As I sat before my computer screen reviewing Ms. W’s electronic medical record, one word gripped my attention, paragraph after paragraph: survivor. Now in her mid-60s, she had survived HIV/AIDS during an era when treatment was non-existent. She was soon after diagnosed with breast cancer, but fought through intensive chemotherapy, and survived. Then, as if one malignancy wasn’t enough, she developed ovarian cancer, and endured surgery and radiation to, yet again, survive. This time, she had come with a complicated urinary tract infection; her ureters had been narrowed by prior courses of radiation therapy, and she would need pelvic surgery to stent open those tiny canals and prevent future infections.
Ms. W’s degree of health complications is not unusual in our public hospital, where many patients show up with more advanced disease compared to patients with adequate insurance coverage and access to preventive healthcare. But it is not so common to see such survival against these great odds. There was something unique about Ms. W’s resilience, and I was interested to understand it more clearly.
Approaching Ms. W’s room, I found her lying in bed, deeply engaged in a crossword puzzle, looking weathered but resilient. Her pale hands, riddled with the small scabs and deep bruises of constant blood draws, gripped the newspaper firmly. Her focus was uninterrupted, despite the moans and mumblings of three other patients reverberating throughout the room, each hidden behind a curtain of faded fabric. Ringed shadows underlined her deep brown eyes, the fatigue of having been through this hospital routine a million times over—the perpetual phlebotomist pricks, the early morning visits from a flurry of doctors, the shrieks of equally sick roommates that prevent a decent night’s sleep. But fixated on her puzzle, she also seemed unbothered; while objectively ill in the moment, she seemed also aware that this was, indeed, just a moment.
I had budgeted more time than usual this morning to chat with Ms. W, as it was our first interaction as doctor and patient. We reviewed the details of her present health dilemma, and after a physical exam (concerning for an enormous lymph node in her groin), I pulled a chair to her bedside, and transitioned our conversation toward another dimension.
“Ms. W, tell me, who is important to you in your life?” It is a question I always like to ask in some form upon meeting a new patient, because social support can often dictate some degree of a patient’s prognosis. Medical doctors are not trained to ask this question, but most doctors who pay attention to their patients as human beings will agree that social support can be just as influential to a patient’s health as their family history, or genetic disposition to disease.
She hesitated, at first, an intimate question from a relative stranger. “Well… I love my dog, Dexter,” she joked, her voice raspy, detailing a funny anecdote about her fluffy pal and his loyal tendencies. As she carried on, I noticed hospital staff bringing in breakfast trays, and was tempted to look at my watch, or hurry along the conversation. But I had set out the time. So I sat, and listened. In that quiet space, Ms. W blossomed. “And then there’s my partner, Rita,” Ms. W finally offered up, pushing her breakfast tray aside as though to not lose focus on her subsequent statements. “I wouldn’t have gotten through this all without her.”
Over the next few minutes I had the privilege of hearing her touching love story. Ms. W and Rita had met as nurses in this very hospital, decades ago, and while they remained committed partners through difficulties like her HIV and cancer diagnoses, they were only able to officially wed in 2011, once marriage equality was legalized in New York. Her tone became somber as she recalled other friends who had also received an HIV diagnosis, but had committed suicide during a time when medication was inaccessible. While she took credit for her own personal resilience in surviving these moments, she also gave thanks to Rita for fostering her through her grimmest obstacles. “I really wouldn’t have gotten through it all without her,” she emphasized, again.
We shared another few moments together reviewing her goals of care: she was ready to fight anything that came her way this hospitalization, she had an unshakeable sense of self, and she considered herself a survivor with many years of life ahead. With a warm smile she pulled her breakfast tray close again, giving me the OK to move on. As she dug into her hospital hash browns, I waved and retreated into the hallway.
