- Click here for explanation of classification scheme.

To reveal other possible results from prior studies related to fluoride content of water and prevalence of Down syndrome births.

Prior research was specifically selected for review in this analysis.

Not specified.

Recruitment

Specific subject data can be found in the original studies but consisted mainly of the following:

• Metropolitan Atlanta Congenital Malformations Surveillance Program (1960 to 1973)
• National Intelligence Service
• 44 large US cities.

Design

The data from prior investigators was re-analyzed in order to determine the following:

1. Quality of data ascertained from birth certificates vs. multiple sources
2. Detection of fluoride linked Down syndrome in younger mothers
3. Potential of daily fluoride intakes as one of the major causes of Down syndrome births
4. Excess Down syndrome births linked to artificial water fluoridation worldwide.

Intervention

Reassessment of statistical analysis techniques resulted in alternate conclusions from the initial study researchers.

Statistical Analysis

• Chi square for 2 X 2 fold tables with combination of subgroups as needed
• Exponential regression models
• Heyroth's formula for calculating total daily intake of fluoride
• Linear regression analysis
• Histogram
• Algebraic averages and means.

Timing of Measurements

Measurements were taken as described in the original studies. 

Dependent Variables

Incidence of Down syndrome births.

Independent Variables

• Maternal age
• Fluoride exposure
• Sources of data collection. 

 

Meta-analysis combination of three study participants provides:

• N=2,170,986 births (maternal age less than 30=1,686,621)
• Age of participants was assessed as less than 30 or other
• All studies were conducted in the US.

Please refer to specific studies for more detail. 

Following is a table overview of the three main studies: [Metropolitan Atlanta Congenital Malformations Surveillance Program (Atlanta), National Intelligence Service (NIS), 44 US Cities (44C)] and descriptions of participants relating to Fluoride (F), Non-fluoride (NF), Down syndrome (DS) and Birth rate per 10,000 births (BR).

	Maternal Age	F  DS/Total Birth	NF DS/total birth	F DS BR	NF DS BR	Difference DS-BR F & NF
Atlanta	All	166 / 166,182	86 / 101,639	10.0	8.5	1.5
Atlanta	Less than 30	94 / 133,942	33 / 82,815	7.0	4.0	3.0 (P<0.005)
NIS	All	115 / 234,300	524 / 1,032,100	4.9	5.1	-0.2
NIS	Less than 30	49 / 173,500	181 / 773,800	2.8	2.3	0.5
44C	All	178 / 432,580	90 / 204,185	4.1	4.4	-0.3
44C	Less than 30	90 / 353,148	48 / 169,416	2.6	2.8	-0.2

 

 

 

Is the presence of fluoridated water associated with the rate of Down syndrome births? Research on this contentious issue is inconsistent. Some studies find an association while others do not find an association. The author of this article examines the research and finds that the data sources and the way that the data is analyzed makes a difference in terms of the finding.

Part 1:  Quality of Down Syndrome Birth Rate Data Ascertained by Birth Certificates vs. Multiple Sources

• Quality of data ascertained by hospital records and multiple sources may be more useful than that obtained from birth certificate alone. For instance, one study that looked at birth certificates only as a way to estimate the rate of Down syndrome births (Erickson/birth) showed a lower rate (0.46 per 1000 births) than a study that examined hospital records (Liliental and Benesch) (1.32 per 1000 births).
• Statistical analysis of Down syndrome birth rate should be performed separately on younger mothers less than 30 (or 32) years of age and on older mothers more than 30 or 32 years of age.

Part 2: Detection of Fluoride Linked to Down Syndrome in Erickson's 1976 Data for Younger Mothers

• When only women less than 30 years of age are analyzed, the method of analysis makes a difference for the findings. For instance, when women younger than 30 are broken down into three age groups (19 years or younger, 20 to 24 years and 25 to 29 years), there was no association between fluoridation and Down syndrome. However, when all women under 30 years are combined, there is a significant association (chi square = 8.04, P<0.005).
• Data obtained by birth certificate alone lacks quantitative and qualitative information necessary to negate a fluoride link to Down syndrome birth. The author suggests drawing on data from hospital records and multiple sources.

Part 3: Daily Fluoride Intakes as One of the Major Causes of Down Syndrome Births

The author then attempts to test the hypothesis of a relationship between level of fluoride ingestion and Down syndrome birth rate. The author uses older (rather than newer) formulas to estimate daily intake level of fluoride, though there is no justification for why this approach is chosen. Since the data did not include the level of fluoride in non-fluoridated areas, a "reasonable" amount of fluoride needed to be calculated. Finally, it is unclear what other confounding factors were included in the regression equation. Given these assumptions and ambiguities, the author reports that the total daily intake of fluoride appears to promote Down syndrome rates, second only to maternal age when applied to the existing data sets.

Part 4:  Excess Down Syndrome Births Linked to Artificial Water Fluoridation Worldwide

Given the above assumptions and analyses, the number of Down syndrome babies linked to water fluoridation worldwide may be estimated to about 3,000 cases per year (range 1,120 to 6,000).

Higher levels of fluoride intake may lead to increased risk of Down syndrome.

University/Hospital:	Tokyo University Medical School
Other:	Not reported

Strengths

Researcher appears to take a thorough look at statistical analysis and data sources for each study, all conducted in the US.

Weaknesses

• Assumptions for reanalysis of the data (why given formulas are used, "reasonable" amounts of the presence of fluoride in water given the absence of this information in the original data and insufficient explanation of the confounders examined in the multiple regression) are not adequately supported or justified
• Some data is ascertained through personal communication and extrapolation
• Limitations of the study briefly discussed, funding source was not identified.