Predicting Rural Women's Breast Cancer Screening Intention in China: A PLS-SEM Approach Based on the Theory of Planned Behavior

BACKGROUND: It was reported that the incidence of breast cancer (BC) was the highest among cancers worldwide. The breast cancer screening (BCS) program is regarded as an effective preventive measure. However, rural women's willingness to participate in the BCS program is relatively low. To provide measures to prevent BC, it is necessary for the government to identify the influencing factors of rural women's BCS intention. METHODS: A cross-sectional study was conducted among 3,011 rural women by a convenience sampling method through face-to-face interviews on a self-designed questionnaire based on the theory of planned behavior (TPB). The partial least square structural equation model (PLS-SEM) was conducted to determine the predictors of BCS intention, and a multi-group analysis (MGA) of age was performed to identify if there were differences in all hypotheses between different age groups. RESULTS: There were still rural women who have not been screened for BC in five years (41.7%). The research model of rural women's intention to accept this prevention against BC was rational. All of the hypotheses are supported. Especially, subjective norm (SN) (β = 0.345, p < 0.001) is found to be the strongest predictor followed by the perceived behavioral control 1 (PBC 1) (personal factors, including distance, transportation, busyness, etc.) (β = 0.165, p < 0.001), attitude (β = 0.152, p < 0.001), past behavior (PB) (β = 0.150, p < 0.001), knowledge (β = 0.121, p < 0.001), and perceived behavioral control 2 (PBC 2) (pain and cultural-social factors including embarrassment from a physician, etc.) (β = 0.042, p < 0.05). The advocacy and education (A&E), medical level and service attitude (ML&SA) of township health centers and village clinics can affect behavior intention (BI) via attitude, SN, and PBC. The results of MGA of age indicate that there are significant differences among rural women of different ages regarding the relationship between A&E and PBC 2 (p < 0.01) and the effect of PB on BI (p < 0.001). CONCLUSION: The TPB with the addition of PB, knowledge, ML&SA, and A&E can provide the theoretical basis for the policy intervention that aims to enhance the rural women's BCS willingness. MGA of age is conducive to promoting the implementation of the BCS policy. The findings are of great significance to improve rural women's health levels.