While one should always be careful when looking at composite data – this dataset is the sum of country sums – the graphical representation of the Ebola number of cases seems to have a confirmed reduced increase as the curve appears to flatten off. The same trend is not as clearly visible in the total number of deaths as yet.
The data series of WHO reported Ebola cases and deaths consolidates its linear increase versus time with little sign of acceleration or flat topping.
In terms of the KPIs for Guinea, Liberia and Sierra Leone, there is progress as the results as of November 23rd 2014 are
3 goals at 100%
8 goals above 50% of target
9 goals below 50% of target.
The November 2nd 2014 WHO Ebola Report is particularly interesting in several respects:
- the total number of cases is revised downwards
the total number of deaths is also revised downwards from 4951 to 4818
there is a clear warning about accuracy of the data reported: “Cases and deaths continue to be under-reported in this outbreak” highlighting the difficulty in obtaining accurate data
Ten key performance indicators (KPI) have been introduced. Glancing at the table of results of these KPIs we see
a. Three KPIs are below 50% of the goal (target)
b. Four KPIs are above 50% of the goal
c. Only 1 KPI has met the 60 days target which is 70% of the goal for that particular KPIs
In other words, across all measurable actions we are significantly short of the plan at the time of the last report.
We now have points of inflection in the WHO data on Ebola (source: “Ebola Response Roadmap Situation Report”), as the graph of total cases and number of deaths is no longer roughly linear versus time:
- the total of cases has peaked upwards
the rate of total number of deaths has flattened.
There was no specific comment in the WHO report about the trend that one can observe from these charts, except the mention that the total number of cases was corrected down in the last report as a number of “suspected” cases (reported earlier) in Guinea have been discarded.
The UN’s WHO report of October 15th 2014 tells us that we have 8977 cases of Ebola with 4493 deaths recorded as of October 12th. The plot below shows that the timeline continues to present a linear trend and there is no significant point of inflection either up or down as far as the globally reported data is concerned.
It appears that, from press reports, the newly affected countries are getting more organised, more communicative and assigning more senior and credible figures to coordinate responses. However, at EU level, it seems that sense of urgency was lacking as the group of responsible public representatives could not agree on implementing temperature measurements systematically at borders. (And contrary to what some ignoramus radio announcers have said, you don’t need to stick a thermometer in every disembarking passengers, a profesional thermal camera does an excellent job monitoring a flow of people with sufficient precision and accuracy, at least for 1st level of detection)
The reporting is getting more structured and a series of acronyms are getting established:
- EVD: Ebola Virus Disease
- ECN: Ebola communication network – a collection of Ebola resources from UNICEF, CDC, USAID, IFRC, WHO
- UNMEER: UN Mission for Ebola Emergency Response, headquartered in Accra (Ghana)
and now 3 groups of countries are appearing for tracking data:
- The widespread and intensive transmission nations: Guinea, Liberia, Sierra Leone
The initial case or localised transmission nations: Nigeria, Senegal, Spain and the United States of America
The neighbours of active transmission nations: Benin, Burkina Faso, Côte d’Ivoire, Guinea-Bissau, Mali and Senegal.
This said, I, for one, still do not understand the absence of decision concerning the restriction of movement to/from areas of intense transmission given the obvious proximity effect in transmission.
While there is no shortage of news coverage of the Ebola virus disease, I found some amazing contradictions. While we hear “we follow protocol” from health authorities, if you look at the data it is hard to see any improvement. This is the data I put together from total cases and total deaths from the Ebola response roadmap updates from United Nations’ WHO. It keeps going up. Yes up.
Now, those of us who practice the “Engineering systems approach” or manage critical operations or are safety executives know well that in the presence of any incident, there is a need to elaborate and execute 3 responses:
- a containment plan – to stop propagation of the damage to other targets
a corrective action plan – to fix the damage that occurred
and a preventive action plan – to avoid reoccurrence of the incident.
What is astounding in terms of containment, is that we have witnessed the oposite of precaution: the flying to Madrid Spain, twice in a row, of confirmed cases. Was Madrid known as a world class Ebola treatment center? No, there was no logic. Now, in spite of “following protocol”, the virus is on the loose with at least one local medical staff confirmed infected and who knows how many suspected.
Now, from years of running operations in “hostile” or “dangerous” circumstances (oil and gas exploration, semiconductor manufacturing), I’ve learned that no serious entity allows you to operate unless you can demonstrate you have redundant lines of defence (3 in offshore operations), so if one fails you have backup and backup. In other words we do not trust one protocol, we need more and multiple layers of safety. Why: 1 – many rules, even in physics or engineering, are simplifications of gaussian distributions (bell curve) meaning that some rare cases will fall outside the range where the “rule” is valid, 2 – noise in systems i.e. there is a difference between what happens on the blackboard and the real world, 3 – accidents, errors and omissions.
Even if the probability of contamination is low, the impact or severity of it is quite high, therefore the mitigation plans should be far more comprehensive than the mere “following of protocol”. This is not enough, we need deciders that have the courage of implementing a series of redundant measures in parallel. Why no travel bans for example?