Our team has examined research into the economic impact of high levels of loneliness and the associated costs and losses. Through this process, we have identified four main categories of socio-economic costs: healthcare, sick days, unemployment and life satisfaction.
For further information, please contact: ce@copenhageneconomics.com
Further references (apart from those in each sub-section):Campaign to end loneliness (2023) , Measuring your impact on loneliness in later life; Christiansen et al. (2023), Loneliness, social isolation, and healthcare utilization in the general population; Defactum (2022), Ensomhed i Danmark - analyse af befolkningsdata fra 2017; Fetzer Institute, UCLA loneliness scale version 3; Frijters and Krekel (2021), A Handbook for Wellbeing Policy-Making. History, Theory, Measurement, Implementation, and Examples; Lawton et al. (2021), Does volunteering make us happier, or are happier people more likely to volunteer? Addressing the problem of reverse causality when estimating the wellbeing impacts of volunteering; Mind Matters (2021), Building Resilience - A prescription for tackling the global mental health crisis one step at a time; New Economics Foundation (2017), The costs of loneliness to UK employers; Population at the first day of the quarter by region, marital status, time, age and sex, last quarter 2017; Population at the first day of the quarter by region, marital status, time, age and sex, last quarter 2021; Sundhedsstyrelsen (2018), Danskernes Sundhed - Den Nationale Sundhedsprofil 2017; Sundhedsstyrelsen (2022), Danskernes Sundhed - Den Nationale Sundhedsprofil 2021.
Loneliness is associated with increased prevalence of various mental and physical illnesses, leading to increased use of medical resources. The Burden of Disease in Denmark - Risk Factors, published by the Danish Health Authority (2022), estimates the total annual increase in healthcare costs in Denmark associated with loneliness. We convert this into an estimate of the annual increase in healthcare costs per lonely person by dividing by the number of lonely people. Increased healthcare costs are then allocated to municipal and regional budgets depending on the specific type of healthcare cost. Using this approach, we find that for every lonely person aged 16-64 years, loneliness is associated with increased annual healthcare costs of DKK 185 for municipal budgets and DKK 7,775 for regional budgets, totaling DKK 7,960.
For further information, please contact: ce@copenhageneconomics.com
Loneliness is associated with an increased number of sick days in the workplace. The Burden of Disease in Denmark - Risk Factors, published by the Danish Health Authority (2016, 2022), estimates the total increase in the number of sick days in Denmark associated with loneliness. We convert this to an estimate of the annual increase in sick days per lonely person by dividing by the number of lonely people in employment. The costs associated with an increased number of sick days are then allocated to municipal budgets (sick pay) and total economic losses (lost output) per lonely person. Using this approach, we find that loneliness is associated with annual sickness costs/losses of DKK 8,819 per lonely person aged 16-64.
For further information, please contact: ce@copenhageneconomics.com
Loneliness is associated with increased unemployment. We take the UK estimates from Morrish et al. (2022), which quantify the probability of being unemployed for lonely compared to non-lonely people, and apply this to a Danish context by accounting for the Danish prevalence of loneliness and the overall level of employment in Denmark. The costs associated with unemployment are then distributed across municipal budgets (cash benefits), state budgets (unemployment benefits) and total economic losses (lost production) per lonely person. Using this approach, we find that loneliness is associated with annual unemployment costs/losses of DKK 16,836 per lonely person aged 16-64.
For further information, please contact: ce@copenhageneconomics.com
Loneliness is associated with reduced life satisfaction. We estimate the economic value associated with reduced life satisfaction by i) calculating an estimate of the value of a WELLBY in Denmark, which is the value of a one-point change in life satisfaction on a scale of 0-10, by scaling the value of a VOLY, which is the value of a year of life, according to a method described by the UK Treasury, and ii) combining this estimate with a value for the effect of loneliness on life satisfaction on a scale of 0-10 (specifically, the change associated with going from being lonely "often/always" to being lonely "some of the time") that comes from the Danish Open Social Value Bank (which is based on Simetrica Jacobs). Using this approach, we find that loneliness is associated with an annual cost in terms of reduced life satisfaction of DKK 89,221 per year per lonely person aged 16 to 64
For further information, please contact: ce@copenhageneconomics.com
Source(s): Chilton et al (2020), A scoping study on the valuation of risks to life and health: the monetary value of a life year (VOLY) Final report; HM Treasury ( 2021), Key Figures Catalog; HM Treasury (2021), Wellbeing Guidance for Appraisal: Supplementary Green Book Guidance; Open Social Value Bank (2023)
To assess the benefits of loneliness interventions, we convert the costs and losses described in the previous sections into points on the UCLA 20-point scale. We assume that the average difference between a lonely person and a non-lonely person corresponds to a 30-point shift on the 20-point UCLA scale, ranging from 20 to 80 points. We identified the following intervention types based on Defacum (2023), Interventions to reduce loneliness.
Social networks: Interventions aimed at expanding participants' social networks and/or include opportunities to interact with other people.
Examples:
Meetings for seniors
Shared activities
Excursions
Social support: Interventions aimed at increasing social support through regular care.
