Colombia Zika virus report, Epidemiological Week No. 33...

UPDATE No.1: 29AUG2016

The latest epidemiological report from Colombia, which includes data on Zika virus disease (ZVD; 14AUG2016-20AUG2016), has been produced by the Colombian National Institute for Health team.[1]
NOTE: While these data were reported the past epidemiological week, they may not be from that week. See earlier post about possible reporting lag.

Graph No.1. The cumulative curve of confirmed ZVD cases
(green circles, left-hand axis) and the change in confirmed ZVD case
numbers when compared to the preceding week's total
(green bars, right-hand axis). Data from [1].
Click on graph to enlarge.

Graph No. 1 shows that 0 new laboratory confirmed cases of ZVD were reported this week; the 6th week of this. The total sits at 8,826 or 9% of all clinically suspected Zika virus (ZIKV) detections.

Graph No.2. The cumulative curve of suspected ZVD cases
(pink circles, left-hand axis) and the change in suspected ZVD case
numbers when compared to the preceding week's total
(red bars, right-hand axis). Data from [1].
Click on graph to enlarge.

Graph No. 2 shows the change in suspected cases. These are not laboratory confirmed. The suspected ZVD cases continue to rise but have been slowing for weeks, adding 270 this week to total 93,551.

Graph No.3. The cumulative curve of clinically suspect (lilac triangles, left-hand axis) and confirmed ZIKV infections (lilac circles, left-hand axis) and the change in
confirmed ZIKV infection numbers when compared to the preceding week's total
(purple bars, right-hand axis). Now added the reported number of microcephaly cases
confirmed as ZIKV infected (yellow bars, right-hand axis). Data from [1].
Click on graph to enlarge.

Graph No. 3 shows that to epidemiological week (EW) No. 33, 11,752 suspected (+30 compared to last week) and 6,054 confirmed ZIKV infections (-2) have been identified in pregnant women.

As of this report, 34 (+5 from last EW) live births have been diagnosed with congenital ZIKV syndrome (CZVS; microcephaly/central nervous system disorder), confirmed as being ZIKV positive. That represents 0.53% of all confirmed ZIKV positive mothers-the 4th consecutive EW in which this proportion has risen.

Some back of napkin calculations looking at these numbers suggest that there are 5-6 deliveries for every 1,000 ZIKV-positive pregnant women that result in a ZIKV infected baby with microcephaly. This assumes each neonate has been tested for ZIKV as [6] suggests. This figure has no clear understanding of the number of aborted or miscarried foetuses that are also occurring from ZIKV-positive pregnant women. Abortions and miscarriages will need a local baseline to understand the scope of this component of the impact of ZIKV infection.

209 other microcephaly diagnoses (down from 254 last week) are now under investigation - this value has also been rising very quickly and suggests suspicious CZVS cases in Colombia are accruing faster than the pace of complete investigation can keep up with. 

It now seems very likely that we can expect those bars to keep rising steeply in the coming weeks. The line is well and truly crossed.

Graph No. 4 below focuses on just the ZIKV-positive cases and those that remain under investigation, highlighting how the investigatory total has changed each week and been trending upwards since Epidemiological Week No. 14 and rising by +5 for 2 weeks in a row.

Graph No.4. The change in confirmed ZIKV infection numbers
when detected in association with a microcephaly diagnosis, compared
to the preceding week's total (yellow bars, left-hand axis). Data are from [1].
Click on graph to enlarge.

It has now been 317 days, or 10 months 12 days, since ZIKV was first confirmed in Colombia on 16th October 2015.[2] Colombia is currently carrying the next biggest load of suspected ZVD cases [8] (although more confirmed in Puerto Rico. NB: not all of the "Casos confirmados" are lab confirmed, but I'm told most are).[3] Keep in mind that when talking about microcephaly - we have to think back in time to what insult or infection might have occurred in the first or second trimester (probably-still not definitive). The counts of virus occurring this week will have zero impact on what happened back then. Also keep in mind that Colombia may be reporting things differently from Brazil.[5,7]

Brazil first reported positive (but unconfirmed) laboratory tests for Zika virus disease on 29th April 2015. Brazil then started to report a rise in foetal anomalies (an initial 141), in the form of microcephaly on 30th October 2015. This was 184 days - or about 6 months later.[4]


References...

