For some with impending Alzheimer’s, a crowded room begins to feel just as lonely as an empty one. Could this perceived social isolation be an early warning sign of plaque build-up in the brain? According to a paper in the December JAMA Psychiatry, perhaps so. Scientists led by Reisa Sperling, Brigham and Women’s Hospital, Boston, report that cognitively normal people with amyloid in their brains are 7.5 times more likely to report feeling lonely. The finding points to socioemotional changes, in addition to cognitive and behavioral ones, that could be associated with Alzheimer’s. If the results are replicated in larger cohorts, screening for loneliness could help identify candidates for amyloid scans, scientists said.
“From a statistical perspective, the findings are quite robust,” said José Luis Molinuevo of Barcelonaβeta Brain Research Center in Spain. “It may be that this is part of the very early symptomatology of the disease that we are not capturing yet.”
Loneliness is a subjective feeling of social and emotional detachment that remains distinct from objective social isolation. In other words, some people may feel alone even when surrounded by people. For instance, geriatric clinician Carla Perissinotto, University of California, San Francisco, recalls a woman who kept weekly social engagements, employed a full-time caregiver, and saw her son every day, but complained that she felt immensely lonely. Conversely, one man who lived alone and saw almost no family during the week said he felt fine. "I’ve seen it go both ways,” wrote Perissinotto to Alzforum. “We can’t make assumptions about who is or isn’t lonely.”
Previous studies have associated feeling lonely with memory problems, functional decline, and progression to AD (Donovan et al., 2016; Perissinotto et al., 2012; Wilson et al., 2007). Given the links to impairment and disease, Sperling and colleagues wondered whether loneliness could be related to one of the principal pathologies of Alzheimer’s disease—the accumulation of amyloid plaques.
First author Nancy Donovan and colleagues assessed 79 cognitively and psychiatrically healthy people aged 60 to 90 from the Harvard Aging Brain Study (HABS). In addition to undergoing a positron emission tomography scan with Pittsburgh Compound B (PiB-PET) to measure brain amyloid, each person took a three-question survey to assess how lonely he or she felt. It asked how often people felt left out, isolated from others, or lacking in companionship. They scored 1, 2, 3, or 4 points for answering never, rarely, sometimes, or often, respectively. A person was considered lonely if they answered any of the three questions “sometimes” or “often.” The researchers also derived a separate social network score, which objectively measured social activity. It accounted for whether a person lived with another, had at least three monthly visitors, and participated in community groups or religious activities. Separate scales assessed anxiety and depression. The researchers also noted whether participants carried the ApoE4 allele.
In this cohort, the average loneliness score was 5.3. The more amyloid a person had in his or her brain, the more likely they were to report feeling lonely. This was even after adjusting for age, sex, ApoE4 status, depression, anxiety, and social network score. When PiB-PET was analyzed as a continuous variable, every 0.1 DVR increase was accompanied by a 75 percent higher chance of feeling lonely. When the researchers divided the group into amyloid-positive and –negative subjects, almost half of the 25 amyloid-positive folks, but only a third of the 40 amyloid-negative participants, reported being socially withdrawn. That gave those with brain amyloid a 7.5 times higher chance of feeling lonely compared to the negative group. The association was strongest for the 22 people who carried the ApoE4 allele. Interestingly, PiB-PET did not correlate with the objective social network score, suggesting a person with amyloid in the brain could be socially active and yet still feel disconnected.
“It’s a strong effect in this sample,” wrote Michael Weiner, University of California, San Francisco, to Alzforum. However, the finding needs to be replicated before it becomes established, said Weiner, who was not involved in the study. He added that researchers currently do not assess feelings of social isolation in the Alzheimer’s Disease Neuroimaging Initiative (ADNI), which he leads.
Scientists from other large cohort studies that include amyloid scans, such as the Australian Imaging, Biomarker & Lifestyle Flagship Study of Ageing (AIBL), also confirmed that they do not collect data on loneliness. There are some international cohorts and large epidemiological studies that have incorporated this loneliness assessment to see how the score changes with age, such as the Ann Arbor, Michigan-based Health and Retirement Study (HRS), said Donovan. These studies, however, do not scan for amyloid.
Donovan proposed that while the field understands very little about the neural basis of loneliness, the link with amyloid could result from functional alterations in the default mode network, which takes an early hit in Alzheimer’s disease. The group is in the process of measuring loneliness in the remaining HABS participants, and will follow scores longitudinally to see which changes first. They also plan to examine whether loneliness correlates with changes in network connectivity or tau buildup using functional magnetic resonance imaging and tau PET. Scientists contacted for this article were unaware of other studies that assess loneliness along with functional brain measures.
If this finding holds up in prospective studies and in other cohorts, loneliness could be added to the algorithms that predict who is most likely to have a positive amyloid PET scan, Molinuevo said. Very few predictive factors are now available—ApoE status, cognitive decline, age—and researchers are actively seeking new ones, he said. Further refining those algorithms would reduce the cost of building trial-ready cohorts, such as the Global Alzheimer’s Platform (GAP) or the European Prevention of Alzheimer’s Dementia (EPAD), which Molinuevo co-leads (Aug 2016 conference news).
Other researchers wrote that loneliness could be the first of many socioemotional factors that characterize the early stages of Alzheimer’s. “Why stop with loneliness?” wrote Paul Rosenberg, Johns Hopkins School of Medicine, Baltimore, in an accompanying commentary. There may be other emotions yet to be discovered—such as fear, existential angst, dread, or even positive feelings—that could reflect amyloid burden or the presence of other AD biomarkers. Assessing various emotions in older adults while measuring relevant biomarkers could yield clues for further study, he wrote.—Gwyneth Dickey Zakaib
- Donovan NJ, Wu Q, Rentz DM, Sperling RA, Marshall GA, Glymour MM. Loneliness, depression and cognitive function in older U.S. adults. Int J Geriatr Psychiatry. 2016 May 9; PubMed.
- Perissinotto CM, Stijacic Cenzer I, Covinsky KE. Loneliness in older persons: a predictor of functional decline and death. Arch Intern Med. 2012 Jul 23;172(14):1078-83. PubMed.
- Wilson RS, Krueger KR, Arnold SE, Schneider JA, Kelly JF, Barnes LL, Tang Y, Bennett DA. Loneliness and risk of Alzheimer disease. Arch Gen Psychiatry. 2007 Feb;64(2):234-40. PubMed.
- Sivam A, Wroblewski KE, Alkorta-Aranburu G, Barnes LL, Wilson RS, Bennett DA, Pinto JM. Olfactory Dysfunction in Older Adults is Associated with Feelings of Depression and Loneliness. Chem Senses. 2016 May;41(4):293-9. Epub 2016 Jan 24 PubMed.
- El Haj M, Jardri R, Larøi F, Antoine P. Hallucinations, loneliness, and social isolation in Alzheimer's disease. Cogn Neuropsychiatry. 2016 Jan;21(1):1-13. Epub 2016 Jan 6 PubMed.
- Donovan NJ, Okereke OI, Vannini P, Amariglio RE, Rentz DM, Marshall GA, Johnson KA, Sperling RA. Association of Higher Cortical Amyloid Burden With Loneliness in Cognitively Normal Older Adults. JAMA Psychiatry. 2016 Dec 1;73(12):1230-1237. PubMed.
- Rosenberg PB. Loneliness as a Marker of Brain Amyloid Burden and Preclinical Alzheimer Disease. JAMA Psychiatry. 2016 Dec 1;73(12):1237-1238. PubMed.