Understanding Change Management in Neonatal Intensive Care

It is a consistent finding from clinical and health services research that attempts to translate new research evidence and knowledge into routine clinical practice often fail. This failure often occurs at the point of implementation and integration of new knowledge into everyday practice. Neonatal medicine offers several examples of the challenges associated with effective implementation of practice change to improve care. In particular, thenutritional care of preterm infants is an area where robust evidence for practice is sometimes lacking and clinical practice is variable. However, there is some evidence for potentially better practices in this area, includingrecommendations for the nutrient intakes for these infants based on research and consensus opinion. There is also evidence that standardising care based on current evidence and consensus 'best practice' can improve outcomes.This thesis aimed to understand the factors affecting the translation of new practices into a complex health care environment and their integration into routine clinical care, and it used the development, implementation and evaluation of a complex intervention aimed at improving the nutrient intakes and growth of preterm infants in NICU as a framework to do this. The complex intervention consisted of comprehensive nutritional guidelines based on current available evidence or consensus based best-practice, a nutritional screening tool to identify infants at high risk of poor growth, a multidisciplinary nutrition team and a weekly nutrition ward round, and nurse 'champions for nutrition' who helped implement the new practices andguidelines. Normalization Process Theory (NPT), a novel sociological framework which focusses on embedding new practices into routine care, was used to develop, guide and monitor the implementation of the intervention, and to help understand the factors affecting implementation.A mixed methods approach was used. Quantitative methods were used to measure infant nutrient intakes and growth, guideline compliance, staff attitudes and intentions towards practice change (using the Theory of Planned Behaviour) and the extent to which the intervention was becoming 'normalised' into routine care (using NPT). Qualitative methods included focus groups to identify barriers prior to implementation and structured staff interviews to assess the factors affecting the normalisation of the new practices following implementation. The intervention was partially implemented during the latter five months of 2011, and then fully implemented during the whole of 2 012. Infant outcomes were compared to a pre-implementation period in the first seven months of 2011.Results demonstrated that both the partial and full implementation of the intervention was associated with modest improvements in daily protein intake and weight gain compared to the pre-implementation period. These were sustained beyond the intervention period into 2013. There was no impact on head growth. Several important processes of nutritional care (such as the timing of commencement of milk feeds and appropriate choice of milk) were not changed by the intervention. Measures of guideline compliance increased during the study period and were related to measures of normalisation using NPT. A theoretical framework outlining the factors affecting implementation was developed from the qualitative interview data, with key elements including an environment ready for change, the distribution of decisions away from doctors and towards nurses, and an emphasis on feedback of the effectiveness of the new practices.Overall, this thesis allowed insight into the process of practice change in a complex environment. The findings suggest that complex interventions using multiple elements have the potential to change practice and improve patient outcomes. NPT may offer a way of enhancing the implementation process and subsequent integration of new practices into routine care. Understanding and adjusting for the factors that influence this process may lead to improvements in clinical practice and patient outcomes.