TABLE 1.

Relationship between social isolation and/or loneliness and cognitive function in older adults.

References	Objective, country and cohort	Characteristics of the sample	Loneliness/social isolation measurement	Cognitive measurement and function	Adjustments for covariates	Results
Evans et al. (2018)	To examine the relationship between social isolation and cognition in later life, and to consider the role of cognitive reserve in this relationship. Baseline and 2-year follow-up data Wales (UK). Cognitive Function and Ageing Study-Wales (CFAS-Wales)	N = 2224 Age ≥ 65 Mean (SD): 73.47 (±6.28) Sex: N (%) F: 1127 (50.67%) M: 1097 (49.33%)	Social isolation: Lubben Social Network Scale-6 (LSNS-6) Cognitive reserve: number of years in full time education, occupational complexity and cognitive activity	Cambridge Cognitive Examination (CAMCOG) Orientation, comprehension, expression, memory (remote, recent, and learning), attention and calculation, praxis, abstract thinking and perception	Age, gender, education (years), and physically limiting health conditions, such as sensory problems (hearing and eyesight), and ability to complete daily tasks alone	Being socially isolated in later life is associated with poor cognitive function, and cognitive reserve moderates this association at 2-year follow-up. Maintaining a socially active lifestyle in later life may enhance cognitive reserve and benefit cognitive function
Griffin et al. (2020)	To jointly examine isolation, loneliness, and cynical hostility as risk factors for cognitive decline in older adults. Follow-up every 2 years during 6 years USA Health and Retirement Study (HRS)	N = 6654 Age ≥ 65 Sex: N (%) F: 6026 (58.08%) M: 4350 (41.92%)	Social isolation: (1) frequency of contact with social network and (2) type of relationship Loneliness: Hughes Loneliness Scale Cynical hostility: Cook–Medley Hostility Inventory	Modified version of the Telephone Interview for Cognitive Status (TICS) Recall (i.e., immediate and delayed word recall) and mental status (i.e., the serial 7 s, backward counting from 20, and object, date, and president/vice president naming)	Age, education, sex, socioeconomic status (SES), and race), health status, and functional limitations	Loneliness and cynical hostility correlated with lower cognitive function, but none predicted change in cognitive function. Social isolation was associated with lower cognitive function and steeper decline in cognitive function Objective social isolation is a predictor of lower cognitive function and faster cognitive decline
Hajek et al. (2020)	To determine the link between perceived social isolation and cognitive functioning longitudinally Germany German Ageing Survey (DEAS)	N = 6420 Age ≥ 40 Mean (SD): 65.0 (±10.7) Sex: N (%) F: 3228 (50.3%) M: 3192 (49.7%)	Perceived social isolation: assessed using a scale by Bude and Lantermann	Digit symbol test, adapted from the digit symbol substitution Test Perceptual motor speed and processing speed of visual perception and information	Age, family status, household net equivalent income, labor force participation, self-rated health, physical functioning, physical illnesses, loneliness (De Jong Gierveld scale) and depressive symptoms (15-item version of the CES-D)	Increases in social isolation were associated with decreases in cognitive functioning longitudinally
Lara et al. (2019)	To examine the association of loneliness and social isolation on cognition over a 3-year follow-up period in middle- and older-aged adults Spain Collaborative Research on Ageing in Europe (COURAGE in Europe) Study	N = 1691 Age ≥ 50 Mean (SD): 64.5 (± 9.8)	Social isolation: measured considering: being married or cohabiting with a partner (or not); having less than monthly contact with children, other immediate family or friends; participating in any organizations, religious groups, or committees Loneliness: 3-item UCLA Loneliness Scale	Forward and backward digit span (Wechsler Adult Intelligence Scale), word list memory tasks (CERAD), animal naming task and a composite cognitive score Immediate recall, delayed recall, verbal fluency and episodic memory	Age, education, sex, level of physical activity, alcohol consumption, and disability. Additionally three chronic conditions were measured: depression in the previous 12 months, diagnosis of stroke and diabetes	Both loneliness and social isolation are associated with decreased cognitive function over a 3-year follow-up period The effect of loneliness and social isolation on cognition remained unchanged after the exclusion of individuals with depression, supporting the notion that loneliness and social isolation are not merely makers of depressive symptoms
