Drinking from the Fire Hose: Odds and Ends on the Gasping Oppression

I spend most of my time taking care of hospitalized patients with acute infections and issues of public health are, outside of infection control, not a high priority. Vaccinations in training were always like clean water and fresh food: their benefit was a given and I never needed to consider the benefits and subtleties of  vaccination. There is just so much time in a day and I was more concerned with AIDS, endocarditis and meningitis to worry about the ins and outs of vaccination.

One of the many benefits of writing for SBM, and being the Chair of Infection Control, is it is a stimulus to keep up on aspects of medicine that I might not otherwise pay close attention to, like vaccines. I have been far more interested in vaccines, especially influenza vaccines, since starting practice in 1990 than I ever was in the decade I spend in training.

Vaccination and the efficacy of vaccines is not as straightforward as I would have thought 30 years ago. It was give a vaccine, generate an antibody, and, viola, the patient is protected. The vagaries of the flu vaccine are even more pronounced, since response to the vaccine is variable and the population has never been vaccinated at levels, more than 90%, where herd immunity would likely kick in.

My ideal flu vaccine study, which would be both impossible and unethical, would be to vaccinate everyone West of the Mississippi and no one to the East (no coincidence that me and mine live in the West) and study the short and long term effects. Until that day, I am stuck with the hodgepodge of medical studies that look at the results of influenza vaccination and add insights into the disease.

I thought this week it would be fun to mention some interesting studies about influenza, the vaccine and flu immunity that have come out in the last 2 years. This is not meant to be anything more than a compilation of articles I thought were interesting, and the only purpose is to give a hint as to the complexities of influenza and  vaccination. How did I choose these articles? Every two weeks I do an infectious disease update podcast (Puscast, on iTunes), and I went through my influenza related pdf’s from the Puscast from the last two years. Much to my surprise I had read over 250 influenza related articles. Is that alot or a little? There were 12,411 influenza articles on Pubmed in the same time frame, so I read about 2% of the influenza literature for my Puscast, but it represents about .6% of the 4000 articles I read in the same period. I am always impressed with the extent of what I have yet to learn compared to what I do know. So much knowledge, so little time to consume it.  Dr Richard Bryant, my attending as a medical student and one of the big reasons I went into ID, said reading the medical literature is like drinking from a fire hose. Oh well, here is what I think is interesting, curious, or important about influenza in the last year.  A sip from the deluge.

Believe is what you do when there is an absence of facts. One of the aggravating issues at the hospital is getting health care providers to get the flu vaccine. The best I can do is about 70% of employees to get vaccinated. Across the country those numbers are the same, although MD and RN vaccination rates hover around 80%.  That is pathetic, since I have the old fashioned idea it is incumbent upon us to protect our patients, especially given that nosocomially acquired flu has a 27% mortality rate and people shed the flu before they are ill or likely to come to work when ill:

“…we estimated that 1%–8% of infectiousness occurs prior to illness onset. Only 14% of infections with detectable shedding at RT-PCR were asymptomatic, and viral shedding was low in these cases.”

Silent shedders may be unimportant as an outpatient, but if it were a nurse or respiratory therapist going from room to room, it could result in rapid spread of disease.

The antivaccine wackaloons see low vaccination rates in HCW’s as an indictment of the vaccine, that HCW’s know something that the rest of us do not. That is not the case. I have wandered the hospital giving flu shots, and there are the same lame reasons given by the HCW’s as the general population, although HCW’s should know better since they ostensibly have access to all the information and the best minds (mine) in the field. I publish an article on Medscape every October on why HCW’s are a dumbass for not getting the flu vaccine.  Oddly, people take exception at being called a dumb ass.

What factors are important in HCWs getting the flu vaccine?

Knowing that the vaccine is effective (mhRR 2.22; 95% CI 1.93 to 2.54), being willing to prevent influenza transmission (mhRR 2.31; 95% CI 1.97 to 2.70), believing that influenza is highly contagious (RR 2.25; 95% CI 1.66 to 3.05), believing that influenza prevention is important (mhRR 3.63; 95% CI 2.87 to 4.59) and having a family that is usually vaccinated (RR 2.32; 95% CI 1.64 to 3.28) were statistically significantly associated with a twofold higher vaccine uptake.

