reported exposure. J Infect Dis Epidemiologic and clinical observations. Afr Health Sci
1999;179:S92–S97. 2001;1:60–65.
24. Dowell SF, Mukunu R, Ksiazek TG, Khan AS, Rollin PE, Peters 41. Nkoghe D, Kone ML, Yada A, Leroy E. A limited outbreak of
C. Transmission of Ebola hemorrhagic fever: a study of risk fac- Ebola haemorrhagic fever in Etoumbi, Republic of Congo, 2005.
tors in family members, Kikwit, Democratic Republic of the Trans R Soc Trop Med Hyg 2011;105:466–72.
Congo, 1995. J Infect Dis 1999;179:S87–S91. 42. Nyenswah T, Blackley DJ, Freeman T et al. Community
25. Baron RC, McCormick JB, Zubeir OA. Ebola virus disease in Quarantine to Interrupt Ebola Virus Transmission—Mawah
southern Sudan: hospital dissemination and intrafamilial spread. Village, Bong County, Liberia, August–October, 2014. MMWR
Bull World Health Organ 1983;61:997. Morb Mortal Wkly Rep 2015;64:179–82.
26. Roddy P, Thomas SL, Jeffs B et al. Factors associated with 43. Nyenswah T, Fallah M, Sieh S et al. Controlling the last known
Marburg hemorrhagic fever: analysis of patient data from Uige, cluster of Ebola virus disease—Liberia, January–February 2015.
Angola. J Infect Dis 2010;201:1909–18. MMWR Morb Mortal Wkly Rep 2015;64:500–4.
14 International Journal of Epidemiology, 2015, Vol. 0, No. 0

44. Nyenswah TG, Fallah M, Calvert GM et al. Cluster of Ebola 61. Legrand J, Grais R, Boelle P, Valleron A, Flahault A.
Virus Disease, Bong and Montserrado Counties, Liberia. Emerg Understanding the dynamics of Ebola epidemics. Epidemiol
Infect Dis 2015;21:1253. Infect 2007;135:610–21.
45. Richards P, Amara J, Ferme MC et al. Social pathways for Ebola 62. Prescott J BT, Fischer R, Miazgowicz K, Judson S, Munster VJ.
virus disease in rural Sierra Leone, and some implications for Postmortem stability of Ebola virus. Emerg Infect Dis
containment. PLoS Negl Trop Dis 2015;9:e0003567. doi: 2015;21:856–9.
10.1371/journal.pntd.0003567. 63. World Health Organization. Field Situation: How to Conduct
46. Olupot-Olupot P. Ebola in children: Epidemiology, clinical fea- Safe and Dignified Burial of a Patient who Has Died from
tures, diagnosis and outcomes. Pediatr Infect Dis J Suspected or Confirmed Ebola Virus Disease. geneva: WHO,
2015;34:314–16. 2014.
47. Dowell SF. Ebola hemorrhagic fever: why were children spared? 64. De Roo A, Ado B, Rose B, Guimard Y, Fonck K, Colebunders R.
Pediatr Infect Dis J 1996;15:189–91. Survey among survivors of the 1995 Ebola epidemic in Kikwit,
48. WHO Ebola Response Team. Ebola virus disease in West Democratic Republic of Congo: their feelings and experiences.
Africa—the first 9 months of the epidemic and forward projec- Trop Med Int Health 1998;3:883_85.
tions. N Engl J Med 2014;371:1481–95. 65. Khan AS, Tshioko FK, Heymann DL et al. The reemergence of
49. Busico KM, Marshall KL, Ksiazek TG et al. Prevalence of IgG Ebola hemorrhagic fever, Democratic Republic of the Congo,
antibodies to Ebola virus in individuals during an Ebola out- 1995. J Infect Dis 1999;179:S76–S86.
break, Democratic Republic of the Congo, 1995. J Infect Dis 66. Kerstiëns B, Matthys F. Interventions to control virus transmis-
1999;179:S102–S07. sion during an outbreak of Ebola hemorrhagic fever: experience

