By Lindsey N. Clark, MD, and Nancy J. Selfridge, MD

Dr. Clark is a Clinical Skills Facilitator, Clinical Foundations Department, Ross University School of Medicine, Barbados, West Indies.
Dr. Selfridge is Professor, Clinical Foundations Department, Ross University School of Medicine, Barbados, West Indies.

SYNOPSIS: Loneliness appears to be an independent risk factor for type 2 diabetes, although further research to identify the causal relationship between loneliness and type 2 diabetes development is needed.

SOURCE: Hackett RA, Hudson JL, Chilcot J. Loneliness and type 2 diabetes incidence: Findings from the English Longitudinal Study of Ageing. Diabetologia 2020;63:2329-2338.

Loneliness, or the perception of unmet social needs and dissatisfying social relationships, is a common experience. Forty percent of adults older than age 65 years report feelings of loneliness, and studies show loneliness tends to increase with advancing age.1

In addition to the feeling of a persistent negative emotional experience, loneliness appears to affect physical and mental health. Various studies have established loneliness can be a predictor for all-cause mortality and has been associated with chronic diseases, such as coronary heart disease, hypertension, metabolic syndrome, cognitive decline, and dementia.1,2

Despite the growing body of literature reporting the associations between loneliness and chronic inflammatory diseases, the relationship between loneliness and type 2 diabetes remains understudied, particularly examinations of loneliness as a potential risk factor for type 2 diabetes. With estimates showing 462 million individuals affected by type 2 diabetes globally, diabetes is ranked as the ninth leading cause of mortality.3,4 At the same time, there is growing global concern over a “loneliness epidemic” that is infiltrating society. Studies have showed social networks and household size are shrinking, one-third of adults in the United States older than age 45 years have reported feeling lonely, and the prevalence of loneliness can be expected to increase as the population ages.5 Additionally, given that loneliness has been linked to several risk factors for type 2 diabetes, such as aging, obesity, and metabolic syndrome, identifying the prospective association between loneliness and type 2 diabetes is of great importance.

Hackett et al used data from the English Longitudinal Study of Ageing (ELSA) to conduct a prospective, longitudinal, observational study focused on assessing loneliness as an independent risk factor for type 2 diabetes. Started in 2002, ELSA collects data on people older than age 50 years living in England with the goal of understanding all aspects of aging. Every two years, data are collected on the same set of ELSA participants. As of 2020, more than 18,000 individuals have participated in ELSA.6 Participants were selected from the ELSA database, and the authors followed the same study design as ELSA by collecting questionnaire data from participants in “waves” occurring every two years.

Overall, there were eight waves of data collection spanning 15 years. Wave 1 data collection began in 2002-2003 to identify potential participants, with wave 2 (2004-2005) used to collect data on participant loneliness and diabetes diagnosis status at baseline, as well as covariate data on age, sex, ethnicity, smoking status, alcohol consumption, frequency of physical activity, body mass index (BMI), hypertension diagnosis, and household non-pension wealth (indicator of socioeconomic status.) Participants in wave 2 also completed a nurse visit during which covariate data were confirmed, blood pressure readings were taken, and HbA1c was measured. Participants who indicated a diagnosis of type 2 diabetes or who recorded an HbA1c in the diabetic range of ≥ 6.5% were excluded from the study. During the follow-up period of wave 3 (2006-2007) through wave 8 (2016-2017), self-reported information on participant incidence of type 2 diabetes was collected. Participants who provided a complete data set on loneliness in wave 2 and type 2 diabetes status in the follow-up period were included in the final analysis.

Loneliness, the primary predictor variable, was assessed using the University of California Los Angeles Loneliness Scale, a 20-item scale presenting various feelings of loneliness and isolation that participants could rate according to frequency.6 Participants were given three numerical options for rating each item: 1 (hardly ever/never), 2 (some of the time), and 3 (often). Participant ratings were averaged, with higher values associated with greater loneliness. Cronbach’s alpha of 0.82 was reported in the study sample. Secondary predictor variables included social isolation, living alone, and depressive symptoms. Social isolation was measured based on frequency of contact with children, family, and friends. Participants were given a social isolation score from 0-4, with higher scores indicating greater isolation. Living alone was based on a self-reported yes/no question. Depressive symptoms were measured using an eight-item Centre for Epidemiological Studies Depression Scale, where participants could score 0-7, with scores ≥ 6 considered signs of severe depression.

Cox proportional hazards regression was used to test the association between loneliness and type 2 diabetes after controlling for age, sex, wealth, ethnicity, smoking, physical activity, alcohol consumption, BMI, hypertension, cardiovascular disease, and HbA1c. Loneliness was inserted as a continuous variable, where the hazard ratio (HR) and 95% confidence intervals (CI) represented a 1 U increase. The secondary analysis consisted of adding covariates and secondary predictor variables to the statistical model to test the independent effect of loneliness on diabetes incidence. Ultimately, five models were created, with covariate data added in Model 1, depression added in Model 2, living alone added in Model 3, and social isolation score indexes added in Model 4. Model 5 was the final model and included loneliness, all covariates, depression, living alone, and social isolation as type 2 diabetes predictors.

