Ask someone how long they have been a caregiver and you often get a pause. Not because they cannot remember, but because there is nothing to remember: no appointment, no form, no one saying the word out loud. The driving to appointments started. Then the pill sorting. Then sleeping with one ear open. Somewhere in there, a husband or a daughter became something the health system has a category for, and usually nobody told them. That gap between the work and the word is the subject of a study published online on August 24, 2026 in the journal Dementia (London) by Ake Grundberg of Nestor FoU-center and colleagues Taina Sormunen and Margareta Westerbotn, who set out to describe, in family members' own accounts, how they came to be identified as a family caregiver to a person living with dementia.
This is qualitative research, and it is worth being exact about what that means. The team conducted individual semi-structured interviews with 21 family caregivers in Sweden, recorded and transcribed, lasting between 18 and 80 minutes with an average of 42. Data collection ran from March 2025 to January 2026. Participants were reached through a dementia nurse who had run discussion groups for family caregivers, through a relatives' association in one municipality whose chair asked members, and then through snowball sampling, with earlier participants suggesting others. The transcripts were analysed with inductive content analysis, with coding discussed among all three authors until they agreed. There are no percentages here and no comparison group. What there is instead is a map of routes.
The researchers describe three: "Self-identification as a caregiver", "Identified before cognitive assessment" and "Identified through cognitive assessment". Holding them together is the main category the authors abstracted from the whole set, "Personal resources and random inclusion during unsystematic processes". Read that phrase slowly, because it is the finding. Some people arrived at the label by working it out themselves. Some were spotted before the memory investigation had even begun. Some were noticed during the assessment of their relative, when they happened to be sitting in the room. Which of those happened to any given family looked, in these accounts, less like a system doing its job and more like weather.
Why the label matters is not sentimental, it is administrative. The authors note that under Swedish legislation the social services should provide support to people who care for a relative who is senior, has a long-term illness or has a disability, and then point to the hole in the middle: it is not clear how those family caregivers are supposed to be identified in the first place so that the support can reach them. In these interviews, the things that decided the outcome included organisational factors such as limited cooperation and collaboration between the health care system and the social services, and also what the family member themselves brought to the encounter: previous experience of health and social care, knowledge of what the law entitled them to, the ability to navigate complex systems, and the psychological capacity to ask for help during a stretch of uncertainty and emotional strain. That last item is the cruel one. The person least able to make the phone call is often the person who most needs the call made.
For a family reading this, the practical translation is narrow but real. Support you are never identified as needing is support that may simply never be offered, and in this sample being identified frequently depended on who you met and what you already knew rather than on any reliable process. The study sits in a Swedish context the authors sketch from national sources: nearly half of people living with dementia in Sweden remain in ordinary housing with little or no formal care, and around one in five Swedish citizens has at some point provided support or care to an older person, a figure the paper draws from the National Board of Health and Welfare. The authors' conclusion is a call on services, not on families: identification, they write, is often fragmented and inconsistent, and what is needed are more proactive, preventive and coordinated approaches across health care and social services.
None of which tells anyone what to do next, and the paper does not pretend to. It does not measure whether being named a caregiver earlier leads to less exhaustion, better sleep, or a relative who stays at home longer; those are different studies. What it offers is smaller and oddly useful: the reassurance that if you have never once called yourself a caregiver, that is not an oversight on your part. In 21 Swedish households, the word tended to arrive late, sideways, and by chance.