Apr 2, 2013

The Ministry of the Environment Thyroid Survey Follow-up: A Comparison Study or Not?


This is a follow-up report to the article below:


On March 29, 2013, the Ministry of the Environment released a detailed report of the “Study of the presence of thyroid findings in three prefectures outside Fukushima Prefecture,”  following a March 8 release of the preliminary report.  This study is referred to as the “MOE thyroid survey” in the analysis section.


The MOE thyroid survey was conducted in order to find out the incidence of thyroid nodules and cysts in children in areas not affected by radiation contamination due to the Fukushima nuclear accident.  


Complete translation of the official report is followed by an analysis of data provided.
Original report in Japanese is found at this link: http://www.env.go.jp/press/press.php?serial=16520

*****
March 29, 2013 (Heisei 25)

Announcement in regards to the result of the study of the presence of thyroid findings in three prefectures outside Fukushima Prefecture.

As the fiscal year Heisei 24 (2012) project, the Ministry of the Environment implemented thyroid ultrasound examinations on a given number of individuals in three prefectures outside Fukushima Prefecture due to the fact that about 40% have findings of small cysts, etc., (so-called A2 assessment) in the thyroid examination of the Prefectural Resident Health Management Survey conducted by Fukushima Prefecture.  We would like to report our results.

Results of the study of the presence of thyroid findings in three prefectures outside Fukushima Prefecture

1. Background and purpose of the study

In the thyroid examination of the Prefectural Resident Health Management Survey by Fukushima Prefecture, about 40% have findings such as small cysts less than 20.0mm in size (*1).  These small cysts (*1) do not need any further detailed examination, but we have been told that the inclusion of such small findings in the result has created anxiety among residents.

This type of a large scale, sophisticated ultrasound examination is the first such attempt in the world, and not much accurate information exists in regards to the frequency of thyroid cysts in children as well as the potential variability that may be seen in the result.

Under such circumstances, the Ministry of the Environment decided to conduct an ultrasound examination survey (*2), similar to the one conducted as part of the Prefectural Resident Health Management Survey, in children from three prefectures outside Fukushima Prefecture.  The purposes of this survey is to promote the understanding by the residents and provide information in regards to the validity of the result.

(*1) Meaning cysts without solid components.
(*2) Thyroid ultrasound examination conducted in this survey is a screening examination, not intended for diagnosis of thyroid cancer.

2. Summary of the survey

(1) Survey period
November 2012 to March 2013

(2) Survey contractor
The Japan Association of Breast and Thyroid Sonology

(3) Survey areas and the number of subjects surveyed


(4) Survey method
○ Conduct thyroid ultrasound examination in the subjects with the same standard as the Prefectural Resident Health Management Survey.
○ Findings will be assessed with the same guideline followed by the Prefectural Resident Health Management Survey.  Frequency of thyroid abnormalities such as cysts will be calculated in surveyed areas.

3. Survey results: summary

(1) All surveyed areas: summary

[1] Number and proportion of subjects by assessment category
Cysts here means cysts with no solid component.

[2] Number and proportion of subjects for each assessment category (by sex and age)

[3] Number and proportion of subjects who had nodules and/or cysts


(2) Result by surveyed area: summary

Contact information


Yasuo Kiryu
Senior official in charge of Radiation Health Management
Environmental Health Department
Environmental Policy Bureau
Ministry of the Environment
Phone 03-5521-9248 x6375
           03-3581-3351

Yoshie Hirose
Assistant to senior official in charge of radiation health department
Phone 03-5521-9248 x6396

*****
Analysis

The MOE thyroid survey was originally intended to be a “comparison” study to the Fukushima thyroid survey, according to a story published by Mainichi Shimbun on August 26, 2012. (A complete translation of this story can be found at this link: http://fukushimavoice-eng.blogspot.com/2012/08/thyroid-examination-cabinet-office.html)

Oddly, there is no mention of the word “comparison study” anywhere in this report, nor the mention of such terms as “control groups.”  In addition, the original story mentioning “comparison study” has been partially erased so that only the last half of the story appears.

Yet the news coverage of the March 9th Asahi Shimbun report mentions, “To obtain control data for comparison,” indicating that in the eyes of the media and the general public, this is clearly perceived as a comparison study.  (Incidentally, despite the photo of a little girl taking liquid stable iodine, which might suggest the Fukushima children were protected by iodine, only a small group of Fukushima residents actually received iodine.  The photo was taken at an evacuation center in Kawamata town, where Futaba town residents had evacuated.  Futaba town already had a supply of iodine as a town very close to Fukushima Dai-ichi nuclear power plant.  Kawamata town residents, on the other hand, were not given any iodine: rather, the town officials were telling people there was nothing to worry about, which was the perception at the time).

