Interest in fluoride’s affect on bone density stems back to the mid-twentieth century, when scientists first started using fluoride as a drug to build bone mass in patients with osteoporosis. From the 1960s to the 1990s, numerous clinical trials examined the effect of high daily doses of fluoride (20-34 mg/day) on bone density and fracture rates. The results were not what fluoride advocates had anticipated: although fluoride did increase spinal and pelvic bone mass, it did little to reduce fracture rates in these bones. Not only did the new bone mass do little to reduce fracture risk in the the spinal area, but the studies repeatedly found that fluoride increased the rate of fracture in the peripheral skeleton (i.e., the arms, wrists, legs, & femoral neck in the hip). The fracture rate in these peripheral bones was increased, in part, because fluoride reduced the density of cortical bone, the primary type of bone in the peripheral skeleton.

The discovery that fluoride was decreasing cortical bone density in the patients caused significant concern because cortical bone plays a key role in the integrity of the femoral neck, a bridge-like bone that connects the leg to the hip and one the key sites for osteoporotic hip fractures.

After it became clear that fluoride therapy caused detrimental effects on bone density and strength, epidemiological studies were conducted to determine if populations with elevated levels of fluoride in the water suffered similar effects. The results of these studies are discussed below.

The Early Studies

Before discussing the results of the modern epidemiological studies, it is important to address the results of similar, but much cruder, studies that were performed back in the 1950s and 1960s. (Bernstein, et al 1966; Leone, et al 1955). These early studies, which reported that communities with high levels of fluoride in the water had lower rates of osteoporosis, were the trigger that prompted the use of fluoride as an osteoporosis treatment. As numerous authors have since pointed out, however,  the studies suffered from several serious flaws. (Phipps, et al 2000; Sowers 1991; Phipps & Burt 1990). For example, (1) the studies “failed to control for important confounding variables,” (Phipps, et al 2000) as “the study populations were not well-characterized as to ethnicity, age, sunlight exposure, and other factors which might confound such a relation [between fluoride and bone mass].” (Sowers 1991). Further, as noted by Phipps & Burt:

“[The] studies in the 1950s and 1960s utilized radiography to measure osteopenia, but a major drawback with radiographs as a measurement technique is that osteopenia is not detectable until approximately 30% of the bone mass has been lost. Single photon absorptimetry, developed subsequent to these earlier studies, is a more sensitive and accurate method for measurement of cortical bone mass. . . . The computer-generated results also prevent the introduction of bias into investigations that are not ‘blind’ in terms of the exposure of interest.”
SOURCE: Phipps KR, Burt BA. (1990). Water-borne fluoride and cortical bone mass: a comparison of two communities. Journal of Dental Research69: 1256-1260.

The Modern Studies: High Fluoride Communities (2.5-4 mg/L)

In contrast to the findings of the early studies, modern studies have consistently found that exposure to higher levels of fluoride in the water (2.5 mg/L to 4 mg/L) is associated with reduced bone mass in cortical bone (as determined through measurements of forearm density).

• In 1990, Phipps & Burt reported that, after controlling for  all known predictors of bone mass, women living in a New Mexico community with 3.5 mg/L fluoride “had approximately 7% less bone mass” in their forearms than women from a community with 0.7 mg/L.
• In 1991, Sowers reported that residence in an Iowa community with 4 mg/L fluoride “was associated with a significantly lower radial bone mass in premenopausal and postmenopausal women, an increased rate of radial bone mass loss in premenopausal women, and significantly more fractures among postmenopausal women.”
• In 1996, Phipps reported that in an Oregon community with 2.5 mg/L fluoride, “each mg increase in daily fluoride intake” was associated with a “significant, although small, negative association with cortical bone BMD.”

The findings of lower cortical bone mass in high-fluoride communities is consistent with the aforementioned results of the clinical trials where high doses of fluoride were found to increase the density of trabecular bone (the primary bone in the spine) at the expense of cortical bone.

The Modern Studies: Fluoridated Communities (1 mg/L)

Research into the effects of artificially fluoridated water on bone density and bone fracture have produced less consistent results than the high-fluoride studies. Importantly, however, the studies have revealed a similar, if subtler, trend towards fluoride having a differential effect on cortical and trabecular bone.

