This thesis evaluated the existing rapid response system (RRS) based on vital signs and New Zealand Early Warning Scores (NZEWS) to recognise and respond to deteriorating patients outside of intensive care settings at Taranaki Base Hospital (TBH) through a series of studies using various methods. As the RRS is a complex system consisting of an afferent limb which captures inputs such as vital signs (VS), Early Warning Scores (EWS) and an efferent limb which acts on a set trigger defined by the afferent limb inputs and sets off Medical Emergency Team (MET) for an appropriate level of care. Firstly, this thesis describes the rate of errors and omissions within the existing RRS at the study site. It was observed that error and omission rate within the existing RRS was unacceptable. The errors and omissions accrue from the starting point (vital signs measurement, 14.9%) through the mid-point of the RRS i.e., calculation of Early Warning Scores (EWS), 25.1%, to the endpoint of the RRS (responding to deteriorating patients, 30.1%). This high error and omission rate contributes to missed opportunities to recognise and respond to deteriorating patients who have shown antecedents of significant deterioration. Secondly, this thesis describes the workload involved in the RRS activities by quantifying the time spent on each set of vital signs observations (afferent RRS). The observations from this study were consistent with the findings of recent research overseas. This thesis extends the body of knowledge by quantifying the workload involved in rapid response (efferent RRS) and offers to extrapolate workforce implications of the workload involved in the RRS activities. Then this thesis explored the staff perceptions about existing RRS and electronic RRS after a demonstration of the electronic RRS. The staff felt that existing RRS activities consumed considerable time and perceived that the manual nature of the task relied heavily on human vigilance, making it error prone. Staff also felt that ‘too many papers’ also contributed to human error. Staff were generally optimistic about electronic RRS and were keen to put it to a trial. Another study included in this thesis established that the incidence rate and survival status of the patients who underwent in-hospital cardiac arrest (IHCA) in the last five years (2016 to 2021) improved year-on-year, but significant improvement was observed after the implementation of Patient-at-Risk (PaR) nurses. This thesis extends the body of knowledge by applying Failure Mode and Effect Analysis (FMEA) methodology to identify the failure modes within existing RRS and to assess whether the likelihood of those failures would reduce using electronic RRS. The FMEA found that the electronic RRS could potentially reduce or eliminate 70.2% of these failures. The FMEA also identified the afferent RRS was more susceptible to failures (61.4%) as compared with the efferent limb of the RRS (38.6%). Based on the findings of this thesis, it is recommended that electronic RRS should replace non-electronic RRS which will not only help reduce the error and omission rate and the potential failures but also reduce the workload involved in the end-to-end RRS activities by communicating rich data between different users in real-time. This will improve user experience, speed up the efferent limb of RRS and therefore may improve the patient outcomes. It is also recommended that New Zealand should establish an IHCA registry at a national level to promote evaluation and research on the IHCA in a systematic way. Other recommendations include supporting the use of FMEA methodology in analysing complex and high-risk healthcare systems and processes.