Loneliness: Time for Medicine to Address This Risk Factor

This transcript has been edited for clarity. 
Indu Subramanian, MD: Welcome, everyone, to Medscape. We’re so excited to cover a topic that’s very near and dear to my heart. We’re talking about loneliness and health — brain health, specifically. I am Dr Indu Subramanian. I am a neurologist at UCLA. I’m very excited to interview Professor Holt-Lunstad today.
Julianne Holt-Lunstad is a professor of psychology and neuroscience, and the director of the Social Connection and Health Lab at Brigham Young University. She’s also the founding scientific chair and board member for the US Foundation for Social Connection and the Global Initiative on Loneliness and Connection. She’s worked very closely with Dr Vivek Murthy, our Surgeon General, on the important topic of loneliness and how it affects all of our health.
Perhaps we could just hear a little bit about your background. 
Julianne Holt-Lunstad, PhD: I started out in my doctoral training looking at laboratory-based studies and how stress impacts various biological and physiological markers, really trying to understand how it is that stress impacts health. Of course, part of that was always social factors, so the extent to which social support could buffer the effects of stress and how relationships can also be sources of stress. 
Then I get into my career, and I’m doing a lot of laboratory-based studies, really taking this farther, but starting to recognize that while I’m trying to elucidate some of these biological pathways, outside of a small group of academics, it’s not really widely recognized that social connection, or lack thereof, is important for health; maybe it’s associated with emotional well-being but perhaps not relevant to physical health. 
That led me to do my seminal meta-analysis looking at, first, all indicators of social connection and how this impacts risk for mortality; and then I followed that up with one looking at isolation, loneliness, and living alone as markers of lacking connection and how that leads to increased risk for earlier death.
It was through those, really looking at this at a big-picture level, seeing the profound effects that it has on even our lifespan, that helped me then go from more of a micro-level focus to more of a macro-level. Much of my work more recently has now shifted toward how do we take this evidence, how do we implement it across various sectors, including the healthcare sector, and how does this relate to public health, so that we can potentially reduce risk.
Subramanian: I give lectures on this topic all the time. It’s one of my big passions. I had no exposure to this in medical school. I did go to medical school, full disclosure, over 20 years ago. I think we had very little on the social determinants of health. We talked about the Krebs cycle and other things ad nauseam, but I think being out in practice — and I work at a VA as well — I just see every day how loneliness can affect our patients and ourselves, honestly, as healthcare providers — all of us. We also had the pandemic enter and show us firsthand how profoundly important this is. 
There are some very profound statistics that are coming from your work. [Loneliness is] worse than obesity, worse than smoking a half a pack of cigarettes a day. Perhaps you can just update us on some of those types of things. 
Holt-Lunstad: When my colleagues and I first published this initial meta-analysis, as I mentioned, the broader public lacked any kind of awareness of this. First off, what we found was across these various indicators of high levels of social connection or low levels, so we aggregated across these different indicators. What we found was that greater social connection was associated with a 50% increased odds of survival. 
We have an odds ratio of 1.5. We could have just reported that, but we realized that the broader public is not going to be able to fully contextualize what that means. We so often hear about the various things that are good or bad for us, and so I thought, how do we benchmark this relative to things that people do care about and that the medical community takes seriously?
We sought then to take the effect size that we got across the 148 different studies and then look at the most robust meta-analyses on other factors. We included things like smoking, alcohol consumption, physical activity, air pollution, and obesity.
We had several in there, and so we benchmarked this relative to these other factors. What we found was that this was on par with these other factors. I think this was a really important first step in helping to raise awareness of just how important this is, that this rivals these other factors that we put a large amount of attention and resources toward, and that we need to be taking this just as seriously.
I will note that there have been several meta-analyses and several large-scale, prospective studies that have been published since my meta-analyses that have replicated this. I think that also further underscores our confidence in this finding because we now have multiple meta-analyses, hundreds of studies, and millions of participants.