Walking away from Ms. W’s room, I was impressed that her medical chart—full of detailed blood pressure readings, medication dosages, and past surgical procedures—never once mentioned Rita, a person quite significant to her survival. The notion of a life partner as a protective factor for our health is one not often investigated in the doctor’s office or in hospital rooms. In fact, doctors often feel so limited by the time they have with patients that we often forget entirely to ask about a patient’s social supports (or lack thereof), weighing it as a lower-yield piece of clinical information compared someone’s family history of disease, or medication compliance. Many doctors also feel that, even if they could identify someone as high-risk for social isolation, there would be few ways to aid that aspect of a patient’s life. We offer lifesaving medications, we conduct transformative surgeries, but it feels utterly beyond a medical profession to approach the seeming impossibility of “prescribing friends”. We know intuitively, in our guts, that people who are alone in the world tend to fare worse than those with a “Rita” around, but asking about social support is simply not a common reflex in modern medicine.
My meeting with Ms. W left the thought buzzing in my mind: but maybe it should be?
For as long as human beings have been philosophers, we have self-identified as, “social beings,” ultimately requiring human connection to survive a lifetime, and advance as a species. Aristotle, well known for labeling man a ‘social animal,’ even went so far as to suggest that, “anyone who either cannot lead the common life or is so self-sufficient as not to need [society] … is either a beast or a god.” The evolutionary expert himself, Charles Darwin, also hypothesized that caring for others (what he called ‘sympathy’) was a positive survival trait, and thus, “those communities, which included the greatest number of the most sympathetic members, would flourish best, and rear the greatest number of offspring.” But for all the goodness that comes with positive human connections, aren’t there plenty of individuals today who manage to survive (if not thrive) in relative isolation, as neither beast nor god, neither giving nor receiving an abundant amount of sympathy?
Most of our fellow modern citizens easily produce work and engage in financial transactions in entirely faceless, digital environments. It’s also common to conduct our social interactions through texting or video applications, and most of us hear our daily ‘news’ through a vast expanse of social media networks. Could it be that technology has become so advanced that we have surpassed an evolutionary ‘need’ for in-person social interaction?
Yet, as human beings, we feel the downsides of digital conveniences, too—the exhaustion of staring at a computer screen for hours on end, the isolating nature of scrolling through Facebook “friend” feeds just to feel envy towards what’s posted by others, and the misunderstanding fueled by unverified, inflammatory ‘fake news’ sources. Even the former U.S. Surgeon General himself, Dr. Vivek Murthy, recently warned Americans of an emerging “loneliness epidemic,” citing studies that up to 60 million Americans are suffering from isolation at this moment in time. It turns out being alone in the world just doesn’t feel quite right in our gut, and there may be reasonable biologic reasons behind our instinct.
For ages, discourse on the ‘dangers of social isolation’ was dominated by an evolutionary hypothesis that socializing was solely a survival mechanism: we need other people to find and share scarce food resources, to alert us of opportunities and dangers, and to collectively defend against predators. But by the mid-20th century, an important distinction was brought to light by researchers studying the effects of social isolation in animals and humans alike. Dr. John Cassel, in his work, The Contribution of the Social Environment To Host Resistance, nurtured the development of the “stress buffering hypothesis,” or as he described, the role of social relationships to provide a “protective factor buffering the individual from the physiologic or psychologic consequences of exposure to [a] stressor.” Cassel’s conclusion was quite a profound change from the prior understanding of why social relationships matter: he implied that our friends and family were not only mechanisms by which we attain resources, or protect the herd, but also therapy in and of themselves, influencing the way our body works from the inside out.
In one study he reviewed, researchers delivered a series of unanticipated electric shocks to rats, a stimulus known to induce stomach ulcers; some of the rats were kept in isolation during the shocks, and others were grouped with their littermates. The rats who endured the shocks in isolation ended up with significantly more ulcers than those who went through the challenge together with a sibling, implying that simply having a friend nearby during a physical challenge can boost physical resilience. Cassel also reviewed a bevy of human sociologic studies, including one of his own that showed, among black men in North Carolina in the 1960s, the likelihood of having a stroke increased proportionately with the level of “social disorganization” in their neighborhood. Neighborhoods with high levels of ‘disorganization’ were defined as those with more single-parent households, higher incarceration rates, and higher rates of separation and divorce, or in other words, more instances of social isolation. Similar to the animal studies, these results suggested that social connection has the power to biologically protect our bodies from the physical side effects of chronic social and emotional stressors.