And infected HCW’s markedly increase your risk of influenza like illness if you are exposed in the hospital.

For patients exposed to at least 1 contagious HCW compared with those with no documented exposure in the hospital, the RR of HA-ILI was 5.48 (95% confidence interval [CI], 2.09-14.37); for patients exposed to at least 1 contagious patient, the RR was 17.96 (95% CI, 10.07-32.03); and for patients exposed to at least 1 contagious patient and 1 contagious HCW, the RR was 34.75.

Part of the problem is there have been no studies to demonstrate that specifically vaccinating hospital workers will decrease influenza in patients. There is certainly buckets of biologic plausibility to suggest that if I do not get flu because of the vaccine, I will not pass it on to you as I make rounds in the ICU. There have been studies to suggest benefit in nursing homes, but of course nursing homes are not the same as acute care hospitals, were the length of stay is increasingly measured in hours rather than days, lessening the risk of exposure.

The literature now has an extremely suboptimal study that suggests vaccinating HCW’s decreases influenza in the hospital.

In summary, our observational study indicates a protective influence of vaccination of more than 35% of HCW on HAI in patients. Other experimentally-designed investigations are needed to demonstrate the effectiveness of HCW vaccination in the control of influenza outbreaks in healthcare settings, and to determine the threshold for vaccinated HCW proportion with more accuracy. Our findings must not be misinterpreted. To date, the HCW vaccination rate of 35% is not optimal to control HAI.

But I will take what I can get, since the more that are vaccinated, the fewer will get disease. What is curious to my mind is why people would think otherwise about the vaccine. My advice to you and yours during the flu season:  If admitted to the hospital, you are probably in a population at risk for dying of influenza, and if your provider has not had the vaccine, they may not be convinced of the contagiousness of flu nor its seriousness. They are likely a dumb ass, and you do not want a dumb ass involved with your health care.  You need to request another HCW and forbid them from entering your hospital room. If anyone from my family is admitted to the hospital, especially my mother, I will not allow anyone in her room who has not had the flu vaccine. I suggest you do the same.

Since the flu vaccine is not close to 100% effective, there are other measures to help decrease the spread. A career in infection control has consistently reinforced the idea that there is never one intervention that completly decreases the risk of infection, and no one, except the antivaccine wackaloons, think that the vaccine is an all or nothing phenomena. Simple mechanical issues often do not demonstrate benefit: good hand washing and masks are of variable help in outbreak situations, if started early.

During the 2009/10 and 2010/11 pandemic seasons, Dr. Buchholz and colleagues compared 84 households in which 1 member had influenza. The households were randomized into 3 groups: facemask use and intensified hand hygiene; facemask use only; and no intervention (control group). Among the 218 noninfected contacts in the 84 households, 35 (16%) developed flu. An intention-to-treat analysis found no benefit from facemasks and handwashing, nor from facemasks alone. However, when the intervention was started within 36 hours of symptom onset in the index case, pooled data from the 2 intervention groups showed an 84% reduction in secondary flu infection. In addition, when facemask alone was considered in a per protocol analysis, there was a 70% reduction in risk.

Other evaluations the effect of mask and hand washing was not as good for flu prevention.

Data from 788 households were collected from April 2008 to February 2011. The median age of the household member with flu was 5 years, and 724 (92%) slept in the same room with their parents. Among the household contacts, 31% became infected. Handwashing was found to have no significant benefit, and facemask use was associated with a modest increase in risk for influenza infection.

So that is why I would prohibit the unvaccinated health care worker from entering my room even if masked, although there are some hospitals in my area are having workers who are not vaccinated wear a mask.  I should probably wear a mask for aesthetic reasons

There is a difference between using a mask and hand washing in the community to prevent acquisition, which is only of perhaps modest benefit, and is probably not the same as masking an infected person to prevent spread in the hospital. Given that hand hygiene is often not 100%, I would be further disinclined to allow the unvaccinated HCW in a mask to touch me and mine, never knowing for sure where those hands may have been.  Actually I do know.  And you should be ashamed.

