The CDC's Ebola Predictions in September Were Reasonable

There is some talk online and in the conservative media that the predictions made by the Centers for Disease Control and Prevention back in September regarding potential Ebola cases by January 2015 were hysterical exaggerations.

Actually, they were not.  For all its failings surrounding the Ebola response – I'd argue the response was too timid, and an under-reaction instead of an over-reaction – the CDC's September projections were entirely reasonable.

Needless to say, back in September all major international media outlets were covering this CDC estimate.

The original CDC prediction from September reads as follows:

By September 30, 2014, CDC estimates that there will be approximately 8,000 cases, or as high as 21,000 cases if corrections for underreporting are made. Without additional interventions or changes in community behavior, CDC estimates that by January 20, 2015, there will be a total of approximately 550,000 Ebola cases in Liberia and Sierra Leone or 1.4 million if corrections for underreporting are made.

This projection came from a CDC report also released in September entitled "Estimating the Future Number of Cases in the Ebola Epidemic – Liberia and Sierra Leone, 2014-2015."  The authors used Ebola case data up to August 31, 2014.

Anyone can download the cumulative reported Ebola case data for Liberia and Sierra Leone from the CDC.  Between July 28 and August 28, the number of Ebola cases in these two countries increased by a factor of 3.1 from 774 to 2,404.  The rate of increase between the end of June and end of July was also about the same.  Thus, when this CDC report was written in early September using data from up to the end of August, Ebola cases in Liberia and Sierra Leone were consistently increasing by a factor of about 3.1 per month.

If you extrapolate this rate of increase (once again: a factor of 3.1 per month), you arrive at a projected number of Ebola cases in Liberia and Sierra Leone of 690,000 by the end of January, or about 550,000 by January 20 – just as the CDC was projecting.

Ebola cases were, and still are, likely being under-reported.  At the time of the CDC's September report, the assumption was that the under-reporting factor was 2.5, which is where the 1.4 million number comes from (i.e., 550,000 multiplied by 2.5).

The growth in Ebola cases during September was starting to slow slightly, but even by mid-September the increase was still at a factor of 2.4-2.6 per month, which is well within the margin of measurement error for the end of August modeling efforts, given the difficulty in getting reliable up-to-date case data out of West Africa.

As it has turned out, the number of Ebola cases in these two nations is well below the worst-case projections (currently at about 16,000, although the real number could be up to severalfold higher).  This is where the CDC's statement that "without additional interventions or changes in community behavior" applies.  There was already a number of "interventions or changes in community behavior" underway.  These responses to the outbreak have continued up to the present and have mitigated the potential worst-case impacts.

There is merit in criticizing many aspects of the CDC's response, and especially the inferior responses of other organizations such as the World Bank and World Health Organization.  But based on the data at hand in early September, the CDC was correct to report that Ebola cases in Liberia and Sierra Leone could – emphasis on "could" – reach upwards of 550,000 to 1.4 million by early 2015.

There is some talk online and in the conservative media that the predictions made by the Centers for Disease Control and Prevention back in September regarding potential Ebola cases by January 2015 were hysterical exaggerations.

Actually, they were not.  For all its failings surrounding the Ebola response – I'd argue the response was too timid, and an under-reaction instead of an over-reaction – the CDC's September projections were entirely reasonable.

Needless to say, back in September all major international media outlets were covering this CDC estimate.

The original CDC prediction from September reads as follows:

By September 30, 2014, CDC estimates that there will be approximately 8,000 cases, or as high as 21,000 cases if corrections for underreporting are made. Without additional interventions or changes in community behavior, CDC estimates that by January 20, 2015, there will be a total of approximately 550,000 Ebola cases in Liberia and Sierra Leone or 1.4 million if corrections for underreporting are made.

This projection came from a CDC report also released in September entitled "Estimating the Future Number of Cases in the Ebola Epidemic – Liberia and Sierra Leone, 2014-2015."  The authors used Ebola case data up to August 31, 2014.

Anyone can download the cumulative reported Ebola case data for Liberia and Sierra Leone from the CDC.  Between July 28 and August 28, the number of Ebola cases in these two countries increased by a factor of 3.1 from 774 to 2,404.  The rate of increase between the end of June and end of July was also about the same.  Thus, when this CDC report was written in early September using data from up to the end of August, Ebola cases in Liberia and Sierra Leone were consistently increasing by a factor of about 3.1 per month.

If you extrapolate this rate of increase (once again: a factor of 3.1 per month), you arrive at a projected number of Ebola cases in Liberia and Sierra Leone of 690,000 by the end of January, or about 550,000 by January 20 – just as the CDC was projecting.

Ebola cases were, and still are, likely being under-reported.  At the time of the CDC's September report, the assumption was that the under-reporting factor was 2.5, which is where the 1.4 million number comes from (i.e., 550,000 multiplied by 2.5).

The growth in Ebola cases during September was starting to slow slightly, but even by mid-September the increase was still at a factor of 2.4-2.6 per month, which is well within the margin of measurement error for the end of August modeling efforts, given the difficulty in getting reliable up-to-date case data out of West Africa.

As it has turned out, the number of Ebola cases in these two nations is well below the worst-case projections (currently at about 16,000, although the real number could be up to severalfold higher).  This is where the CDC's statement that "without additional interventions or changes in community behavior" applies.  There was already a number of "interventions or changes in community behavior" underway.  These responses to the outbreak have continued up to the present and have mitigated the potential worst-case impacts.

There is merit in criticizing many aspects of the CDC's response, and especially the inferior responses of other organizations such as the World Bank and World Health Organization.  But based on the data at hand in early September, the CDC was correct to report that Ebola cases in Liberia and Sierra Leone could – emphasis on "could" – reach upwards of 550,000 to 1.4 million by early 2015.