Workstream 1

Workstream 1, inclusive of three studies, will build upon existing evidence to understand the landscape of loneliness and social isolation in Australia. We will use the principles of a Group Model Building approach to develop a conceptual model of loneliness and social isolation within Australia (Study 1). Utilising existing large datasets, we will examine the predictors of loneliness and social isolation across geographical units, understand how loneliness and social isolation lead and contribute to chronic diseases (including specific populations at risk), and identify geographical ‘Hotspots’ of risk of loneliness or social isolation (Study 2). Finally, the costs and quality of life impacts associated with loneliness and social isolation will be determined in those living with and without chronic diseases (Study 3).

Study 1: Group Model Building approach to understand the complexity of loneliness and social isolation for chronic disease management

Associate Professor Michelle Lim has previously developed a conceptual model that illustrates the complexity of loneliness, and how known and emerging correlates and risk factors can interact. This conceptual model will be expanded (particularly the generic categories of mental disorders and physical disorders) with input from key project partners and stakeholders adopting a whole systems perspective. A Group Model Building (GMB) approach will be employed, which is a process by which a group of stakeholders systematically engages with modellers in model construction to solve a focused problem within a complex system.A workshop will be conducted with 30 key stakeholders including decision-makers, chronic disease experts, health and community service providers, and people with lived experience. Recruitment will be facilitated via our established project partners. During the workshop, a facilitator and modeller will visually represent the problem, focusing on the casual mechanisms and develop a preliminary concept map. Stakeholders (consumers and partners) will then refine the concept map. Following the workshop, the modeller will construct a map on agreed parameters and relevant connections. This concept map will inform Workstream 2 and guide which epidemiological and economic data are required. Active participation of stakeholders will increase the likelihood that recommendations from the model developed in W3 will be accepted and implemented.

Study 2: Predicting loneliness and social isolation across geographical units for targeted chronic disease management

The drivers of loneliness and social isolation and the bi-directional relationship with chronic disease is poorly understood because these drivers and relationships are complex and dynamic. The aim of this study is to interrogate large national longitudinal datasets to answer distinct and critical research questions. Via three sub-studies we will: (i) identify the predictors of loneliness and social isolation across the lifespan; (ii) examine how loneliness and social isolation lead to development of chronic diseases; and (iii) develop a ‘Heat map’ to detect geographical areas of greater risk of loneliness and social isolation for effective chronic disease management. Relevant datasets that measure loneliness and social isolation, collect information on the presence of chronic diseases and other information that are known to be risk factors of loneliness or social isolation include: Household Income and Labour Dynamics of Australia (HILDA), 45 and up Study, Longitudinal Study of Australian Children (LSAC), Ten to Men and the Australian Longitudinal Study on Women’s Health (ALSWH). Although loneliness and social isolation are measured differently across datasets, what is important is the identification of predictors and impacts, which we will synthesise across datasets to inform our economic model in Workstream 3. Predictors to be explored will be determined by existing literature and the concept map developed in Study 1. These relative risks will be standardised for age and sex using population data from the Australian Bureau of Statistics in order to map the prevalence of loneliness and social isolation spatially by developing a ‘Heat map’. This study will provide the evidence base on the causes, correlates, and consequences of loneliness/social isolation and the mechanisms that drive a decline in health. The heatmap, which will be available via the Ending Loneliness Together website, can be used to identify hotspots to help local decision makers to direct interventions where it is most needed, by offering tailored interventions that address characteristics behind prevalence of loneliness and social isolation in each area.

Study 3: Determining the cost and quality of life impacts associated with loneliness and social isolation adjusted for chronic conditions

Costs and quality of life are key input parameters for economic modelling. The aim of this study is to determine the cost and quality of life impacts associated with loneliness and social isolation, adjusting for demographic characteristics and existing health conditions. The analyses will utilise the datasets from Study 2. These measures allow the calculation of Quality-Adjusted Life Years (QALYs); a generic measure of disease burden that combines longevity and the quality of life during those years. We will estimate age-specific utility scores by levels of loneliness and social isolation using generalised linear modelling, controlling for sociodemographic variables and chronic diseases. In the next step, we will estimate utilities for each chronic disease and for multimorbid conditions using four different estimators, from which we will develop a national catalogue of utility values for chronic diseases. Cost information will be obtained by linking all datasets (except HILDA) to the Medicare Benefits Schedule (MBS) for health professional visits and diagnostic tests and the Pharmaceutical Benefits Schedule (PBS) for medication. Costs for services not captured by MBS/PBS (e.g. hospitalisations, productivity) will be sourced from self-reported surveys that capture information on health service use. Cost and quality of life impacts associated with loneliness and social isolation will inform the model input parameters and will be a vital source for evaluators and policy-makers in future work.