However, the Ministry of the Environment made no effort whatsoever to compare their results to the Fukushima thyroid survey results.  Moreover, their presentation of data is not very clear in some cases, despite their alleged intention “to promote the understanding by the residents and provide information in regards to the validity of the result.”  An effort was made here to present the MOE data along with the Fukushima data for the ease of comparison in order to help the general public understand what the numbers may or may not mean.

A word of caution is warranted here:  whether the MOE thyroid survey constitutes a truly good set of comparison data is another matter.  Nevertheless, the data is presented here in a comparative manner for further analysis and interpretation by whoever is interested in this subject matter.

Another critical piece of information regarding the MOE thyroid survey is that the MOE official in charge, Yasuo Kiryu, has an interesting track record, according to a freelance Japanese journalist, Oshidori Mako.  Kiryu was in charge of the Special Disease Management Division at the MOE handling the Minamata lawsuits in 2011.  Apparently his division demanded that the physician who was asked to be an expert witness for the government lie about the plaintiff having Minamata disease.  In other words, the physician intended to testify that the plaintiff indeed had Minamata disease, but Kiryu’s division repeatedly asked the physician to state otherwise.  The physician refused and was taken off as an expert witness.  The government found another physician to testify, and the plaintiff lost.  This was in June 2011.  Kiryu became a senior official in charge of Radiation Health Management as of September 2012.


*****
1. Background and purpose of the study

MOE’s mentioning of “small cysts less than 20.0 mm” is rather misleading.  There is a big difference between a 3.0 mm cyst and an 18.0 mm cyst, and it seems strange to call them “small” collectively.  Moreover, MOE did not provide the size distribution of nodules and cysts as the Fukushima thyroid examination does on page 4 of the PDF shown in the link below.  

Proceedings of the 10th Committee Meeting for Fukushima Health Management Survey
Thyroid ultrasound examination results

2. Summary of the survey

(1) Survey period

November 2012 to March 2013

The actual thyroid ultrasound examination was apparently finished by late January 2013, but it took MOE some time to compile the data to present the results to the parents of the subjects in three different locations.

(2) Survey contractor

The Japan Association of Breast and Thyroid Sonology

The survey was commissioned to the Japan Association of Breast and Thyroid Sonology (JABTS) apparently after an open bidding process.  However, JABTS is one of the seven organizations that were originally involved with setting the guidelines for the Fukushima thyroid ultrasound examination.  This was a process that lacked transparency, as explained halfway into this article, immediately below the first two tables:  http://fukushimavoice-eng.blogspot.com/2012/09/regarding-comments-made-by-ippnw.html

Additionally, Shinichi Suzuki, a physician in charge of the Fukushima thyroid examination, is on the board of the JABTS.  Does this not represent a conflict of interest?

In addition, according to Oshidori Mako, the actual survey was subcontracted to medical schools and/or hospitals associated with national universities.  http://news-log.jp/archives/7348/3

In case of Nagasaki-city, it was confirmed to be Nagasaki University, which is a home base for Shunichi Yamashita, a former vice president of Fukushima Medical University.

(3) Survey areas and the number of subjects surveyed

It is unclear how the three prefectures, Aomori, Yamanashi and Nagasaki, were selected.  
For information, this map shows geographical relationships between those prefectures and Fukushima.

As for the selection of the subjects, this also seems to lack transparency, but information from the Internet revealed that they were students at schools associated with national universities.  There are kindergartens (age 3-5), elementary schools and middle schools, usually attached to the school of education at some universities.  The three prefectures selected have no high schools at national universities, so it is not clear how 16-18 age groups were assembled.  Nevertheless, this does not seem to represent a balanced and generalized sample population, not to mention not matching the age groups in Fukushima thyroid examination in terms of the youngest age group not including those under age 3 like Fukushima.


The number of subjects in each age group and the proportion in the subject population is tabulated below along with similar information from Fukushima.


As you can see, total number of the subjects in each of the three prefectures varies.  In addition, the age distribution varies widely amongst the three prefectures in the MOE survey and also between them and Fukushima.  In particular, age groups 11 to 15 and 16 to 18 are in higher proportions than Fukushima in each of the three, and these are the age groups which have higher proportions of abnormalities.

Information in regards to the sex distribution was not provided for each of the three prefectures in the MOE survey, although it was provided for all three combined.

(4) Survey method

○ Conduct thyroid ultrasound examination in the subjects with the same standard as the Prefectural Resident Health Management Survey.
○ Findings will be assessed with the same diagnostic guideline followed by the Prefectural Resident Health Management Survey.  Frequency of thyroid abnormalities such as cysts will be calculated in surveyed areas.

