Water fluoridation is widely seen as one of the great public health achievements of the 20th century, credited with substantially reducing tooth decay. But there has been growing controversy among scientists about whether fluoride may be linked to lower I.Q. scores in children.

A comprehensive federal analysis of scores of previous studies, published this week in JAMA Pediatrics, has added to those concerns. It found a significant inverse relationship between exposure levels and cognitive function in children.

Higher fluoride exposures were linked to lower I.Q. scores, concluded researchers working for the National Institute of Environmental Health Sciences.

None of the studies included in the analysis were conducted in the United States, where recommended fluoridation levels in drinking water are very low. At those amounts, evidence was too limited to draw definitive conclusions.

Observational studies cannot prove a cause-and-effect relationship. Yet in countries with much higher levels of fluoridation, the analysis also found evidence of what scientists call a dose-response relationship, with I.Q. scores falling in lock step with increasing fluoride exposure.

Children are exposed to fluoride through many sources other than drinking water: toothpaste, dental treatments and some mouthwashes, as well as black tea, coffee and certain foods, such as shrimp and raisins. Some drugs and industrial emissions also contain fluoride.