Fast-forward to the present time, and decades of subsequent research have elicited precisely how our cells respond to chronic stressors like social isolation. Dr. Steven Cole, Director of the UCLA Social Genomics Core Laboratory, is a pioneer in the mapping of pathways by which social environments change our very genes. Cole has discovered that when our brains perceive a social threat like loneliness, especially if the threat is present day after day, our “fight or flight” sympathetic nervous system (SNS) becomes chronically overactive. The human SNS was designed to respond primarily to emergencies, by releasing hormones that cause our heart to race and blood pressure to elevate. But when we crank those hormones up all the time—as happens with daily, chronic stressors like isolation—this can cause our bodies to churn out more pro-inflammatory proteins, and suppress long-term immunity, leaving us more likely to develop illnesses like heart disease, Alzheimer’s, diabetes, and even metastatic cancer, while stunting our ability to fight off viruses.
An important detail of this pathology is the concept of chronic loneliness, the expertise of Dr. John Cacioppo, Director of the Center for Cognitive and Social Neuroscience at the University of Chicago. In Cacioppo’s book Loneliness, he explains that the experience of loneliness is not necessarily a bad one; we may all feel lonely during times of isolation, appropriately, in the same way we feel hungry when we’ve gone without food for too long. Hunger only kills, after all, when it’s a never-ending, day in, day out experience. Similarly, loneliness becomes destructive to our bodies only when we can’t find the social connections we need to satisfy our biologic need.
Cole and Cacioppo have both hypothesized about the basic science behind loneliness leading to early death, and their profound experiments have even given proof of how these pathways work at the most microscopic level. In one collaborative experiment, they examined over 22,000 gene transcripts from the white blood cells, the cells that make up our immune system, of healthy adults who self-identified as either perpetually isolated or well-connected. When they compared the genetic material of lonely respondents to non-lonely folks, they found that an entire group of genes controlling inflammation were overactive in lonely people, and two groups of genes responsible for immune system function were suppressed in lonely people. Their finding provides descriptive scientific reasoning behind decades of socio-medical observations, suggesting that loneliness doesn’t just alter the way we interact with the world around us—like not having someone to bring us to a doctor’s appointment—but also changes the way our body responds to stress from the inside out, making us more susceptible to heart attacks, cancers, and other inflammatory conditions, and decreasing our ability to fight off certain infections.
Cole and Cacioppo’s research offers new insights into understanding my patient, Ms. W, the Survivor. Did Rita’s accompanying presence really play a therapeutic role all those years when she was fighting breast cancer? It is the most common malignancy among American women, but not necessarily the most deadly. While a host of biologic factors contribute to predicting survival—the tumor stage at time of discovery, the expression of different hormone receptors on the tumor—is there any scientific evidence to suggest that a solid social network—whether a partner or a best friend or a support group—contributes to improved survival as well?
Dr. Candyce Kroenke, who has dedicated her career to studying the influence of social networks on breast cancer treatment and prognosis, found that women who were socially isolated prior to breast cancer diagnosis were twice as likely to die of their breast cancer, compared to peers with strong social support. She also found that women who had small social networks and low social support were more likely to die of their breast cancer, but women with small social networks and high social support were more likely—like Ms. W— to be survivors. This important detail teased out that it wasn’t so much quantity that mattered as quality, and a tiny group of truly supportive loved ones was more important than a large group of superficially committed friends.
But is this “friend effect” entirely attributable to the old logical argument, that having devoted caregivers means getting to appointments on time and not missing any medications? Or does social connection positively change the way our body responds to malignancy, from the inside out? Several recent cancer studies in mice suggest that the latter is true, evidencing that isolation ramps up “fight or flight” hormones, which then directly quicken the growth of existing tumors. Other mouse studies have found that the hormone “oxytocin”—often called the “tend-and-befriend” hormone, because it’s found at higher levels in moms during maternal-child attachment and in all people during social bonding—has the power to put a halt on the hormones that speed up tumor growth, and stop breast cancer in its tracks. Though various cancers are uniquely insidious in their pathology, with many intricate reasons for growth or suppression, this growing body of social isolation research offers an additional ray of insight into Ms. W’s longevity—that perhaps having a stalwart supporter by her side protected her from worsening disease.
