Saturday, 4 July 2015

Editor's Rant: Communicating the data and about the data...

It is pretty damn hard work trying to get hold of data on virus outbreaks around the world. 

When it is, it may be available in unfriendly formats. It may not be made public at all. When it is available, it is often slow to appear or it may have random reporting gaps, or be partially incomplete. The style of the released data can change overnight as well, sometimes going from detail to summary.

So why bother about trying to get hold of these numbers at all? It's not like I work in the field. Well, that is a question I'm increasingly asking myself of late too. My personal reason has been because I think there need to be more voices in the vacuum between the numbers being reported and the often dry public health reports. I think scientists, even if they are not lifetime experts on a given virus or outbreak, still have much to offer when they come out from behind their manuscripts and apply their skills to interpreting what's happening. Well, many do anyway. And they should do it more. Now, perhaps more than ever, science needs steer away from its cold, dense and boring niche writing to a chattier, more helpful and community-based style of engagement. It astonishes me how often the public's interpretation of outbreak numbers must come from the media or from hobbyists, or even professionals who work in other areas and give of their own time to help explain something to us in their personal time. Helpful and engaging information and better access should come from the source of the data.

So it becomes really annoying (you would have to know me quite well to know how many times I just rewrote those words) when data are given out for public use that are a total mess...and there is not one tiny mote of explanation for it. I called it appalling on Twitter tonight. And at other times there are no explanations for why there are gaps, why data are delayed, why the format may have changed today compared to last week, why a line list is missing a case, using a new and totally independent numbering scheme or suddenly reshuffled, why there is no news about a new outbreak. No word. No contact. No-one taking the lead. No...communication.

I have met a lot of people since I have been blogging who, in various ways, have put in their own personal time to help out bigPublicHealth, to help take up the slack in communicating to the media and to the public. It is hard to quantify the impact of that combined help-but I can assure you that it reaches far and wide and is not insignificant. One would think that it should be easier to provide this help when one is willing to make use of their own time and use their own resources, or that those people should be shown enough respect to be able to simply find and apply reliable raw data so they can help out. But one would be an idiot. I very clearly remember a time when I could send a public Tweet to WHO's Head of Public Relations, Gregory Haertl, and get an informed reply. Those days have passed. I remember there being an #AskEbola channel on Twitter that gave answers. That engagement is just not there anymore. I'm sure its funding and resources and blah blah...but not as sure as I could be if that were spoken about in public. Communication. Someone needs to step up on this. As the quotes above allude to, 2015 is not 2014. And one of those differences is that everyone wants timely and comprehensive information they can rely on during times of outbreak. This hasn't been discusses enough but it should be.

Friday, 3 July 2015

Ebola virus: wild and domestic animals, plants and insects...

Initial Ebola virus (EBOV) infection of humans is a rare zoonotic spillover event.  

Hypsignathus monstrosus, Epomops franqueti and Myonycteris torquatebats, all fruit-eating megabats of the family Pteropodidae, are considered to be important reservoir hosts, yet they do not show signs of disease.1 

While a great deal remains unknown about the identity and spectrum of natural ebolavirus hosts,1 zoonoses appear to co-occur with bat pregnancy.2

Animals that have died from ebolavirus infections include 3,4:

  • Duiker (Cephalophus sp.; an antelope) 
  • Gorilla (Gorilla gorilla) 
  •  Chimpanzee (Pan troglodytes)

Living animals found to harbour infectious EBOV include:

  • Cynomolgus macaque monkey (Macaca fascicularis
  • Franquet’s epauletted fruit bat (Epomops franqueti) 
  • Hammer-headed bat (Hypsignathus monstrosus
  • Little collared fruit bat (Myonycteris torquata)
Those animals with only antibodies to EBOV in the absence of infectious virus, suggesting past exposure include 5,6:

  • Domestic dogs (Canis lupus familiaris
  • Peter’s lesser epauletted fruit bat (Micropterus pusillus; fruit-eating) 
  • Angolan free-tailed bat (Mops condylurus; insect-eating) 
  • Giant roundleaf bat (Hipposideros gigas; insect-eating) 
  • Egyptian fruit bat (Roussetus aegyptiacus; fruit-eating) 
  • Geoffrey’s rousette (Rousettus amplexicaudatus; a bat species; fruit-eating) 
  • Lord Derby’s scaly-tailed squirrel (Anomalurus derbianus)

Porcupines (Hystrix cristata) have been implicated as a source for human EBOV exposure but virus-positive animals have not been documented.4 
Between nine and 25% of 337 domestic dogs from various towns and villages in Gabon during an EBOV outbreak in 2001-2002 were identified as possible hosts for EBOV when found to be seropositive. 7,8 It was not known when they became seropositive nor has it been experimentally determined that dogs are able to host an active EBOV infection.9,10 Dogs were observed in contact with suspected virus-laden fluids and with other animals during the Gabon outbreak but seropositive dog specimens did not contain EBOV antigen or viral RNA. Three specimens from these seropositive dogs did not yield infectious virus in cell culture either and thus there remains no documented evidence for a canine source of human EBOV infection. In 2014, two dogs owned by human cases of EBOV/Mak in Spain (euthanized without testing 11) and the United States of America (tested negative for EBOV 12,13) did not exhibit any signs of disease. 
Domestic pigs have been found to be a natural host for the Reston ebolavirus 9,14 and antibodies to EBOV have also been found in guinea pigs, an animal that can also be experimentally infected.15 Domestic dogs and guinea pigs appear to become infected without symptoms.6,7 Horses, mice, guinea pigs and goats have been experimentally inoculated with EBOV to produce antisera or test therapeutic preparations.16,17 
Pigs experimentally infected with a member of the Zaire ebolavirus become symptomatic.8 NHP, guinea pigs and mice have been used to examine aspects of disease progression and exhibit various degrees of disease when experimentally infected.18,19 
On a few occasions in one study into possible hosts, a low viral load of EBOV could be sporadically recovered after inoculation of a snake (up to 11 days post inoculation), a mouse (up to nine days later) and a spider (21 days later) but the authors of this study concluded that these results could have represented residual inoculum.21
Plants, arthropods, cows, cats and sheep have not been found to naturally carry or host ebolavirus infection but only small numbers of some species have been examined. 3,20-22


