Institute for Ethics and Emerging Technologies
IEET > Rights > HealthLongevity > GlobalDemocracySecurity > Vision > Bioculture > Contributors > Andrew Maynard > Biosecurity

Print Email permalink (1) Comments (9039) Hits •  subscribe Share on facebook Stumble This submit to reddit submit to digg


Measles mortality rates – 2008-2011 outbreak, France


Andrew Maynard
By Andrew Maynard
riskscience.umich.edu

Posted: Feb 5, 2015

Yesterday, I posted a piece examining the oft-quoted mortality rate for measles of one to two deaths per thousand cases of infection.  Today, I want to look at what can be learned from more recent and more comprehensive dataset – this one from the 2008-2011 measles outbreak in France.

In the early 2000’s, measles was a relatively rare occurrence in France.  From 2008 to 2011 though, there was a dramatic increase in cases – peaking at over 3,000 new cases per month being recorded in 2011.  Because the outbreak occurred in a developed country where the disease was no longer considered a pressing pubic health issue, it provides a unique opportunity to estimate mortality rates following infection by the virus in economies with robust healthcare systems.

In 2013, Denise Antona and co-authors published a comprehensive assessment of the outbreak in the Centers for Disease Control and Prevention journal Emerging Infectious Diseases.   Over the four year study period, there were 22,178 documented cases of measles.  11.6% of cases (2,582) involved complications , including pneumonia (1,375 cases, 6.2%), acute otitis media (321 cases, 1.4%), and hepatitis or pancreatitis (248 cases, 1.1%). According to the paper’s authors, diarrhea was reported in 100 cases (0.4%). Overall, there were ten deaths reported (0.05%).

The data are particularly useful for examining morbidity and mortality rates associated with measles in a developed country like France, as with the relative novelty of the disease, the number of reported cases is likely to have been substantially higher than in earlier decades when the disease was commonplace.

In table 1 below (based on Antona et al.’s paper), the number of measles-related complications per 10,000 cases of infection is given for different health impacts and age ranges, based on individuals who were hospitalized.

Focusing specifically on mortality, the overall rate was 4.5 deaths per 10,000 documented cases of measles. The rate was highest amongst individuals 30 years old or more (7.2 deaths per 10,000 documented cases), with those between 15-29 years old having a mortality rate of 6.9 per 10,000 documented infections.  There were no deaths recorded amongst the 1,663 children under the age of one who were documented as contracting measles.

The paper’s authors also split the documented deaths between patients who were immunodeficient, and those who were not.  Of the ten patients who died, one had congenital immunodeficiency and six had acquired immunodeficiency (e.g., Hodgkin’s lymphoma, Crohn’s disease, HIV, immunosuppressive treatment).  In other words, only 3 out of the 10 deaths recorded were associated with non-immunocompromised individuals.

The overall mortality rate of around 4-5 deaths per 10,000 cases of measles for all individuals is slightly higher than the estimate based on historic US data.  However, it doesn’t account for underreporting of measles infections, which the paper’s authors estimate at something over 50%.

Adjusting the data in table 1 by assuming only 50% of measles cases were reported gives the morbidity and mortality rates estimated in table 2 (below).

The overall mortality rate here is reduced to 2.3 deaths per 10,000 cases of measles.  This value still reflects considerable uncertainty – there are likely to have been more non-fatal complications than were recorded with hospitalized patients for instance, meaning that estimated morbidity rates are probably substantially underestimated.  However, it is less likely that there were substantially more measles-related deaths than were reported by Antona et al.  It is also consistent with analysis of US data which suggest a mortality rate of the order of 1 in 10,000 for infected individuals.

That said, the morbidity and mortality rates presented here only tell part of the story. As was seen with the immunocompromised patients in the study, susceptible sub-populations may face a significantly greater risk from measles.  This becomes a particular issue where there is an increasing probability of them coming into contact with infected individuals – as is the case where immunization isn’t widespread.

Nevertheless, the evidence from the French outbreak is strongly suggestive that, for infected individuals living in a developed economy with a robust healthcare system, the mortality rate is most likely less than 2-3 deaths per 10,000 cases.


February 3, 2015 by Andrew Maynard

What is the risk of dying if you catch measles?

If you catch measles, what are your chances of dying?

When I was a kid, measles was one of those things you were expected to catch.  I had it when I was five, and must confess, I don’t remember much about the experience.  I do remember being confined to bed.  And I also remember being told that measles could cause blindness – as a budding reader, this scared me.  But I don’t recall anyone hinting at anything worse.  If my parents were worried, they didn’t show it. And I’d certainly never heard of kids who had died – even in playground rumors.