These studies also suggest, at a bare minimum, that doctors should be paying far more attention to the level of social support in the lives of their patients, particularly patients like Ms. W. Social connection is not only a proven correlate of longevity and health, but also a powerful human experience shown to promote long-term health and immunity a cellular level. But if this information is so evident, why aren’t more doctors jumping to assess their patients for isolation, and make interventions that help them get more socially connected?
Herein lies one of the main obstacles to advancing this knowledge into action: how exactly do we measure social isolation in the doctor’s office? Animal studies have been able to go where human studies, ethically, cannot, by measuring how behaviors and bodies change in response to time in an isolation chamber. We cannot reasonably ‘randomly assign’ one group of people to years of isolation, so instead human studies survey people about their loneliness, then examine these people over time—either their genetic makeup, or their hormone levels, or their health outcomes—to see what occurs differently in lonely versus well-connected people. But surveys rely on a wide variety of self-reported measures, both objective and subjective. Do you live alone, have a partner, and have friends? Even when surrounded by people, do you “feel” alone, or feel welcome in social environments? Not all studies account for the ways depression, anxiety, and personality disorders influence the way we form and keep relationships, or even how personality alone aligns with loneliness: what of the social butterflies who find relationships superficial, or the introverted individuals who interact less frequently, but more deeply?
In my own small panel of patients, I’ve witnessed a wide diversity of medically-manifested social isolation, each with different driving factors to their loneliness. There’s my middle-aged Dominican woman with uncontrolled diabetes who came to me with decaying teeth falling straight from her eroding gums, a side effect of sky-high sugar levels seeping into saliva attracting destructive bacteria. Upon further questioning, I learned that she was also unemployed, raising two developmentally disabled children, living in fear that her husband was cheating on her, and seeing zero friends on a regular basis. Not exactly the questions a doctor asks of the average diabetic, but in this case, it could help explain both how her disease became so disastrous in the first place (no one to help her check sugars or go see the doctor), and how her diabetes likely progressed more rapidly than other patients’ (stress signals cortisol hormones to surge, which leads to overproduction of glucose in the liver).
Then there’s my older, undocumented Chinese gentleman who complained to me last week in a flurry of Fuzhou phrases (translated into Mandarin by his son, then into English by an interpreter) about a few weeks of intense, aching pains in his right pelvis and left knee—an unusual pattern for any common arthritis, even considering his long hours of hard labor in a restaurant kitchen. I ordered a set of x-rays to investigate further and was found his pelvis littered with small bone lesions, an omen of the cancer multiple myeloma (MM), which was confirmed on later lab testing. Though MM is a cancer with largely unknown causes, recent immunologic studies have shown high levels of noradrenaline—one of the main hormones churned out by stressed-out, socially isolated people—can stimulate the growth of tumor cells that advance MM. Was my patient’s social isolation as an undocumented immigrant—illiterate in his new country’s language and culture, required to work endless hours ins solitude doing menial work to survive economically—a source of stress that caused his MM to advance faster than it might have otherwise?
One of the most classic cases of isolation I see in my office is perhaps the most deceptive: the isolation of the full-time caregiver. Mothers and grandparents caring for a houseful of rambunctious children, elderly spouses caring for their dying partner, distraught children caring for terminally-ill parents. Though we see a ‘caring’ family relationship and think it a proxy for ‘good connection,’ it is well documented in medical literature that caregivers can suffer the most debilitating feelings of isolation, and endure the deepest associated stresses.