    1. Leroy EM, Kumulungui B, Pourrut X, et al. Fruit bats as reservoirs of Ebola virus. Nature 2005;438:575-6. 
    2. Plowright RK, Eby P, Hudson PJ, et al. Ecological dynamics of emerging bat virus spillover. Proc Biol Sci 2015;282:20142124.
    3. Olson SH, Reed P, Cameron KN, et al. Dead or alive: animal sampling during Ebola hemorrhagic fever outbreaks in humans. Emerg Health Threats J 2012;5
    4. Lahm SA, Kombila M, Swanepoel R, Barnes RF. Morbidity and mortality of wild animals in relation to outbreaks of Ebola haemorrhagic fever in Gabon, 1994-2003. Trans R Soc Trop Med Hyg 2007;101:64-78.
    5. Marsh GA, Haining J, Robinson R, et al. Ebola Reston virus infection of pigs: clinical significance and transmission potential. J Infect Dis 2011;204 Suppl 3:S804-9.
    6. Gonzalez JP, Herbreteau V, Morvan J, Leroy EM. Ebola virus circulation in Africa: a balance between clinical expression and epidemiological silence. Bull Soc Pathol Exot 2005;98:210-7.
    7. Allela L, Boury O, Pouillot R, et al. Ebola virus antibody prevalence in dogs and human risk. Emerg Infect Dis 2005;11:385-90.
    8. Weingartl HM, Nfon C, Kobinger G. Review of Ebola virus infections in domestic animals. Dev Biol (Basel) 2013;135:211-8.
    9. Stansfield SK, Scribner CL, Kaminski RM, Cairns T, McCormick JB, Johnson KM. Antibody to Ebola virus in guinea pigs: Tandala, Zaire. J Infect Dis 1982;146:483-6.
    10. Connolly BM, Steele KE, Davis KJ, et al. Pathogenesis of experimental Ebola virus infection in guinea pigs. J Infect Dis 1999;179 Suppl 1:S203-17.
    11. Why Dallas Won't Kill The Dog Of The Texas Nurse With Ebola. Business Insider, 2014. (Accessed 27/4/2015, at )
    12. Starting today, Dallas Animal Services will begin testing Nina Pham’s year-old dog Bentley for Ebola. The Dallas Morning News, 2014. (Accessed 17/4/2015, at
    13. EBOLAVIRUS, ANIMAL RESERVOIR (05): USA, DOG, NOT. 2014. (Accessed 01/05/2015, at )
    14. Barrette RW, Metwally SA, Rowland JM, et al. Discovery of swine as a host for the Reston ebolavirus. Science 2009;325:204-6.
    15. Rouquet P, Froment JM, Bermejo M, et al. Wild animal mortality monitoring and human Ebola outbreaks, Gabon and Republic of Congo, 2001-2003. Emerg Infect Dis 2005;11:283-90.
    16. Kudoyarova-Zubavichene NM, Sergeyev NN, Chepurnov AA, Netesov SV. Preparation and use of hyperimmune serum for prophylaxis and therapy of Ebola virus infections. J Infect Dis 1999;179 Suppl 1:S218-23.
    17. Bray M, Davis K, Geisbert T, Schmaljohn C, Huggins J. A mouse model for evaluation of prophylaxis and therapy of Ebola hemorrhagic fever. J Infect Dis 1998;178:651-61.
    18. Ebihara H, Takada A, Kobasa D, et al. Molecular determinants of Ebola virus virulence in mice. PLoS Pathog 2006;2:e73.
    19. Geisbert TW, Young HA, Jahrling PB, Davis KJ, Kagan E, Hensley LE. Mechanisms underlying coagulation abnormalities in ebola hemorrhagic fever: overexpression of tissue factor in primate monocytes/macrophages is a key event. J Infect Dis 2003;188:1618-29.
    20. Turell MJ, Bressler DS, Rossi CA. Short report: lack of virus replication in arthropods after intrathoracic inoculation of Ebola Reston virus. Am J Trop Med Hyg 1996;55:89-90.
    21. Swanepoel R, Leman PA, Burt FJ, et al. Experimental inoculation of plants and animals with Ebola virus. Emerg Infect Dis 1996;2:321-5.
    22. Ebola haemorrhagic fever in Sudan, 1976. Report of a WHO/International Study Team. Bull World Health Organ 1978;56:247-70.

      Tuesday, 30 June 2015

      Ebola mysteriously returns to Liberia...[UPDATED]

      v2-1JULY2015 AEST
      In a gut-wrenching, but not wholly unexpected event, a new case of Ebola virus disease (EVD) has popped up in a town called Nedowein (or Nedowian [8]), about 50km south west of Liberia's capital, Monrovia.

      Liberia had been declared a country free of EVD on 9-May-2015 - 52 days ago, or 1 month, 21-days, or 1248 hours.  

      The 17 year old male (17M) died on Wednesday (about 6 days ago) and has already been buried by all accounts. Samples from his corpse tested positive at least twice.[3]

      It's not an unexpected event because both Guinea and Sierra Leone, adjoining countries, continue to struggle with EVD and have been unable to stop the disease from spreading, even though in relatively small numbers compared to what was occurring in 2014. 

