Remote work and mental health are back in the spotlight after a new study was used to make a familiar argument: working from home makes people depressed. Remote work isolates employees. The answer is to get people back into the office.
That is not what the study proves.
The research points to something important: people who work fully remotely, particularly those living alone, can experience more isolation and poorer mental health over time.
That deserves attention.
But the leap from “some remote workers are becoming isolated” to “the office is the answer” is exactly the kind of category error that creates poor workplace policy.
Because isolation is not the same thing as remote work.
And attendance is not the same thing as connection.
What remote work and mental health research should make employers ask
The important finding is not that everyone needs to return to an office.
It is that, for some people, work has been operating as social infrastructure.
It has provided routine. Incidental interaction. Informal learning. A reason to leave the house. Someone noticing when they are not quite themselves. The small, ordinary moments that can make a person feel part of something.
When that infrastructure disappears without being intentionally replaced, isolation can increase.
That is not an argument against flexibility.
It is an argument against treating flexibility as a complete workplace strategy.
A person working remotely five days a week, living alone and without meaningful social connection is in a very different position from someone working flexibly across home, office, community, family, coworking spaces and relationships outside work.
These are not interchangeable arrangements.
Yet many employers are already using one finding about isolation to justify blanket return-to-office policies.
That is not evidence-led management.
It is policy by headline.
The office was never automatically a cure for loneliness
This is the part missing from most of the commentary.
People can be physically present in an office and still be isolated.
They can commute for two hours, sit in an overstimulating open-plan environment, make polite conversation, avoid asking for help, eat lunch alone and go home exhausted.
They can be visible and still feel unseen.
They can be surrounded by colleagues and still have no psychological safety, no sense of belonging and no one they trust enough to say: “I am struggling.”
For many disabled, neurodivergent, chronically ill, caring and geographically excluded workers, remote work did not remove connection.
It removed barriers.
It reduced the energy cost of participation.
It gave people more control over sensory needs, concentration, medication, food, movement, recovery and communication.
For some people, this is not a lifestyle preference.
It is the adjustment that makes sustained work possible.
That does not always mean a complex or formal process. Often, the first useful response is to look at the working environment, ask what is creating the barrier and make a practical change before delay becomes the problem. In this open-plan office reasonable-adjustments example, the difference was not a new policy or a six-week referral. It was a manager recognising the decision point, acting on an environmental barrier and recording what changed.
A blanket office-return policy that ignores this does not solve isolation. It simply transfers the cost of organisational uncertainty back onto the employee.
The opposite of isolation is not attendance. It is belonging.
Start with design. Then respond to what remains.
The first responsibility is not to wait for someone to disclose a condition, explain a barrier or make a formal request.
It is to design working practices that make connection, flexibility and participation easier for the widest range of people from the outset.
That includes purposeful in-person time, clear communication, flexible ways to contribute, predictable team rhythms and access to the information people need to do good work.
Reasonable adjustments still matter where a person experiences a barrier that the shared design does not remove.
For some employees, flexibility is not a perk. It is an adjustment.
This is where blanket return-to-office policies become riskier than they first appear.
For a disabled employee, working from home or using a hybrid arrangement may reduce a real workplace disadvantage: sensory overload, fatigue, pain, anxiety, migraine triggers, inaccessible travel, the energy cost of masking, or the cumulative impact of managing a condition alongside work.
That does not mean every remote-working request must be agreed.
It means employers need to examine the barrier before deciding that office attendance is the solution.
A new study about isolation cannot replace an individual conversation about what helps someone do their job sustainably.
The question is not:
“Does this person prefer to work from home?”
It is:
“What disadvantage are they experiencing, what adjustment could reduce it, and what evidence are we using to decide?”
That is the difference between a flexible-working policy and a reasonable-adjustments process.
The problem is not remote work and mental health in isolation. It is whether the organisation has designed connection, clarity and flexibility intentionally enough for people to sustain good work.
The actual risk is unintentional isolation
Remote work becomes harmful when organisations assume physical proximity was doing more work than it actually was.
In-office working used to create accidental connection:
- the conversation before a meeting starts
- someone noticing when you seem unlike yourself
- informal learning through observation
- shared routines
- quick access to context and support
- the moments that make people feel part of a team
When people move remote, those things do not disappear because employees have failed to make an effort.
They disappear because the organisation has not deliberately replaced them.
That is a work-design problem.
Not an employee-location problem.
The leadership question is not:
“How many days should everyone be in the office?”
It is:
“What would make connection, flexibility and participation easier to access by default — before anyone has to ask individually?”
What Heads of People should examine before changing policy
1. Is connection designed into the way work happens?
Do not assume the answer is more office attendance.
Look at whether people can build relationships, learn informally, access context, contribute in different ways and stay connected without relying on chance conversations or proximity to a senior person.
Connection should not depend on who feels confident enough to speak up, who can attend an office most easily, or who has disclosed a need.
Build in clear team rhythms, purposeful collaboration, accessible onboarding, multiple ways to contribute and in-person time with a reason behind it.
Then review where the design still leaves people behind.
2. Is this a connection problem or a management problem?
A manager who only contacts remote employees to chase updates is not creating connection.
A team with vague priorities, unclear decisions, poor onboarding and constant reactive messaging will feel disconnected whether people work from home or from the same building.
If the only route to being visible is sitting near a senior person, the organisation does not have a remote-work problem.
It has a management-system problem.
3. Is in-person time purposeful?
People do not need to commute simply to sit on video calls from another location.
In-person time has value when it is designed around a purpose:
- relationship building
- collaborative problem-solving
- mentoring
- learning
- difficult conversations
- decisions that genuinely benefit from being together
“Because it is Tuesday” is not a purpose.
4. Are you creating connection without making it performative?
Not everyone wants after-work drinks, compulsory team-building or high-energy social events.
Connection can be quieter than that.
It can mean structured mentoring, peer spaces, accessible onboarding, optional coworking, clearer communication, team agreements and managers who know how to include people who do not naturally dominate the room.
The goal is not to make everyone socially identical.
The goal is to make belonging possible.
5. Are you confusing attendance with trust?
This sits underneath many return-to-office debates.
When leaders cannot see people working, some become anxious that work is not happening. Attendance becomes a proxy for reassurance.
But presence is not performance.
Someone can be at a desk for eight hours and be unclear, overloaded, interrupted or disengaged.
Someone working remotely can be delivering exceptional outcomes while managing their energy in a way that makes sustained work possible.
Leadership is not about recreating visibility.
It is about creating clarity around outcomes, ownership, support and follow-through.
Remote work is not the issue. Disconnection is.
The study should make employers more thoughtful about how remote work is designed.
It should not give them permission to retreat into a simplistic office-versus-home argument.
Some people need more in-person connection.
Some people need more control over where and how they work.
Many need both.
The organisational task is not to choose one model and impose it on everyone.
It is to build working practices that make connection intentional, expectations clear and flexibility equitable.
Because the question is not where people are sitting.
The question is whether the way work is designed allows them to stay connected, contribute well and sustain themselves over time.
Before you change your remote-work policy
A headline is not a decision framework.
Before introducing or tightening attendance expectations, identify what is actually happening.
Where does the way work is organised make connection, flexibility or participation harder than it needs to be? Are managers relying on visibility because outcomes are unclear? Does the policy create barriers that better design could remove by default?
ANCHOR helps People teams surface the assumptions inside workplace decisions before they become disengagement, adjustment disputes or avoidable turnover.
Surface the Decisions Driving Turnover →
Not every workplace problem needs more effort.
Sometimes the missing piece is better structure.
Sometimes it is clearer strategy.
Sometimes it is rebuilding self-trust after a way of working has stopped fitting.
In Community is a free space to find your next right step — without pressure to perform.
Start with The Missing Piece, a short experience designed to help you identify whether you need more Structure, Strategy or Self-Trust.
Research note
This article responds to recent research on remote work, isolation and mental health. Its findings on isolation, especially for people living alone, are worth taking seriously. The leadership risk is using those findings to justify blanket attendance rules rather than examining the design of connection, management and flexibility in the organisation.
Research referenced: Emanuel, N., Harrington, E. and Pallais, A. (2026), Home alone: Remote work, isolation, and mental health, published in Science.
The study analysed five nationally representative US surveys covering more than 568,000 respondents. It compared pre-pandemic data from 2011–2019 with data from 2022–2024, excluding the acute pandemic years of 2020–21.
It found that the rise in remote work was associated with more time spent alone and poorer mental-health outcomes, particularly for people living alone. It does not establish that office attendance is the solution, nor does it separately test the effect of hybrid work.
