Fig. 1

Flow diagram for cross-sectional survey. Ineligible resident will be replaced with next eligible individual within 10 households from the original unit in a descending order. Should no eligible individual be found, interviewer will recruit any eligible individual within the vicinity (100 m radius). Should more than one eligible individual be identified in one household, the KISH grid will be used to randomly select a participant. Of the 717 households, 12.1% were uncontactable after 3 attempts and 29.4% were incomplete contact attempts. A total of 270 surveys were analyzed for ever gone for mammogram due to one missing value of mammogram status and 230 surveys were analyzed for regular mammogram as 41 participants were ineligible for regular mammogram