1. http://www.ins.gov.co/boletin-epidemiologico/Boletn%20Epidemiolgico/2016%20Boletin%20epidemiologico%20semana%2033.pdf
2. http://www.who.int/bulletin/online_first/16-171082/en/
3. http://www.nature.com/news/first-zika-linked-birth-defects-detected-in-colombia-1.19502
4. http://who.int/bulletin/online_first/16-171082/en/
5. http://virologydownunder.blogspot.com.au/2016/06/the-elephant-squeaked.html
6. Zika Virus Disease in Colombia — Preliminary Reporthttp://www.nejm.org/doi/full/10.1056/NEJMoa1604037#t=article
7. https://www.statnews.com/2016/06/22/zika-abortion-latin-america/
8. http://www.salud.gov.pr/Estadisticas-Registros-y-Publicaciones/Pages/Informe-Arboviral.aspx

Updates...

1. Clarified Puerto Rico has more confirmed cases than Colombia but fewer suspected

Brazil's microcephaly and CNS disorder (M&CD) monitoring: Report No. 27, 2016-Week No. 20...

These graphs are made by me using data obtained from epidemiological week (EW) number 20's Brazil Ministry of Health microcephaly and foetal and infant microcephaly and central nervous system (CNS) disorders (M&CD) report.[1]

Brazil last reported a total of 120,161 suspected Zika virus detections some weeks back. Around one thousand of these have been portability confirmed.[2]

Suspected M&CD cases...

The total number of suspected M&CD cases increased by 89 to 7,623 this EW (compared to last).

The graph above shows the number of suspected M&CD diagnoses in Brazil up to 21-May-2016. The cumulative curve (yellow dots; left hand axis) is growing, but slowly.

This was another slower weekly rise (orange bars; right-hand axis). These bars are based on the difference in total suspected cases reported this EW compared to that reported in the last EW. This method may not reflect the diagnoses that occurred during the past EW (some may have come from days or weeks earlier), but that level of detail is not available in the MOH report.

Confirmed and discarded M&CD diagnoses...

M&CD cases under investigation decreased by 75 to 3,257 this week - the ninth consecutive decrease.

In the graph above, we can see that 114 (blue bars; right hand axis) suspected M&CD diagnoses were discarded upon closer investigation with a current total of 2,932 removed.

The rate of these resolved diagnoses (line with blue dots, left-hand axis) continues to outpace the rate of the smaller overall number of confirmed M&CD diagnoses (red dots, left-hand axis).

As of this EW, 19% of suspected M&CD diagnoses have been confirmed while 38% of suspected diagnoses have been discarded-a percentage that has increased for 16 weeks.

The cumulative number of confirmed M&CD diagnoses does continue its climb this EW, growing by 50 new diagnoses (red bars; right-hand axis) to total 1,434.



The number of these M&CD diagnoses to be confirmed with a Zika virus infection also grows (green dots; left-hand axis) by 1 new detection (green bars; right-hand axis) to 208 this EW.

Those confirmed Zika virus infections represent 15% of all confirmed M&CD diagnoses and 3% of all suspect diagnoses.

References...

1. http://portalsaude.saude.gov.br/images/pdf/2016/maio/24/Informe-Epidemiol--gico-n---27--SE-20-2016--24mai2016-12h00.pdf
2. http://combateaedes.saude.gov.br/images/sala-de-situacao/informe_microcefalia_epidemiologico26.pdf

Brazil's microcephaly and CNS disorder (M&CD) monitoring: Report No. 24, 2016-Week No. 17...

These graphs are made by me using data obtained from epidemiological week (EW) number 17's Brazil Ministry of Health microcephaly and foetal and infant microcephaly and central nervous system (CNS) disorders (M&CD) report.[1]

Brazil reports a total of 120,161 suspected Zika virus detections this EW.[1] Around one thousand of these have been laboratory confirmed.

Suspected M&CD cases...

The total number of suspected M&CD cases increased by 115 to 7,343 this EW.

The graph above shows the number of suspected M&CD diagnoses in Brazil up to 30-April-2016. The cumulative curve (yellow dots; left hand axis) is growing, but slowly. 

