FITS into Practice: Translating Research into Practice in Reducing the Use of Anti-psychotic Medication for People with Dementia Living in Care Homes

Objectives: This paper reports on the acceptability and effectiveness of the FITS (Focussed Intervention Training and Support) into Practice Programme. This intervention was scaled up from an earlier cluster randomised-controlled trial that had proven successful in significantly decreasing antipsychotic prescribing in care homes. Method: An in depth 10-day education course in person-centred care was delivered over a three-month period, followed by six supervision sessions. Participants were care-home staff designated as Dementia Care Coaches (DCCs) responsible for implementing interventions in 1 or 2 care homes. The course and supervision was provided by educators called Dementia Practice Development Coaches (DPDCs). Effectiveness data included monitoring antipsychotic prescriptions, goal attainment, knowledge, attitudes and implementation questionnaires. Qualitative data included case studies and reflective journals to elucidate issues of implementation. Results: Of the 100 DCCs recruited, 66 DCCs completed the programme. Pre-post questionnaires demonstrated increased knowledge and confidence and improved attitudes to dementia. Twenty per cent of residents were prescribed antipsychotics at baseline which reduced to 14% (31% reduction) with additional dose reductions being reported alongside improved personalised goal attainment. Crucial for FITS into Practice to succeed was the allocation and protection of time for the DCC to attend training and supervision and to carry out implementation tasks in addition to their existing job role. Evaluation data showed that this was a substantial barrier to implementation in a small number of homes. Discussion and conclusions: The FITS into practice programme was well evaluated and resulted in reduction in inappropriate anti-psychotic prescribing. Revisions to the intervention are suggested to maximise successful implementation.

Automatically Evolving Rotation-invariant Texture Image Descriptors by Genetic Programming

In computer vision, training a model that performs classification effectively is highly dependent on the extracted features, and the number of training instances. Conventionally, feature detection and extraction are performed by a domain-expert who, in many cases, is expensive to employ and hard to find. Therefore, image descriptors have emerged to automate these tasks. However, designing an image descriptor still requires domain-expert intervention. Moreover, the majority of machine learning algorithms require a large number of training examples to perform well. However, labelled data is not always available or easy to acquire, and dealing with a large dataset can dramatically slow down the training process. In this paper, we propose a novel Genetic Programming based method that automatically synthesises a descriptor using only two training instances per class. The proposed method combines arithmetic operators to evolve a model that takes an image and generates a feature vector. The performance of the proposed method is assessed using six datasets for texture classification with different degrees of rotation, and is compared with seven domain-expert designed descriptors. The results show that the proposed method is robust to rotation, and has significantly outperformed, or achieved a comparable performance to, the baseline methods.