Abstract
<title>Abstract</title> <p>Background Population aging has created an urgent need for accurate identification of heterogeneous Older adults care service needs. Current assessment methods rely predominantly on large-scale questionnaires, which are costly, subjective, and difficult to implement at the grassroots level. Routine physical examination data, particularly two-cancer (breast and cervical cancer) screening results, offer a promising low-cost alternative, yet their secondary use in care needs assessment remains underexplored. Methods This study utilized two-cancer screening data from 810 women aged 55–80 years across 6 districts and 12 counties in Guilin, China. Breast ultrasonography (BI-RADS grading) and liquid-based cytology (TBS classification) were employed as core indicators. Descriptive statistics, chi-square tests, and Pearson correlation analyses were conducted to examine age-specific distributions and associations. A three-stage mapping framework — "screening results → health risks → care needs" — was constructed as a theoretical lens to translate objective health indicators into stratified care need inferences. Results The overall health status of the sample was favorable, but the sample was heavily skewed toward younger age groups: 80.4% were aged 55–64 years, whereas the 75–80 age group accounted for only 1.9%. Notably, this oldest-old group, despite showing the highest LBC positivity rate (26.7%) and dual-abnormality rate (6.7%) in our data, had the lowest screening participation rate, revealing a pronounced "high-risk, low-participation" phenomenon. However, given the extremely small sample size (n = 15) in this age group, these risk estimates should be interpreted with caution and may reflect selection bias (the "healthy worker" effect). Age was significantly associated with LBC classification (P = 0.003) and comprehensive risk level (P = 0.002), but not with BI-RADS grade (P = 0.184). Breast ultrasound and LBC indicators were independent (r = − 0.004, P = 0.920), and combined assessment demonstrated superior risk identification value compared to single-indicator evaluation. Based on the mapping framework, three distinct care need patterns emerged as theoretical inferences: younger-old women (55–64 years) primarily require preventive health management; middle-old women (65–74 years) need health monitoring and facilitated medical access; and the oldest-old (75–80 years) may demand intensive medical care and integrated medical-Older adults services, though this inference awaits verification with larger, more representative samples. Conclusions Routine two-cancer screening data can serve as a valid, low-cost, and objective supplementary source for older adults care needs assessment. Combined analysis of breast and cervical screening indicators effectively identifies high-risk groups and can help infer differentiated care needs, providing empirical support for precision allocation of older adults care resources at the grassroots level. Critically, the extremely low participation of the oldest-old underscores the urgent need to improve service accessibility for this vulnerable group rather than simply labeling them as high-risk based on limited data.</p>