Unlike objective tests such as blood tests which can be standardized by using similar machines, similar methods and similar calibrations, ultrasound examination interpretation might be more influenced by examiners’ subjectivity, experience and training.  Even using the same diagnostic guideline might not be foolproof if the findings are not consistently defined amongst examiners.

Moreover, Oshidori Mako, who has been in direct contact with many Fukushima residents since the disaster began, frequently visiting highly contaminated areas such as Iitate village, has been told by some parents about the short length of the Fukushima ultrasound examination.

Some Fukushima parents who took their children to facilities outside Fukushima Prefecture for second opinions were surprised how much longer the physician/technician spent conducting each ultrasound examination.  Often the findings were more extensive, such as a larger number or size of cysts and nodules.

Thus, it is possible to have different sets of findings on an individual depending on the examiner.

3. Survey results: summary

(1) All surveyed areas: summary

[1] Number and proportion of subjects by assessment category

For ease of comparison with the Fukushima thyroid examination, MOE results were tabulated together with the Fukushima results as below.
A word of caution is warranted here not to be fooled by just looking at percentages.  It is not fair to compare things with different conditions.  As the MOE survey population appears to differ from the Fukushima survey population in terms of age distribution and proportion, and for a reason to be explained in the next section, it might not be appropriate to compare the MOE results with the Fukushima results.

However, differences between Fukushima results from FYH23 (Fiscal Year Heisei 23 or 2011) and FYH24 (Fiscal Year Heisei 24 or 2012) might be worth noting.  Both A2 and B assessments have higher proportions in FYH24.  This might be explained by progression of abnormalities in thyroid gland with time and/or continuous radiation exposure these Fukushima children are still subject to.

As for the assessment category B, it may not be appropriate simply to compare percentages of those who “qualify” for secondary examination.  Sometimes the lesions might be totally benign, so it is more meaningful to compare the “results of the secondary examination.”

Proceedings of the 10th Committee Meeting for Fukushima Health Management Survey, held on February 13, 2013, revealed that of the 186 children in category B from FYH23, 162 underwent secondary examination which consists of more detailed ultrasound examination, blood and urine tests, and biopsy if warranted.  76 out of the 162 had a fine-needle aspiration biopsy, and 10 were suspected of having cancer.  As of February 13, 2013, 3 out of the 10 had been confirmed to have cancer and already undergone surgery.  The remaining 7 were awaiting an excisional biopsy, with 80% chance of having cancer.  An average age of these 10 children was 15, an average size of the lesion was 15 mm and 3 were boys and 7 were girls.

These cancers, according to Fukushima Medical University, are not unlikely to be related to radiation exposure, since it has only been two years since the accident and the latency for pediatric thyroid cancer was at least four years for Chernobyl.  They claim these cancers were merely “latent cancers” which were already present and might have eventually be found at a later time, which happened to be discovered due to a mass-scale ultrasound screening.  

If Fukushima’s results are no different from other prefectures’ results, one would expect to find an equivalent proportion of the B assessment to have cancer.  In Fukushima, it is 10 suspicious cases (3 confirmed) out of 76 who had fine-needle aspiration biopsy, out of 162 who actually underwent secondary examination, out of 186 who qualified for secondary examination.  This means 1.6 to 4.3% case of “latent” cancer, with 4.3% based on about 8 total cases of cancer as the unconfirmed 7 have 80% chance of having cancer.  In the MOE study, then, there could be 1 to 2 “latent” cancer cases.

[2] Number and proportion of subjects for each assessment category (by sex and age)

The table provided in the original MOE report for the number of subjects for each assessment category by sex and age was retabulated in the same format as the similar tables in the Fukushima thyroid examination to follow.


MOE: number of subjects for each assessment category by age and sex




Fukushima FYH23 (top) and FYH24 (bottom): proportion of subjects for each assessment category by age and sex


The table provided in the original MOE report for the proportion of subjects for each assessment category by sex and age was also tabulated in a similar manner as the table for the number of subjects as above, although this type of table was not provided in the Fukushima results.

MOE: proportion of subjects for each assessment category by age and sex

Fukushima FYH23: proportion of subjects for each assessment category by age and sex

Fukushima FYH24: proportion of subjects for each assessment category by age and sex

These “proportion of subjects for each assessment category by age and sex” tables turned out to be quite revealing.

The very right column in each table shows a proportion of male vs female for each age group.  The figures in the right bottom corner shows a proportion of male vs female for each survey population group.  As you can see, the female:male ratio is about 1:1 in both Fukushima groups but 1.1:1 in the MOE group.

When you look at the age 11-15 group, the female:male ratio is about 1:1 in all three survey groups, but the age 16-18 group show a much higher female:male ratio of 1.6:1 in the MOE group, compared to 1.05:1 to 1.1:1 in the Fukushima groups.