Luchetti et al. (2020)	To test whether loneliness is associated with the risk of cognitive impairment up to 11 years later in a European sample of middle-aged and older adults. Austria, Belgium, Denmark, France, Germany, Greece, Italy, Spain, Sweden, Switzerland, Netherlands, and Israel. Survey of Health, Ageing and Retirement in Europe (SHARE)	N = 14114 Age ≥ 50 Mean (SD): 63.61 (±9.33) Sex: N (%) F: 7720 (54.7%) M: 6394 (45.3%)	Perceived loneliness: single-item measure of loneliness (CES-D) at baseline and three-item version of the UCLA loneliness scale at follow-ups from 2011	Immediate and delayed recall of 10 common words and naming as many animals as possible in 60 s. Memory recall task and fluency	Age, sex, educational level, social isolation, clinical and behavioral covariates, health-related activity limitations, and depression symptoms (using the EURO-D scale)	Feeling lonely increased the risk of developing cognitive impairment up to 11 years later The association remained significant in accounting for age, sex, education, clinical and behavioral risk factors, health-related activity limitations, social isolation and depressive symptoms
McHugh Power et al. (2019)	To investigate potential cross-lagged associations between sustained attention and loneliness, measured at baseline and again after 4 years Ireland Irish Longitudinal Study of Ageing (TILDA)	N = 6239 Age ≥ 50 Mean (SD): 63.05 (±9.22) Sex:% F: 54.6% M: 45.4%	Loneliness: specified as a latent factor with three indicators: the items constituting the 3-item version of the UCLA Loneliness Scale: “I feel left out,” “I feel isolated,” and “I lack companionship”	Sustained Attention to Response Task (SART) sustained attention	Age, sex, education (“no qualification,” “intermediate qualification,” and “degree qualification or higher”), depressive symptomatology (CES-D, with the item inquiring about loneliness removed)	While sustained attention at baseline predicted loneliness 4 years later, the converse, that loneliness would predict sustained attention, was not supported
McHugh Power et al. (2020)	To evaluate the relationship between loneliness and cognitive functioning, and whether depressive and anxiety symptoms have intermediate roles therein Ireland Irish Longitudinal Study of Ageing (TILDA). Data was collected at three time-points 2 years apart	N = 7433 Age ≥ 50 Mean (SD): 63.99 (±9.83) Sex: N (%) F: 3966 (53.36%) M: 3467 (46.64%)	Loneliness: 5-item version of the UCLA Loneliness Scale Depressive symptoms: 20-item CES-D scale Anxiety symptoms: Hospital Anxiety and Depression Scale Anxiety (HADS-A)	Measured as a latent factor, with four indicators: measures of immediate and delayed word recall (word list), verbal fluency (animal naming task) and a global measure Immediate and delayed word recall, verbal fluency, attention, orientation, memory, registration, calculation, language and praxis	Age, education level, sex and physical health (number of cardiovascular conditions, including angina, heart attack, heart failure, stroke, and abnormal heart rhythm etc.; and number of chronic conditions, including the above cardiovascular conditions, hypertension, diabetes, asthma, bronchitis, cancer, arthritis etc.)	Loneliness at time-point 1 predicted cognitive functioning at time-point 3, and anxiety symptoms at time-point 2. Depressive but not anxiety symptoms mediated the relationship between loneliness and cognitive functioning. However, the indirect effect of loneliness on cognitive functioning via depressive symptoms was small relative to the direct effect
Read et al. (2020)	To investigate associations between level and changes in social isolation and in memory in older men and women. Six measurement occasions every 2 years were conducted England (UK). English Longitudinal Study of Aging (ELSA)	N = 11233 Age ≥ 50 Mean (SD): 65.1 (±10.1) Sex: N (%) F: 6,123 (54.5%) M: 5,110 (45.5%)	Social isolation: index derived from five binary items	Word list recall in which the participant was asked to learn 10 common unrelated words Memory	Age, indicators of socioeconomic status (education, wealth, home ownership), and health-related behaviors (smoking, physical activity): all treated as time-invariant using values Limiting long-term illness, depressive symptoms, and whether working or doing voluntary work: all treated as time-varying covariates	Social isolation increased and memory decreased over time. The association between social isolation and memory decline is driven by the effect of social isolation on memory, rather than the reverse.
