Rapid Cancer Registration Data dashboards

Rapid Cancer Registration Data (RCRD) provides a quick, indicative source of cancer data. It is provided to support the planning and provision of cancer services.

The data is based on a rapid processing of cancer registration data sources, in particular the Cancer Outcomes and Services Dataset (COSD) information. In comparison, National Cancer Registration Data (NCRD) relies on additional data sources, enhanced follow-up with trusts and expert processing by cancer registration officers.

These dashboards provides RCRD-based cancer incidence, early-stage proportion and treatment proportion data (for surgical tumour resections, chemotherapy and radiotherapy), by cancer group and demographic factors. There are two versions of the dashboard:

1. Public-access version - available from the links above, which includes national breakdowns by demographic factors, or data for cancer alliances and integrated care boards (ICBs).
2. Restricted-NHS-access - available from the Cancerstats2 platform* (opens in a new window), which additionally includes provider-level data, and breakdowns by demographic factors for all geographies.

*Please note that this CancerStats2 platform requires an N3/HSCN secure network connection. 

Acknowledgement

This work uses data that has been provided by patients and collected by the NHS as part of their care and support. The data are collated, maintained and quality assured by the National Disease Registration Service, which is part of NHS England.

The Rapid Cancer Registration Data (RCRD), including summary data available here, may be useful for service improvement projects including healthcare planning and prioritisation. However, it is poorly suited for epidemiological research due to limitations in the data quality and completeness (see the latest RCRD data quality document for more details).

The data could be used to aid the planning of cancer services, providing teams with recent data to try to monitor activity changes, either nationally or locally. It may also help to highlight areas of concern, to direct future service or research activities. However, the data should only be treated as indicative of the total numbers and trends, as it is not complete and it is subject to change, for example as more data becomes available.

This data is not suitable for making straightforward 'like-for-like' comparisons between providers, integrated care boards (ICBs) and cancer alliances and should not be used as a performance metric to compare activity between areas. In addition to the numerous clinical factors which typically inform a treatment decision, data for individual cancer alliances, ICBs and providers will be affected by a range of other factors, including local systems and processes; data quality and completeness; and the patient case-mix, for example in terms of cancer types or socio-demographic characteristics.

When using the Rapid Cancer Registration Data (RCRD) dashboards, it is important to be aware of the RCRD data quality and also how the measures presented, based on the RCRD data, compare to other sources of similar data. The sections below provide more information about this. 

The Rapid Cancer Registration Data (RCRD) data is derived from a range of datasets, with diagnoses derived largely from the Cancer Outcomes and Services Data (COSD), and to a smaller extent from the Cancer Waiting Times (CWT) or Hospital Episodes Statistics (HES) data. Hospital trusts submit data for these datasets to the relevant different systems, but the provision of this data can be affected by a variety of local issues or pressures. The data used to create RCRD is continually submitted and updated by trusts and regularly refreshed by NDRS, and therefore the number of diagnoses from RCRD is subject to change in future updates, as more data becomes available. This is most likely to affect the most recent few months included in each release.

Detailed assessments of the data quality of RCRD are published in monthly RCRD data quality reports.                    

The treatment proportions data published in these dashboards for more recent months will often be lower than previous months. This data is based on diagnosis date and the time between diagnosis and treatment means they will not have complete follow-up in the available data; some treatments may not yet have occurred, and others will not yet be included in the available data. However, some information is included broken down by time to treatment which will allow better comparability for recent months with only partial follow-up. The dashboards try to highlight data with incomplete follow-up using dashed lines, a reduced intensity of colouring, using an asterisk or a separate selection for data with complete follow-up only. The data is determined as complete or incomplete based on the latest near-complete month for the relevant treatment dataset(s) (see the Data timeliness section below) and the agreed treatment interval for the cancer group (see the Treatment definitions and proportions section below).

