We examined the predictive validity of a segmentation strategy based on intention and cancer worry for mammography adoption and explored key factors for promoting mammography adoption in each segment. A questionnaire survey was completed by 641 women aged 40-59 years. Among them, 559 answered a follow-up survey after 15 months. They were categorized into five segments: maintenance group (S5), higher implementation intention group (S4), higher goal intention group (S3), higher worry group (S2), or lower worry group (S1). The odds of participants in each segment adopting mammography during the follow-up period were calculated. Logistic regression analysis was conducted to identify psychological predictors (five attitudes to mammography and perceived health competence) of transition to upper segments (S1 vs. S2, S2 vs. S3, S3 vs. S4, S4 vs. S5). Compared to S5, other segments did not undertake mammography at significant rates during the follow-up. The following were significant predictors for inclusion in upper segments: Lack of importance and perceived health competence were associated with inclusion in S2; lack of importance and barriers to screening were associated with inclusion in S3; perceived health competence was associated with inclusion in S4; and lack of importance was associated with inclusion in S5. These results confirm the predictive validity of a segmentation strategy, and indicate that there might be specific key factors for each segment in promoting mammography adoption.