It is doubly interesting since there was a proof of concept  study to demonstrate the route of flu infection may be through the eyes. Volunteers were squirted with live flu vaccine while wearing various protective measures, and as long as there was no eye protection they  became infected with the vaccine strain. I suppose if my unvaccinated HCW wore a spacesuit I would let them in my mothers room, but otherwise no way.  See the addendum at the end for a recent real world, i.e. messy and incomplete, example.

The two groups who had inordinate mortality from the H1N1 pandemic were the obese and the pregnant, and both were targets for vaccination. It is harder to get a pregnant female (but not a pregnant male) to take anything for fear of affecting the fetus.

Flu is bad for the baby. Babies in utero during the 191p pandemic had

reduced educational attainment, increased rates of physical disability, lower income

and babies who were from vaccinated mothers were larger:

During this period, the proportion of infants who were small for gestational age was lower in the influenza vaccine group than in the control group (25.9% v. 44.8%; p = 0.03). The mean birth weight was higher among infants whose mothers received the influenza vaccine than among those who received the control vaccine during this period (3178 g v. 2978 g; p = 0.02).

Given that epidemiologic data suggests “pandemic influenza caused first trimester miscarriages in ∼1 in 10 pregnant women”, besides covering your own ass for dying from flu while pregnant, not getting the flu would be of great benefit for your unborn child. And the benefit continues after birth

Results. The mothers of 2 (2.2%) of 91 case subjects and 31 (19.9%) of 156 control subjects aged !6 months, and 1 (4.6%) of 22 case subjects and 2 (5.6%) of 36 control subjects aged > 6 months, had received influenza vaccine during pregnancy. The effectiveness of influenza vaccine given to mothers during pregnancy in preventing hospitalization among their infants, adjusted for potential confounders, was 91.5% (95% confidence interval [CI], 61.7%–98.1%; P p .001) for infants aged !6 months. The unadjusted effectiveness was 90.7% (95% CI, 59.9%– 97.8%; P p .001).

Conclusions. Influenza vaccine given to pregnant women is 91.5% effective in preventing hospitalization of their infants for influenza in the first 6 months of life.

I always assume that various biases result in overestimation of therapeutic benefit in studies. That is part of the decline effect in clinical medicine. As you weed out the bias the efficacy of treatments often declines. I had never considered that bias underestimates effectiveness, since there is an inherent urge in researchers to find positive results.

I have mentioned many times that statistical analysis make me feel like Mr. Gumby, and I dropped my statistics class every year for four years in college. One they got past the Bell shaped curve, they lost me to brain hurt. To this day I have to take the results of statistical analysis at face value. It turns out that, at least with the flu vaccine, bias may be underestimating the effectiveness by about 10%

By simulating a case-control study of influenza VE in children, we found that the sources of bias we included, when considered together, more often led to an underestimation rather than an overestimation of true vaccine benefit.

How widely the results are applicable to other interventions or studies that are not case control, I lack the intellectual wherewithal to say.

Association is not causation, but one of the effects of vaccination is prevention of complications of flu: bacterial infections and vascular events

There was strong evidence for a link between influenza and MI both in England and Wales, where 3.1%–3.4% of MI-associated deaths (P < .001) and 0.7%–1.2% of MI-associated hospitalizations (P < .001) were attributable to influenza, and in Hong Kong, where the corresponding figures were 3.9%–5.6% (P = .018) and 3.0%–3.3% (P = .002).


In a retrospective chart review of 119 individuals admitted to the hospital with H1N1 virus infection, 7 patients (5.9%) were found to have experienced thrombotic vascular events.

But those who listen to my Puscast know the bee in my bonnet of infection leading to inflammation which is prothrombotic which leads all manner of vascular events. Inflammation is bad, even when you are infected, but is the price we pay to survive infections.

Blood stream infections go up with flu season,

Regular influenza seasons were characterized by increased rates of S. pneumoniae BSI but with no increase in S. aureus and S. pyogenes BSI rates. The pH1N1/2009–2010 influenza outbreak was characterized by (1) higher rates of S. pneumoniae bacteremia among children but not among adults (IRRs for S. pneumoniae BSI among children aged 0–4 years during the summer and winter of 2009–2010 were 14.8 [95% confidence interval {CI}, 5–43.7] and 6.5 [95% CI, 3.6–11.8], compared with 2006–2009 summers and influenza-active winter weeks, respectively ]), higher rates of S. aureus BSI in all age groups (IRRs during the summer and winter of 2009–2010 were 1.6 [95% CI, 1.4–1.9] and 1.5 [95% CI, 1.2–1.7], compared with 2006–2009 summers and influenza-active weeks, respectively [P , .0001]), higher rates of S. pyogenes BSI during 2009–2010 influenza season (IRR 2.7 [95% CI, 1.6–4.6] and 3.3 [95% CI, 1.9–5.8] during the summer and winter of 2009–2010, compared with 2006–2009 summers and influenza-active weeks, respectively ).

and pneumonia

 During ∼185 million person-years of surveillance, we observed 21,239 episodes of invasive pneumococcal pneumonia; 485,691 specimens were tested for influenza. Influenza circulation was associated with 11%– 14% of pneumococcal pneumonia during periods of influenza circulation and 5%–6% overall. In 2 of 3 regions, the association was strongest when influenza circulation data were lagged by 1 week.

A lot of badness is associated with the flu; at least the heart attacks may be prevented:

The incidence of AMI was significantly reduced in the 60 days following vaccination (compared with the baseline period), ranging from a reduction of 32% (IRR 0.68; 95% CI 0.60–0.78) at 1–14 days after vaccination, to 18% (IRR 0.82; 95% CI 0.75–0.90) at 29–59 days after vaccination.

The coolest article of the year suggests we may be on the path to a universal flu vaccine. Every year the structure of the flu proteins drifts and every 25 years or so there is a entirely new strain from genetic reassortment, or a shift. It is why we need a new vaccination every year. They have discovered a site on the influenza hemagglutinin protein, FI6, that has

broadly neutralizing antibody that recognizes all 16 HA subtypes, including emerging ones, such as H5N1.

If you are lucky, after the vaccine or the disease, you could develop an antibody against the FI6 site and be immune to all subsequent strains of flu. I say luck, as the researcher screened “104,000 peripheral-blood plasma cells from eight recently infected or vaccinated donors” to find one cell making that antibody.

Why the production of broadly neutralizing antibody occurs rarely is an unanswered question, although some are lucky enough to make such antibody. It might partially explain why seasonal influenza infection conferred cross-protection against pandemic influenza while vaccination does not.

It does make the whole issue of efficacy of vaccine that much more interesting. You get the vaccine, and depending on what parts of the vaccine you react to, you could have more or less chance of disease or disease severity.  Hopefully they will be able to force the immune system to recognize the FI6 epitope more reliably and we will may be able to eradicate influenza. Or not. Not with the current vaccination rates. But I can dream. First smallpox, then rinderpest (good thing Wakefield was not a veterinarian), then influenza?

Complications of the vaccine are always a concern and the big one with H1N1 was Guillain-Barre syndrome (GBS), since after the 1979 1976  flu vaccine there was an uptick in cases of the disease, a complication never adequately explained and never repeated. Not only was the H1N1 vaccine not associated with GBS, but the flu itself had a higher association

Adjusted odds ratios for GBS occurrence within 6 weeks after seasonal and A/H1N1 vaccination were 1.3 (95% CI: 0.4, 4.1) and 0.9 (95% CI: 0.1, 7.6), respectively. Study results confirm that influenza virus is a likely risk factor for GBS. Conversely, no new concerns have arisen regarding influenza vaccination.

And you will be relieved to know that if you had GBS in the past, you can be vaccinated safely.

Results. We identified 550 cases of GBS over 33 million person-years. Following their GBS diagnoses, 989 vaccines were given to 279 of these individuals, including 405 trivalent inactivated influenza vaccines (TIV) administered to 107 individuals with a prior diagnosis of GBS. Among the 550 total cases of GBS, 18 initially had onset within 6 weeks of TIV; of these, 2 were revaccinated with TIV without a recurrence of GBS. Only 6 individuals of 550 (1.1%) had a second (recurrent) diagnosis of GBS. Among these 6 individuals, none had any vaccine exposure at all in the 2 months prior to the second onset of GBS.

Conclusions. In our population of over 3 million members, during an 11-year period, risk of GBS recurrence was low. There were no cases of recurrent GBS after influenza vaccination and none within 6 weeks after any vaccine.

Right. I’m impressed with the power of Kaiser Permanente; I could never convince a GBS patient to receive a vaccine.

None of the studies quoted above are definitive, but if you are a fan of the concept of preponderance of data, it would seem that influenza is bad and the vaccine has benefits. There are numerous other interesting articles on antibiotic resistance, efficacy of vaccine, and the 500th anniversary of the gasping oppression I did not mention.  There is always more to learn.


A true story.  When I first read it today, I thought it was made up to make a point.  I did not write the text, but it occurred at a Portland hospital in mid March and I reproduce it with permission.

Patient Story: Spreading the Flu 4-2012

As a reminder, we have added patient stories to meetings as a way of “bringing the patient into the room”, clarifying the context for our quality plan, and emphasizing the complexities and the importance of the work we are undertaking.

Today’s story is about a group of patients, a nurse, and influenza.  It starts with Patient #1, a 47 year-old woman admitted through the emergency department (ED) in mid-March with fever and shortness of breath.  She was transferred to an inpatient unit with a mask on, which triggered the staff on the receiving unit to implement droplet precautions.  Initially thought to have pneumonia, testing confirmed her symptoms were the result of influenza type A, H1N1.  After four nights in the hospital, she was discharged home after an uneventful hospital stay and a flu shot.

Patient #2, next door to Patient #1, is a 61 year-old man who was admitted in early March for a GI bleed with multiple co-morbidities.  His progress was steady until nine days after admission, when he developed a new fever and respiratory symptoms.  These symptoms developed on the same day of Patient #1’s admission.  Influenza was suspected two days following the development of his fever, and staff implemented droplet precautions.  Lab testing confirmed influenza type A.  He remained hospitalized for two more days and received a flu shot before being transferred to a skilled nursing facility.

Down the hall, Patient #3, a 71 year-old man, was admitted two days after Patient #1 for acute stroke and urinary tract infection.  On day 3 of his hospitalization, he developed a fever and cough.  Lab testing confirmed influenza type A.  Droplet precautions were ordered with the lab test for influenza.  He remained hospitalized an additional four nights and received a flu shot before being discharged.

Patient #4, a 73 year old man, down the hall from the first two patients and around the corner from Patient #3, was admitted on the same day as Patient #1 following a fainting event at home.  Due to his long-standing heart issues, he was kept overnight for observation and discharged the following morning.  However, he returned to the ED three days later with continued symptoms.  He was discharged from the ED only to return the next day with shortness of breath.  Six hours after being readmitted, staff suspected influenza and ordered droplet precautions.  His lab tests returned positive for influenza type A.  After spending three nights in the hospital, he was discharged home after receiving a flu shot.  The following day, he was admitted to the intensive care unit and continued receiving treatment as an inpatient for secondary pneumonia, a complication of his influenza type A infection.

The fifth person in our story is a nurse on the unit where these four patients were admitted.  She works on a nursing unit whose hand hygiene performance is currently 67%, and where 85% of the unit staff were vaccinated for this year’s seasonal flu.  The particular nurse in this case, however, was 1 of only 9 on the unit who chose not to be vaccinated.  Her manager stated that the reason the nurse gave for not receiving the vaccine was that she “was not convinced of the evidence that the vaccine protects patients from transmission … she said she would get the vaccine if she truly believed it protected her patients, but that she didn’t.”

This nurse cared for Patient #1 on her first day of admission.  She cared for Patient #2 on the eighth and ninth day (when he developed flu symptoms) of his stay.  She also cared for Patient #3 on the first two days of his inpatient stay.  There does not appear to be any direct contact with this nurse and Patient #4.

The nurse in our story developed symptoms consistent with influenza three days after working with Patient #1 and Patient #2 (which is the usual 1- to 4-day incubation period for influenza).  Due to symptoms, she only worked a partial shift that day.  Suspecting her symptoms may be influenza, she used a mask until relief staff was available.  She returned home and was able to care for herself without medical intervention.  She was not tested for influenza and remained off work for one week.  She is still undecided about receiving the flu vaccine.

Oregon is one of only two states in the US that prohibits employers from requiring vaccines as a condition of employment.  Our organization is actively working with colleagues throughout the state to try to change this.


Posted in: Basic Science, Science and Medicine, Vaccines

Leave a Comment (42) ↓