This was another slower weekly rise (orange bars; right-hand axis). These bars are based on the difference in total suspected cases reported this EW compared to that reported in the last EW. This method may not reflect the diagnoses that occurred during the past EW (some may have come from days or weeks earlier), but that level of detail is not available in the MOH report.

Confirmed and discarded M&CD diagnoses...

M&CD cases under investigation decreased by 130 to 3,580 this week - the sixth consecutive decrease. This may be occurring for a range of (guessed) reasons including:

1. improved capacity to address a large volume of request for clinical classification of suspect diagnoses.
2. streamlining the confirmation processes
3. retrospective application of changed head circumference definitions resulting in reductions to the number of the microcephaly diagnoses.
   This aspect will not reduce the number of cases with structural brain changes that occur in the absence of a decreased head size - which are reportedly also related to Zika virus and/or other causal influences.
   There is time involved in making these diagnoses because of the need for detailed ultrasound and other diagnostic investigations to identify congenital infection outcomes including intracranial calcifications, dilation of cerebral ventricles or posterior fossa changes and other issues.

In the graph above, we can see that 172 (blue bars; right hand axis) suspected M&CD diagnoses were discarded upon closer investigation with a current total of 2,492 removed.

The rate of these resolved diagnoses (line with blue dots, left-hand axis) continues to outpace the rate of the smaller overall number of confirmed M&CD diagnoses (red dots, left-hand axis). 

As of this EW, 17% of suspected M&CD diagnoses have been confirmed while 34% of suspected diagnoses have been discarded. 

The cumulative number of M&CD diagnoses does continue its climb this EW, growing by 73 new diagnoses (red bars; right-hand axis) to total 1,271. 

The number of these M&CD diagnoses to be confirmed with a Zika virus infection also grows (green dots; left-hand axis) by 9 new detections to 203 this EW (green bars; right-hand axis). Those confirmed Zika virus infections represent 16% of all confirmed M&CD diagnoses and 3% of all suspect diagnoses - but these are not fair comparisons for a range of reasons I won't go on about here.

References...

1. http://portalsaude.saude.gov.br/images/pdf/2016/maio/04/coes-microcefalia-informe-epi-24-se17-2016.pdf

MERS-CoV detections by week...

Yeah, I got bored waiting. 

I finally came around to the fact that the data for MERS-CoV have always been so inconsistent that why worry about that now. So what about dates? I have no idea how long it will be until we see the WHO post data with dates for the hundreds of Kingdom of Saudi Arabia (KSA) cases we know nothing much about that occurred in the Jeddah-2014 healthcare outbreak. The WHO without Gregory Hartl is a much less interactive entity on social media these days. One person can make such a difference.

As it stands, the KSA Ministry of Health has clearly decided to hold fast on not releasing dates for symptom onset, hospitalisation and death for new cases; the new and improved Command & Control Center (CCC) colour scheme speaks for the data....instead of having the data....it seems.

Click on chart to enlarge.

So, the chart. It shows the end of the outbreak in western KSA...still...having been that way for a few weeks. Its hard to tell exactly when the cases have fallen out for the last 13 days or so, since we don't know when they became ill (which is what I use most often when plotting cases-and will again should those data ever become available). So take those dots over the last month with a grain of salt for now. I expect they will reduce in number as they sort into the preceding week(s). 

As I note in the chart - I do not include the "found113" cases here. They remain dodgy data for now. I'll also wait for the WHO to make sense of them.

H7N9 cases by week...the other bird flu is done nesting [UPDATED]

Click on image to enlarge.
H7N9 cases by week, worldwide, from Week
beginning Monday 30.09.13 to 06.01.14.
Sources are publicly available data. 

A quick plot of case occurrence to the best of my ability given the variability in the data. Follow the left-hand column down then back up to the top of the right-hand column and down again to follow the 22 cases and 2 deaths through time.

I started with Week #33 (33rd week of H7N9; given that Week#1 starts from Feb-18; the week in which symptoms were reported for the 87-year old male index case on Feb-19 [updated]). Prior to Week #33, things were very quiet going back to April-2013. A couple of cases in July. 

These charts will change as we get more cases and more details and perhaps, more fatalities among these cases. There are between 152-158 H7N9 lab-confirmed cases, including fatalities.

The main message here is that H7N9 case numbers are on the rise again. 

It was in March and April 2013 when we saw the greatest number of H7N9 cases, but given that we now have specific PCR-tests and other assays available combined with a far greater number of "launch sites" this year, I expect we'll see increased numbers of human cases from here on in. That could change if drastic control measures are taken of course.

We have also learned from the recent case in Canada, that H5N1 is probably entrenched in areas that are not being tested/reported (see CIDRAP article). There is no doubt in my mind that H7N9 will be in the same nest, circulating in at least the 12 provinces or municipalities of mainland China from which we have seen human cases arise over the past 49-weeks. 

Speaking of which, there have been very few deaths in this approximately 15-week block. The overall proportion of fatal H7N9 cases was around 30% but among the subset of cases here, only 9.1% have died after H7N9 infection. 

Have things changed? 

I can't imagine the virus has changed since it seems most acquisitions are from poultry and there would not be evolutionary pressure in a bird to transmit more efficiently on the off-chance it was acquired by a human! So does this mean...

• Case management has improved?
• Are more cases recovering and being discharged?
• Are we still in that horrible waiting period when cases are in intensive care awaiting their fate?
• Are we simply not getting details about deaths and discharges from China any longer? 

We were certainly missing information about which cases died towards the end of the first wave of H7N9 - my database can only identify 37 of the predicted 48 deaths. These details have not yet appeared in the scientific literature so any averages (sex, or age for example) calculated from public sources are far from perfect. I presume Chinese researchers have those data.

MERS-CoV by the numbers: recent weekly case activity...

Click on image to enlarge.

Confirmed MERS-CoV cases including 

fatal infections (green) and deaths (red) 

each week. Case numbers are listed on the 

y-axis (side), days of each week  along 

the x-axis (bottom). Case dates are 

derived from announced date of onset 

but if absent, on the date of reporting. 

Deaths are listed by date of death.

This follows on from my previous post (you can track links to earlier weekly charts) about lab-confirmed Middle East respiratory syndrome coronavirus (MERS-CoV) cases, plotted by week.

Approximately 85% of all cases have come from the Kingdom of Saudi Arabia. Around 63% of all cases with sex data are male (1:1.73, M:F). Among the fatal cases with data it's 75% male (1:3.06, M:F).

I've added in some previous charts because as the new cases and case details appear, so do the placement of the cases alter slightly. Hopefully, with WHO doing such a good job in providing details, these graphs will solidify and we can move on in the next post. 

I've marked in the Hajj week and the 14-day outer limit of the incubation period. Nothing much to see from that; no spike in cases, even after time has passed to allow the case data to catch up.

My tally suggests 161 cases (still awaiting the Spanish case to be confirmed or not) with 68 deaths, a proportion of fatal cases sitting at 42%.

 

A couple of things stand out to me from these charts...

1. What is the lag between illness onset and MERS-CoV case announcement like? For example the recent 37-year old man who died was reported on 20-Nov, but became ill 9-Nov. Obviously there is time required to reach hospital (13-Nov) and then be tested and re-tested [confirmed] but this case died 18-Nov and was not reported as MERS-CoV case for 2-days. The case before that, a 65-year old man became ill 4-Nov, was in hospital 14-Nov and was reported 19-Nov. Before that the 73-year old woman became ill 13-Nov, hospitalized 14-Nov, died 19-Nov and was also reported 19-Nov.
   I presume that this indicates there is no active MERS-CoV PCR screening of influenza-like illness, but rather for  in the Kingdom of Saudi Arabia?
2. There have been 5-8 fatal cases per month since June and 16-25 cases per month in total. However, with that lag, there may be more to come from November. 

In particular, point #2 makes me wonder if the KSA is settling in to stable (albeit very small numbers of total cases) transmission or acquisitions of MERS-CoV?

MERS-CoV by the numbers: recent weekly case activity...

Click to enlarge.

Confirmed MERS-CoV cases (green) and

deaths (red) each week. Case numbers

are listed on the y-axis (side), days of each

week along the x-axis (bottom). Case dates

are derived from announced date of onset

but if absent, on the date of reporting.