Results showed 8,780 participants identified as eligible at the conclusion of wave 2, with 4,112 participants providing a complete data set that could be used in the final analysis. In the follow-up period, 262 participants reported developing type 2 diabetes. Cox regression modeling showed loneliness to be a significant predictor of type 2 diabetes incidence (HR, 1.46; 95% CI, 1.15-1.84; P = 0.027) independent of covariates, including age, sex, ethnicity, wealth, smoking, physical activity, alcohol consumption, BMI, HbA1c, hypertension, and cardiovascular disease. Additionally, Model 2 through Model 4 showed loneliness as an independent predictor; depressive symptoms, living alone, and social isolation were not significant predictors of type 2 diabetes incidence. Model 5 (final results) also continued to show loneliness to be an independent predictor of type 2 diabetes (HR, 1.41; 95% CI, 1.04-1.90; P = 0.027). An additional analysis showed loneliness was associated with a greater likelihood of smoking and physical inactivity, and a reduced likelihood of regular alcohol consumption.


Hackett et al presented a first-of-its-kind study that prospectively examined loneliness as a risk factor for type 2 diabetes. To date, other researchers have conducted cross-sectional analyses demonstrating a relationship between loneliness and type 2 diabetes. However, these studies were limited in their ability to determine whether loneliness stimulates the development of type 2 diabetes or if type 2 diabetes onset and management lead to a strain on the quality of social relationships, ultimately resulting in loneliness.7,8 Based on their work, Hackett et al concluded loneliness can be considered a predictor of type 2 diabetes independent of other social and mental health variables, such as depressive symptoms, living alone, and social isolation.

Potential limitations in the study design include selection bias and the inability to generalize findings to non-white populations (the authors noted the ELSA database, from which participants were selected, contains few ethnic minority participants). An analysis of the baseline covariate data of participants lost to follow-up showed poorer health, lower financial status, and higher rates of loneliness when compared to participants who completed the study. Thus, selection bias caused by non-random exclusion may have occurred.

Patients identified as experiencing loneliness may benefit from interventions that address improving the quality of social relationships and treating maladaptive thought processes. Literature reviews identify a key characteristic of loneliness as an individual’s hypervigilance to perceived social threats and negative social information. Interventions focused on providing patients with therapy to recognize the internal cognitive biases caused by loneliness, along with tactics to improve perceptions of social interactions, may be more beneficial than interventions aimed at improving social skills or simply increasing opportunities for social interaction.1,2

Additionally, a recent study showed online cognitive behavioral therapy for individuals experiencing loneliness showed a decrease in loneliness and anxiety and an increase in quality of life, with benefits sustained at a two-year follow-up.9 While further research is needed to determine the physiological mechanisms by which loneliness may lead to the development of type 2 diabetes, physicians now can consider loneliness as an independent risk factor for type 2 diabetes, adding to the knowledge of the effects of loneliness on the development of chronic inflammatory diseases. Hackett et al provide strong evidence for clinicians taking time to ask patients about their sense of social support, loneliness, and feelings of isolation, as well as considering referrals to psychologists for appropriate therapy to help alleviate loneliness. 


  1. Hawkley LC, Cacioopo JT. Loneliness matters: A theoretical and empirical review of consequences and mechanisms. Ann Behav Med 2010; 40:218-227.
  2. Henriksen J, Larsen ER, Mattisson C, Andersson NW. Loneliness, health, and mortality. Epidemiol Psychiatr Sci 2019;28:234-239.
  3. International Diabetes Federation. IDF Diabetes Atlas. 9th ed. International Diabetes Federation;2019.
  4. Khan MAB, Hashim MJ, King JK, et al. Epidemiology of type 2 diabetes — global burden of disease and forecasted trends. J Epidemiol Glob Health 2020;10:107-111.
  5. Holt-Lunstad J. The potential public health relevance of social isolation and loneliness: Prevalence, epidemiology, and risk factors. Public Policy & Aging Report 2017;7:127-130.
  6. Banks J, Batty GD, Coughlin K, et al. English Longitudinal Study of Ageing: Waves 0-9, 1998-2019 [data collection]. UK Data Service doi: 10.5255/UKDA-SN-5050-20
  7. Russell D, Peplau LA, Ferguson ML. Developing a measure of loneliness. J Pers Assess 1978;42:290-294.
  8. Christiansen J, Larsen FB, Lasgaard M. Do stress, health behavior, and sleep mediate the association between loneliness and adverse health conditions among older people? Soc Sci Med 2016;152:80-86.
  9. Richard A, Rohrmann S, Vandeleur CL, et al. Loneliness is adversely associated with physical and mental health and lifestyle factors: Results from a Swiss national survey. PLoS One 2017;12:e0181442.