Loneliness as a Public Health Problem

Transcript

[Sandro Galea]
Welcome to Ideas Matter, a podcast hosted by WashU. I’m Sandro Galea. In 2010, researchers showed that lack of social connection carries a health risk similar to smoking up to 15 cigarettes a day. Loneliness, in other words, is a public health problem. Poor social connection can increase the risk of a range of health challenges, including heart disease, stroke, and depression. A 2023 Surgeon General’s report reported that half of U.S. adults experience loneliness. Other countries have acknowledged the loneliness crisis. In 2018, the U.K. appointed a minister of loneliness as part of a suite of policies meant to address social isolation. Given the scale of loneliness in the U.S. and globally, our conversation about health is incomplete if it does not engage with this problem. Today’s guest is uniquely qualified to help lead this conversation. That finding I mentioned about loneliness and smoking? That was her work.

Julianne Holt-Lunstad was the first U.S. researcher to conduct a large-scale study of evidence linking poor social support to morbidity risk. She’s a professor of psychology and neuroscience at Brigham Young University, where she leads the Social Connection & Health Lab. She’s also the founding scientific chair and board member for the U.S. Foundation for Social Connection and the Global Initiative on Loneliness and Connection. I’m delighted to be speaking with her today, Julianne, welcome.

[Julianne Holt-Lunstad]
Thanks. Glad to be here.

[Sandro Galea]
Great to have you. Let’s start with a question about you. How did you come to be doing this work?

[Julianne Holt-Lunstad]
Yeah, I mean I’m sure, like many scientists, it was a bit of an evolution over time. I started out in grad school doing a lot of psychophysiology studies, and I’m doing laboratory-based experiments where we’re looking at stress and how it impacts our physiology. And in the context of this was always social relationships. So looking at how social support would buffer the negative health effects of stress and dampen these physiological responses. But we were also looking at how poor-quality relationships could actually be sources of stress. And so, from that work, I started thinking, and my work started evolving around not only how do our social connections impact our health outside of the context of stress, but also recognizing that outside of a small group of academics, the rest of the world really did not see or view this as a health issue. You know, maybe a well-being issue, but not as a health issue.

And so that’s where I started, um, that was really the impetus of that original meta-analysis and looking at this a little bit more robustly, because of course we knew the epidemiological studies that had existed, but really just needing to synthesize this data in a way that would show very robustly, um, or at least test robustly, whether or not our connections truly matter to our health, and particularly the most objective outcome out there, mortality. Because there’s nothing subjective about death.

[Sandro Galea]
So, what do you think it is about our interpersonal relationships that makes them so important for our physical health? And if they’re so important, why are they so understudied?

[Julianne Holt-Lunstad]
I think it’s probably been overlooked because it’s one of these things that we take for granted. We know that it impacts our quality of life, but have taken for granted the fact that it might actually impact our actual life and lifespan. So we know throughout human history that humans have needed other humans for survival. And if you look across social species, being part of a group is adaptive for survival. So whether it’s evading predators or pooling of resources, obtaining food, being part of a group is adaptive for survival. And being outside of that group can threaten survival. It’s a very vulnerable place to be. And so it’s not surprising that we see, for instance, neuroscience data that suggests that our brains actually are most well-regulated when we are social and function optimally. And when we are not, and when we’re in isolation, this can lead to dysregulation. And so whenever there’s dysregulation, of course our bodies adapt in the short term, but when that adaptation leads to dysregulation more chronically, that can put us at increased risk for a variety of outcomes that of course we’re now seeing in the research play out, from physical health to mental health to cognitive health.

[Sandro Galea]
You know, you had this really compelling, I think, finding, observation, that loneliness has a similar health risk to smoking up to 15 cigarettes a day. I’m going to get to that in a second, but first, tell us about the study from which it emerged. What were you looking for in that research? What were you doing?

