Part of Central Transformation Principles
ASCOF 2B/2C
The number of adults whose long-term support needs are met by admission to residential and nursing care homes (per 100,000 population).
Set cohort of interest
Client Type is a service user
Event type is service
Service type is one of long term support: Nursing care or residential care,
Derive age at period of interest end date see summary
⇩
Give all clients a new Identifier using firstly the NHS Number when available, followed by the local authority identifier to fill any gaps wherever possible to allocate a unique ID to each client in the cohort. See summary for further details
⇩
Create bespoke Event ID
Using a prescribed list of unique fields, create a new Event ID field made from a concatenation of these fields to help identify distinct service events for each client
Event ID = concatenation of Local Authority (LA) Code, Client ID, Event-Start Date, Service Type, Service Component and Delivery Mechanism
⇩
Remove open records subsequently recorded as closed
Using the new Event ID field to pin-point specific Events within the data, run a process to remove any previously submitted ‘Open’ events (i.e. no Event End Date) that have subsequently appeared in more recent submissions with an Event End Date before period of interest (for example 1/04/2024)
⇩
Create a subset to include any previous admissions that would have been included in previous reporting periods, where the event start date is before 1 April 2023
⇩
Create a separate subset to include any admissions that have started since 1 April 2023, and so therefore are new for this reporting period.
⇩
Join the two tables and only retain rows for those individuals who started a nursing/residential service during the period in question but didn't have any records of previous admissions
⇩
Numerator: Using this, count the number of new admissions by age band
Denominator: ONS population estimate by age band
Known limitations
Identifying if an admission is brand new is predicated on historical data being available since a new event can legitimately be created every time there is a change (even if the residential stay itself is long-standing) so there may be some over-counting compared with the current methodology.
The specification does not differentiate between permanent and temporary nursing and residential stays, and as this methodology is dependent on central transformation, without local triangulation, temporary admissions may be included going forwards.
As with all measures, the process is reliant on local authorities accurately capturing fields as per the relevant specification defined lists. Any fields necessary for the derivation that are invalid as per the CLD specification are removed from the analysis source data will not be corrected and invalid field entries cannot be mapped to the specification. All invalid field entries are flagged and captured in the data quality reports received by local authorities to highlight areas to be corrected in future submissions.
For the purposes of replicating SALT-derived ASCOF measures, which involve an age breakdown, where a client has missing age information, they would not be included as they cannot be mapped to an age band.
NHS Number is used as a unique identifier for each client wherever possible. Where NHS number is not populated the local authority unique ID is used instead, if this can be done without compromising accuracy. In instances where no ID can be attributed to an event row without introducing the risk of either double-counting or incorrect allocation of identifiers to individuals, these event rows will be removed from the headcount (see summary of main concepts for further information).
Last edited: 5 June 2024 3:51 pm