Every single hour, loneliness is linked to nearly a hundred deaths.
Not metaphorical deaths. Not “dying inside”, or the kind people joke about in group chats. The World Health Organization’s Commission on Social Connection published its flagship report last year, and the math is beyond jarring: about 871,000 deaths annually associated with loneliness, driven by elevated risks of heart disease, stroke, diabetes, cognitive decline, or depression. And perhaps even more worrisome, one in six people on Earth reports feeling lonely.
If a virus did this, we would shut down airports, we would go into quarantine, we would have the most capable minds on earth finding a solution.
Instead, we do almost nothing because loneliness doesn’t behave like any other disease. It has no fever, no test, no visible wound. It arrives quietly, in furnished apartments and full lecture halls, and it compounds like nothing we have seen before. We’ve been trained to think of it as a mood. But it’s a lot more than that.
And don’t get me wrong, I don’t want to scare you. I’m just saying that for something that is clearly this big of a problem, we’re missing a prescription.
And the loneliest people are often the youngest people.
Here’s the stat that should reorganize how we think about this: loneliness is not primarily an old-age problem. The WHO report found the highest rates among teenagers, about one in five people aged 13 to 17, with young adults close behind. The generation with the most connective technology in human history reports the most disconnection. Interesting.
Sit with the strangeness of that. Every previous era of loneliness could be explained by distance – a person in a remote village or the immigrant an ocean away from home. Today’s loneliness happens at zero distance. It’s the dorm room: a building holding five hundred people, and a 19 year-old inside it who hasn’t had a substantive conversation in a week. The gap isn’t just geographic anymore. It’s the space between the connections we have and the connections we actually feel, which is, in fact, exactly how the WHO defines loneliness: the distress in that gap.
What have we done about it?
Looking at the infrastructure we’ve constructed in response, its easy to notice what all of it has in common: it activates after the fire starts.
- Therapy: genuinely valuable and genuinely scarce, expensive, and waitlisted.
- Crisis lines: heroic, and by definition reserved for the worst moment of someone’s life.
- Diagnosis, referral, treatment: an entire system, and nearly all of it gated behind an admission most people will never make out loud. To access almost any of it, you must first stand up and declare I am not okay, even in cultures, families, and languages where that sentence can feel impossible to say.
- So the system works beautifully for the person already in crisis and willing to admit it, but reaches almost no one drifting toward it, or with a little less acceptance. Public health people have a name for this pattern in other domains: we built emergency rooms and forgot to build sidewalks. We treat heart attacks, but and ignore diet. In mental health, we fund the fire department but act baffled that the city keeps burning.
The question nobody is funding: what does prevention look like for loneliness?
I want to propose an unglamorous answer: companionship.
Companionship is not therapy. That’s the point.
Not treatment. Not a clinical intervention. Not a chatbot pretending to be your psychologist. Companionship is the old, boring, pre-medical technology of regular, low-stakes presence/ It’s the roommate who notices you skipped dinner, the standing Sunday call with a friend, the ritual, the object on the desk that reminds you someone thought of you, the community where showing up requires no confession first. It is everything that keeps the gap between wanted, actual connection from widening into something with a mortality statistic attached.
We dismiss this layer because it doesn’t look like medicine. But why does it have to? Because if you run it through a public-health lens, it starts to look like the highest-leverage layer we have:
It’s cheap, has no per-hour clinical cost, and no insurance approval. It also scales, through culture, ritual, and communities that reach people no appointment system ever will. It’s ungated, with no diagnosis required, no waitlist, no declaration of brokenness at the door. And it works upstream, at the drifting stage.
So, the new worldview, the one I believe will improve this issue, is that connection isn’t the soft stuff around health. It is health itself. And this isn’t my opinion, it is what researchers writing in The Lancet Public Health have started calling the neglected third pillar, alongside physical and mental.
Why am I talking about this in the first place?
I’ll be totally transparent about why I’m saying all this. The reality is that I’m 19, in the age group that forms part of this new demographic, and I’m building a company called Bemellou around exactly this thesis: that the first step toward feeling better should be small enough to actually take.
We start with something almost embarrassingly simple, a soft plush, an object on the desk that says someone thought of you, and it opens the door to true wellness and companionship through sessions, programs, people, tools, and a community for whoever wants to walk further. No diagnosis at the door. No confession required. Not therapy, just something to make it all easier.
I’m not telling you this because I think my company solves loneliness. Nothing solves loneliness alone; that’s the entire point of this blog. I’m telling you because I don’t think you can spend years watching people you love say they’re “okay” while quietly drowning, and then write about companionship as prevention, without trying to build some of that infrastructure yourself. So, I more than welcome everyone to judge the thesis on the evidence. Bemellou is just where I think I can put in my little grain of salt.
Companionship won’t fix everything that’s broken. But as medicine goes, it’s the easiest solution we’ve got, and we’ve barely begun to use it. So let’s start.