So as the current outbreak of measles in the US continues to spread, I’ve been intrigued by statements that the disease has a mortality rate of somewhere between one and three young children per thousand infected.

Of course I know as a public health academic that measles is highly infective and can cause severe harm – even death.  But there was a dissonance between what I was reading and what I felt was correct. Surely if one out of every few hundred kids died as a result of measles as I was growing up, I’d have got wind of it?

The mortality rate of around 1 in 1000 though comes with a sound provenance.  It’s there in black and white on the Centers for Disease Control and Prevention (CDC) web pages:

“For every 1,000 children who get measles, one or two will die from it”

A 2004 review in the Journal of Infectious Diseases provides further insight.  Using CDC data on reported measles cases in the US between 1989 and 2000,Orenstein, Perry and Halsey indicated that approximately three children under the age of five died for every thousand that caught measles, and that the overall mortality rate for all ages was also around 3 per thousand people infected – the table below gives the data they used in deaths per thousand cases.

This seems pretty convincing – maybe measles is more dangerous than we used to think back in the 70’s.

But there’s a catch.

The CDC dataset that Orenstein, Perry and Halsey used specifically refers to reported cases of measles.  The derived mortality rate is for cases that are serious enough to have been flagged and logged by the agency.  The question then becomes, how many cases occurred that weren’t reported, and how (if at all) do these alter the estimated mortality rate?

In a 2004 review entitled “Measles Eradication in the United States” (an optimistic title, in the light of current events), Orenstein, Papania and Warton make the point that not every case of measles in the US is reported, or at least it wasn’t, when the disease was more common.

According to their paper, from 1956 to 1960, there were an average of 450 measles-related deaths reported each year in the US, or approximately 1 death per 1000 reported cases.  At the time though, it was estimated that more than 90% of Americans had been infected by measles by the age of 15 – equivalent to roughly 4 million children and teens per year. (Langmuir, A.D. (1962), Medical Importance of Measles. Am J Dis Child 103(3):224-226.)

These data suggest that the chances of dying from measles in the US in the late 1950’s was probably closer to 1 in 10,000.

Accounting for non-reported cases that led to death, and some uncertainty in the numbers, the mortality rate is realistically likely to be around one in a few thousand.  But based on the data, it’s not likely to be as high the one or two deaths per 1,000 that’s being widely cited.

That shouldn’t detract from how important it is to prevent measles, and especially to protect young children, the elderly and other susceptible groups from infection.  This is not a disease to be dismissed or taken lightly.  It still kills nearly 150,000 people a year around the world according to the World Health Organization.  It is highly infectious.  And for communities to be adequately protected, there need to be high levels of immunization.

Yet from the available evidence, claiming that one or two children out of every 1,000 infected in the current US outbreak will die seems far fetched.

Sadly, using this mortality rate to hammer home the importance of getting kids vaccinated could well backfire.  Like myself, many parents from my generation haven’t seen evidence for such a high chance of dying from the disease.  And to use data that not only feel wrong, but are not backed up with evidence, only serves to undermines trust in public health experts.

Anti-vaccine proponents are smart enough to realize this.  Each time the data on infectious diseases and risk are spun beyond their legitimate bounds, anti-vaccine proponents are given a helping hand in winning the hearts and minds of concerned parents.

Instead, public health experts and their advocates need to remain true to the data that support their message.  Granted they won’t always be clear and compelling.  But this should never be an excuse to spin the data to fit the story.

Ironically, for many people it’s not the numbers that count when it comes to making decisions over whether to vaccinate or not, so much as who to trust.  But in deciding whom to place that trust in, numbers – and how they are used – can be very important indeed.

 


Andrew Maynard is Director of the Risk Science Center at the University of Michigan School of Public Health.
Print Email permalink (1) Comments (9040) Hits •  subscribe Share on facebook Stumble This submit to reddit submit to digg


COMMENTS


Thank you for this very informative article. Is there any information as to how many of the measles outbreaks you examine in Table 1 were among vaccinated individuals vs unvaxxed (presumably for MMR)?





YOUR COMMENT (IEET's comment policy)

Login or Register to post a comment.

Next entry: Nanotechnology: New Polymerization Methods for Plastic / Organic Electronics (1hr)

Previous entry: 13 Questions: About Greece, Europe, Austerity – and Us

HOME | ABOUT | STAFF | EVENTS | SUPPORT  | CONTACT US
JOURNAL OF EVOLUTION AND TECHNOLOGY

RSSIEET Blog | email list | newsletter |
The IEET is a 501(c)3 non-profit, tax-exempt organization registered in the State of Connecticut in the United States.

Executive Director, Dr. James J. Hughes,
35 Harbor Point Blvd, #404, Boston, MA 02125-3242 USA
Email: director @ ieet.org