      What makes this new case in Liberia a little mysterious is that 17M died far from the border with either of these countries; approximately 150km from Sierra Leone's south-eastern border and about 200km from the nearest Guinean border. Sure, these are not insurmountable distances to travel while incubating an Ebola virus infection, but it would have been a simpler call that this was an imported case if it had occurred on or nearer to the border of one of the two countries with ongoing disease. However, it seems the young man did not travel outside Liberia.[8]

      Hopefully the contact tracing and investigations that are going on now will find that 17M simply made contact with someone who had traveled from outside of Liberia, perhaps to Nedowein, which is described as the home town of 17M.[2,7] If this is not an imported case then one is left to wonder about various other scenarios including:

      1. sexual or other less common transmission of Ebola virus from an as yet undiscovered convalescent EVD case
      2. contact with an unknown case who had traveled across the border from a country with EVD
      3. a new zoonotic acquisition of a different Ebola virus variant
      4. there may still be clusters of EVD within Liberia that have been smouldering on without the knowledge of any authorities

      Time and further hard work will no doubt tell.

      UPDATE: A second  case, associated with 17M ("Abraham") has been diagnosed.[9] Some discussion is evolving around the consumption of dog meat by 17M,[11] however, the same questions around how a dog would become infected (no record of the detection of active replication in a dog have been recorded to date, although antibodies have suggested the possibility in earlier outbreaks) will apply.

      Further reading...

      Version history..
      1. New links added; town name variation added from Science report; hypothesis of contact with another imported case - #2; note on lack of travel outside of Liberia; announcmene tof a 2nd case

      Sunday, 21 June 2015

      Matching MERS case identification numbers from two differing sources...

      Update #1 27JUNE2015
      Update #2 28JUNE2015
      Sometimes people work from different playbooks.
      Figure 1. MERS in South Korea. Most cases now plotted
      on graph using their dates of illness onset.
      An outbreak in decline. This is up-to-date - 3 new cases
      from 21JUN2015 added (column at right hand side -
      onset dates unknown)
      Click on graph to enlarge

      In this instance, the data from the World Health Organization's (WHO) new list of Middle East respiratory syndrome (MERS) cases - with extra detail - uses a case identification key that's out of synchrony with that produced by the South Korean (SK) Health Authority which can be found in each of its posts announcing new MERS cases and deaths. 

      Attempting to link the two lists has mostly been an exercise in pedantry, but sometimes it is useful to know which case one is talking about when discussing an outbreak or cluster of disease....'Hey Bill, what didja think of that 70 year old MERS case who drove the ambulance carrying that infected 75 year old MERS case and then those others got MERS as well..?' doesn't really roll off the tongue does it?

      Embedded image permalink
      Figure 2. What the graph above looked like before
      we had dates of illness onset. Many cases
      were 'moved' to earlier time points because
      report dates always follow onset of illness dates
      and they can follow by varying periods of time -
      sometimes a day, sometimes a week or more.
      Two file formats are in the folder I've linked to below. This is my first attempt - yes, it is a work in progress - to match up the new WHO case list from the 19th June which includes the highly prized date of illness onset (DOOs) for most cases - with that of the South Korean (SK) Health Authority. They do differ. Quite a bit. And in several ways. For example, the numbering scheme is off by one or more, SK69 seems to be missing from the WHO list, there is a question mark hanging over SK152 & SK156 and the WHO data seem to have a number of different ages from the SK data- mostly differing by one year (presumably someone is rounding up or down). 

      If I've stuffed anything up or if you can solve my problem cases - please pass that info along and I'll update the files on this page. Hopefully the next WHO version will have addressed all of this anyway (it didn't but perhaps a future one will).

      These are publicly available and you can download them for your own interest.

      There is a download arrow at the top of the Google Drive page.

      1. Google Drive folder with MERS data files
      1. With the help of FluTrackers updated line list to cross check against, the first half of my list has been updated - some bugs fixed. 
      2. After about 5 hours - on and off - FluTrackers helped me sort out a few errors and the latest version of my list has been uploaded into the the folder linked above. Some typos corrected.

      Saturday, 20 June 2015

      MERS-CoV in South Korea - other data formats for data provided by WHO...

      The following links take you to different version of the PDF if you want to harvest those data...

