Background: Nurses represent the largest proportion of healthcare workers and are in close contact with patients. Each nurse has a unique set of attitudes towards their duties and roles. Understanding nurses’ attitudes towards safety allows organisations to identify individual practices and behaviours that may have an impact on patient outcomes. Nursing-sensitive patient outcomes, such as patient falls and pressure injuries, significantly impact patients and healthcare organisations. Therefore, nursing unit characteristics, such as safety attitudes, quality of care, missed care, and staffing levels, might improve patient outcomes. To date, evidence on the association between these characteristics and nursing-sensitive patient outcomes, using empirical patient data, is limited and inconsistent. Aim: The aim of this project was to examine the impact of nurses’ attitudes towards safety, quality of care, missed care and staffing levels, on four nursing-sensitive patient outcomes, collected from two sources of data in five acute Ministry of Health (MOH) hospitals in the Kingdom of Saudi Arabia (KSA). The four nursing-sensitive patient outcomes included patient falls, pressure injuries, healthcare-associated infections, and inpatient mortality. The project also explored the relationship between nursing-sensitive patient outcomes using two data sources, namely, nurses’ perceptions of the frequency of adverse patient outcomes in their units reported in a cross-sectional survey, and the incidence of nursing-sensitive patient outcomes reported in administrative data. Methods: A multi-site multi-source, cross-sectional project was conducted. The data were collected from 35 nursing units in five acute hospitals in two regions of the Kingdom of Saudi Arabia. In phase one, an online nurse survey gathered a comprehensive range of data, including the Safety Attitudes Questionnaire (SAQ), quality of care, missed care, nurse staffing levels, and nurses’ perceptions of the frequency of adverse patient outcomes. In phase two, secondary administrative datasets on patient falls, pressure injuries, healthcare-associated infections, and inpatient mortality for the years 2018, 2019, 2020, and 2021 were analysed for each participating hospital. Data from phases one and two were merged in phase three to address the project’s aim. A variety of descriptive analyses were performed, and inferential statistics,including logistic regression and Generalized Linear Models (GLM), were fitted to examine the association between project variables. Results: The SAQ mean score was 61.48 (SD = 12.9), suggesting that nurses had less than positive attitudes towards safety. The highest mean score was in safety behaviour (Mean = 66.25, SD = 18.6), and the lowest mean score was in working conditions (Mean = 52.2, SD