Attitudes of pregnant women toward AI-enabled robotic nursing in obstetric and newborn care: A mixed-methods study with implications for healthcare work

Background The integration of artificial intelligence (AI) and robotic technologies into healthcare is reshaping nursing work, particularly in emotionally sensitive areas such as childbirth and newborn care. Understanding how pregnant women perceive such technologies is essential for guiding the responsible redesign of nursing roles and human–robot collaboration in obstetric care. Objective This study examined pregnant women's attitudes toward AI-enabled robotic nursing practices and explored the implications of these attitudes for the future of nursing work in obstetric and newborn care. Methods A convergent mixed-methods design was employed. Quantitative data were collected from 114 pregnant women attending the obstetrics outpatient clinics of a single public hospital in Istanbul, Türkiye, using the Artificial Intelligence Attitude Scale in Healthcare and a sociodemographic information form. In the same session, all participants took part in a semi-structured qualitative interview lasting 20–45 min. Quantitative data were analysed using descriptive statistics, independent samples t-tests, one-way ANOVA, and Pearson correlation, with effect sizes and 95% confidence intervals reported alongside p-values. Qualitative data were analysed using inductive thematic analysis performed independently by two researchers, with reporting guided by the Consolidated Criteria for Reporting Qualitative Research (COREQ). Quantitative and qualitative strands were integrated through a joint display. Results Negative attitude subscale scores (3.12 ± 1.14) were slightly higher than positive attitude scores (3.09 ± 1.13), with the general attitude mean situated at 3.10 ± 0.80; the scale does not generate a single total score, and the subscales are interpreted comparatively. Once effect sizes (Cohen's d) and 95% confidence intervals were inspected and Levene's tests for equality of variances were re-examined, none of the pairwise comparisons between subgroups reached statistical significance in the independent samples t-tests. In the one-way ANOVA, only educational level was associated with a significant difference in positive attitude scores (p = 0.038), with university-educated women reporting more positive attitudes than those educated to middle-school level or below. Three main themes were identified: (1) attitudes toward robotic nurses in delivery rooms (sense of unease; perception of privacy); (2) perspectives on robotic nurses in newborn care (trust issues; effects on mother–infant bonding); and (3) the role of robotic nurses in healthcare (assisting healthcare personnel; appropriateness for specific tasks). The joint display showed that the close but slightly negative-leaning subscale means were consistent with the qualitative pattern of simultaneous recognition of technological usefulness in selected supportive tasks and substantial reservations about the emotional, relational, and privacy-related dimensions of nursing care in obstetric and newborn settings. Conclusions Pregnant women in this sample expressed cautious attitudes toward AI-enabled robotic nursing in obstetric and newborn care, with the balance of evidence leaning marginally toward negative rather than positive evaluations. They accepted supportive, task-oriented roles for robots but resisted their substitution for relational and emotional aspects of human nursing care. These findings suggest that the integration of robotic technologies into obstetric nursing should be approached as workflow redesign that preserves emotional labour and human caring as core nursing responsibilities, while reallocating standardised and time-consuming tasks to robotic support.