Before I studied any of this, I watched it happen. The person I loved was living with addiction and serious mental-health difficulties, and on paper there was help around him: clinicians, referral routes, crisis services, different parts of a system that were all meant to do something. He still fell through the gaps between them, and he died.
I went to Cambridge in 2019 wanting to understand, properly, how that happens. I did not think a degree would explain one life, but I had become preoccupied with a larger question. How can support exist, sometimes in considerable quantity, and still fail to reach the person it was built for?
What My Research Looked At
My dissertation looked at sexual health and HIV prevention, because it offered a very clear example of the difference between making help available and making it genuinely accessible.
The UK has made enormous progress in HIV prevention. PrEP, testing and treatment have changed the landscape considerably, but progress has not been experienced evenly across every group. I was interested in what happens when the route through which public-health information is delivered is itself a place where some of the intended audience has learned not to feel welcome.
A great deal of HIV-prevention messaging aimed at gay and bisexual men has appeared within dating apps. The accounts I studied described racism and fetishisation on those platforms as routine experiences for some of those men. If somebody withdraws from a space because being there repeatedly comes at a cost, then the health information placed inside that space becomes less useful to them, however good the information itself may be. The help had not disappeared; the person had moved somewhere it was no longer reaching.
Those experiences were not mine, and I was careful in the research about the limits of what I could claim to know from outside them. That boundary mattered then and still does.
The Where, Not Only the Why
There is already a substantial literature on why people may disengage from services. Distrust can come from previous experience. Shame and stigma matter. So does the cost of being seen entering a particular service or admitting something in a setting where disclosure carries consequences.
What stayed with me was the importance of where. A service can be well designed, well funded and genuinely well meant, but still sit somewhere the people it hopes to reach do not go, no longer trust, or cannot safely be seen. Accessibility depends as much on whether somebody can imagine themselves walking through the door as on whether the door is open.
There was an irony in the research itself. The ethics process concluded that interviewing the group directly was not appropriate for an undergraduate project, so I worked from existing published accounts instead. I understood the decision, but it also illustrated something interesting: even research concerned with why institutions fail to reach people can find itself unable to reach those people directly.
One Box at a Time
The framework I used was intersectionality, a term introduced by the legal scholar Kimberlé Crenshaw. The basic insight is that systems built around one category at a time can miss what happens when several identities or circumstances overlap.
A service designed around sexuality may not fully account for race. One designed around race may not account for sexuality. Add family expectations, faith, migration, professional status, money, gender or the consequences of disclosure, and the neat categories used by institutions can become increasingly unlike the life of the person sitting in front of them.
Disability and neurodivergence can matter in the same way, as can money, housing, immigration status, geography and previous contact with institutions. Having to repeatedly explain why a standard environment is difficult, correct assumptions about yourself, or decide how much of your identity is safe to disclose adds work before the work has even begun. None of this means that people from a particular group will respond in a particular way. It means access is not experienced evenly, and the fact that a service technically accepts somebody does not tell us whether that person feels able to use it.
Not every service has to be everything to everybody, but a technically appropriate service can still be the wrong doorway for a particular person. I see versions of that problem in recovery work all the time.
When the next step is too much
Those questions are not only academic for me. In recovery work I see the same gap every week. Someone can have several professionals around them and still be struggling badly. They can understand perfectly well what has been recommended and still find themselves unable to make the appointment, answer the phone, complete the form, explain what is happening again or walk through another unfamiliar door. That does not necessarily mean the service is wrong. It may mean that, at that moment, the distance between the person and the service is still too great.
During periods of distress, instability or overwhelm, even straightforward tasks can become difficult to organise. A referral may involve phone calls, forms, assessments, appointments, travel, waiting lists and repeated explanations before any actual support begins. Each part may be reasonable on its own, but put together they can become one more system somebody has to manage at precisely the point when managing anything feels difficult.
This is easy to underestimate when you are standing outside the situation. “There is a service for that” may be factually correct while doing very little to describe whether the person can reach it today, so sometimes the useful question is not simply what help is available? but what would make that help possible to use?
Trust has a history
Difficulty engaging with services does not always begin with the service currently being offered. Someone may have asked for help before and waited months. They may have been assessed repeatedly, passed between teams, discharged because they missed appointments, or found themselves telling the same difficult story to a succession of strangers. A previous service may simply have been the wrong fit, or it may have failed them at a time when they needed it to work.
Experiences like that accumulate. The next professional may be thoughtful, competent and entirely different, but trust does not necessarily reset because the organisation or name on the door has changed, and what looks like reluctance from the outside can sometimes be caution learned through experience.
That does not mean mistrust should simply be accepted as permanent, or that somebody should be encouraged to withdraw from appropriate care. It means there may be work to do before another referral can become a relationship the person is able to use.