One eye-opening study monitored elderly people in caregiver roles to their sick spouses, monitoring several years of interleukin-6 (IL-6) levels, a tiny protein that jump-starts clogged blood vessels and subsequent heart attacks and strokes, and advances disease like diabetes and some cancers. Researchers found IL-6 rates to be four times higher amongst people caring for an ailing spouse, versus elderly people not in a caregiver role, and sadly those elevated inflammation levels didn’t even dissipate after their loved one passed away. When we think to our own experiences of loss—whether through death after long illness or unexpected accident, or through romantic break-ups or familial exiling—we’ve all felt that pain of heartbreak, which creates not only a devastating emotion, but also a very real biologic reaction. It lasts not only for the moment of our loss, but for months, even years after, as though seeing faces or hearing songs that remind us of the painful loss set in motion the inflammatory process again and again.
The variations on social isolation that come through my clinic—and that we all experiences in our lives—seems endless. For some people, their skin color or gender or religious dress causes them to endure daily discrimination that puts them on edge, worrying about rejection (or violence, or deportation). Many others have a history of endured abused, or witnessed abuse, and their post-traumatic stress provides ongoing over-activation of their sympathetic nervous system. Still many others spend hours consumed by their telephone screens, scrolling through posts and pictures, finding a digital multitude of ‘connections’ as in-person relationships atrophy offline (it’s a process that seems benign, but correlates with high levels of isolation and loneliness). Do these different conditions all cause an equal magnitude of social isolation, or loneliness, and the subsequent stress that comes with it? Do some carry more weight than others in how they affect our bodies?
Refining these questions will be crucial in the translation of scientific evidence—that isolation and loneliness lead to poor health outcomes—into tools that doctors will use more regularly in the clinic and hospital to identify—and provide support for—patients at the highest risk. Doctors are accustomed to using to tools that have been tested against a ‘gold standard’ for accurate measurement; for example the brief “PHQ-9” set of questions is used to assess for clinical depression, but only came into practice after a multitude of studies tested it against the gold standard of a patient spending an hour in conversation with a psychiatrist, and showed it to be just as accurate in making the diagnosis. In my own patient encounters I include a mix of subjective and objective questions about of loneliness, drawing from the abundance of surveys that loneliness researchers have employed in their studies, but to this day there is no single, well-studied survey, and no clear ‘gold standard’ for identifying loneliness to test it against.
For the time being, many different different question sets exist. Researchers at UCLA have worked for nearly 40 years to refine a robust 20-question “loneliness” survey, asking respondents to what degree they agree with statements like “I feel left out,” or, “I have nobody to talk to,” or “I am unhappy doing so many things alone.” Many other studies focus more on objective levels of connection, like the Interpersonal Support Evaluation List (ISEL) which asks respondents for agreement on statements like “When I need suggestions on how to deal with a personal problem, I know someone I can turn to,” or the Lubben Social Network Scale (L-SNS) which asks respondents to tally up numbers like, “How many relatives/friends you feel close to such that you could call on them for help?” Yet another survey tool for isolation—the Berkman-Syme Social Network Index (SNI)—was used in one of the largest studies to date on isolation, published in the American Journal of Public Health in 2013. Researchers tapped into a national database of nearly 17,000 citizens to study social isolation as a risk factor for death, compared to more traditional risk factors like smoking, high blood pressure, and obesity. Their SNI tool constructed a score for loneliness, ranging from 0 (totally isolated) to 4 (not isolated), based on how respondents answered the following four questions in surveys over prior decades:
- Are you married or living together with someone in a partnership?
- In a typical week, do you speak with or get together with loved ones at least 3 times?
- Do you attend religious services 4 or more times per year?
- Do you belong to a club or organization?
The AJPH study found unquestionably that a low SNI score—extreme isolation—was associated with risk of death similar to that of heavy smoking, being obese, and having high blood pressure. These results have since been echoed in popular media warnings about our “loneliness epidemic,” and the very real health dangers associated with an increasingly disconnected society. These findings also were significant enough to be codified in a recent set of recommendations on “social determinants of health” from the Institute of Medicine, and published in the prestigious New England Journal of Medicine. These guidelines recommend that doctors start asking patients more regularly about their socio-economic status, education level, and—of course—their level of social isolation.