For the ethnicity, stage at diagnosis and Route to Diagnosis breakdowns, data completeness and quality is generally lower for recent months, than for earlier months in the time-series. This reflects the current availability of the data used for these breakdowns. For a given month, the completeness of these breakdowns is likely to increase over time, with updates to the dashboard, as more data becomes available. As a result of these data quality issues, it should particularly be noted that:

• to avoid providing potentially misleading data, the Routes to Diagnosis data will often be censored to an earlier date than the incidence data, as it is only provided for months where the required HES and CWT data are sufficiently available for this breakdown to be reliably derived
• previous work showed that stage at diagnosis completeness increased by around 10 to 15% from first publication to more stable data around 4 to 6 months later
• we have published some additional guidance on using RCRD to measure progress against the 75% staging ambition.

The Rapid Cancer Registration Data (RCRD) is derived from routine data sources, without manual input from expert cancer registration officers. It relies heavily on the quality and completeness of submitted data, without information being cross-checked or additional data sources being followed-up. It is therefore not as complete as the full National Cancer Registration Data (NCRD) and has not been quality assured to the same degree.

Comparison of the diagnoses recorded in RCRD and in NCRD, for April to September 2018, has shown that:

• 12.0% of NCRD registrations were not identified within RCRD (missed registrations, false negative error of RCRD)
• 6.0% of cases identified in RCRD cannot be linked to a diagnosis in the NCRD (incorrect rapid registrations, false positive error of RCRD)

These error rates vary between cancer groups, with lower rates of missed registrations or incorrect rapid registrations for sites such as breast and prostate, but higher error rates for bone and soft tissue and unknown tumours. It is known that these latter types of tumours can be more difficult to diagnose than others, often requiring additional investigations or tests; this may mean that it takes longer for full diagnosis details to be received in routine data sources. The final number of NCRD diagnoses are likely to be more different to RCRD numbers for these cancer groups than for others. Particular caution is therefore advised when using data in these dashboards for these cancer groups and the data quality should be carefully considered to determine whether it is suitable for the purpose of the work.

For bladder cancer, comparison to NCRD indicates that individual diagnoses identified as malignant (C67) in RCRD correspond to a mixture of malignant and in-situ (D09) diagnoses in NCRD. However, there are also NCRD malignant bladder cancer cases which are identified as non-malignant in RCRD and so not included in numbers within this tool. The opposing impact of these differences largely cancel each other out, meaning that the overall number of malignant bladder cancer diagnoses identified from RCRD corresponds well to NCRD. However, caution is advised when considering detailed demographic and geographic breakdowns of the bladder cancer data, as there may be additional differences between RCRD and NCRD in the characteristics for malignant and non-malignant bladder cancers.

NCRD attempts to stage all stageable cancers (some cancer types do not have a specific staging system that allows their stage to be classified). However, in RCRD, stage information is only available for 13 specific cancers, which represent approximately three quarters of all malignant cancers (excluding non-melanoma skin cancer). Details of these cancer sites and how they compare to similar sites from in the NCRD can be found in the monthly RCRD data quality reports. RCRD does not attempt to derive a stage for other cancer sites for several reasons, including complex and/or multiple relevant staging systems, for example which may require more detailed ICD-10 coding than is possible with RCRD, or poor stage completeness in NCRD.

Treatment proportion data will be lower than similar proportions published based on NCRD. The rapid treatment data presented here is based on surgery recorded in COSD or inpatient HES data, chemotherapy recorded in SACT data and radiotherapy recorded in RTDS data. The NCRD standard treatment methodology additionally uses information from the cancer registration recorded AT_TREATMENT_ENGLAND data. This data includes treatment information from a range of other sources which has been recorded and linked to an individual diagnosis record by cancer registration officers. A review of the data sources used for existing treatment proportions information indicates that the AT_TREATMENT_ENGLAND data alone provides about 12% of surgery data, 18% of chemotherapy data and 10% of radiotherapy data.

Stage at diagnosis is one of the most important factors affecting cancer outcomes, and promoting earlier stage at diagnosis is a key aim of the NHS Long-Term plan. Measuring and monitoring national staging data is crucial to understand variation and deliver evidence-based decisions.

We have published some additional guidance on using RCRD to measure progress against the 75% staging ambition. In line with this guidance, the early-stage proportion data in the public-access version of the dashboard is presented by rolling 12-month period, although data for individual months can be obtained from the data downloads available from the dashboard.

