An analysis of patient dependency data, utilizing the TrendCare system
thesisposted on 05.01.2017 by Plummer, Virginia Margaret
In order to distinguish essays and pre-prints from academic theses, we have a separate category. These are often much longer text based documents than a paper.
The allocation of nursing resources in hospitals is a major policy issue and there are controversies about whether a system based on equity ratios or one based on measurement of patient dependency is more accurate. This study is the first empirical analysis of nurse patient ratio and patient dependency data on the same patient and staff cohort. The analysis was of 103, 269 valid shifts of care representing 1,998,902 nursing hours. It is expected to be of interest to all stakeholders and notably to funding agencies that have established nursing policy using either of these two systems to measure and allocate nursing workloads. Examples of these policies include the introduction of mandated nurse patient ratios in Victoria by the Department of Human Services (DHS), the Safe Staffing Law governing hospitals and nurse patient ratios in California, USA and a staffing by TrendCare agreement incorporated in the Enterprise Bargaining Agreements of some regional Victorian public hospitals. There is a keen interest in the outcome of these policies by observers in other Australian states, New Zealand, the USA and various other international settings as many of the difficulties of accounting for nursing remain unresolved. This study was designed to inform the debate about future policy directions. The TrendCare system was selected to facilitate this analysis because it is a computerized system which has the capacity to simultaneously measure nursing workloads by a dependency method of nursing hours per patient day (HPPD) by various patient types and by nurse patient ratios. A statistical analysis of nursing hours and patient types was undertaken through retrospective analysis of existing administrative data, provided by 22 acute care public and private hospitals in Australia, New Zealand and Thailand. The results showed that both ratios and TrendCare can predict a fair allocation of nursing resources to patients. Further, the results showed that TrendCare predicts actual direct nursing care requirements with greater accuracy than ratios for the full range of settings and patient types and this facilitates better allocation of nursing resources. TrendCare predicts more of the variance than ratios, for each hospital level, public and private hospitals, for Australian and New Zealand hospitals, metropolitan and rural hospitals, all patient type categories and morning, evening and night shift. There is no category of variable where ratios predict more of the variability than TrendCare. The results also demonstrated that the cost of nursing care would be less for hospitals using TrendCare than for ratios, providing the same quality of care by the same nurses to the same patient cohort. In some cases the quality of care may be improved using the TrendCare system since it was designed to ensure quality can be maintained using predicted acuity-based resource allocation requirements. The measurement of quality was outside the scope of the thesis but this is an important outcome for the costs of care and for distribution of the limited nursing resources experienced by most countries in a worldwide shortage of working nurses.