Hospital Admitted Patient Care Activity 2019-20

Data for paediatric and neonatal care is collected principally through separate data sets which are not comparable to the adult critical care data set; analysis is therefore limited to those records which identify the critical care unit function (the principal clinical service provided in the ward) as predominately one for the treatment of adults. Nevertheless, some neonatal and paediatric patients will have been treated in these critical care units and their data are included for analysis.

For coherence with other HES-derived publications, records have been included for analysis only where the discharge date is within the 2019-20 financial year.

Most raw critical care records in the data set are associated with only one record of an APC episode  but some are associated with multiple episodes in an APC spell; to avoid multi-counting these critical care records, an algorithm has been applied to link each critical care record to the APC episode record against which it is the ‘best match’. Because of limitations in the algorithm, a small number of raw critical care records are identified as a ‘best match’ with more than one APC episode and the final presented numbers of critical care records slightly over-count the true number of critical care periods.

The best match methodology and its impact are laid out in more detail in the Best Match Methodology section. 

Certain records have been excluded from analysis, including:

• records exhibiting illogical critical care dates (i.e. an end date prior to the start date); and
• critical care records which have not been able to be linked to a hospital APC record  owing to its having been removed from the APC data-set during a data cleaning process. These cleaning processes may include the removal of duplicate records, activity which occurred outside of the relevant date range or removal on specific request of a data provider.

The first step is an examination of the relationship between each critical care to APC association and to assign these relationships a ranking in a thirteen-point hierarchy (see Figure 2). A determination has been made that the lower the cardinal number of the rank in the hierarchy, the better the relationship between the critical care record or records with the most senior rank.

Where step 1 of the algorithm identified more than one APC episode per critical care record as a best match candidate, the record or records with the maximum number of critical care support activities are retained.

Where two or more records remain as best match candidates, the record with the lowest record ID (a unique but arbitrary record identifier) is retained.

Where a single critical care period record is associated with more than one APC episode with different identifiers and with equally good hierarchy rank, each such association will be flagged by the match algorithm as a best match. As a result, a small number of duplicates will remain in the final data-set for analysis; a summary is shown in Appendix 1 of the Summary Report. These duplicates can arise from there being duplicate episodes in the APC data-set or from the APC records with which the critical care period was associated being of equally poor quality.