Despite the substantial literature on how loneliness is associated with poor health and premature mortality, there is little detailed research on the extent of its economic gradients. We provide this evidence using a sample of around 400,000 respondents aged 40-70 years from the UK Biobank, who were assessed between 2006 and 2010. We focus on differences in loneliness, as well as social isolation and a lack of social support, across educational attainment, household income, local area deprivation, and recent experience of financial stress. We employ two statistical approaches, the first exploiting the large sample size and detailed geographical information about where respondents live, so we compare individuals who differ in their economic status but reside within the same postcode district. The second approach exploits the fact that for around 36,000 respondents we observe their social health and economic circumstances at two points in time (second wave of assessment conducted between 2014 and 2020), so we conduct a panel analysis that accounts for intercorrelations between the social health measures, and controls for incomplete follow-up of panel members. Across both approaches, we find a substantially higher probability of reporting loneliness, social isolation and a lack of social support, for men and women with lower economic status. Together with the existing health-loneliness literature, these findings establish a 'loneliness pathway' contributing to health inequalities, and consequently a need for effective interventions that might address loneliness and social isolation as part of a broad policy initiative on health inequalities.

The COVID-19 pandemic brought about an increased reliance on the Internet for various daily activities. Given the known digital divide, it is important to understand whether older adults changed their Internet use patterns, but current evidence is limited to cross-sectional studies. This study documents changes in frequency and types of Internet use among older adults from before to shortly after the outbreak of the COVID-19 pandemic (2018/2019 to June/July 2020), and the factors predicting regular use during these early days of the pandemic. Using data on 6,840 adults aged 50 + from the nationally representative English Longitudinal Study of Ageing, we apply longitudinal fixed-effects models to examine within-individual changes in Internet use behaviour. There was no change in the likelihood of daily Internet use between 2018/2019 and June/July 2020, despite the increased digitalisation of services over the pandemic. Daily use in June/July 2020 was negatively related to age, neighbourhood deprivation, and loneliness, and positively related to partnership status, education, employment, income, and organisation membership. Using the Internet for making calls and getting information about Government services increased, which was important given the social restrictions and overall uncertainty. However, Internet use for finding health-related information decreased. As the world moves towards digital alternatives post-pandemic, it is important to continually ensure older adults are not at risk of exclusion.

We highlight the problem of loneliness, and argue that it is not only a public health issue but also an economic problem. We provide a brief review of findings from the key literature on the associations between loneliness, mental and physical health, and healthcare costs; and then present some evidence on its trends, the extent of socioeconomic inequalities and its links with health and healthcare usage, in Australia. We hope to encourage further economics research on loneliness, and related issues of social isolation and poor social support, to aid the design of policies and interventions to reduce loneliness.