What It Looks Like in Practice
It might be an executive who is very uncomfortable being seen entering a service associated with addiction. It might be a gay man whose drug use is bound up with sex, intimacy and shame, and who does not recognise himself in a service designed around a very different presentation. It might be a young person whose difficulties sit across neurodivergence, identity, family and substance use, while each professional around them is holding only one part of the picture.
It can also be much more ordinary. Somebody leaves a good treatment programme with a sensible discharge plan and returns to a home, relationship or working life in which almost none of the surrounding conditions have changed. The treatment and the aftercare both exist; the question is whether either of them survives contact with the actual life.
Exclusion can happen inside apparently appropriate services too. Somebody can be sitting in the room and still feel that the conversation is not built for them. It might be the group in which they are the only person with their background, the form that has nowhere sensible to put their circumstances, or the well-intentioned professional who begins with a question that tells them immediately how much explaining will be required. None of this means the service is bad, only that fit matters.
Falling between the cracks
This is one of the ways people fall between services without necessarily being abandoned by any individual professional. From the system’s point of view, several things may already be in place: a referral exists, a number has been supplied, an appointment has been offered and advice has been given. From the person’s point of view, none of it may yet feel reachable.
The gap between those two perspectives matters. If we measure support only by whether it has been offered, it becomes very easy to conclude that somebody is refusing help when the more useful question may be why the help has not yet become usable. Sometimes the answer is practical and sometimes it is relational, but often it is both.
Pace matters
There is a difference between avoiding something indefinitely and needing the next step broken down into a size that can actually be managed. Pressure sometimes helps, and sometimes it makes engagement less likely; the useful response depends on the person, the situation and what is at stake.
Working at a manageable pace does not mean pretending urgent problems are not urgent. There are situations where immediate clinical, safeguarding or emergency action is necessary, and a relational approach does not remove that responsibility.
Outside those situations, however, another demand is not always what moves somebody closer to help. Sometimes it is a smaller step: a clearer explanation, one conversation rather than five, somebody staying while the phone call is made, or enough consistency that the person does not have to start again from the beginning each time.
Another referral is not always the missing piece
Modern systems are very good at referral. A difficulty is identified and the person is moved towards the professional or service designed to address it, and often that is exactly what should happen.
But referral itself is not support, and referral does not guarantee engagement. There are moments when adding another service, another number or another professional to the list simply creates another thing the person cannot yet manage, and at those points the more useful intervention may be much less impressive.
It may be helping somebody understand the options they already have, or sitting with uncertainty long enough for them to decide what they can do next. It may be practical structure around sleep, food, appointments or the day ahead, or staying alongside them while they reconnect with services at a pace that feels manageable. None of those things replaces specialist help, but sometimes they are what allow specialist help to become useful.
What That Means for How I Work
I have been working on my own sobriety since 2017. I spent twenty years in industries and social circles where drinking was deeply embedded in the culture. None of that makes me an expert in somebody else’s life, and I do not think shared experience gives anyone that right. What it has given me is a habit of starting with the person’s actual circumstances rather than with the category I think they belong in.
My work is non-clinical. I do not diagnose, assess medically or offer treatment in place of clinicians. Where clinical care is needed, I say so, and where a psychiatrist, therapist, addiction physician or treatment team is already involved, I can work alongside them with the client’s consent. The point is not to recreate a clinical service in somebody’s home.
What I can sometimes do is work in the space between a person and the help that already exists. That might mean slowing the situation down enough for somebody to understand what has actually been offered, helping organise the practical next step, being alongside them before or after an appointment, or helping them put into words what they need to say when the conversation feels too difficult to begin. It can also mean being consistent for long enough that another relationship starts to feel possible, so that useful support has somewhere to land.
One practitioner means the person does not have to begin again with a new face every week. There is no requirement to accept a particular label before the conversation can start, and while I do not create a clinical record, I do keep the limited administrative and factual records described in my Privacy & Records policy.
The principle is simple, even if the situations are not: a referral is not the same as somebody receiving help, and a service existing is not the same as somebody being able to use it. A technically correct pathway is not much use if the person cannot, will not or does not know how to enter it.
Help has to arrive
We should be careful about describing people as difficult to engage when the route into engagement may itself be difficult. A service can be excellent and still be unusable to somebody at a particular moment. A referral can be appropriate and still fail to become a relationship, and a person can need help and still distrust the people offering it. Those contradictions are not unusual; they are part of working with human beings rather than systems on paper.
The question, then, is not only whether help exists. It is whether enough trust, stability and practical possibility exist for that help to arrive. Sometimes the most useful intervention is not another referral, but having someone alongside you who can help make the next step feel possible.
Help that exists is not the same as help that arrives. Much of my work lives in the distance between the two.