It underscores just how important it is to prioritize social connection for health.
Subramanian: Just looking at your body of work, 20 years ago, you were studying this way before most people cared. Then you’ve had this evidence in the cardiovascular universe of outcomes and mortality from cardiovascular events. I think there’s been work in suicide risk, depression, anxiety, and many things that affect populations, and certainly people living with neurologic disease. 
In the recent literature, we have evidence that being lonely makes Parkinson’s worse for our patients. These are profound problems. We also have increased risk of getting Alzheimer’s disease, dementia, Parkinson’s — you name the disease. 
We have the evidence, but what can we do to really change outcomes for our patients? 
EAR Framework and How to Respond
Holt-Lunstad: There have been a few efforts that I’ve been involved in that I think have some important practical clinical relevance. I served as a member of a consensus committee for the National Academy of Sciences that issued a report with recommendations on the medical implications of isolation and loneliness. That report is freely available. 
From that work, consistent with these recommendations, my colleague, Carla Perissinotto, and I published a piece in The New England Journal of Medicine that introduces a framework for how to address this in clinical settings and even how it could be used in other settings. We recognized that we need practical kinds of solutions. 
One of the barriers that I faced as I was trying to publish some of my meta-analyses was not even recognizing the relevance to the readership. I distinctly remember one of my reviewers saying something to the effect of, “Even if we recognize that this is important, what can we actually do about it?” There is a perception that this may be out of their realm of expertise or ability to deal with.
What we created was a very simple framework that’s very flexible. We call this the EAR framework. EAR is an acronym for educate, assess, and respond. “Educate” is that just as clinicians might educate their patients on various lifestyle factors that either can increase their risk for health outcomes or can exacerbate their existing health conditions, that there’s real importance for clinicians to educate patients around the importance and relevance of social connection to their health.
“Assess” is the importance of assessing this in the clinic, and recognizing and identifying those across the risk trajectory. During the pandemic, I served as a subject matter expert for the Gravity project, where we were able to establish loneliness, isolation, and social support in the electronic health record. 
There are now ICD-10 codes associated with this. There are options for this. I do recognize that different kinds of systems may build it in differently, but it is available in Epic, which is one of the largest ones. 
We often may not necessarily know who is lonely and who is not. We maybe have this caricature of what we think a lonely person looks like. We often assume this person’s an older adult, or maybe you’ve got this image of a teenager in their basement on their phone in the dark. But really, loneliness can look like you and me.
We see celebrities, CEOs, even physicians, reporting that they are lonely. Really, we need to assess all of our patients, and in fact, that’s a recommendation by the National Academies of Science and Medicine. 
Assessing patients, recording that; and then the final step is R, respond. That response can be either integrating that into their care, perhaps through involving family members in the management of their disease, or finding support groups, or even support among an integrative team that may be able to provide support. Or it might be referrals out —what is often called social prescribing — and referring to resources in one’s clinic.
Far too often the thought is, “I don’t have time to deal with this so I’m just going to refer out.” I have another meta-analysis of 106 randomized controlled trials, all of which were done in medical settings where patients were randomly assigned to either standard medical treatment alone or standard medical treatment plus a psychosocial intervention. Those in the intervention had a 20% increased odds of survival and a 29% increased length of survival.
There are practical ways that clinicians can address this in the clinic. It doesn’t always mean — although in some cases it really is appropriate — to refer them to outside sources. 
Subramanian: Patients are not going to come to you and tell you that they’re lonely. They don’t even understand that this is something that they should be voicing to us as clinicians because they don’t see it as part of their medical environment.
I think we need to change the concept that this can be profoundly important for every health outcome that they may be battling in in the clinic. As clinicians, I don’t even think that we have realized that this matters for health, so to expect the patient to bring this to us is not going to happen.
I think it’s a highly stigmatized issue as well. People feel lonely and then they feel alone in being lonely, like they’re the loser at the lunch table sitting by themselves. They really don’t see that this is such a common issue. I think your own statistics note that in aging populations, over 25% are identified as being lonely, and this was prepandemic data.
