Systems resilience in home care for older adults: A human factors perspective

Home care support for older adults in England is not ready for the challenges and pressures an ageing population is placing on society. Home care services provide practical care support with assessed eligible activities of daily living. It is primarily outsourced by local authorities to the private sector using a ‘time and task’ commissioning model limited to client contact time. Workforce turnover and vacancy rates are the highest in the health and social care sector, and home care workers are amongst the lowest paid occupation in England.

This PhD uses a Human Factors Ergonomics (HFE) systems approach to answer the research question; ‘What are the factors that challenge the quality and safety standards of local authority commissioned home care?’ Three empirical studies were conducted with participants working in adult social care and the home care sector using a mixed methods research design and two HFE theoretical frameworks; Systems Engineering Initiative for Patient Safety (SEIPS 2.0) and Concepts for the Application for Resilience Engineering Quality Improvement (CARE QI). SEIPS 2.0 was used to identify work system barriers and facilitators which shape work performance and care outcomes. CARE QI was used to describe the misalignments between work-as-imagined demand and capacities to promote understanding of the adaptations these gaps trigger at the point of care delivery (work-as-done).

The results show that time factors trigger resilience adaptability behaviours. These occur as frequent modifications to care support processes and tasks due to misalignments between commissioning arrangements (demand) and work organisation policies and practices (capacities). Analysis of the work-system identified five key performance shaping factors impacting care quality and safety outcomes; time factors, job design, inequality, organisational practices and resilience behaviours. These are all traceable to the commissioning arrangement ‘time and task’ model.