So what is the “right” set of questions for a doctor to ask at the bedside, or in an already time-crunched clinic visit? Does objective isolation matter more? Or can a mere ‘feeling’ of loneliness better capture the risk a patient has for worsening health? The UCLA score, used in studies that uncovered the biologic mechanisms of loneliness as a disease, doesn’t include questions about objective isolation at all. Would the results have been different between people who were lonely and lived alone, versus lonely and lived with others? And the widely cited AJPH study did not have data from respondents about the quality of their relationships, or their feelings about connection. Could their results also be skewed? Couldn’t there be unmarried people who still feel completely socially connected, or those who cringe at chatting on the phone or going to church, but have a few satisfying social connections otherwise? Or conversely, couldn’t someone feel socially isolated even while married, if it is an unhappy pairing? Or socially dissatisfied despite frequent engagements?
The question remains incompletely answered then, how, in a limited set of minutes in clinic or at the bedside, can a healthcare provider know which questions to ask about social isolation and loneliness? Which will yield more insight into the pathologic influence of social isolation?
Perhaps even more importantly, even if we could learn to correctly identify social isolation and loneliness, what steps could we then meaningfully take to reduce the subsequent associated health risks? What would it look like to see meaningful human connection in action?
Eventually, I had the pleasure of finally meeting Rita at Ms. W’s bedside. She would diligently visit Ms. W every evening, after work, and they would deliberate over cross-word puzzles together, flip through mindless reality TV episodes, or just sit and talk, all the while their own conversations drowning out the beeps of heart monitors and chattering of other patients around them. Rita was sturdy and tall, yet unassuming; she never interrupted conversations, instead sitting and listening while Ms. W spoke urgently and at length with her doctors and nurses, as if an invisible timer was running out to share all she had to offer. In contrast, Rita’s presence was quiet, though powerful in the way she could create a separate world for Ms. W, as if they weren’t sitting in a busy public hospital but were relaxing on the couch at home.
Her peaceful presence was crucial when, over these same few days, we decided to biopsy the large, lumpy lymph node I palpated the first day I met Ms. W. When the results came back, it showed evidence of a third cancer, bladder cancer, another life-shattering diagnosis. Sharing this bad news with Ms. W was a difficult conversation, though it was made far less painful having Rita by her side to cushion the blow. Ms. W sighed, and looked first to Rita, who offered her a hand squeeze and a “we-can-do-this” look of confidence in return. She then turned to me, and paused. “My body makes so much noise all the time,” she finally said, softly. “I’ve been hearing it, lately, between all the noise, whispering: ‘Get ready.’ ” Her eyes wandered around, and she closed them with a sigh. “I’m really not totally surprised.” Her hand, still bruised and scabbed, clutched Rita’s tightly. “I’m ready to fight. We’re ready to fight.”
And Ms. W certainly fought. She, and Rita, of course. Throughout my one-month rotation on Ms. W’s floor, she recovered from her infection, stabilized enough to receive her urinary stents, and began speaking with an oncologist to help tackle this new beast. All with Rita by her side.
Practicing medicine well requires making decisions everyday about therapies and treatments that you calculate will give someone the greatest odds of survival. Though philosophically we know how much survival requires interdependence, most of our medical pedagogy remains constructed on the notion that a (wo)man is an island, an entity unto itself, a unique combination of inherited and acquired risk factors that determine a person’s ultimate fate. As a medical student I was rarely ever asked to think of my patients as inherently ‘social animals’; in fact, this type of thinking was often brushed aside as something extra, something ‘nice’ but non-essential, something that belonged to the separate field of medical humanities, not to the realm of basic science. And yet we have robust scientific evidence that our longevity and health are intricately dependent on having reliable, meaningful social relationships over long periods of time.
Why do doctors continue to practice medicine on a false notion that their patients are isolated specimens who can survive in isolation? What will it take for doctors to begin systematically integrating into their care the scientific tenant that has been thoroughly proven that human beings survive longer when they are socially supported? Most doctors, understandably, focus on the risk factors we can change with medications, like blood pressures and sugar levels. Many feel unwelcome in or unfamiliar with the territory further upstream, the external factors like chronic social isolation that cause our blood pressures to rocket and our sugar levels to rise. It is intimidating to intervene in complex, murky notions of “community” and “connection” and “belonging”. But as the data continues to build showing the very real genetic damage done by social isolation—and the therapeutic effects of devoted social connection—it will be crucial for practitioners to integrate an understanding of this science into their daily medical practice.