In 2000, for example, Phipps found that exposure to fluoridated water was associated with an increase in the density of the trabecular-rich spine, but a decrease in the density of the cortical-rich forearm. (Phipps 2000). Similarly, a 1997 study by Arnold reported an increased density in the lumbar spine among young adults exposed to fluoridated water, but no increase in density in the total body or proximal femur. (Arnold 1997).

Iowa Fluoride Study (2009): Fluoride’s Effect on Bone Density in Children

More recently, a team of enthusiastically pro-fluoridation researchers from the University of Iowa published the first-ever study on the relationship between individual fluoride intake, bone mineral content, and bone density in children. (Levy, et al 2009). Although it has been known for over 30 years that “fluoride accumulates in the developing skeleton of children at a much faster rate than in adults,” (Levy 2009) and although a Surgeon General’s panel of doctors expressed significant concerns about this issue back in 1983, there has been strikingly little research on the impact of fluoride on developing bones.

In the Iowa study, the researchers monitored the fluoride intake for each child from birth to 11 years of age, and periodically measured the mineral content and density of the children’s bones. Although the authors did not find a statistically significant relationship, their data shows found that — at all ages — the highest exposed girls had lower bone mineral content and density than girls from the lowest exposed group. At 8.5 years of age, the highest exposed girls had 6.4% less bone mineral content in their hips (p = 0.01) and 4% less bone mineral content in their whole body (p = 0.02). At 11 years of age, the highest exposed girls continued to have lower mineral content, with 5.7% less mineral content in the hip and 4.3% less mineral content in the whole body (p = 0.02).

By contrast, fluoride intake was generally associated with higher mineral content and density in the boys, although the association was not as strong as the negative association detected in girls. Interestingly, the spine was the bone site that showed that greatest increase in bone mineral content and density in the boys. At 11 years of age, for example, the spines of boys with the highest fluoride exposure had 4.5% more bone mineral content (p = 0.07) and 4.4% more bone mineral density (p = 0.05).

The pattern of bone mineral changes identified by the Iowa researchers is thus roughly similar to the pattern seen with high-fluoride exposures. Specifically, the trabecular-rich spine was the bone site in boys with the strongest fluoride-induced increases in mineral content, and was the only reported bone site from girls that did not experience a decrease in mineral content.

These findings suggest that, at least for girls, fluoride’s differential effect on bone density is found among highly exposed girls in fluoridated communities, which is particularly significant when considering that reductions in cortical bone density are a key mechanism by which fluoride can increase fracture rates.

Finally, it is important to consider an often overlooked limitation with the Iowa team’s study. Although the study was designed to provide guidance about the safety of fluoride exposures in general, its study population was not representative of the general population. As the authors note, most of the children in the study were from higher socio-economic backgrounds. The children in the study were thus relatively well nourished and, hence, less susceptible to fluoride’s toxic effects. In addition, 97% of the study population was white, and just 1% were black. This is significant because studies have repeatedly shown that black children suffer significantly higher rates of dental fluorosis than white children, and thus may be more susceptible to fluoride’s toxic effects on mineralized tissues.  Accordingly, the Iowa study’s findings may have been even more pronounced if vulnerable subsets of the population had been included. As is so often the case, however, vulnerable subsets of the population have no voice when it comes to fluoride policies.

References:

• Arnold CM, et al. (1997). The effect of water fluoridation on the bone mineral density of young women. Can J Public Health. 88(6):388-91.
• Bernstein DS, et al. (1966). Prevalence of osteoporosis in high- and low-fluoride areas in North Dakota. JAMA. 198(5):499-504.
• Cauley JA, et al. (1995). Effects of fluoridated drinking water on bone mass and fractures: the study of osteoporotic fractures. J Bone Miner Res. 10(7):1076-86.
• Kröger H, et al. (1994). The effect of fluoridated drinking water on axial bone mineral density–a population-based study. Bone Miner. 27(1):33-41.
• Leone NC, et al. (1955). A roentgenologic study of a human population exposed to high-fluoride domestic water; a ten-year study. Am J Roentgenol Radium Ther Nucl Med. 74(5):874-85.
• Levy SM, et al. (2009). Associations of fluoride intake with children’s bone measures at age 11. Community Dent Oral Epidemiol. 37(5):416-26.
• Phipps KR, et al. (2000). Community water fluoridation, bone mineral density, and fractures: prospective study of effects in older women. BMJ. 321(7265):860-4.
• Phipps KR, et al. (1998). The association between water-borne fluoride and bone mineral density in older adults. Journal of Dental Research 77:1739-1748.
• Phipps KR, Burt BA. (1990). Water-borne fluoride and cortical bone mass: a comparison of two communities. Journal of Dental Research69: 1256-1260.
• Sowers MR, et al. (1991). A prospective study of bone mineral content and fracture in communities with differential fluoride exposure.American Journal of Epidemiology 133: 649-660.

Excerpts from the modern studies:

1) Water Fluoride (2.5 to 4 mg/l) & Bone Density in Adults

“The aim was to study the effects of a high level of exposure to fluoride (F) on bone metabolism in women, aged 33 to 45 years, living in the grassland area of Inner Mongolia. We measured the concentrations of F in urine and drinking water from the household of each subject. . . . The F level in the drinking water was positively correlated with the level of the markers for bone resorption. A significant acceleration of bone resorption was seen when the water F was more than 4 ppm.”
SOURCE: Zhang M, et al. (2001). Positive correlation between fluoride in drinking water and markers of bone resorption (abstract). Fluoride 34:214.

“Multiple-regression analyses were used to evaluate the effect of an individual’s daily fluoride intake (mg/day) on BMD, while controlling for significant confounding variables. Daily fluoride intake was not a significant predictor of lumbar spine BMD (parameter estimate = 0.002, p = 0.607), or proximal femur BMD (parameter estimate = 0.001, p = 0.797). Daily fluoride intake was, however, a significant predictor of forearm BMD (parameter estimate = -0.004, p = 0.015). After adjusting for all of the significant confounders, we associated each mg increase in daily fluoride intake with a 0.004 g/cm3 decrease in forearm BMD. . . .  Depending upon which fluoride exposure method is used, two different sets of conclusions can be drawn from this study. If the ecologic measure (city of residence) is used, then exposure to higher levels [2.5 mg/L] of fluoride in community water systems increases lumbar spine and proximal femur BMD. If the individual level measure (daily fluoride intake) is used, then exposure to higher levels of fluoride in community water systems decreases forearm BMD.”
SOURCE: Phipps KR, et al. (1998). The association between water-borne fluoride and bone mineral density in older adults. Journal of Dental Research 77:1739-1748.

“young women in the higher-fluoride community had significantly lower mean bone mass than did women in the control and higher-calcium communities. Furthermore, the mean loss of radial bone (primarily cortical), expressed as absolute difference or percentage of loss, was greater in women of the higher-fluoride community than in women of the control and higher-calcium communities… We could determine no reason, apart from the higher fluoride exposure, why women in the higher-fluoride community should have greater loss of bone mass than women in the other two communities.”
SOURCE: Sowers MR, et al. (1991). A prospective study of bone mineral content and fracture in communities with differential fluoride exposure.American Journal of Epidemiology 133: 649-660.

“This study investigated the relationship between cortical bone mass in an older female population and their ingestion of fluoride from community water supplies. The study was conducted among lifelong female residents in Lordsburg (3.5 ppm fluoride) and Deming (0.7 ppm fluoride), New Mexico. Although bivariate analyses showed no difference in cortical bone mass between women in the two communities, with multiple regression analyses, significant predictors of bone mass (p<0.05) were weight, years since menopause, current estrogen supplementation, diabetes, and fluoride exposure status. Based on a model containing all of these variables, women living in the high-fluoride community had a bone mass ranging from 0.004 to 0.039 g/cm2 less than that of a similar women living in the optimum-fluoride community. . . . The effect of city of residence can also be expressed as a percentage reduction in bone mass; a Lordsburg woman (high-fluoride community) had approximately 7% less bone mass than a Deming (low-fluoride community) peer of similar weight and years since menopause. . . . The negative association we found between fluoride exposure and bone mass was not an anticipated result, since this study was stimulated by the hypothesis that fluoride may actually prevent overall skeletal osteopenia by increasing cortical bone mass. However, we cannot attribute this result to bias or random error, since the other significant findings were consistent with theoretical considerations and prior research.”
SOURCE: Phipps KR, Burt BA. (1990). Water-borne fluoride and cortical bone mass: a comparison of two communities. Journal of Dental Research69: 1256-1260.

