Hospitals are learning to address the discrepancy between the number of admissions and the limited number of hospital beds. The increase in demand for hospital beds and the urgency to move patients out of the Emergency Department means that patients admitted can be placed in other departments' wards. These patients are called outliers. Investigating their outcomes of care using administrative data is not a straightforward process. Aggregate statistical information alone may not be able to discern some of the essential characteristics of the data. To discover insights especially when using secondary data collected for other purposes requires extensive processing of data. This study applied data science concepts to investigate the outcomes of care of outlier patients. It also constructed start-to-end patient journeys for some of the patients to give a holistic view of their journey. This paper describes the process involved in pre-processing and making the data ready for analysis. It presents results which has already enabled the discovery of knowledge on the outlier situation at the hospital. The study was done in collaboration with Nelson Marlborough Health (NMH) in New Zealand.