One simple place to start—until we have better tools to accurately measure isolation and better interventions to ‘solve’ it—is to begin making questions on human connection a standard line in our questioning, and to begin advising patients on the importance of connection with a trusted partner as an intervention that could literally save their life. From this point lies enormous opportunity for physicians and patients alike to re-envision an entire healthcare system that investigates social circumstances, and treats them as skillfully as we do physical symptoms. A system that maximizes points of human contact rather than leaving patients stranded with robotic phone menu options and long waits to see a provider, through home- and community-based health workers, and collaborations with community organizations.
After my month on Ms. W’s floor came to an end, it was time for me to sign-out my list of patients to the next physician taking over. I reviewed the key medical details of her case in front of our dusty, decades-old computer monitors, and answered questions about her next steps in recovery and survival, realizing quickly I had left out a crucial detail. “You also have to meet Rita,” I insisted. “You might not survive it all without her.”
 Aristotle, Benjamin Jowett, and H W. C. Davis. Aristotle’s Politics. Oxford: At the Clarendon Press, 1920. Print.
 Darwin, C. R. The descent of man, and selection in relation to sex. London: John Murray. Volume 1. 1st edition. 1871. Page 72.
 Cassel J. The contribution of the social environment to host resistance: the Fourth Wade Hampton Frost Lecture. Am J Epidemiol. 1976 Aug;104(2):107-23.
 Conger JJ, Sawrey W, Turrell ES: The role of social experience in the production of gastric ulcers in hooded rats placed in a conflict situation. J Abnorm Psychol 57:214-220, 1958.
 Neser WB, Tyroler HA, Cassel JC: Social disorganization and stroke mortality in the Black population of North Carolina. Am J Epidemiol 93:166-175, 1971.
 Cole, Steven W. et al. “Myeloid Differentiation Architecture of Leukocyte Transcriptome Dynamics in Perceived Social Isolation.” Proceedings of the National Academy of Sciences of the United States of America 112.49 (2015): 15142–15147.
 Kroenke, Candyce H et al. “Social Networks, Social Support and Burden in Relationships, and Mortality after Breast Cancer Diagnosis in the Life After Breast Cancer Epidemiology (LACE) Study.” Breast cancer research and treatment 137.1 (2013): 261–271.
 Qin, Jun-fang et al. “Adrenergic Receptor β2 Activation by Stress Promotes Breast Cancer Progression through Macrophages M2 Polarization in Tumor Microenvironment.” BMB Reports 48.5 (2015): 295–300.
 Cassoni, P., Sapino, A., Fortunati, N., Munaron, L., Chini, B. and Bussolati, G. (1997), Oxytocin inhibits the proliferation of MDA-MB231 human breast-cancer cells via cyclic adenosine monophosphate and protein kinase A. Int. J. Cancer, 72: 340–344.
 Yang, Eric V. et al. “VEGF Is Differentially Regulated in Multiple Myeloma-Derived Cell Lines by Norepinephrine.” Brain, behavior, and immunity 22.3 (2008): 318–323.
 Kiecolt-Glaser, Janice K. et al. “Chronic Stress and Age-Related Increases in the Proinflammatory Cytokine IL-6.” Proceedings of the National Academy of Sciences of the United States of America 100.15 (2003): 9090–9095.
 Primack, Brian A. et al. Social Media Use and Perceived Social Isolation Among Young Adults in the U.S. American Journal of Preventive Medicine, Volume 53, Issue 1, 1-8.
 Pantell, Matthew et al. “Social Isolation: A Predictor of Mortality Comparable to Traditional Clinical Risk Factors.” American Journal of Public Health 103.11 (2013): 2056–2062. PMC. Web. 19 Oct. 2017.
 Nancy E. Adler, Ph.D., and William W. Stead, M.D. Patients in Context — EHR Capture of Social and Behavioral Determinants of Health. N Engl J Med 2015; 372:698-701.