Could you tell us the current statistics? 
Holt-Lunstad: It can vary depending on whether the data include those that report moderate and severe levels of loneliness or just severe levels of loneliness. I’ll share just a couple relevant prevalence rates that we often draw upon.
One is the Global State of Social Connections survey that was done in partnership between Meta and Gallup and looked at data across 142 countries. A quarter of the global population reports feeling fairly or very lonely. I share that one because what we found is that this was truly a global issue, not just a US issue; 22 of the highest-prevalence countries were in Africa. This really is a global issue. 
Here in the United States, a study based on Cigna data showed that as many as 1 in 2 Americans — half of Americans — are reporting that they are lonely. Regardless of how it’s measured, one thing I should note is that we see that it’s present across all ages and all demographics. No one is immune from this. 
That’s not to say that we don’t see variability. In fact, what we find is that some of the highest rates are among youth, so adolescents and young adults. Again, we have this perception that it’s primarily an older-adult issue, but actually the highest rates seem to be among younger populations. That’s not to say that it isn’t also a serious issue among older adults. 
We find higher rates among those who report living alone, those who report struggling financially, and — I think this is particularly important to recognize for clinicians — those who report either mental or physical health issues and concerns.
Any time someone is dealing with a health issue, they’re at increased risk of becoming isolated and lonely. Not only do we know that being isolated and lonely can lead to and increase poor health outcomes, but poor health outcomes also can increase risk for isolation and loneliness. That’s a really important trigger point for clinicians to potentially intervene. 
The other risk factors that we sometimes see are those that identify with a marginalized or minoritized community. This could be based on race, sexual orientation, or identity. Because of that, I think it’s also really important for clinicians to perhaps recognize how this might be playing a role for their patients.
Subramanian: I was reading your work, and the striking thing was us as clinicians being possibly the only touchpoint for a lonely individual in their lives. Missing that ability to screen and to possibly intervene really hit home for me. I do ask quite briefly about this to anyone that I meet. 
I think people feel like if you’re going to start the conversation, it’s going to be a hugely long answer, and you only have 15 minutes. Once we’ve identified people — and I really do take the concept of social prescribing pretty seriously as a neurologist who runs a center of excellence — I ask how they’re doing, and then I go on to ask if they feel isolated or left out — these sorts of questions.
Then I’ll say, “This week, I really want you to call your best friend from high school,” who they may have identified as the only person that they would reach out to, and try to make that part of the health plan. As I would write to increase their Parkinson’s medicine or to exercise, I now write that as something to do. 
It’s something that I, as a healthcare provider, have identified that could matter for their disease outcome. I think that makes me feel empowered, and the patient to then feel empowered, to actually do something about this.
Holt-Lunstad: I want to just acknowledge that, first, the medical community can’t solve this alone but it does play an important role. I think the example you just shared is so important and I think it also is a good reminder of that EAR framework.
Although it’s educate, assess, respond, EAR also really signifies listening, right? As we respond and provide those kinds of recommendations, referrals, and prescriptions, really listening to your patient and understanding their needs and the barriers that they face is important.
What might be a barrier for one patient might be very different for a different patient. Those underlying causes and barriers may vary, so it’s important to really listen to your patients’ circumstances and the context in which this might be occurring. 
For some, the recommendation that you have might be critically important. For another, it might be another approach because they have a different set of circumstances that have led to their loneliness which could be addressed. Really, when we think about “respond,” it’s all about being responsive to their needs and the barriers that they face, too.
Subramanian: What are some tips that you can give our listeners out there to implement today in their clinic that can make a difference?
Holt-Lunstad: I would first recommend thinking about how you can use that EAR framework and implement that in your own practice.
Part of that is becoming educated yourself. A large portion of providers don’t recognize how important this is for physical health and see this more as a mental health issue. Recognizing just how important this is can be one of those first steps.
