Objective To evaluate the intervention effects of low-fluoride brick tea in the population, and to provide data for the prevention and control of the brick-tea type fluorosis. Methods Eighty-six Kazakh families with 5-12 years old children were selected and divided into two groups in the severe brick-tea type fluorosis areas of Akesai County of Gansu Province. Forty-six households were intervened by drinking low-fluoride brick tea as intervention group and another 40 households drank general brick tea as control group. The fluoride content in water, tea and urine was monitored and the total daily fluoride intake of adults and children was calculated by the fluoride content of the tea before and during intervention. The baseline prevalence of dental fluorosis was surveyed in all Kazakh school students aged 5 - 12 years before intervention, dental fluorosis prevalence were surveyed in two groups after the intervention. The fluoride content in water, urine,tea, and brick-tea samples was detected by iron electrode method, and dental fluorosis was diagnosed by Dean's method. Results The fluoride content of water were 0.36,0.50 mg/L respectively before and 42 months after intervention. The total daily fluoride intake of adults and children in the intervention group (being 4.39,5.12,5.38,4.49 mg in adults and 1.90,2.33 in children, 2.33, 1.94 mg for four calculations) were lower than those in control group (8.42,9.07,8.35,7.92 and 3.65,3.93, 3.62,3.43 mg). Except the second batch (530.4 mg/kg), the average fluoride content of the other 3 batches of low-fluoride brick tea(239.3,222.88,154.7 mg/kg) was lower than that of 4 batches of market brick tea(366.9,412.2, 286.0,379.6 mg/kg). The fluoride content of low-fluoride brick tea samples was in accordance with the national standard(< 300 mg/kg) in 16 of 21 samples in 4 the batches, and the qualifying rate was 76.19%(16/21). Only 5 of 21 market brick tea samples in 4 batches was qualified, accounting for 23.80%(5/21), both were significantly different(χ2= 11.52, P < 0.01). In 12, 36, 42 months after intervention, urine fluoride content in the intervention group of adult(1.84,1.23,1.77 mg/L) and children(1.55,0.65,1.10 mg/L) was less than that of the control group (adults: 3.37,3.68,3.02 mg/L, children: 2.64,1.64,2.62 mg/L), both being statistically significant (t value were 2.94,2.43,3.91,3.29,2.31,4.42, P < 0.01 or 0.05). The detective rate of dental fluorosis was 69.02%(127/184)at baseline among children. After the intervention, it lowered to [44.83% (13/29) in the intervention group, significantly lower than that in the control group[71.88%(23/32), χ2 = 4.60, P < 0.05]. Conclusion Low-fluoride brick tea can reduce the fluoride intake of the residents who drink brick tea, and alleviate excessive fluoride and the damage of high-fluoride. Key words: Fluoride poisoning; Tea; Intervention studies; Outcome assessment