Downloaded from http://ije.oxfordjournals.org/ by guest on December 9, 2015

50. Bausch DG, Borchert M, Grein T et al. Risk factors for Marburg from Kikwit, Democratic Republic of the Congo, 1995. J Infect
hemorrhagic fever, Democratic Republic of the Congo. Emerg Dis 1999;179:S263–S67.
Infect Dis 2003;9:1531. 67. Emond R, Evans B, Bowen E, Lloyd G. A case of Ebola virus in-
51. Rowe AK, Bertolli J, Khan AS et al. Clinical, virologic, and fection. Br Med J 1977;2:541–44.
immunologic follow-up of convalescent Ebola hemorrhagic 68. Fasina F, Shittu A, Lazarus D et al. Transmission dynamics and
fever patients and their household contacts, Kikwit, control of Ebola virus disease outbreak in Nigeria, July to
Democratic Republic of the Congo. J Infect Dis September 2014. Euro Surveill 2014;19:20920.
1999;179:S28–S35. 69. Borchert M, Mutyaba I, Van Kerkhove MD et al. Ebola haemor-
52. Bausch DG, Towner JS, Dowell SF et al. Assessment of the risk rhagic fever outbreak in Masindi District, Uganda: outbreak de-
of Ebola virus transmission from bodily fluids and fomites. scription and lessons learned. BMC Infect Dis 2011;11:357.
J Infect Dis 2007;196:S142–S47. 70. Baize S, Pannetier D, Oestereich L et al. Emergence of
53. Liddell AM, Davey RT Jr, Mehta AK et al. Characteristics and Zaire Ebola virus disease in Guinea. N Engl J Med
clinical management of a cluster of 3 patients with Ebola 2014;371:1418–25.
virus disease, including the first domestically acquired cases in 71. Hall R. The 1995 Kikwit Ebola outbreak – model of virus prop-
the United States. Ann Intern Med 2015;163:81–90. erties on system capacity and function: a lesson for future viral
54. Mora-Rillo M, Arsuaga M, Ramı́rez-Olivencia G et al. Acute re- epidemics. Am J Disaster Med 2006;2:270–76.
spiratory distress syndrome after convalescent plasma use: treat- 72. Ikegami T, Niikura M, Saijo M et al. Antigen capture enzyme-
ment of a patient with Ebola virus disease contracted in Madrid, linked immunosorbent assay for specific detection of Reston
Spain. Lancet Respir Med 2015;3:554–62. Ebola virus nucleoprotein. Clin Diagn Lab Immunol
55. Schibler M, Vetter P, Cherpillod P et al. Clinical features and 2003;10:552–57.
viral kinetics in a rapidly cured patient with Ebola virus disease: 73. Heffernan RT, Pambo B, Hatchett RJ, Leman PA, Swanepoel R,
a case report. Lancet Infect Dis 2015;15:103–40. Ryder RW. Low seroprevalence of IgG antibodies to Ebola virus
56. Mackay IM, Arden KE. Ebola virus in the semen of convalescent in an epidemic zone: Ogooue-Ivindo region, northeastern
men. Lancet Infect Dis 2014;15:149–50. Gabon, 1997. J Infect Dis 2005;191:964–68.
57. Rodriguez L, De Roo A, Guimard Y et al. Persistence and genetic 74. Heymann D, Weisfeld J, Webb P, Johnson K, Cairns T, Berquist
stability of Ebola virus during the outbreak in Kikwit, H. Ebola hemorrhagic fever: Tandala, Zaire, 1977–1978.
Democratic Republic of the Congo, 1995. J Infect Dis J Infect Dis 1980;142:372–76.
1999;179:S170–S76. 75. Park DJ, Dudas G, Wohl S et al. Ebola virus epidemiology, trans-
58. Bonnet M-J, Akamituna P, Mazaya A. Unrecognized Ebola hem- mission, and evolution during seven months in Sierra Leone. Cell
orrhagic fever at Mosango Hospital during the 1995 epidemic in 2015;161:1516–26.
Kikwit, Democratic Republic of the Congo. Emerg Infect Dis 76. Knobloch J, Albiez E, Schmitz H. A Serological Survey on Viral
1998;4:508. Haemorrhagic Fevers in Liberia. paris: Institut Pasteur/
59. Changula K, Kajihara M, Mweene AS, Takada A. Ebola and Virologie, 1982.
Marburg virus diseases in Africa: Increased risk of outbreaks 77. Towner JS, Sealy TK, Khristova ML et al. Newly discovered
in previously unaffected areas? Microbiol Immunol Ebola virus associated with hemorrhagic fever outbreak in
2014;58:483–91. Uganda. PLoS Pathog 2008;4:e1000212.
60. McElroy AK, Erickson BR, Flietstra TD et al. Biomarker correl- 78. MacNeil A, Reed Z, Rollin PE. Serologic cross-reactivity of
ates of survival in pediatric patients with Ebola virus disease. human IgM and IgG antibodies to five species of Ebola virus.
Emerg Infect Dis 2014;20:1683. PLoS Negl Trop Dis 2011;5:e1175.
International Journal of Epidemiology, 2015, Vol. 0, No. 0 15

79. Perkins MD, Kessel M. What Ebola tells us about outbreak diag- 82. McSpadden K. The Ebola Virus is Mutating, Say Scientists.
nostic readiness. Nat Biotechnol 2015;33:464–69. Time, 29 January 2015.
80. Vladyko A, Zaı̆tseva V, Trofimov N et al. [False-positive 83. Nyenswah T, Massaquoi M, Gbanya M et al. Initiation of a ring
reactions in laboratory diagnosis of Lassa, Marburg, and approach to infection prevention and control at non-ebola health
Ebola viral hemorrhagic fevers and AIDS]. Vopr Virusol care facilities – Liberia, January-February 2015. MMWR Morb
1996;42:66–70. Mortal Wkly Rep 2015;64:505–08.
81. Olabode AS, Jiang X, Robertson DL, Lovell SC. Ebola virus is 84. Klompas M, Yokoe DS. The Ebola transmission paradox. Am J
evolving but not changing: no evidence for functional change in Infect Control 2015;43:786.
EBOV from 1976 to the 2014 outbreak. Virology 85. Gear S, Cassel GA, Gear AJ et al. Outbreak of Marburg virus dis-
2015;482:202_07. ease in Johannesburg. Br Med J 1975;4:489–93.

Downloaded from http://ije.oxfordjournals.org/ by guest on December 9, 2015