Tuesday, September 30, 2014

Ebola cases exceed 1000 new cases per week

Based on World Health Organization  (WHO) data, in early September I projected that new Ebola cases would exceed 1000 cases per week by September 21, 2014 (link).  As of the most recent WHO update on September 26, 2014 ( with data current through September 23, 2014), the number of new cases exceeded 1000 in week number 39, starting on September 21. See graph below. 

WHO has not provided a more recent update of the Ebola cases counts. But given the multitude of media reports about many additional uncounted cases and deaths in several of these West African nations, the WHO numbers are probably very low and don't reflect the nature and severity of the Ebola outbreak that is evolving in West Africa.

Tuesday, September 16, 2014

Ebola infections continue to grow in West Africa

The Ebola epidemic in the three West African countries of Guinea, Liberia, and Sierra Leone continues to grow. As depicted in the graph below,  there have been erratic increases in the number of new Ebola cases in these counties since week number 20 (starting May 11, 2014).

The next graph below, based on the average number of new Ebola cases per week, indicates that the longer term statistical trend in the number of cases is continued exponential increase. The variability in case data from these three countries for week numbers 37 and 38 suggests that local public health officials in these countries can no longer accurately track and count Ebola cases.  Without international support there is no reason to expect that the infection rate will decline any time soon in these countries.

Graph notes: The data used in this graph are derived from World Health Organization and are current through September 16, 2014.  Only cases from the Ebola outbreak in Guinea, Liberia, and Sierra Leone are used in this graph.

Tuesday, September 9, 2014

Is the Next Wave of MERS starting in Eastern Saudi Arabia?

The Ministry of Health in Saudi Arabia has announced two new cases of Middle East Respiratory Syndrome (MERS), both from the city of Al Jubail in eastern Saudi Arabia. The first is a 34 year-old male reported on August 29, 2014 and the second is a 60 year-old male reported on September 8, 2014. The older man is intensive care. Notably, both of these cases are health care workers which suggests that there is another hospital outbreak of MERS in Al Jubail.   Al Jubail is located about 30 miles north of Dammam where two other MERS cases have been recently reported. These 4 cases could indicate that MERS is again active in the Eastern Province of Saudi Arabia.


Monday, September 8, 2014

Seasonality Cycles of Novel Influenza Strains

It is well known that non-pandemic influenza has a seasonal repeating periodicity, especially in temperate climates [1]. The causes of seasonal cycles of influenza infections are not well understood. However, an analysis of seasonality of influenza around the world in a recent PLOS article indicates that cold-dry and humid-rainy conditions are associated with peaks in the frequency of seasonal influenza cases in different regions [2]. The authors in this article suggest that “these two distinct mechanisms account for influenza seasonality in temperate and tropical climates, perhaps due to changes in the dominant mode of transmission.”

In the past decade there have been several outbreaks of novel influenza infections, including (A)H5N1, (A)H7N9, and (A)H10N8. Based on limited data, it does appear that novel influenza infections follow the same seasonal pattern as non-pandemic influenza.

For example, H7N9 was first reported by the Republic of China to the World Health Organization (WHO) in early 2013. Since then more than 440 cases have been reported, all originating in China. The graph below shows the frequency of more than 400 WHO confirmed H7N9 human cases by month from 2013 and 2014 based on symptom onset date. From about 20 months of data, H7N9 shows a seasonal increase between December and May in China.

Case data for H5N1 infections has been accumulating for more than a decade from 15 countries. Based on WHO data, the distribution of symptom onset dates for more than 600 H5N1 cases since 2003 also shows a periodic seasonal increase in cases between December and May as shown in the graph below.

Currently, only three cases of a H10N8 have been reported, all from China (two confirmed). This is an insufficient number of cases to plot on a seasonal basis however, all three these cases were reported in period from November 2013 to February 2014.

The data suggest that even novel influenza infections seem to be constrained by the same environmental factors that control the infection cycle of seasonal non-pandemic influenza.