The very bottom row in each table, named “Total,” shows female:male ratios for each assessment category with all age groups included.  In all three survey groups, the female:male ratio is not too different at 1.1:1 to 1.2:1 in A2 and 1.9:1 to 2.3:1 in the B assessment, with the MOE group on the higher end.  However, if you focus on the age 16-18 group, the female:male ratio is approximately 2:1 in both Fukushima groups, but it was significantly higher at 4.7:1 in the MOE group.  In the age 16-18 group, the female:male ratio in the A2 assessment is 1.2:1 in both Fukushima groups and 1.7:1 in the`` MOE group.

In summary, the MOE survey group has a higher female:male ratio than the Fukushima groups as a whole, but it is especially high in the age 16-18 group.  In addition, the female ratios in the A2 and B assessment categories are higher than the Fukushima groups.  This might have resulted in the MOE’s A2 and B assessments exceeding Fukushima’s.

Can you truly call such a survey population a “control group” and its data “baseline” as a general guideline?

On the other hand, according to the Swiss study shown in the link below, it is known that thyroid gland goes through a rapid growth in puberty.  In pubertal female, an increased activity of the thyroid gland is related to menarche.  In addition, female hormones might have stimulating and male hormones inhibitory effects on thyroid gland and its endocrine control axis.

This suggests it might not be unusual to find thyroid abnormalities in the age 16-18 female group, which could be called a true “screening effect.”  However, if that is the case, the age 16-18 female group in Fukushima should also show a similar degree of thyroid abnormalities.

The table provided in the original MOE report for the proportion of subjects for each assessment category by sex and age was also made into bar graphs showing the proportion of each assessment category by age group for each sex, in the same format as in the Fukushima thyroid examination.

MOE survey population: male

Fukushima survey population: male


MOE survey population:

female


Fukushima survey population: female



MOE survey population: male and female combined
In the MOE survey group, A2 ratio in each group is higher in both sexes than the Fukushima survey groups, but it is especially higher in the age 11 to 15 and the age 16 to 18 groups.
As can be seen in the first table showing the number of subjects in each age group and the proportion in the subject population, shown again below, these two age groups occupy higher ratios in the MOE survey group (45.7% for the age 11 to 15 group and 20.8% for the age 16 to 18) than the Fukushima survey groups (30-35% for the age 11 to 15 group and 10.6-16.0% for the age 16 to 18 group).


This confirms what was previously mentioned:  the MOE survey group has a higher proportion of the age groups which have higher proportions of the A2 assessment as well as higher proportions of females.

[3] Number and proportion of subjects who had nodules and/or cysts

The table provided by the MOE report was tabulated together with two sets of results from Fukushima thyroid examination.
MOE did not provide the size distribution for the nodules and cysts in either A2 or B categories.  It does not seem reasonable to simply compare proportions without putting the numbers in proper context, as previously stated throughout this analysis.

(2) Result by surveyed area: summary

The table provided in the official MOE report was colored in a similar way to the table with Fukushima results, which is shown below.





It is unclear why Yamanashi Prefecture has a much higher proportion of the A2 category than other two prefectures and Nagasaki has a much lower proportion of the B category than the other two.  Sex distribution for each of prefectures in the MOE survey was not provided.  

Rather than simply drawing a conclusion of “Fukushima results are about the same as other prefectures,” isn’t it more important to investigate why such variabilities are seen in the MOE survey prefectures as well as whether if the MOE survey truly represents “normal” incidence of pediatric thyroid abnormalities?

The A2 assessment of 56.5 % in the MOE survey group leads to several hypotheses.

1. This is the normal incidence of childhood thyroid abnormalities.
2. Something is causing thyroid abnormalities in children all over Japan.
3. The MOE number is higher due to biases such as uneven age and sex distribution of the surveyed population.
4. There is some error in the study design and/or parameters.

It may be that no one know the full truth.  At least it will be difficult to figure it out without more detailed information such as the sex distribution of the each MOE prefecture, the amount of time spent on ultrasound examination for each MOE survey subject, etc.  It might be necessary to compare this data to a similar survey outside Japan, which might not be easy.

For now, we await the final word on the fate of the seven suspected cases of thyroid cancer from the FYH23 Fukushima group, the results of secondary examination from over 700 children from the FYH24 Fukushima group, and at last the results of the first screening examination from the remaining 158,783 children from the FYH25 Fukushima group which will be ongoing until March 2014.  

Meanwhile, the ones from FYH23 with the A2 assessment, who did not require secondary examination, are still living with radiation in Fukushima yet not getting follow-up thyroid ultrasound to monitor their nodules and/or cysts until FYH26, which begins in April 2016.






No comments:

Post a Comment