Wang et al. (2020)	To test the potential impact of loneliness amongst older old people on their cognitive function over a 20-year period UK Cambridge City over-75 s Cohort (CC75C) Study	N = 713 Age ≥ 75 Mean (SD): 86 (±4) Sex:% F: 71% M: 29%	Loneliness: single-item scale “Do you feel lonely?”; with response options “not at all lonely,” “slightly lonely,” “lonely” and “very lonely”	Mini-Mental State Examination (MMSE). Orientation, memory recall, working memory, attention, language, visual-spatial skills.	Age, sex, and education	Feeling slightly lonely and lonely were both associated with decline in cognitive function but neither of these associations were significant Loneliness was not a risk factor for cognitive function decline over a 20-year period. Loneliness did not exert long-term harmful effects on cognitive function in the oldest old
Yin et al. (2019)	To examine whether there is a bidirectional relationship between loneliness and cognitive function over a 10-year follow-up England (UK). English Longitudinal Study of Aging (ELSA)	N = 5885 Age ≥ 50 Mean (SD): 65.3 (±9.0) Sex: N (%) F: 3401 (55.4%) M: 2734 (44.6%)	Loneliness: abridged version of the revised UCLA Loneliness Scale	Word recall and verbal fluency tests Memory and verbal fluency	Age and sex, educational level, wealth, illness or disability that impaired their everyday life over an extended period Depression (using a combined algorithm of physician diagnosis and a positive score on the seven items of the CES-D scale, after excluding the loneliness item from the standard eight-item CES-D)	Higher loneliness is associated with poorer cognitive function at baseline and contributes to a worsening in memory and verbal fluency over a decade. These factors seem, however, to be partially intertwined, since baseline memory and its rate of decline also contribute to an increase in loneliness over time
Yu et al. (2020)	To examine the relationships of social isolation and loneliness on cognitive function among Chinese older adults over a 4-year follow-up period China China Health and Retirement Longitudinal Study (CHARLS)	N = 7761 Age ≥ 50 Mean (SD): 60.97 (±7.31) Sex:% F: 49.2% M: 50.8%	Social isolation: three items were combined to create an index of social isolation: married/not married; weekly contact with children; and participating in any social activities over the last month Loneliness: one single item included in the CES-D: “In the last week, how often did you feel lonely?”	Telephone Interview for Cognitive Status (TICS) and immediate word recall followed by delayed recall Mental status (orientation, visuospatial ability and numeric ability) and episodic memory	Demographic variables and behavioral, psychological, and clinical risk factors, age, gender, education, and area of residence (urban/rural). Depressive symptoms (measured with CES-D-9 (a modified CES-D-10 excluding the loneliness item). Chronic diseases (including hypertension, diabetes, and heart diseases), health behaviors/habits (including drinking and smoking), and disabilities (functional limitations in activities of daily living (ADL)	Social isolation was significantly associated with decreases in all cognitive function measures at follow-up even after controlling for loneliness and all confounding variables. Loneliness was significantly associated with cognitive decline at follow-up in the partially adjusted models. However, these associations became insignificant after additional confounding variables (chronic diseases, health behaviors, disabilities and depressive symptoms) were taken into account
Zhou et al. (2019)	To investigate the association between loneliness and cognitive impairment among older men and women in China over a 3-year follow-up period China Chinese Longitudinal Healthy Longevity Survey (CLHLS)	N = 6898 Age ≥ 65	Loneliness: one single question: “Do you feel lonely?” (item extracted from the CES-D)	Self or proxy-report at follow-up using a culturally adapted, Chinese version of the MMSE Orientation, reaction, calculation ability, recall, and language ability	Social-demographic variables, lifestyles, health status and social isolation: Age, education level, employment status, and body mass index (BMI), current smoking and current drinking, physical exercise. Health status, such as CVD, diabetes and activities of daily living (ADL) disability, both instrumental and basic ADL. Social isolation: assessed using the following three separate items: Living alone (yes or no), being married (yes or no) and having social support	Although elderly women more frequently reported feelings of loneliness, the impact of loneliness on cognitive impairment was significant among elderly men but not elderly women