Total Daily Fluoride Intake & Bone Density in Children:

The bone density data from the Levy (2009) study is available in the following links:

• Bone Density Data for Girls
• Bone Density Data for Boys

Water Fluoride (1 mg/L) & Bone Density in Adults:

“OBJECTIVE: To determine whether fluoridation influences bone mineral density and fractures in older women. . . . Women were classified as exposed or not exposed or having unknown exposure to fluoride for each year from 1950 to 1994. Outcomes were compared in women with continuous exposure to fluoridated water for the past 20 years (n=3218) and women with no exposure during the past 20 years (n=2563). In women with continuous exposure mean bone mineral density was 2.6% higher at the femoral neck (0.017 g/cm(2), P<0.001), 2.5% higher at the lumbar spine (0.022 g/cm(2), P<0.001), and 1.9% lower at the distal radius (0.007 g/cm(2), P=0.002).”
SOURCE: Phipps KR, et al. (2000). Community water fluoridation, bone mineral density, and fractures: prospective study of effects in older women. BMJ. 321(7265):860-4.

“Osteogenic effects of therapeutic fluoride have been reported; however, the impact of exposure to low level water fluoridation on bone density is not clear. We investigated the effect of long-term exposure to fluoridated water from growth to young adulthood on bone mineral density (BMD). METHODS: BMD was measured in 24 healthy women from Regina (fluoride 0.1 mg/L) and 33 from Saskatoon (fluoride 1.0 mg/L), with no differences between groups for height, weight, lifestyle or dietary factors. RESULTS: Saskatoon women had significantly higher mean BMD at total anterior-posterior lumbar spine (APS) and estimated volumetric 1.3 (VLS), with no difference at total body (TB) or proximal femur (PF). CONCLUSION: Exposure to water fluoridation during the growing years may have a positive impact on axial spine bone density in young women.”
SOURCE: Arnold CM, et al. (1997). The effect of water fluoridation on the bone mineral density of young women. Can J Public Health. 88(6):388-91.

“In the current study, we found no evidence that exposure to fluoridated water at the optimal level of 1 mg/l had any impact on bone mass. Bone mass was measured in both the axial and appendicular skeleton and included sites of bone mass which contained predominantly trabecular or cortical bone.”
SOURCE: Cauley JA, et al. (1995). Effects of fluoridated drinking water on bone mass and fractures: the study of osteoporotic fractures. J Bone Miner Res. 10(7):1076-86.

“Bone mineral density (BMD) of the spine and femoral neck was measured in a random stratified sample of 3222 perimenopausal women aged 47-59 years. A total of 969 women had used fluoridated drinking water (1.0-1.2 mg/l) for over 10 years. These women were compared with 2253 women with low levels of fluoride in drinking water (< 0.3 mg/l). BMD of the spine was significantly higher in the fluoride group than in the non-fluoride group (1.138 +/- 0.165 vs. 1.123 +/- 0.156 g/cm2, P = 0.026). Femoral neck BMDs did not differ between the groups. When the BMD values were adjusted for confounding factors (age, weight, menopausal status, calcium intake, physical activity level, deliveries, alcohol consumption and estrogen use), the differences between the groups increased (P < 0.001 for the spine and P = 0.004 for the femoral neck, respectively). There was no significant difference between the groups in the prevalence of self-reported fractures sustained during 1980-1989. We propose that the fluoridation of drinking water has a slight increasing effect on axial BMD in women in low fluoride areas.”
SOURCE: Kröger H, et al. (1994). The effect of fluoridated drinking water on axial bone mineral density–a population-based study. Bone Miner. 27(1):33-41.