The promise of digital health has always been told in the language of convenience: book in seconds, check results on your phone, never wait on hold again. A new qualitative longitudinal study of nine women experiencing homelessness in London documents the other side of that ledger. Its finding is blunt but carefully argued: digital health systems can create new layers of exclusion for people least equipped to absorb them, and the researchers behind the work close with a proposal structured in three layers to address it.

Why the method matters

Nine participants is a small number by the standards of a survey, and some readers will be tempted to dismiss the study on that basis alone. That would miss the point of the method. Qualitative longitudinal research follows people across time rather than sampling a single moment, which is precisely what you need if you want to understand how access to care actually fails — not in the aggregate, but in the sequence of small events that make up a life without stable housing. A snapshot can tell you how many people report difficulty booking an appointment. Only sustained contact with people's lives can show you when, and in what order, the difficulty arrives.

That matters because the assumptions baked into digital-first care are stacked, and homelessness breaks them in sequence. Registration typically presumes an address. Booking presumes a phone with charge, credit and data. Reminders presume a number or inbox that survives a move. Using the app at all presumes a degree of literacy, privacy and calm that most people take for granted and cannot imagine losing. Each assumption holds for the settled majority nearly all of the time, so each one goes unexamined. The study's use of the word "new" in describing these layers of exclusion is doing real work: digitization here is not simply failing to solve old barriers, it is manufacturing fresh ones of its own.

Three layers, one argument

The researchers pair their findings with proposals arranged in three layers, and the layering itself is the argument. A problem that lives simultaneously in the design of a system, in the practice of frontline staff, and in the rules that govern access cannot be fixed at any one of those levels alone. An app redesign does not help the woman who has no phone at all that week. A sympathetic receptionist cannot override a form that demands documents the person has no way to produce. A policy memo changes nothing if the interface still assumes a permanence of address that the user's life does not have. Any honest response has to reach all three at once, which is presumably why the authors chose the structure they did.

There is a broader pattern here worth naming. Digital services tend to be judged by how they perform for the majority, and the people who fall through are filed away as edge cases — a word that quietly means "not us." In most commercial contexts that habit is merely sloppy. In health care it is where the cost lands heaviest, because the person who finds the front door locked is usually the one with the least slack to come back and try again tomorrow. The study does not need to make sweeping claims about outcomes to make this point; the documentation of exclusion, described in the participants' own circumstances over time, is enough.

What would it look like to design for the margins instead? Offline routes that function as more than a token alternative. Staff with the time and institutional permission to walk someone through a process rather than point at a QR code. Systems that tolerate a changed phone number, a missing postcard of proof, a life in motion. None of this is anti-technology, and the study should not be read as a call to abandon digital tools. Its quieter point is more uncomfortable: exclusion is not a bug that crept in unnoticed. It was designed in, by people who never had the nine women of this study in mind — which means it can, with the same deliberateness, be designed out.