Today’s interview is with Julianne Holt-Lunstad, PhD, professor of psychology and neuroscience at Brigham Young University, USA. Her research is focused on the long-term health effects of social connection and has been seminal in the recognition of social isolation and loneliness as risk factors for early mortality. Professor Holt-Lunstad is presenting the plenary lecture “The meta-analysis on loneliness” at September’s ECNP Congress.

Here she is interviewed by Tom Parkhill, ECNP press officer.

How big is the problem of loneliness?
That question needs to include not only how serious the problem is, but also how many people are affected. We have good evidence not only of serious health effects, but that a significant portion of the population may be at risk or affected. My meta-analysis shows that, looking at the data available world-wide, loneliness was associated with a 26% increased risk of earlier death, with social isolation and living along also showing similar or greater risks. This is data, controlling for age, initial health status, and other factors. So this suggests that these are independent risk factors for premature mortality. We also have evidence that a significant proportion of the population may be socially isolated or living alone. Of course this varies with which country we are considering, data tends to be collected differently from country to country. I’ll bring out this data more in my presentation.
We find that approximately a quarter of Americans live alone and the rates are higher in many parts of Europe, but again this varies from country to country.
Loneliness also varies from country to country and survey to survey of course. I’ve seen reports ranging from 20% to 60%, depending on how the questions are asked. It’s hard to get an exact prevalence rate, as this is not routinely collected at a population level. Nevertheless even the most conservative estimates suggest that a significant proportion of the population may be affected. So this suggests that it’s a serious problem.

You say that the surveys give quite a prevalence range; does this also imply that loneliness means different things to different people, and maybe that there are cultural differences?
Even within the US we have significant variation; even within the same year we’ve seen responses ranging from 22%, another at 33%, another at 46%. When I looked closely to see what might be different, part of it may be down to how loneliness was measured. One was specific to older adults, another was looking at a broader age range. So the measurements do make a difference, and we see that in the differences in prevalence rates between older and younger adults. There’s some data which suggest that younger people may have higher loneliness prevalence than older people.

Have people been apprehensive about admitting they are lonely? Are people now willing to talk about it more?
Certainly there has been some stigma in talking about loneliness. One of the reasons that prevalence rates can vary is whether people are asked direct or indirect questions about loneliness. A direct question for example may be “Do you feel lonely”, which can produce a certain response. But if people are asked an indirect question like “Do you lack companionship”, or “Do you feel left out”, people may be more willing to respond to these indirect questions. And so yes indeed there is significant stigma around that.
I’m hoping that with COVID-19, people will feel more open in talking about this. The global pandemic is a shared experience; we’ve all had to limit our contact with others, and so there may be less of a sense that in admitting to loneliness you are somehow a social failure. Perhaps we’ll have more of a sense that this is something we can all talk openly about.

Do you see any early evidence that COVID-19 is increasing problems?
There are several surveys, and in fact I’m part of two where we are still collecting data. A couple of early surveys have shown a 20 to 30% increase in loneliness. Now these were not using the same validated, standardised questionnaire, so we need to compare prevalence rates using the same methods. We’re also trying to use multiple measurement point to see if the perception of loneliness changes over time; do people feel better or worse as time goes on, it’s still an evolving situation.

You have previously spoken about national strategies. For example the UK launched a national loneliness strategy in 2019. Is there anything you think that society, or the clinical community, can do to address the problem of loneliness?
It’s become readily apparent during the pandemic just how fundamental human contact is to every single aspect of our lives. And it’s become apparent because every single aspect has had to change, from work, education, transport, entertainment. There is social relevance to every single one of these sectors. The policies we implement can have drastic influences on us both for good and for bad, and so as we think about policies, we really need to evaluate whether a policy might create greater social isolation, or will it foster greater connection? In particular we need to be cautious about any policy in response to the immediate threat of the pandemic which may necessarily impose physical distance, until we find a way of contain the virus, perhaps via a vaccine. We need to be very careful about these measures becoming permanent. These may have far-reaching implications on long-term public health, well beyond this initial crisis.

Plenary Lecture
PL.06 – The meta-analysis on loneliness
Tuesday 15 September, 13:00-13:30

Register to the 33rd ECNP Congress Virtual via the button below: