You are here

Trust Me: I Have The Statistics To Prove It

By: Dr. F. Edward Yazbak

Pronouncements by the Centers for Disease Control and Prevention (CDC) are accepted as absolute truth by the media and by most Americans. Should they be?

On July 28, 2005 at 4:02 a.m. EDT, published a long article (1) that started, "Almost 81 percent of the nation’s toddlers are getting vaccinated on time, a record level that comes five years ahead of government expectations, federal health officials reported Tuesday.

The CDC informed the media and the nation of the "good news" even before its official publication in the Mortality and Morbidity Weekly Report (MMWR) of July 29. (2) The latter consisted of two tables and 16 lines of text, followed by a longer editorial note, which started, "The National Immunization Survey (NIS) provides vaccination coverage estimates for children aged 19-35 months for each of the 50 states and 28 selected urban areas."

In the second paragraph, authors N. Darling, T. Santibanez and J. Santoli described how their data was obtained. "To collect vaccination data for age-eligible children, NIS uses a quarterly random-digit-dialing sample of telephone numbers for each of the 78 survey areas. NIS methodology, including the weighting of responses to represent the entire population of children aged 19-35 months, has been described previously. During 2004, health-care-provider vaccination records were obtained for 21,998 children. The overall survey response rate for eligible households was 67.4 percent."

Note: The investigators mentioned that they used a random-digit-dialing sample for each of the 78 survey areas, but did not mention the actual or estimated number of telephone attempts. They seemed reassured that their "weighting of responses" would accurately represent the entire population of children aged 19-35 months and just casually mentioned in passing that one out of three eligible households in the limited survey areas did not respond.

Their methodology was described in an article published in the American Journal of Preventive Medicine in 2001. The Medline abstract states, "The National Immunization Survey is a large federally funded survey designed to estimate vaccination coverage rates for children residing in the United States aged 19 to 35 months. In 1999, over 8 million telephone call attempts were made to obtain provider-reported vaccination histories on 22,521 children in the age range of interest."                                                                                 

Note: There were 8 million attempts to obtain information on 22,521 children. The recent 2004 NIS survey yielded information on only 21,998 children.

In 2000, the estimated population of the United States was 282,125,000. Considering that the 0-4 population was estimated at 19,218,000, one can suggest that the 19- to 35-months cohort was approximately 6.6 million children in 2000 and more by 2004. Retrieving information on 22,000 children (on average) simply means that the CDC's much publicized figures were based on information obtained on less than 0.3 percent of the children 19 to 35 months of age in 50 states. (3)

 Under ideal circumstances, and with perfect matching of ethnic, social and other parameters, such a small sample could have yielded estimates within certain established limits - if it had a near 100-percent-response rate. It is impossible to know how reliable those estimates were with response rates of 67.4 percent of the chosen sample. Even with intense weighting of responses, they could not be guaranteed to result in perfect representation. For that, one would also have needed a thorough knowledge - at the local level - of the population make-up at each telephone prefix and the inclusion of wireless phoneprefixes because thousands of Americans do not have hard-wired phones anymore.

Confounders in telephone surveys are many. Parents who are "on board" and have fully vaccinated children are more likely to respond, creating a response bias. More importantly, those who did not vaccinate their children, may feel guilty, fear reprisals and either refuse to participate or provide less than truthful information. The available report does not disclose what steps were taken to limit false reporting or the size of the sample checked for accuracy by reviewing actual medical records. In addition, whenever weighting is applied, there is always a range of uncertainty and confidence limits, which may widen the possibilities and yield low and high results

As in real estate, location is an important factor in any survey. Years ago, a review of doctors' records in a capital city revealed that vaccination rates were higher in a certain census tract than they were in the next.

So were the results reliable? Were they worth the cost and the effort? Were they meaningful or were they just "good news" propaganda? The most charitable answer to these questions is probably "Who knows?"

Having reviewed many CDC and CDC-sponsored studies, I have never found one that did not "support" the agency's agenda and produce the required result. Finding high vaccination rates will be advantageous when the next budget is discussed. Whereas, finding that Italy, for example, had a better vaccination rate would be disastrous. The CDC has initiated unparalleled vaccination programs; surveys such as this recent one are essential to guarantee support on all fronts for those programs and the future proliferation of new ones. The CDC approves and supports those studies that will yield favorable results and generate positive propaganda for the battery of vaccines which need funding through the Federal Vaccine for Children program. The CDC will evidently limit access to those scientists who may decide to pursue avenues of research, or test hypotheses, likely to lead to negative findings.

When vaccination rates are mentioned, the issue of "herd immunity" is often discussed. To be precise, what is called herd immunity is really "herd protection." Originally, it meant that susceptible people, living in areas where the great majority had had a certain disease and acquired long-lasting immunity, were less likely to come down with said disease than others living in communities which had not been affected on a large scale. Currently, herd protection refers to the presumed relative protection of citizens living in areas where an overwhelming majority of the children and adults have been vaccinated - usually at rates as high as 95 percent. Literally, herd immunity should imply that the minority of people who were not vaccinated actually gained a measure of immunity from the vaccinated majority.

Herd Immunity was not mentioned in the CDC's recently published report possibly because the achieved vaccination rates, though they surpassed target levels, were still a long way from 95 percent. 

The results of the survey - projected estimates at best - were acclaimed as a wonderful achievement and considered rock-solid by all, starting with the unnamed MMWR editor who stated assuredly: "The findings in this report indicate that, for the first time, vaccination coverage (80.9%) for the 4:3:1:3:3 series exceeded the Healthy People 2010 goal (objective 14-24a) to increase to at least 80% the proportion of children aged 19-35 months who receive all vaccines recommended for universal administration for at least 5 years."

Inside the CDC, this survey must have generated the legendary "good news/bad news" reactions. It was great news for the National Immunization Program, which is really the National Vaccination Program. It was not such wonderful news for Jose Cordero, now director of the National Center on Births Defects and Developmental Disabilities, the CDC division in charge of autism. Now, he will have to produce more studies to convince parents that the ever increasing number of mandated vaccines and the success of vaccination programs are not causing the present increase in autism.

Let's look at one of those cases of autism. Like many others, nine-year-old Johnny has been evaluated by a psychiatrist, a neurologist or a developmental pediatrician, in addition to a half a dozen professionals in the special education department at school. All of them agree that he has autism. He rides to school in a yellow minivan with an aid watching his every move. In the classroom, he needs constant supervision and consumes 15 to 25 percent of his teacher's time and attention. He makes continuous "engine noises," obsesses over different things all day long and freaks out upon hearing a vacuum cleaner start. And yet, experts at the CDC, which published the recent vaccination figures as solid facts, insist that Johnny may not have autism.   


With the flu vaccination campaign starting soon, a famous CDC "rounded-figure" statistic is certain to be floated: Influenza kills 30,000 Americans every year.  

Last year, the director of the vaccine research group at a renowned medical institution, kicked it up a notch by telling an annual session of the American College of Physicians that influenza is the sixth leading cause of death for older Americans and infects 5 percent to 10 percent of elderly Americans every year. The flu leads to 300,000 hospitalizations, he said, and kills 30,000 to 40,000 Americans every year. (4)

Talk about elastic statistics. (5)

So, how many people really die of Influenza every year?

Following are the CDC's own figures:

In 2002: 753 (p.16)

In 2001: 267 (p.16)

In 2000: 2,175 (p.15)

In 1999: 1,685 (p.28)

The upcoming flu season promises to be as eventful as last year's.

The European Vaccine Manufacturers just announced that there would be three- to four-week production delays because of late delivery of one of the three viral seed strains chosen by the World Health Organization. (6)

Once again, the CDC will probably cite "convincing studies" to prove that healthy infants aged six months and over should receive the influenza vaccine. One of those studies was a CDC-funded retrospective study of some 30,000 children enrolled in a Colorado health plan. The children were not randomly assigned to the vaccine or placebo groups and the study results were only published in the CDC's MMWR. When asked whether the flu vaccine caused any adverse reactions, the lead investigator stated, "Hospital admissions were not tracked, and the parents were not interviewed." (7)           

For years, it has been recommended that infants and children with asthma and congenital heart disease receive an annual dose of influenza vaccine. Interestingly, a study from the Strong Children’s Research Center in Rochester, which was published in the prestigious Archives of Diseases of Children, is rarely mentioned. The authors of that study compared outcomes in vaccinated and unvaccinated asthmatic children, as regards to their asthma symptoms and complications. After adjusting for other variables, the authors found that the vaccinated group had a significantly increased risk of asthma-related clinic and emergency room visits. (8)


CDC reports about the varicella (chicken pox) vaccine are all "good news". The CDC web site (9) on the subject carries the following banner in bold red characters:                        

"Varicella vaccine is 85 percent effective in preventing disease."

The trade name of the chicken pox vaccine by Merck is Varivax. It was licensed in 1995 and has been widely used since then.

So, what is the rest of the story?

  • Using the same database, the vaccine efficacy declined by 20 percent between 1997 and 2001.
  • Chickenpox and shingles (or herpes zoster) are related and are caused by the same varicella-zoster virus.

The morbidity and mortality due to chickenpox are lower as a result of Varivax

After a person has had chickenpox, the virus remains in the body usually for decades and then may reactivate as shingles

We used to believe that chickenpox conferred lifetime immunity. We now know that the immunity was long lasting because of periodic exposures to the disease.

This natural boosting due to exposure to children with chickenpox helped suppress the reactivation of shingles in adults.

Because of the effectiveness of the chicken pox vaccine, the clinical disease has practically disappeared and with it, the natural immunity boost.Now, with the widespread use of Varivax, the incidence of shingles among adults (and children) has increased significantly.

The available abstracts and manuscripts about the incidence of herpes zoster fail to correct for under-reporting, which could be as high as 50 percent. They also usually present a single incidence rate among children for a bimodal trend. The high incidence rate of shingles among unvaccinated children with a previous history of chicken pox is averaged along with the relatively low incidence rate among vaccinated children.

Shingles is a much more debilitating illness and results in nearly five times as many hospitalizations and three times as many deaths as chickenpox. The rash is extremely painful at times - It is called St. Anthony's fire in Italy - and the disease has long-lasting and serious complications.   

R. Edgar Hope-Simpson, a British general practitioner, was the first to propose in 1965, "The peculiar age distribution of [shingles] may in part reflect the frequency with which the different age groups encounter cases of varicella and because of the ensuing boost to their antibody protection have their attacks of [shingles] postponed."

Gary S. Goldman, Ph.D, has extensively researched the increased incidence of shingles and its cost in pain and dollars after the introduction of Varivax in 1995. (10, 11, 12, 13) For years, the CDC disagreed with him.

Nevertheless, just a couple of years after the licensing of the chickenpox vaccine, the CDC and Merck, the vaccine manufacturer, started planning a large scale trial of a Shingles vaccine for adults, which was 14 times more potent than the pediatric vaccine. The results of the clinical trial were recently published in the New England Journal of Medicine. (14)

According to Merck, morbidity associated with herpes zoster and postherpetic neuralgia (PHN) was greatly reduced in immunocompetent older adults who were vaccinated with live attenuated zoster vaccine in a randomized, placebo-controlled, double-blind study.

U.S. researchers enrolled 38,546 adults aged 60 years and older in the shingles prevention study at 22 sites between November 1998 and September 2001. Participants received either one 0.5 mL subcutaneous injection of the investigational varicella-zoster vaccine (n=19,270) or placebo (n=19,276) with a median 3.12-year duration of herpes zoster surveillance.

A total of 957 confirmed herpes zoster cases were reported for those receiving the vaccine (315 patients) and placebo (642 patients). There were 107 cases of PHN among those participants receiving vaccine (27 patients) and placebo (80 patients).

Zoster vaccine use significantly decreased burden-of-illness due to herpes zoster by 61.1 percent and the incidence of herpes zoster and PHN by 51.3 percent and 66.5 percent, respectively. (15)

Also in June 2005, the Massachusetts Department of Health and the CDC published a study titled "The incidence of varicella and herpes zoster in Massachusetts as measured by the Behavioral Risk Factor Surveillance System (BRFSS) during a period of increasing varicella vaccine coverage, 1998-2003".  

Their results: Between 1998 and 2003, varicella incidence declined from 16.5/1,000 to 3.5/1,000 (79 percent) overall with >65 percent decreases for all age groups except adults (27 percent decrease). Age-standardized estimates of overall herpes zosteroccurrence increased from 2.77/1,000 to 5.25/1,000 (90 percent) in the period 1999-2003, and the trend in both crude and adjusted rates was highly significant (p<0.001). Annual age-specific rates were somewhat unstable, but all increased, and the trend was significant for the 25-44 year and 65+ year age groups.                                                                                       

Their conclusions:As varicella vaccine coverage in children increased, the incidence of varicella decreased and the occurrence of herpes zoster increased. If the observed increase in herpes zoster incidence is real, widespread vaccination of children is only one of several possible explanations. Further studies are needed to understand secular trends in herpes zoster before and after use of varicella vaccine in the U.S. and other countries.

An increase from 2.77 to 5.24 per thousand is, indeed, a 90 percent increase in the overall occurrence of shingles in just four years.

So, to recapitulate, the CDC on the recommendation of the Advisory Committee on Immunization Practices recommended the administration of Varivax (varicella live vaccine) to children starting at the age of one year at a cost of $52 per dose.

As coverage in children increased, the incidence of varicella (chicken pox) decreased and the occurrence of herpes zoster (shingles) increased substantially in adults (and also in children).

Merck then helped fund research to test Zostavax (zoster live vaccine from  Oka/Merck), a higher potency vaccine intended to prevent shingles, a disease that increased by 90 percent over four years after the introduction of Varivax, according to the CDC. After a three-year trial, it appears that Zostavax can reduce the incidence of shingles by 50 percent.

In April 2005, Merck submitted an application for Zostavax to the Federal Drug Administration. It is more than likely that the vaccine will be approved and recommended before long. The cost per dose will be $100.

Apparently, all this seems to make sense to the experts. For the rest of us, it is a lesson that even if "varicella vaccine is 85 percent effective in preventing disease" is the truth, it is not the whole truth and nothing but the truth.



When reviewing "official" medical statistics:

If estimates sound too good to be true, they may be.

If figures are approximated to the nearest four zeroes, don't believe them.

If any number is the magic 30,000, it came out of a hat.

If the numbers make sense, what else has not been revealed?

If they do not make sense, you are probably right.

And certainly if anyone tells you, "Trust me," watch out.



  8. Christy C, Aligne CA, Auinger P, Pulcino T, Weitzman M. Effectiveness of influenza vaccine for the prevention of asthma exacerbations. Arch Dis Child. 2004 Aug;89(8):734-5
  10. Goldman GS. Varicella susceptibility and incidence of herpes zoster among children and adolescents in a community under active surveillance.
    Vaccine. 2003 Oct 1;21(27-30):4238-42
  11. Goldman GS. Incidence of herpes zoster among children and adolescents in a community with moderate varicella vaccination coverage.
    Vaccine. 2003 Oct 1;21(27-30):4243-9
  12. Goldman GS. Using capture-recapture methods to assess varicella incidence in a community under active surveillance.
    Vaccine. 2003 Oct 1;21(27-30):4250-5.
  13. Goldman GS. Cost-benefit analysis of universal varicella vaccination in the U.S. taking into account the closely related herpes-zoster epidemiology.
    Vaccine. 2005 May 9;23(25):3349-55.
  14. Oxman MN, Levin MJ, Johnson GR, et al. A Vaccine to Prevent Herpes Zoster and Postherpetic Neuralgia in Older adults. New Engl J Med 2005;352(22),2271-84.

Daily News Navigator