[1] Influenza Seasonality: Underlying Causes and Modeling Theories

[2] Environmental Predictors of Seasonal Influenza Epidemics across Temperate and Tropical Climates

Friday, September 5, 2014

By September 21, 2014 Ebola could be infecting more than 1000 people a week in Africa

A World Health Organization (WHO) official made a dire prediction about the Ebola outbreak today. Christopher Dye, director of strategy in the WHO Director General’s office, stated in a media interview

"If we make a simple projection on what has happened over the last, say, 10 weeks ... and make a projection forward, then what we're faced with is not hundreds of cases a week, which is what we see at the moment, but thousands of cases a week going into next month.”

Based on WHO data, graphing the number of new Ebola cases for each of the last 10 weeks does show the increasing rate of cases each week. After applying a trend line to the data, week number 39 starting on September 21, 2014 appears to be the week when the 1000-cases-per-week milestone will be passed (see graph below). Given the reporting delays in case numbers, this milestone may not be confirmed until a week or two later. 

While media pundits like round numbers and milestones to emphasize talking points, there is nothing unique about a week with 1000 Ebola cases versus 800 cases the week before or perhaps 1300 cases a week later. The focus should be on implementing a plan to reduce the transmission of Ebola in these areas and provide care and services to the unfortunate victims. 

Graph: Projection of the number of new Ebola cases in Africa within the next several weeks, September 5, 2014  

Graph notes: Only cases from the Ebola outbreak in Guinea, Liberia, and Sierra Leone are used in this graph. The cases from Nigeria are not included in the week totals, but would not appreciably affect the projected rate of growth of new cases. All data are from WHO.

Thursday, September 4, 2014

Ebola infections in West Africa continue to grow

In a previous post (Ebola is not yet under control in any West African country), I criticized an article that suggested that Ebola infections were declining in several countries in West Africa because the effective reproductive rate was less than 1. 

The graph below depicts the number of new weekly infections in the three West African countries (Guinea, Liberia, and Sierra Leone).  Cumulatively, the number of Ebola infections is continuing to grow. The World Health Organization has indicated today (link) that these countries have widespread intense transmission of Ebola. 

Graph notes: The data used in this graph are derived from World Health Organization and are current through August 31, 2014. Graph notes: Only cases from the Ebola outbreak in Guinea, Liberia, and Sierra Leone are used in this graph. The cases from Nigeria are not included in the week totals, but would not appreciably affect the projected rate of growth of new cases. All data are from WHO.

Tuesday, September 2, 2014

Ebola is not yet under control in any West African country

In a recent article entitled Estimating the reproduction number of Ebola virus (EBOV) during the 2014 outbreak in West Africa, Christian L. Althaus, a post-doctoral research fellow at Institute of Social and Preventive Medicine at the University of Bern, provides an optimistic and unrealistic assessment of the Ebola outbreak in West Africa. In the abstract, Althaus implies that the Ebola outbreaks in Guinea and Sierra Leone were under control by the end of May and July 2014, respectively. Elsewhere in the text he contradicts the abstract and states:
"This results in a different decrease of the effective reproduction number, Re, after the outbreaks started in each country [Guinea. Liberia, Sierra Leone]. While Re dropped below unity in Guinea and Sierra Leone by end-August 2014, the model suggests that control interventions were not successful in reducing Re in Liberia."
Based on the World Health Organization (WHO) comments that the number of Ebola cases are grossly underestimated (WHO link, WHO link), there is no evidence to suggest that as of September 1, 2014 the effective reproductive rate has dropped below 1 in either Guinea or Liberia nor is there evidence of a concomitant decrease in new cases.

Althaus also states:
"In particular, I assumed that the transmission rate decays exponentially due to control measures after the appearance of the first infectious case."
With more than 240 health care workers infected during this outbreak (WHO link) and health care facilities overwhelmed, there is no reason to assume that adequate control measures are in place. Without adequate control measures, transmission rates will not decrease exponentially in these countries in the near future and the current data do not suggest that the reproductive number in any of these countries is less than 1. At this time, the Ebola outbreak in West Africa is continuing to grow and is not yet under control.