[Julianne Holt-Lunstad]
Yeah, so it was a meta-analysis, which really means we gathered all of the available evidence at the time, all the studies that had examined this, and so part of our criteria was that they needed to have measured some aspect of social connection and followed people over time to see whether that predicted who was alive and who was dead at the end of the follow-up period, and so we ended up with data from 148 prospective studies. So these studies that follow people over time, and we really were trying to be very careful about this, and also trying to account for all of the alternative explanations that could exist. So, of course, age and health status are the two biggies, right? Because it’s possible, so we’re trying to rule out, you know, reverse causality and all of these sorts of things, but it’s possible that people who are in worse health to begin with are less social and therefore more likely to die earlier. Similarly, older people might also be less social, but when you’re older, you’re closer to death, right?

And so we needed to rule out those first off. And what we found was, regardless of age, regardless of initial health status, those who are more socially connected lived longer. We also looked to see whether or not studies accounted for other kinds of lifestyle factors, like smoking or physical activity or other things. And so, what we did was then we looked at what the overall magnitude of this was, adjusting for these other things so that we could get at you know what is the independent effect of social connection on our mortality. And so it was a long, arduous task. It took several years to do, but that’s kind of what went into the process.

[Sandro Galea]
What is driving this loneliness? What are the key structural drivers of loneliness in today’s world?

[Julianne Holt-Lunstad]
Well, of course, there are likely many. And I’ll also just say, as a population, we’re socially disconnected in lots of ways. So, loneliness is just one way in which we can be disconnected. And so I think it’s helpful to distinguish this because it kind of matters. So, loneliness is really the more subjective, distressing feeling. It’s more the feeling alone rather than objectively being alone. And so, this feeling or subjective feeling comes from the discrepancy between our desired level of connection and our actual level of connection. Whereas social isolation is more that objective. It’s being alone, having few relationships, infrequent social contact, and so these are both really important, and of course can go hand in hand.

So objective isolation increases your likelihood of feeling lonely, but you can be alone and not feel lonely. You might actually enjoy your time alone, and you can be surrounded by other people and still feel lonely. Now, the reason I bring that distinction up is because both have been linked to very objective health outcomes, including mortality itself. But the way in which you would go about them to reduce risk might be slightly different. Those are really critical distinctions. But getting back to your question about why are we disconnected, I think that there are a number of factors that are contributing to this. So what’s interesting is we can see these trends over time, and of course, the most recent world event that people think and look to is the pandemic. But what we see is these trends started long before the pandemic. We also look to maybe phones or social media. And while both the pandemic and at least around the year of when these technologies were introduced, yes, we see some exacerbation of these indicators of disconnection, but it didn’t begin there. And so we start seeing, you know, we can even look to the work of Bob Putnam around social capital, and we’ve seen things like declines in trust and belonging to groups and things like that.

Those kinds of declines really started decades ago, and so we’re seeing these things now. But depending on which indicator you’re looking at, some have shown more of a decline than others, and to some extent, it may be related to the kinds of things in our modern society that, in essence, give us this more of an illusion that we don’t need each other anymore, um, and that perhaps even provide barriers to connecting. And so, whether it’s the convenience of being able to order everything online, get our entertainment streaming, that we don’t have to go out into our communities to advances that have happened much farther back in our history with the invention of the television, when entertainment was now in in the home rather than participating out in the community, or even a washing machine where where you can do these kinds of tasks on your own in the comfort of your home, rather than a shared kind of responsibility. And so we see these kinds of modern advancements that not only reduce the likelihood that we are going to meet each other, but even the likelihood that we will even run into each other, and making it that much harder to connect with each other.

[Sandro Galea]
So let me stay with this theme for a second, and let’s talk about social media. You and I have previously intersected on some work around social media and health. So, social media, digital communication, how has it made it easier or harder to maintain meaningful connections? And maybe I’ll toss in there. Aren’t AI friends going to solve all this?

[Julianne Holt-Lunstad]
Well, first I’ll go with the social media, and then I’ll get into AI. I mean, social media can make it easier and harder. And so in some ways, we can think about, and I’m sure anyone listening to this can think of examples where they were able to stay in touch with childhood friends that they would have otherwise lost touch with, or somehow you know, learn about what’s happening. But of course, social media has changed a lot. We’re no longer seeing a lot of the kinds of content where it’s actually people that we even know, and can actually also be associated with displacement of our time, including socializing with others. Now, when it comes to AI, as we know, this is something that is rapidly emerging, rapidly changing, and becoming embedded in almost everything. And when it comes to particularly, so I mean, we can think about all the ways it might impact our social connections from no longer asking advice or collaborating with someone because our AI tool is now our collaborator.

Students no longer join study groups because they can just ask ChatGPT or Claude. So there are lots of ways that it may displace the kinds of interactions that we would otherwise have with people. But then there are AI companions that we’re starting to hear a lot about as well, and even you know news reports of AI boyfriends and girlfriends and people even marrying their AI companions. You know, I was at a recent gathering where there were experts around the field of AI and really thinking about what are the concerns and dangers we should be worried about? And someone said, “Well, I think one of the positive use cases is that this could solve the loneliness epidemic.” And I about fell out of my chair.

[Sandro Galea]
I can imagine.

[Julianne Holt-Lunstad]
Because you know, of course, data will be able to tell us this, but I think it’s the wrong question. And so remember my definition of loneliness: it’s that discrepancy between our desired level of connection and our actual level of connection. So, I want to also kind of go back, and the neuroscientists have also really thought of loneliness as the biological signal of an unmet biological need. So much like hunger or thirst is an uncomfortable sensation that motivates us to meet a biological need. And so if an AI companion can alleviate our loneliness, and it may very well could. Although we’re seeing some preliminary data to suggest that it might only be short term, not long term. But let’s say we live in a world where somehow it does get to a point where it can alleviate the experience of loneliness. Imagine a population that no longer feels thirsty, but isn’t getting adequate water. And that’s something that we could very well face, where people are no longer feeling lonely, but are not actually engaging with each other, not actually getting human connection. And so I think that there’s a real danger there.

[Sandro Galea]
Yeah, that’s a great metaphor. I never thought of it that way. Thank you. Before I forget, let me ask you a bit of a parallel question. We’ve been talking about loneliness and health, but of course, there are many other consequences of loneliness, economic costs of loneliness and social isolation. There are implications of loneliness for, let’s say, healthcare systems. Can you talk us through that for a second?

[Julianne Holt-Lunstad]
Yeah. So the consequences of disconnection don’t stop at just the health outcomes, and certainly not just the health of individuals, but of course health of communities, health of institutions, and society. And part of those are economic kinds of outcomes. And so, for instance, if we look at workplaces, we see that when people are more socially connected, they are more productive. They have lower absenteeism. They come up with more novel and innovative kinds of solutions. And yet, on the flip side, there can be some real economic cost to that. So, greater absenteeism, less productivity. And so there’s those kinds of economic outcomes. We also see from the point of healthcare, there was a study done by AARP where they looked at the Medicare costs of isolation. Needless to say, it was in the billions of dollars of economic costs due to Medicare costs. And so, if we just think conceptually about this, patients who are more isolated may require more time in the hospital because if they don’t have support at home to recover, they may need to stay in the hospital longer.

We also see that patients who are isolated and lonely also are more likely to use emergency services. So these are just a couple of the ways in which this might lead to some of the medical costs. Another economic outcome is that we’re seeing how it relates to prosperity. And so individuals who are more connected to folks who are of different economic classes than themselves, they are more likely to have greater earnings, particularly those who come from a lower socioeconomic status. The more connections they have with folks of higher socioeconomic status, this can influence later prosperity outcomes. So we see that this is just a highlight of some of the ways in which we’re seeing this show up in economics.

[Sandro Galea]
Let’s move to, um, age. We often think of loneliness as associated with older adults, but we also know that loneliness affects younger people, particularly adolescents. Guide us through how one thinks about loneliness, perhaps at both these ends of life.