Examples:
Visiting friend
Contact person
Mentor
Social and emotional skills training: Interventions aimed at training social and emotional skills.
Examples:
Social skills training
Role-playing
Psychological treatment: Interventions that aim to provide insight into and change unhelpful thought patterns or negative emotions.
Example:
Cognitive behavioral therapy
Mindfulness
For all intervention types, we use the estimated effect sizes of randomized control trials (RCTs). Note that we have omitted psychoeducational interventions due to significant differences between effect sizes reported in RCTs and multi-cohort studies.
For further information, please contact: ce@copenhageneconomics.com
Sources: Buckle (2015), Effects of an animal visitation intervention on the depression, loneliness, and quality of life of older people: A randomised controlled study; Creswell et al. (2012), Mindfulness-Based Stress Reduction training reduces loneliness and pro-inflammatory gene expression in older adults: a small randomized controlled trial; Defactum (2021), Interventioner, der skal mindske ensomhed; Homes and Communities Agency, Additionality Guide, Fourth Edition 2014; Larsson et al. (2016), Effects of a social internet-based intervention programme for older adults: An explorative randomised crossover study; Robinson et al. (2013), The psychosocial effects of a companion robot: a randomized controlled trial; Tabrizi et al. (2016), Effects of supportive-expressive discussion groups on loneliness, hope and quality of life in breast cancer survivors: a randomized control trial.
Long-term effects
The model is based on the assumption that the benefits of interventions last exactly one year. If the benefits are persistent, either directly, in that an intervention reduces loneliness for a period of more than one year after the intervention, or indirectly, e.g. by increasing long-term employment potential even though the effect on loneliness has diminished, the model will underestimate the total benefits of the intervention. If the benefits are less persistent, such that an intervention only reduces loneliness for, say, six months, our model overestimates the overall benefits of the loneliness intervention.
Degree of loneliness
The model is based on the assumption that interventions are equally effective for severely lonely and mildly lonely people. It is possible that the benefits of participating in a loneliness intervention depend on how lonely a participant is before the start of the intervention. It is likely that the model is most relevant for interventions targeting people who are considered relatively lonely to begin with. It is possible that the model does not accurately estimate the impact of the interventions. Depending on the participants' initial level of loneliness, it may either overestimate or underestimate the impact.
For further information, please contact: ce@copenhageneconomics.com
Social multiplier effects
We assume an approach that does not take into account the impact on spouses and/or people in close contact with participants. It is likely that less loneliness has positive spillover effects, see Frijters and Krekel 2021.
Benefits for volunteers
Our approach does not take into account potential gains for the volunteers involved in loneliness interventions. These gains may be substantial, although it is unclear whether the impact can be considered causal, see Lawton et al. 2021.
Presenteeism (presenteeism)
Our approach does not take into account the effects of presenteeism, i.e. reduced productivity in the workplace, and the benefits of increased productivity. According to the New Economics Foundation 2017, loneliness reduces productivity by 1.3% due to reduced job satisfaction. Related to this, the Mind Matters 2021 report suggests that the cost of poor mental health to employers in Europe is primarily due to sickness absence, see MindMatters (2021)
Effects due to other factors
Our approach may underestimate the reduction in health expenditure that would be associated with education, BMI, smoking, alcohol consumption, physical activity and dietary patterns because the health expenditure associated with these variables is already corrected for in the underlying data source, see Burden of Disease 2022.
For further information, please contact: ce@copenhageneconomics.com
Our model is based on studies of how loneliness affects people's health and thus healthcare costs. Several of the studies are based on correlations, and although most studies correct for observable differences between people who are lonely and people who are not lonely, such as gender, age and/or education, it cannot be unequivocally concluded that loneliness is the cause of changes in e.g. health and healthcare costs. The studies show that loneliness is the only observable factor explaining e.g. health-related outcomes, but it cannot be ruled out that the changes are due to other unobservable factors that correlate with loneliness.
The estimates for health costs and sick days associated with loneliness are based on the 2016 and 2022 Burden of Disease in Denmark - Risk Factors, which quantifies the costs/losses of often being "involuntarily alone" rather than being severely lonely. We hypothesize that being involuntarily alone can be used as a proxy for loneliness. Some studies suggest that the link between being involuntarily lonely and loneliness is weak, partly because some people may feel lonely even when they are with other people. Our estimates of the share of healthcare costs attributable to loneliness may therefore be imprecise. See for example Defactum (2022), Loneliness in Denmark - for an analysis of population data from 2017. Being involuntarily alone frequently is the best available proxy to estimate the cost of loneliness, and the evidence points to a positive association between loneliness and healthcare utilization, see e.g. Christiansen et al. (2023), Loneliness, social isolation, and healthcare utilization in the general population.
Several of the studies on which the model is based have limitations that may affect generalizability. In Defactum (2021), Interventions to reduce loneliness, which estimates the effectiveness of different intervention types, the analyses are based on studies with small samples and an older target group.
For further information, please contact: ce@copenhageneconomics.com
