I’m a public health researcher focused on HIV, with a background in Social and Infectious Disease Epidemiology.  I am currently a postdoctoral fellow at Harvard T.H. Chan School of Public Health, but this blog is my personal space and should not be construed to represent HSPH in any way.  Rather, this is an outlet for my surfing of PubMed, with occasional digressions into actual thoughts on topics.  And recently quite a lot of material on Ebola.

My primary research focus is on how social and economic structure limits choices and pushes (I guess nudges is the sexy term these days) people towards poor health outcomes.  Hence the blog title.  Of course, in a former life I used to work in Economics.

If you would like to learn more about my work, you might like to visit my website.

And if you happen to find materials on this site useful, and would not mind dropping me a line, it would be lovely to know what people are doing with the information I throw up here. Such feedback might even improve future content, you never know.

Guy Harling

2 thoughts on “About

  1. Hello I enjoyed your article concerning the cdc report and their attempt at estimating the under reporting factor. I could well be missing something but I dont interpret the cdc report in the same way that your article and every other Ive seen on the subject does. As this is a very important issue I thought that it must be important to correclty understand this critical issue. As far as I can see the cdc report first models the outbreak assuming that everyone infected shows up for treatment. It calculates how many beds would be required to treat those patients. It then looks at the number of beds that experts estimate are currently in use/ occupied and based on that number of beds it calculates how many people are currently infected. Then working backwards it can estimate the cummulative of “known” cases to date. It compares this to the official figures and finds it to be to “out” by a factor of 2.5. This is important becasue what its saying is that what is being under reported is the number of “known” cases. That the reported government figures are too low by a factor of 2.5. That many cases dont get access to treatment, through choice or capacity is an entirely different issue and in that regard the key table in the report states that this number is simply “unknown”. This would mean that there is a much larger under-reporting factor comprising the “known” cases that arent being reported (the cdc 2.5 figure) plus “unknown” cases that have never got inside a treatment centre. Lets face it, the reason that we arent able to get on top of this outbreak is that we have continually under estimated its scale and apparently continue to do so. Also consider this, that the maximum number of reported cases under the current scheme will be limited by the capacity of the treatment centres. i.e. if the method of reporting is linked to cases being accomodated in treatment centres then all we are reporting is some function of the increase in capacity.

    An example of the failure to report community cases – the apparently 90 bodies found during the recent SL lock down, most likely ebola victims- those deaths havent showed up on the SL tally. All very worring on so many levels.

    • Paul, thank you for your comment. I think you are right that the CDC approach does not consider “unknown” cases, but it is hard to know how much adjustment to make for this, since by definition the size of the problem is unknown. For example, there were 92 (or maybe more?) bodies found during the lock-down, but we don’t have data on how many of these were due to Ebola, how many to other illnesses that couldn’t be treated because of Ebola, etc.

      Additionally, there are many other possible biases just in estimating the size of the “known” unreported cases. And these biases are almost certainly changing over time as the epidemic progresses: which is actually the worst thing, since it makes it hard to tell if the epidemic is truly getting worse or just reporting is getting better (i.e. if we knew we were always out by a factor of 2.5, at least we could tell how the situation is changing over time).

      For me, we do the best we can, we try and state our assumptions, and then others can test those assumptions and make better predictions. And it’s great when researchers provide open-source or tweakable models (as the CDC does) so we can see how things would look under our best-guess scenarios.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s