Deconstructing the KAP Linear Model: Evidence from Maternal Education, Age Outweigh Knowledge and Attitude in Antenatal Care Utilisation in Northern Nigeria

Abstract

Background: The Knowledge-Attitude-Practice (KAP) framework assumes that knowledge shapes attitude and that attitude subsequently influences practice. Aims: This study examined whether maternal knowledge and attitude predicted antenatal care (ANC) utilisation more strongly than education and age among recently delivered women in Bauchi Local Government Area, northern Nigeria. Study Design: A community-based, cross-sectional, analytical survey. Place and Duration of Study: Six randomly selected wards in Bauchi LGA, Bauchi State, Nigeria. Data were collected as part of a Master of Public Health research project at Ahmadu Bello University, Zaria, and the study report was submitted in May 2026. Methodology: A structured, interviewer-administered questionnaire was used to collect socio-demographic, knowledge, attitude, and ANC utilisation data from 405 women who had delivered within the preceding 6–12 months and were recruited through multistage random sampling. Knowledge and attitude were classified using standard cut-offs, and adequate ANC utilisation was defined as four or more visits. The predictive contributions of knowledge, attitude, education, and age were assessed using chi-square analysis, and independent predictors were identified using multivariable binary logistic regression. Results: Knowledge was classified as Poor in 71.4% of women, and attitude was classified as Positive in 95.3%; however, neither knowledge category (χ² = 3.07, P = .08) nor attitude category (χ² = 0.22, P = .64) was significantly associated with adequate ANC utilisation. In contrast, educational attainment (P = .006) and maternal age (P < .001) were significantly associated with utilisation. In the adjusted model, tertiary education (AOR = 3.46, 95% CI: 1.71–6.99) and age ≥25 years (AOR = 1.89, 95% CI: 1.10–3.24) independently predicted adequate utilisation, whereas attitude (AOR = 1.12) and knowledge (AOR = 1.36) did not reach statistical significance. Conclusion: The findings do not support the linear KAP assumption that knowledge influences attitude, which, in turn, influences practice. In this setting, socio-demographic characteristics were stronger predictors of ANC utilisation than knowledge and attitude, indicating that structural models of health-seeking behaviour may usefully complement the classical KAP framework.