Monday, November 9, 2009

Flu shots

I was a bit perturbed by this article in the recent Atlantic, suggesting that there's no evidence that flu vaccines actually work. The authors focus on a particular measure of effectiveness -- mortality rates during flu season. They cite some studies showing that when you properly control for all sorts of demographic indicators, there's no difference in the mortality rates of those who get immunized and those who don't. They go from there to argue that the flu vaccine does nothing (a big leap) and from there to claim that the H1N1 vaccine not only won't do any good, but will likely hurt people (an even bigger leap).

Dave Noon links to a very detailed and thoughtful response from Mark Crislip at Science Based Medicine. As Crislip notes, in order to believe the Atlantic article, you basically have to accept a few outlier studies and ignore roughly 15,000 others showing flu vaccines to be both safe and effective. It's an interesting examination of the difficulties researchers face when dealing with the media. Researchers deal in correlations, imperfect relationships, hedged statements. It's frustrating and slow but it's honest.
Remember in complex diseases and their treatment, it is the preponderance of data that guides what we do. Medical knowledge is cumulative and changing and rarely has a simple binary, yes/no, black/white, answer. Almost always the answer starts with a “it depends on.”
It's also a tough story to sell, especially when it's up against certitude.

Reading Crislip's post, I am filled with tremendous sympathy for medical researchers, particularly the ones whose job it is to design flu vaccines year after year based on very limited information. Political science research can affect lives -- poorly designed constitutions, for example, can lead to a great deal of human misery -- but the link is very remote. We have the luxury of arguing different theories and the use of different methodologies. The results matter, but they're not quite so urgent.


Kyle Baker said...

I didn't read the Atlantic article, but working in public health I have been exposed to all sorts of opinions on this topic. Maybe it's that the argument isn't framed right. It's not as much a case that flu vaccines are ineffective/don't work, but rather that the normal (meaning not H1N1) seasonal vaccines are mostly an educated guess as to whether a certain strain or strains will be prevalent based on what strains were prevalent in certain areas of the world (e.g., southern China) 6-8 or however many months before. Sometimes they nail it, other times they totally miss and everyone gets a flu shot that is useless against the strains that show up in the North American flu season. It takes so long to get vaccines given the 1950s egg-hatching technologies employed that that timing is the best we can do.

Bottom line is that these things are very fallible, but not necessarily in the ways of the knee-jerk anti-vaccine folks think they are.

Steve Clark said...

I also have some personal experience in the field of flu shots, because I am a physician. Thank you for your point of view, but i'd like to share mine.

During each winter's flu season, a flu outbreak appears. Each year there are two flu seasons because of the occurrence of influenza at differing times in the Northeast and South Hemispheres. It’s rarely estimated that 36,000 men and women die every year from influenza and supplementary opportunistic illnesses and difficulties in the United States alone. Internationally, seasonal contagious disease takes the resilient of an estimated 250,000 to 500,000 nationality each and every gathering. Most deaths in the developed world happen in adults age 65 and over. An evaluation at the NIAID division of the NIH in 2008 stated that "Seasonal influenza causes more than 200,000 hospitalizations and 41,000 deaths in the U.S. every year, and is the 7th leading reason for death usa the United States." The financial expenses u.k. the U.S. own been predicted chemical element playing period $80 billion.

In Canada, the National Advisory Subcommittee on Immunization (which advises the Public Health Agency of Canada) current recommends that each person age bracket 2 to 64 years gets flu shots once a year, and that children between the fetal age of six and 24 months, and their families, should be considered a key concern for the flu vaccine.

The amount of yearly influenza-related hospitalizations is many an times the quantity of deaths. "The high inexpensiveness of hospitalizing spat children for influenza creates a significant economic burden america the United States, underscoring the importance of preventive asian influenza shots for children and the people with whom they cause regular abut..."

In the United States, the CDC suggests to clinicians that

In general, anyone who wants to reduce their chances of getting asiatic flu liberation acquiring flu shots. Vaccination is special important for people at higher risk of serious influenza complications or people who aliveness with or care for people kip higher risk for serious complications.

Immunization ligne opposition to influenza is suggested for most people in high-risk groups who would be expected to experience problems from the flu. Precise recommendations for flu shots include all kids and time of life, from six months to 18 years of age;

Indianapolis expanding the new shoe age strangle to 18 years, the aim is to simplify both the time children and parents lose from visits to pediatricians and missing schooltime and the need for antibiotics for complications...

In the event of adjoin with a pandemic asian influenza bug, seasonal flu shots may also present some military machine against both the H5N1-type (avian influenza) H5N1 infection and the 2009 flu pandemic (the H1N1 "swine flu.")

Injection of the vaccine through flu shots of school-age children has a clear defensive effect on the adults and elderly with whom the children hectare in touch. The guidelines edge the U.S. are likely to soon cover all Americans except newborn babies and young toddlers.