[Julianne Holt-Lunstad]
So, first off, I should say that everyone of all ages may experience loneliness, and it’s part of being human. And so there’s no age group that’s immune to this, but we do tend to see the highest rates among young people. Particularly the most studied young group of people, is Gen Z. Now, that’s not to say that younger ages might not also be affected, but we have fewer studies looking at children. And so adolescents and young adults are where we see these prevalence rates the highest, not just in the United States, but across countries. And so, for instance, in various global studies, we see this across regions of the world. And yet, as you say, this has really been thought of as an older adult issue in the past.

So it’s helpful, I think, to really think about some of the risk factors and how they might impact different age groups differently, and how the same risk factors might show up in both age groups. And so, for example, a major life transition that disrupts your social network can increase your risk for for loneliness. And so, for older adults, that might be retirement. It might be widowhood, becoming an empty nester. These are all life transitions that are quite common in older age. Now, when we think about younger adults, that life transition of leaving home for the first time, your home of origin, whether that is you know getting your first apartment on your own and a job, or going off to college, that is another major disruption to your social network. And so, these are the sorts of things that can increase risk. We also know that poor health can increase your risk for loneliness. And so in older adults, that might be physical health.

In younger adults, of course, young adults can experience chronic health conditions. But we also know that that’s also the time when mental health often emerges. Mental health concerns either one can influence and increase one’s risk for loneliness. And so, if we start thinking about the kinds of things that might be contributing to that, I’ll also just mention one more because you know as our work overlapped, one of the factors that may be contributing is social comparison. And so social media can elevate the kinds of social comparisons that we make, and when the definition of loneliness is that discrepancy between your desired level of connection and your actual level of connection, viewing people who seem to have a lot more exciting social life than you do, or even seeing how folks that you know are all off on vacation together or at a party together, and you weren’t invited, these perceptions and that social comparison can potentially exacerbate that discrepancy, that perceived discrepancy.

[Sandro Galea]
So you’ve done a terrific job of laying out what we know about loneliness and the universe. Now, interventions. So what do we know about from data interventions that are effective at reducing loneliness.

[Julianne Holt-Lunstad]
It’s probably the biggest question and challenge that we’re facing right now because we now have very robust evidence of the kinds of consequences that are associated with this, and now the challenge is how do we reduce that risk? The challenging part is that there is somewhat of a disconnect between the kinds of evidence that we have about the health effects and the kinds of evidence that we have around interventions. And so, for instance, in terms of the health effects, we use very robust methodologies, including you know, longitudinal studies, experiments, randomized controlled trials, whereas a lot of the interventions are often, and in fact, various meta-analyses, systematic reviews, gap map analyses have rated the evidence as relatively poor in quality, primarily because a lot of interventions A, are either not actually evaluating them in the way that shows up in the academic literature.

So there may be all sorts of community programs out there, but they’re not connected with academics, so it’s not showing up in the scientific literature, or it’s looking at just post participation surveys, or maybe even pre-post, very short term. You know, that doesn’t mean that there aren’t randomized controlled trials out there, there are. But the evidence is much more limited in a few ways. I’ll just highlight a few. One is that when we look at the kinds of effects that social connections have, we’re looking at folks’ existing social networks, and yet a lot of these interventions are using volunteers or hired professionals that are complete strangers. And so, you know, it may not be surprising that the magnitude of effect is weaker when you’re dealing with, oh, you’re engaging short term with a stranger versus engaging with a close confidant or friend, um, that those might be qualitatively very different. The other is that the majority of the studies look at individual-based kinds of intervention.

In fact, a lot of them are focused on, for instance, cognitive behavioral therapy. And so it’s focusing on what the individual can do, how the individual can change, with less emphasis across the socio-ecological model, which includes how individuals are shaped by the interpersonal relationships that they’re in the communities that they’re in, the institutions, you know, societal norms, and so we see the kinds of barriers and influences that exist across these levels, and yet most of the interventions are not targeting any of these other levels of influence that may be contributing to this. And so we really have a long way to go in terms of really boosting the kinds of evidence that goes beyond just changing the individual. And in some ways, it puts the burden on the individual and more stigma on the individual as well, because it kind of then gets viewed as this is your you know a personal issue that you need to change. It’s your, you know, your responsibility. When we know that there are many external factors that are contributing factors as well.

