How the forecasts were created
Forecasting uses patterns in recorded data to estimate what could happen in the future. It can help us explore the possible direction and scale of future demand, but it cannot tell us exactly what will happen.
What is forecasting?
Forecasting is a way of using information about the past to explore what the future might look like. These forecasts examine how the number of people identified by adult social care and related welfare services changed between 2018 and 2025.
Statistical models were used to identify patterns in these recorded data and extend them to 2035. The forecasts focus on what is likely to happen in adult social care services based on previous adult social care trends and what happens across related welfare services.
From recorded data to forecasts
The forecasts were developed through four main stages.
-
Recorded data were collected
Published government data from 2018 to 2025 were brought together for adult social care, Personal Independence Payment, Disability Living Allowance, Education, Health and Care plans and Children in Need.
-
Previous patterns were examined
The data were analysed to understand how the number of people identified by each service had changed over time. Adult social care expenditure was also examined to understand how recorded spending had changed.
-
Statistical models were tested
Different statistical models designed for analysing numbers of people were tested. The models that best represented the patterns in the recorded data were selected.
-
Patterns were extended to 2035
The selected models were used to estimate how demand and expenditure could change up to 2035. The forecasts are updated when new recorded data become available.
Forecasts for different areas
Separate forecasts were created because patterns of demand can differ between national, regional and local areas.
What are prediction intervals?
Forecasts cannot tell us exactly what will happen. Each forecast is therefore accompanied by an 80% prediction interval, shown as a shaded area on the charts.
This is the range within which future values are expected to fall, based on the model and the available data. A wider shaded area means that there is more uncertainty.
Prediction intervals describe statistical uncertainty. They cannot account for every future change, such as new legislation, changes to eligibility rules, unexpected events or major changes in how services are funded.
How to read the charts
How should the forecasts be used?
The forecasts are intended to support long-term planning and discussion. They can help local authorities and other organisations consider the possible direction and scale of future demand.
They should not be treated as exact targets or guaranteed future figures. Local knowledge and other evidence should also be considered when planning services or setting budgets.
Important limitations
- The data show people who have been identified and recorded within public services. They do not include everyone with a learning disability or everyone who may need support.
- Each dataset has a different purpose, age range and definition of learning disability.
- A person may appear in more than one dataset, so the figures cannot be added together to estimate the total number of people with a learning disability.
- Prediction intervals cannot account for every future policy, funding, population or service change.
Why will the forecasts change?
The forecasts are based on previous patterns. They are updated when new information becomes available because adult social care, education and the benefits system can change over time.
A change in a future forecast does not necessarily mean that an earlier forecast was incorrect. It may reflect new data, changing patterns or improved information about what is happening.