
Home, hospital, hospice or nursing home – every country has described these places in its own way, so data from different countries could not be compared reliably. The team behind the European EOLinPLACE project developed the International Classification of Dying Places, which the United Nations Statistical Commission has adopted. See how it is built and what the data collected with it will be used for – and at the end take a short test to check how much you remember.
Where do people spend the last days of their lives, and where do they die? At first glance this seems a sad but obvious question. In reality, the answer matters a great deal to governments, because it shapes the planning of hospitals, hospices, home care and social support. If it is known where people most often need care at the end of life, it is easier to decide where money, staff and places are needed.
The trouble was that there was no common way of recording such information. The EU-funded EOLinPLACE project set out to improve the quality of data on places of care and death and to make it comparable between countries. Its result is the International Classification of Dying Places, known as the ICP, which has been adopted by the United Nations Statistical Commission. Countries around the world have thus gained the first-ever international classification of the places where people are cared for at the end of life and where they die.
The way place of death is recorded differs from country to country and depends on local circumstances. Words such as “home”, “hospital”, “hospice” or “nursing home” sound unambiguous, but they are defined differently across health and social care systems. National categories and definitions also vary considerably, so the same situation can end up in one box in one country and in a completely different one in another.
For a statistician, this is a serious problem. If two countries count the same thing by different rules, comparing their results can lead to wrong conclusions. Andrew Hancock, Chair of the United Nations Committee of Experts on International Statistical Classifications, puts it plainly: “Establishing the international standard for dying places fills a significant gap in international official statistics and data”.
A statistical classification is an ordered list of categories to which phenomena are assigned, so that everyone counts them in the same way. Familiar examples are classifications of diseases or occupations. Thanks to them, a doctor in one country and an official in another record the same situation with the same code, and the data can then be added up and compared.
The ICP has a hierarchical structure. This means that broad main categories sit at the top, with increasingly detailed subcategories beneath them. Such a standardised framework can be adapted to the data systems each country uses. A country that collects less detailed data can work with the main categories, while one that wants to know more can draw on the subcategories.
According to its description, the classification distinguishes five major types of dying places:
Each of these types is divided into detailed categories and subcategories. The authors separate a person’s own residence from other dwelling types, which include refugee camps, retirement homes and correctional facilities. They also distinguish between hospitals, nursing homes and hospices, as well as other facilities, such as hotels and churches.
Fun fact: in the ICP, deaths in ambulances are counted separately from deaths in other vehicles. For people who die outdoors, it can in turn be recorded whether this happened, for example, in a forest or on a public road.
A good classification must also provide for situations that fit no box. That is why the creators of the ICP prepared a separate category, “elsewhere”, for cases where the place of death does not match any of the descriptions. There is also a category for situations in which the place is unknown.
The classification has wide uses. Doctors can use it to record the place of death on a death certificate, the official document confirming a death. It can also be applied in electronic death registration systems, clinical records, surveys and research. In all of these, the point is the same: one situation should be recorded with one common code.
According to the team, the main innovation is that the ICP does not serve only to record where someone died. It can be applied at multiple points along a person’s end-of-life pathway, to document where they receive care, where they would prefer to be cared for and where they wish to die. This information can then be compared with the place where death actually occurs.
This allows researchers to examine not only where people die, but also how they move between settings in their final days, for example from home to hospital and back again. They can also study how far reality differs from what people wished for. Barbara Gomes, project lead of EOLinPLACE at the University of Coimbra in Portugal, puts it this way: “Where people are cared for and where they die are fundamental indicators of how societies care for people towards the end of life”.
Now that the United Nations has adopted the classification, the project team is focusing on implementation. Wider use of the ICP would provide a common international metric and make it possible to collect accurate cross-sectional data on places of death, meaning data that describe the situation across many places and groups at the same time. According to the team, such data could ultimately inform the planning of:
Hancock points to the wider picture: “It supports the Sustainable Development Goals (SDGs), will inform policy and decision-making to better understand and support differing and distinct population groups, and enables more consistent and comparable mortality statistics.” The Sustainable Development Goals are a set of goals adopted by the United Nations member states that concern, among other things, health, reducing inequalities and quality of life.
The ICP is a good example of how statistics is not a dry table of numbers but a tool that influences decisions about health and care. The topic brings together several fields: public health, demography, sociology, statistics and data analysis, as well as medicine, nursing and social work. People studying these subjects may in the future design research, build data registration systems or plan care services. Sensitivity matters here too, because behind every number there is a real person and the people close to them.
For years, countries recorded places of death in different ways, so their data could not be compared reliably. The EOLinPLACE project developed the International Classification of Dying Places, adopted by the United Nations Statistical Commission, with five major types and detailed subcategories. It makes it possible to describe not only the place of death but also places of care and the wishes of patients, and the data collected are meant to help plan better care at the end of life.
5 questions · one minute