[Sandro Galea]
So let me pick up on responsibility. Do governments have a responsibility? What can governments do? And maybe specifically, are there governments that have done things that you think are effective, or maybe the evidence shows they’re effective.

[Julianne Holt-Lunstad]
Yeah, yeah. So I’ll first start by saying governments absolutely play a role. I, you know, I want to be careful because this isn’t something that can be solved alone by government. But we also need government to contribute and certainly play their part. We need solutions from all levels of society. So I have certainly spent quite a bit of time really carefully thinking about this, particularly as I worked on the U.S. Surgeon General’s advisory, and I worked on the national strategy framework for this, really thinking about what is the government’s role, what can be done. And part of that is we reviewed a lot of the kinds of work that is happening around the globe. We know that other countries have leadership positions. So, for instance, the UK has a minister for loneliness. So does Japan.

There are policies from many countries, and even here in the United States, we have various policies that are related. And so I want to just share an example that to kind of make this real because I think a lot of times when I speak to people and talk about the role of government, the knee-jerk response might be like government has no and should not play any role in my personal life, and so I want to share an example of where this could be useful. So several years ago, I testified in a Senate hearing on aging, and there was a bill introduced, interestingly, by Senator Warren to make hearing aids more affordable, and what does this have to do with isolation and loneliness? Hearing and any kind of sensory impairment can become a barrier to connecting with other people, and is a risk factor for not only loneliness, but it’s a risk factor for other health outcomes, including dementia. And so this was introduced by Senator Warren, signed into law by President Trump. So first off, it shows an example of where there can be bipartisan support on these things, and secondly, how in essence it is addressing something that is a barrier for people. So not only is the sensory impairment a barrier, but cost is a barrier, and so making those more affordable could help in one way to make that more accessible to folks.

We also know that policies can be barriers to connection. Now, of course, the most obvious, and this is no statement on the pandemic itself, but rather we saw that when we needed to disconnect from or distance ourselves from others, it impacted literally every aspect of society. And so, what we can see is that social contact is related to transportation. It’s related to housing. It’s related to employment. What that suggested that there are policies across these sectors of society that can either hinder our social connections or potentially facilitate our social connection. And so, if we can start evaluating our policies and think about the ways in which policies might be even inadvertently getting in the way of connection, maybe they’re aimed at something entirely different. But as a byproduct, our limiting social connection, what we can do is then start to think about how do we revise these in a way that still preserves its intended outcome without the unintended consequence for social connection, just like we would evaluate policies for their potential economic consequences or environmental consequences.

[Sandro Galea]
My last question: So, what gives you hope that we can actually move in the right direction on this issue? That we can build a world with more opportunities for meaningful social connection.

[Julianne Holt-Lunstad]
I know this is going to sound strange to anyone who you know might be looking at all the headlines, and that is my hope is our young people. And despite all the headlines of how much, you know, Gen Z and young people are struggling, I also see so many incredible rays of hope. You know, I’m on a university campus. I see young people every day, young people who I live with, and I see the desire for change. I see young people really kind of wanting to pave their own path, kind of rejecting some of the norms and societal pressures that we see. I see the incredible passion that they have, and so when I see so many examples of this day after day, I think about not only how they might show up, not only for each other in their environments, but like the next generation of scientists, the next generation of leaders, the next generation of innovators. And so while I don’t want to diminish all of the, you know, and and we do have very robust evidence on many ways in which young people are struggling, we do also see young people who are thriving, and I really do take hope in them.

[Sandro Galea]
Thank you. I’m Sandro Galea. I have been talking with Julianne Holt-Lunstad about loneliness. Julianne, thank you. Thank you for joining us, and thank you to everyone who joined us for Ideas Matter. I look forward to continuing the conversation.