Summary
The Health Action Resource Platform (HARP) data displays are a pictorial display of the datasets available.
They do not explain the underlying reasons behind the results. For example, the length of stay for patients with Diabetes or Stroke in an Integrated Care System (ICS) will be affected by multiple factors including local service models and provision, protocols, staffing levels in health and care services and wider determinants of health. Therefore, the HARP should be used as a catalyst and stimulate more in-depth discussions locally between commissioners, providers, service users, public health experts and other stakeholders about the results, reasons for any disparities highlighted by the HARP, local factors that may affect the data, local service development and resourcing, access to services, patient or user feedback and actions to reduce inequalities.
Data quality around ethnicity recording in the NHS is variable, with significant gaps and inaccurate or conflicting information in national datasets. The pandemic highlighted the urgent need to improve the quality and consistency of data as these data limitations impacted analysis. NHS England has asked all NHS organisations to proactively review and ensure the completeness of patient ethnicity data. Ensuring datasets are complete and timely is essential to underpin an understanding of and response to health inequalities, and a continued focus on improving data quality is needed.
For the data on the HARP, ethnicity data has been collected for each patient contact – the most frequently recorded ethnicity is then being used. Where there is more than one ethnicity recorded at the same frequency, the most recently recorded ethnicity has been used.
Metric Types
Within the HARP there are three key types of metrics;
- Admission and Attendance metrics look at activity for a given condition. This activity is identified by a given ethnicity and an age-standardised rate per 100,000 is calculated based on the population make-up of associated population.
- Re-admission metrics identify activity which is flagged in the Secondary Uses Service (SUS) tables as a re-admission – this is when a patient has been admitted as an emergency within 30 days of being discharged. Rates for this activity are given by ethnicity as the number of re-admissions compared to the number of admissions for the condition.
- Thirdly, Length of Stay (LOS) metrics are given as an average length of stay. These metrics are calculated by looking at the total days in hospital and using the number of admissions as a denominator to give an average by ethnicity and geography. This metric makes no adjustments for outliers, and makes no compensations between acuity of patients, so is broadly indicative only.
These metrics are all shown by ethnicity and at different geographical levels – ICB, Region and England.
Methodology, data extraction and sources
- Secondary Care Activity has been sourced from NHSE held – Secondary Uses Service (SUS) Data. The Secondary Uses Service (SUS) is the single, comprehensive repository for healthcare data in England.
- For inpatient metrics, Admitted Patient Care Spells (APCS) SUS tables have been utilised, and International Classification of Diseases (ICD-10) codes have been used to identify activity under the different metrics using the recorded diagnoses associated with the activity.
- SUS Outpatient Activity where patients attend a planned clinic does not nationally have data that identifies a patient diagnosis; therefore, outpatient metrics look at the treatment function code (TFC) which identified the type of clinic attended – e.g Cardiology, Diabetes etc.
- Emergency Care Data Set (ECDS) activity has been identified using Systematised Nomenclature of Medicine Clinical Terms ( SNOMED CT) codes. SNOMED CT is a structured clinical vocabulary for use in electronic health records).
- Activity has been extracted at an aggregate level by registered Integrated Care System (ICS), ethnicity and metric.
- Additional metrics have been created using national datasets for Mental Health Activity and Maternity.
- Mental Health activity metrics have been created using data held in the Mental Health Services Data Set (MHSDS) tables. These have been identified using the parameters held in the Technical Output Specification V5.0.
- Maternity activity metrics have been identified from the Maternity Services Data Set (MSDS) v2.0.
- To create rates across the different geographical areas we have utilised national ethnicity population data available at a commissioner/ICB level. This is collated by UK Health Facts & Dimensions and refreshed approximately quarterly.
Caveats for domains
Cardiovascular Disease
- These metrics look specifically at where there has been coded activity for either cardiac arrest, atrial fibrillation, coronary heart disease, heart failure or stroke at an emergency department or an emergency/urgent admission. Outpatient Activity has been identified under the TFC Cardiology.
Diabetes
- Outpatients’ activity is identified where the TFC is Diabetic Medicine.
- Admissions and Re-admissions metrics are identified for all admissions with a diagnosis of diabetes. These patients may have been admitted for a different primary reason. This metric is an indicator of activity for diabetic patients, rather than treatment specifically for treating the diabetic condition. Reasons behind pulling the data in this way is so that it can reflect the wider health implications of the diabetic condition, for example a patient admitted for an amputation caused by a complication due to diabetes will be captured in these figures.
- Attendances to an Emergency Department are identified using SNOMED codes identifying hyperglycaemia or hypoglycaemia.
Maternal and Neonatal Health
- Activity metrics included cover both pre- and post-natal activity, as well as birth episodes.
- Activity in secondary care is identified with diagnoses codes (ICD-10) for inpatient activity, SNOMED codes for Emergency Department Attendances.
- Further maternity metrics are being created and checked through using the Maternity Service Data Set (MSDS). These will look specifically at certain incidences recorded in maternity records.
Mental Health
- Current indicators for Mental Health focus on Self-Harm attendances and admissions in secondary care.
- Further metrics around access to services and section orders in the Mental Health Services Data Set (MHSDS) are currently being developed. Ethnicity coding is less well developed in this dataset.
Cancer
- Indicators for Cancer focus on admission to hospital and are based on ICD-10 Diagnoses codes identifying activity for Invasive Cancer Admissions not including Melanoma Skin Cancer.
Respiratory Disease
- This domain focuses on identifying admissions in secondary care for pneumonia, COPD, asthma and respiratory arrest and dyspnoea attendances to ED.
COVID
- This domain highlights activity under outpatient setting for patients attending Post Covid clinics (identified by TFC).
Sickle Cell
Activity for sickle cell disease has been identified from admissions data using ICD-10 diagnoses.
[1] A picture of health: determining the core population served by an urban NHS hospital trust and understanding the key health needs | BMC Public Health | Full Text (biomedcentral.com)