      1. Excel version, Office 2010
      2. MS DOS CSV file
      3. Original WHO PDF
      Its taken a week to get these data, and its unclear why detail data dried up from the 12th June to the 19th, or why the data have appeared now, but they are here at last and that's great.

      Saturday, 13 June 2015

      Middle East respiratory syndrome (MERS) coronavirus: Outbreak in South Korea

      This data visualization was created using data from FluTrackers, the World Health Organization and the Korean Broadcasting System (KBS).

      Wherever possible, the dates are those for a case's onset of illness, but if that is unavailable then the date for hospitalization or if that is unclear, the date that case was reported.

      Because the WHO obviously have more detailed information, I recommend you keep an eye on their site, so long as it is being maintained, for more accurate epidemic curves.


      2. The Korean Broadcasting System's News websites &
      3.  World Health Organization pages
      Main landing page on MERS and MERS-CoV (includes link to WHO line list under General Information)
      Disease Outbreak News pages
      News on the current situation
      Summary of Korean statistics-rapidly updated

      Translating from the Korean for MERS epidemiology...

      I absolutely love the graphic the Korean Broadcasting Service (KBS) have been putting up. Love 'em. Great job!  I just wish they were in English too. I understand that the first priority is to your own population - totally get that. But I do wish that, like the Saudi Ministry of Health, every country could do an English version as well as the native language. 

      But processes, time money and stuff...

      Anyway - to help me try and get more detail on the South Korean clusters I have added some translation to one of the KBS pages which, when you click on the icon of a person, gives you a popup box with some detail. I've added translation to the contents of that box (see the figure below). 

      I don't know how accurate the "Date of infection" field is but will be comparing it to the World Health Organization (WHO) data from earlier to see if it can be useful. This is all because the date of reporting is almost always different from the date of illness onset - and the latter are much better to plot to get an idea of whether an outbreak is rising, peaking or slowing. Having the ability to crowd-plot these numbers is great and (I think) useful to inform the public and our clinical and scientific peers when included alongside some discussion about trends and reason for changes, risk etc. Often (always?) more personable banter, and engagement, seems to be lacking from 'bigPublicHealth' sources.

      Click on image to enlarge.
      Adapted from[1]

      This sort of data mining seems essential if the recent WHO publication of a minimalist 33-word summary of the last 62 cases is now the norm - or perhaps the provision of information from the South Korean health authority to the WHO has changed in format. And that has become acceptable to the WHO, who have not commented on the change. Such summaries, and lack of comment, are also business-as-usual for 'updating' us on human cases of influenza A(H7N9) virus in China.


      Wednesday, 10 June 2015

      Tracing the MERS-CoV cases in South Korea...

      The Korean Broadcasting System (KBS) News channel has made an awesome "bump map" or force diagram (thanks @Casillic) showing the layout of cases and the hospitals they passed through or were admitted to or transferred to.[1] It shows sex, deaths (9 across multiple facilities), profession, rounds of transmission and some dates (written as

      Unfortunately for some of us, it's in Korean. If you, like me, are having trouble remembering your grade school Korean, I'm here to help (a little).

      I've put together a screenshot of this awesome map as of today's count of  - no doubt the values will change tomorrow - and added onto it the hospital names in English. I highly recommend you visit the actual site though - the map is scalable and interactive. This snapshot does not do it justice and won't be updated like the map seems to be.

      I've interpreted the hospital names by eye from the Hong Kong Centre for Health Protection's (CHP) excellent multi-lingual list of all the "MERS hospitals" released by the Korean health authority.[2,3] They may not be perfect and I'd be happy to take any suggestions and corrections (including what is in the orange boxes).

      The index case, #1, is shown in green and is linked here to 2 clinics and a hospital - which differs a bit from the WHO story which includes a final move..."whereupon he was transferred to the nationally designated treatment facility for isolation".

      The most recent additions seem to have a pulsing arrow (only visible at the source).

      Made using a combination of sources. [1,2]
      Click on image to enlarge.