The Development and Feasibility of an Ethnically Tailored Intervention to Prevent Physical Risk Factors for Falls in Older Adults: a Mixed Methods Exploratory Design

Falls are a common problem for Older Adults (OAs) throughout the world, often with very serious consequences. While risk factors such as increasing age, female gender, and low socioeconomic status are well documented in the literature, less is known about the effects of ethnicity on fall prevalence. Substantial health inequalities are known to exist between different ethnic groups, particularly among OAs, but how cultural attitudes and preferences may influence interventions designed to reduce falls is currently not well understood. This thesis follows the first two steps of the Medical Research Council (MRC) guidance for the development of complex health interventions, and using the Behaviour Change Wheel, designs an appropriate intervention to reduce fall risk in OA from a particular ethnic group.

The first objective systematically reviewed the existing literature on reported fall prevalence in different ethnic groups. Twenty-three articles were included in the systematic review, and 16 in the meta-analysis of the 16 retrospective studies that reported falls in the previous 12 months. Differences were found in fall prevalence between ethnic groups, with the Asian group showing the lowest rates at 13.9% (10.9, 16.9). The Hispanic and Black groups were similar with levels of 18.5% (13.0, 24.1), and 18.6% (13.3, 23.9) respectively. The highest rates were seen in the White group at 23.8% (18.7, 28.9). In studies that provided adjusted estimates of effect sizes for the odds/risk of falls, differences still existed between some ethnic groups, even after adjusting for other risk factors. However, heterogeneity was high, and although differences in reported falls rates were found, the reasons for these differences require further research and are likely to be complex and multi-dimensional.

The second objective analysed data from Electronic Health Records (EHR) and the Census to decide which ethnic group to choose for this study. When compared to the White group, the highest fall prevalence was seen in Bangladeshi and Caribbean OAs at 1.26 (1.05, 1.51) and 1.21 (1.04, 1.41) respectively. Census data showed that the population of over 65s in the East London boroughs of Newham, Hackney, Barking & Dagenham, and Tower Hamlets, was greater for Caribbeans than Bangladeshis, and thus Caribbean OAs were selected as the target for this thesis.

The third objective was to perform a qualitative analysis of Caribbean attitudes to ageing, falling, and fall prevention interventions. 19 semi-structured interviews were carried out with participants from the Caribbean OA community, as well as with their family members, community leaders, and medical professionals, in order to identify some of the facilitators and barriers this community have towards a fall intervention programme. Seven main themes were identified from analysis of the interviews: attitudes to ageing, the impact of Caribbean life and culture on OAs attitudes, facilitators of PA, barriers of PA, understanding of PA and health, group leader, and preferred activities.

The fourth objective was to use the results from objective three in a behavioural analysis, employing the Behaviour Change Wheel (BCW). Physical activity (PA) is a complex issue, and thus all the components of the COM-B model were found to influence behaviour. The BCW then led to the identification of 39 Behaviour Change Techniques (BCTs), of which 28 were selected as being likely to be effective after application of the APEASE criteria. These BCTs were used as the foundation of the intervention design.

The final objective was to design a protocol for an RCT used to evaluate the feasibility of the multicomponent intervention. The 12-week intervention protocol includes educational and workshop sessions, weekly supervised sessions with an instructor followed by social sessions, and PA “homework” to be carried out throughout the week. Outcomes of this study will include feasibility of recruitment, percentage of participants completing the programme, its acceptability by participants, and its successful delivery. Overall, its success will determine the feasibility of moving forward to a large-scale trial of the intervention, and the sample size necessary for this trial.

Overall, this thesis identifies differences in the fall prevalence of different ethnic groups, and presents a protocol for feasibility testing of a fall prevention intervention designed for a specific ethnic group of Caribbean OAs. The intervention is based on their particular cultural attitudes and preferences regarding ageing, falling, and PA, and by using the BCW, it is hypothesised that acceptability will be enhanced, therefore increasing adherence and eventual success in reducing falls.