Les Roberts – Freetown, Sierra Leone – April 4, 2015
I had a call from our alum Laura Miller this morning. She was bummed… and a little freaked out about an infant case that arose in Kailahun where there has not been a case for quite a while. This is where the outbreak seems to have entered Sierra Leone. The child’s parents are EVD negative. Something similar (psychologically but not epidemiologically) has happened in Liberia in recent days, and if I believe the newspapers, perhaps those are linked to sexual transmission. Laura and thousands of others have been really demoralized by embers of Ebola arising where we thought the outbreak had come and gone.
I think in part this is because everyone around me has been promoting this phrase “Getting to Zero.” It is driving me crazy, as it is probably an impossible concept. It strongly appears that Ebola has been present in the region for years and was simply unrecognized. Evidence of this includes:
Analysis of hundreds of frozen blood samples from 2011 – 2014 at the Kenema Hospital Lassa Fever Ward found 22% of Lassa Negative patients were EVD +. See: Boisen ML et al. Multiple circulating infections can mimic the early stages of viral hemorrhagic fevers and possible human exposure to filoviruses in Sierra Leone prior to the 2014 outbreak. Viral Immunol. 2015 Feb;28(1):19-31. doi: 10.1089/vim.2014.0108.
A 1996 WHO press release described a 25 year old male on the Liberia / Cote d’Ivoire border “was confirmed by serological tests carried out at the Institute Pasteur in Paris” as EVD positive. See: http://www.who.int/csr/don/1996_01_22c/en/
A population-based random blood sample of 433 residents “in rural areas” of Liberia in 1978-’79 found 6% to be EVD positive. They tested for anti-bodies (range of positives 1/16 to 1/1024). See: Knobloch J. A serological Survey on Viral Hemorrhagic Fevers in Liberia. Ann. Virol. 1982, 133E, 125 – 128.
Frozen sera from 592 residents of mining company settlements were taken in 1973 and analyzed in 1986 and 14% were EVD positive and 21% positive for Marburg. See Neppert J et al. No evidence of LAV infection in Liberia, West Africa, in the year 1973. Blut (1986) 53: 115-117.
While I know little about virology and I know there was a lot of kibitzing about the precision/specificity /cross-reactivity of those indirect immunofluorescence anti-body tests back in the ‘80’s, I think the 1996 case and the recent Kenema data make it probable that there have been tens or hundreds of cases in Sierra Leone in recent years and that getting to a non-outbreak baseline is not the same as getting to zero. Surveillance has been very very poor in the region.
Back in a 2001, a gang of us from the Ministry and IRC did a 500 households random survey about mortality in Kenema District. Only around 16 of 78 reported febrile deaths had even gone for any medical care before dying. And, we all know, a health post or a Primary Health Unit (PHU… a fancier clinic with beds for in-patients usually staffed by a nurse) would have never diagnosed EVD before this past year. (An aside, back then IRC had a network of hundreds of TBA’s and several ambulances who were proud that there were zero known household maternal deaths in Kenema District in 2000, only those few hundred at the hospital; that is, they were perfect in their complicated birth referral record! We estimated there were 1900 and discovered that IRC’s TBA’s were actively hiding them because their training had emphasized how terrible and unacceptable maternal deaths were, creating stigma about reporting. If stigma about reporting maternal deaths arose so easily, think of what household and village quarantine is doing to Ebola surveillance now!)
Adding to the counter-productivity of the “Getting to Zero” concept is the notion of ecological equilibrium. If in past years there was a certain viral load in the animal community, ecological equilibrium would suggest that typically a certain number of human cases would be induced based on how often people killed game, were shat upon by passing bats, had a dog bring back an infected squirrel… Perhaps it was just purely human-related luck in 2014 that triggered these favorable human transmission events (like just by chance a couple of the victims had large body-rinsing funerals), but it is more likely the animal load is now higher. Thus (and I am making this up for the sake of discussion), if nationally the baseline was 10 animal based transmission events for an average of 30 cases a year, we should expect far more sporadic cases this year… and now we can test for it and now we are looking!
I hate campaign slogans predestined to fail. “Yes we can” is the perfect slogan because it is so ambiguous you are never sure if it succeeded or failed. “Getting to zero” is anything but ambiguous. Talk about setting yourself up for failure!