Sudhir 2009.

Methods	FLUOROSIS STUDY Country of study: India Geographic location: Andhra Pradesh Year of study: 2006‐2007 Year of change in fluoridation status: NA Study design: cross‐sectional
Participants	Inclusion criteria: school children aged 13‐15 years; lifelong residence of the region; use of the same source of drinking water from birth to 10 years of age; having permanent teeth with at least > 50% of the crown erupted and no fillings on facial surface Exclusion criteria: migration from some other place; change of source of drinking water; drinking water from more than 1 source; having orthodontic brackets; having teeth with severe extrinsic stains Other sources of fluoride: information was collected on aids used for oral hygiene maintenance (fluoridated or non‐fluoridated); no data on aids used for oral hygiene maintenance reported Ethnicity: not stated Social class: not stated Residential history: lifetime residents Other confounding factors: the questionnaire consisted of information in 2 parts: the first part consisted of information on demographic data, permanent residential address, source of drinking water, duration of use of present source of drinking water, staple food, liquids routinely consumed
Interventions	All natural fluoridation Group 1: < 0.7 ppm Group 2: 0.7‐1.2 ppm Group 3: 1.3‐4 ppm Group 4: > 4 ppm
Outcomes	Outcome: fluorosis prevalence (TF Index); Age at assessment: 13‐15 years
Funding	Not stated
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Sampling	Low risk	Used a stratified random sampling technique. The entire geographical area of Nalgonda district was divided into 4 strata based on different levels of naturally occurring fluoride in drinking water supply. So in each stratum, or for each level, several villages were involved. Sample size was divided equally among all the 4 strata, and representation from both the sexes was included in the sampling
Confounding	High risk	Data were collected on aids used for oral hygiene maintenance (fluoridated or non‐fluoridated) but not reported
Blinding of outcome assessment (detection bias) All outcomes	High risk	Insufficient information
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Data presented for all participants
Selective reporting (reporting bias)	High risk	Data not in suitable format for analysis
Other bias	Low risk	No other apparent bias