We also recognize how challenging this is for healthcare providers. Time is a real barrier. Identify ways to integrate this within your practice in a way that can help you facilitate patient care — how can this actually make your job easier? 
That might be identifying patients’ social resources. How can they get the level of support they need that might help lead to better adherence to the medications they might be prescribed, or other kinds of practices? 
Also, outside the clinic, it’s taking care of your own social health. In our efforts to care for others, we also need to care for ourselves. If we’re burned out and we are struggling ourselves, it’s hard to care for others. For some people, that might be setting aside time for nurturing their own social connections, or in other cases it might be deepening that.
By strengthening ourselves, we can then help others in our various spheres of influence, whether that is in the role as a clinician, or maybe in your role as a parent or as a friend. We all have multiple social roles that we play and a variety of ways.
If you’re an administrator, the ways in which you can potentially influence policy, practice, and operational norms within your organization can be incredibly powerful as well.
Subramanian: I couldn’t agree with you more. For me, it’s been a buffer for so many of my stress points throughout my life as a mom, a daughter, a wife, and a doctor. I think we forget how impactful social connections are for each and every one of us every day. 
I give lectures about this because I’ve been personally transformed through some of your work and others’. There are often clinicians who come up to me or write to me to say that something I said really struck a chord with them and that they realized how lonely and isolated they were. They made a change that saved them from, perhaps, even suicide. 
It is something that I think, in this day and age, is huge. I think we have been able to normalize it in some ways due to the pandemic. I continue to take that opportunity to ask about this freely because I think it affected each of us in different ways.
Holt-Lunstad: I think it’s important to help clinicians recognize that we have data to suggest that there are high rates of burnout. I don’t want anyone listening to this to walk away thinking it’s just one more thing I have to do that I don’t have time for, recognizing that that is a real, significant issue. 
It also requires that the healthcare systems and medical community take a larger role in supporting this and providing adequate training, resources, and support for healthcare providers so that it’s not just piling on one more responsibility, but rather providing the kinds of resources that are needed to support them and their work.
Subramanian: Absolutely. When we’re talking about screening, the UCLA 3 questions that we use take less than a minute. Even that in a 15-minute encounter can eat into the time that you have to be able to spend with the patient. I think adding on yet another screen or reminder ends up adding to burnout sometimes, and depending on who the person is who’s doing the screen, if they really don’t care about the patient, it just becomes part of the paperwork.
I, myself, have had some issues health wise and I was asked some questions about suicidality by somebody who was literally also entering my insurance paperwork information. That person honestly could not have cared less. I was taken aback by who was asking.
I think the person who has the rapport with the patient, who can then actually do something about it meaningfully and can then incorporate it in the healthcare encounter, should, in my opinion, be the one who is actually screening for this and then making a difference — hopefully — in that health outcome if it exists.
I would love to see this as something that continues to be talked about, continues to be paid for in medical encounters, and to be given its rightful place as a hugely, profoundly impactful determinant of all of our health by healthcare systems and insurers as well.
Holt-Lunstad: One of the things that we really need more data on is the most effective ways to do these assessments. How might we best initiate that conversation in a nonawkward way that leads to an openness? For instance, just asking, “Have there been any changes in your social life or connections?” could be really powerful because we know that can be a trigger for loneliness.
Even if they may not be scoring high right now, let’s say they just retired or their kids just left the house, or they just moved to a new area, or they just got a divorce — all of these major changes could be the start of what could put them at risk.
That’s a really important point — to intervene before things get serious. We all know that it’s so much easier to get them early on. It’s so much harder once an issue becomes severe. That’s true for physical health, cognitive health, and mental health, and the same is true for our social health.
Subramanian: One hundred percent. This has been a beautiful conversation. Thank you so much. I’ve learned a lot, and hopefully our viewers will take away some pearls as well. We’ll keep the conversation going.
Thank you so much, Julianne, for spending time with us today. 
Holt-Lunstad: My pleasure. Thank you so much for having me.