This follows on from my previous post (you can follow its links to earlier weekly charts) about lab-confirmed Middle East respiratory syndrome coronavirus (MERS-CoV) cases, plotted by week.

These charts are based on the reports that get into the public domain. Those of us trying to follow and deconvolute MERS-CoV case information pretty much all agree that the data are terrible, but they are what they are. For example, the data are currently absent details on 2 MERS-related deaths for which no links to our case lists can be found. There are 85 cases missing date of disease onset data (which makes these charts imperfect), 8 without an age, 10 without a sex and most have no date of hospitalisation or date of lab confirmation. As I've bemoaned before, a standardised numbering of cases would be helpful too. Among other things.

The charts suggest there is not a lot of activity in terms of new cases, and the number of deaths, thankfully, remain much lower than during the weeks preceding my post in early September.

You can see that during the Hajj (13th-18th of October), there were 3 cases and a death described but in the weeks immediately afterwards, there has been no spike in cases of MERS. What makes this a significant development is that, for the second time since we learned of MERS-CoV, countries outside the Kingdom of Saudi Arabia (KSA) have had a direct hand in the observation and testing of pilgrims. This adds some confidence that severe symptomatic MERS is a relatively rare disease and one that does not spread quickly and efficiently. We have no real data to say that virus doesn't spread quick, widely and efficiently however, just that the severe infection outcomes don't.

So, no sign of a major jump in new cases. In this 4-week period there have been 13 cases which is up 2 from the 4-weeks before that. There have been 3 deaths (PFC of 23.1%, well below the total average of 41.3%) in this period, down from 4 in the previous month.

These numbers still have to be considered with care. This week's Spanish case really shone a light on the issue of laboratory unconfirmed cases of clinically diagnosed pneumonia circulating in the KSA. And where there is one such case there are likely to be others. Many others? We don't know.

So with 155 cases and 64 deaths in 87-weeks, MERS seems to be ticking along, but it shows no signs of becoming a widespread health issue. While it has a PFC of 41%, that is a meaningless number until we start testing more widely than is being done now. There are still many questions to answer about its host, how humans acquire it, whether its widespread in the community - but on the topic of transmission, MERS does not look likely to become a pandemic any time soon.

Recent MERS-CoV weekly case activity...

This follows on from my last MERS-CoV-by-week post, September 19th. 

As you'll see, its been a comparatively quiet 3-weeks (3rd week is not yet over of course).

With the hajj starting, that should be good news.

These data include today's (the 10th of October for me Down Under) 2 new cases (both fatal; FT#142 and FT#143; see Mike Coston's post) which raises the proportion of fatal cases to 43%. 

MERS-CoV detections over the past 6-weeks: 38 cases, 13 deaths.

Click to enlarge.
Laboratory confirmed MERS-CoV
cases (including deaths; green) and
deaths (red) by day (bottom, x-axis), per
week. Number of cases on
the left hand (y) axis peak at 8/week

Updating the Middle East respiratory syndrome coronavirus (MERS-CoV) graphs from just over 2-weeks ago and adding in recent weeks, we see how the cases have been accruing. 

Last week was a big week; 42% of cases from the past 6-weeks occurred then.

There are some differences in some charts when comparing to the earlier post with some of these; put that down to updated dates due to extra data being released and some cases being reported the week(s) after they occurred. I'll keep updating this figure. Those changes may keep happening.

MERS-CoV by the numbers: recent global case numbers by week

Click to enlarge.
Confirmed MERS-CoV cases (green) and 
deaths (red) each week. Case numbers are
listed on the y-axis, days of each week along
the x-axis.

While the week has not ended, it is interesting to sum up where we are in terms of MERS case announcements. During the week beginning August 11th, there were no new MERS-CoV case reports (I'm using reporting data not disease onset because the latter is appalling sparse).

As you can see on the right, in the following 2.5-weeks, things have heated up a lot. 17 cases so far. 

Why is this so? Completely unknown. Could it be clusters due to a few single but distinct spillovers from the natural host or perhaps transmission events from a common/shared source (recent umrah)? We can only guess.

If the local public health officials can identify this sort of info, it could provide a way to put a lid on current and future rounds of MERS cases and outbreaks.