Sunday, 15 June 2014

Epidemiology without dates is just -ology...

Click on image to enlarge.
Data from the Kingdom of Saudi Arabia Ministry of Health MERS website.[1]

The above images give an indication of what the Kingdom of Saudi Arabia's Ministry of Health (MOH) considers to be case detail of immediate relevance to the public according to Prof. Tariq Madani, head of the scientific advisory board within the MOH's Command and Control Centre (CCC). 


Since the newly revised CCC MOH website came online, dates that describe key information have been absent (red dashed outline in the screen capture above). These dates include:
  • Date when symptoms began
  • Date of hospital admission
  • Date when a new fatal case was first reported (allowing a link to be made)
  • Click on image to enlarge.
    Where the re-defined 113 cases fall out across 2013 to
    1st week of May, 2014.[3]
  • Date when a recovered case was first reported (allowing a link to be made; actually this one hasn't ever been present)
Sure, we only had these dates for a short period, and relied heavily on the World Health Organization's (WHO) Disease Outbreak Notifications (DONs) to fill in and "scrub" the data once it was submitted to them, but it was so great while it lasted. 

I have had many emails and Tweets noting how useful it's been to others to see those data distilled into more digestible graphs and charts. These notes have come from both the public and from other scientists too.

This latest change to the MERS information provided us, came after a report on 3-June [2,3]. That is also the entry between where there were dates included [orange] and where dates stopped being included [red] again, in Part A above. The report described a review which started in May and found 113 un(publicly)reported MERS cases (see the bar chart above). Interestingly, 55 of these cases were either not sent to, or were not confirmed by, KSA governmental laboratories (thus should be better listed as probable cases IMO) while 58 results just hadn't been sent to the MOH, who "sign off" on the final reporting of MERS-CoV detections. Great that the CCC is able to track these down and admit to them. Good work. 

Not so great that 33 of the 55 cases are included in this tally despite not being able to be confirmed. 

In response to the "found 113", I halted my charting activities for MERS-CoV that week. I won't be resuming charting until the very long-awaited WHO's DON fills in the gaps on over 400 cases[6].

Much has been made by the media, some of it with comments from me[2,8], of these events. Some have noted that there is now a new sense of transparency about. While I don't argue with reports and comments about a range of collaborative efforts now/already underway (which is great news) and the need to acknowledge things when they improve, I do question how the retraction of some key data, essential for public epidemiology efforts, data that were fine to be included (inconsistent as they were even then) previously, can be labelled as transparency. I certainly don't think those actions meet up to this statement (bolding is mine)[3]...
"Based on the findings of the review, the Ministry has already put in place a number of measures to ensure that best practices of data gathering, reporting, transparency are being strictly observed.."
I've stated before, for H7N9[4] and for MERS[5] that we, the public, do not have a right to such data, even though its deidentified and the privacy of the patient is protected. We may feel entitled, but we are not. As long as the WHO and appropriate Committees or experts are aware of the facts and can judge the risk to the world, then that is the main issue.

Of course, I'd much rather we lived in a world where such relatively innocuous data were available, and complete. But whether or not we get to play with useful and rich data is a separate issue to the difference between saying something is so and it actually being the case. 

There has been an increase in the presentation quality (prettiness) of data but a decrease in the data presented, since the CCC website came online. That's a fact.