In line with our other stage publications, cancers diagnosed at stage 1 or 2 are considered to be diagnosed at an early-stage, as opposed to those diagnosed at stage 3 or 4. Only cancers with a known staging value are included in the denominator of this proportion; this is known as a complete-case approach.

The early-stage proportion results presented in the dashboards are not case-mix adjusted and do not account for differences between populations. The case-mix of cancers diagnosed impacts the percentage of early-stage cancers. For example, breast cancer is far more likely to be diagnosed at stage 1 and 2 than lung cancer. Therefore, providers, ICBs or cancer alliances with a higher-than-average occurrence of breast cancer tend to have a higher percentage of early-stage cancers compared to providers, ICBs or alliances with a higher-than-average occurrence of lung cancer.

A tumour resection is an attempt to surgically remove the whole of the primary tumour. These have been identified using lists of OPCS-4 procedure codes which were agreed through consultation with site-specific clinicians. These lists of procedure codes are available from our Cancer Treatments GitHub repository, or from the Standard Operating Procedure (Linking treatment data for cancer diagnoses CAS SOP v 4.9) available alongside our Cancer treatments dashboard,.

Surgical tumour resection procedures have not been defined for all cancer groups and are therefore only reported for cancer groups with defined lists, as above. Some tumours of other cancer groups could also have been treated with surgery.

For some cancer groups, some procedures (for example, endoscopic resections) may be appropriate for early-stage disease only. However, as stage data is currently less complete in RCRD, this tool only includes data on surgical tumour resection procedures appropriate for all stages of diagnoses; this means the number of surgical tumour resection procedures and the surgical tumour resection proportions for some cancer groups may be lower than expected and lower than published elsewhere (for example, in the Cancer treatments dashboard, which is based on NCRD).

Radiotherapy includes both curative and palliative teletherapy procedures; brachytherapy procedures (including molecular radiotherapy) are excluded.

Chemotherapy includes both curative and palliative chemotherapy but excludes hormonal therapy and other supportive drugs such as zoledronic acid, pamidronate, and denosumab.

The tumours in the ‘Other Care’ category may have received treatment other than chemotherapy, radiotherapy and surgical tumour resection, such as hormonal therapy or management of symptoms; received treatment outside of the time frame assessed; received treatment in a private capacity (i.e. treated in a private setting which was not NHS commissioned); or there may be data missing from the datasets used.

The data on treatments proportions includes data on treatments occurring in the period from 1 month before diagnosis to either 6, 9, 12, 15 or 18 months after diagnosis. The time-period within which the majority of first course of treatments occurred varies by cancer group and treatment type. Therefore, an appropriate time-period for each cancer group has been chosen using a data-driven approach in consultation with clinicians.

The treatment proportion data published in these dashboards is based on the methodology and treatment intervals detailed in the Standard Operating Procedure (Linking treatment data for cancer diagnoses CAS SOP v 4.9) available alongside our Cancer treatments dashboard, applied to the RCRD cohort and using only COSD surgery, HES, SACT and RTDS data to identify treatment.

The surgical tumour resection activity data is based on the date of the surgical procedure. Many of the patients receiving surgery will have been diagnosed several months prior to this surgery. Some patients may also be included more than once, where they had one/more of the included surgical procedures on more than one occasion.

This section provides technical information about the what data is presented on the Rapid Cancer Registration Data (RCRD) dashboards and how it is presented.

Incidence data – the dashboards report the number of new cancer diagnoses of malignant cancer excluding non-melanoma skin cancer (ICD10 C00-C97, excluding C44), from RCRD, by date of diagnosis.

Early-stage proportion – the dashboards report the proportion of staged cancers diagnosed at stage 1 or 2, from RCRD, by date of diagnosis. In the public-access version, this data is presented by rolling 12-month period, which means, for example, that data presented for January 2019 is based on data for the 12-month period from February 2018 to January 2019.

Treatment proportions – the dashboards report the proportion of patients in the RCRD incidence data who are recorded as having received treatment for their tumour, for each of chemotherapy, radiotherapy and surgical tumour resections, by date of diagnosis.

Surgical tumour resection activity – the dashboards also report the total number of surgical tumour resection procedures undertaken by date of activity, as a broader measure of surgical activity.

Patients - the dashboards include all patients diagnosed or treated in England from all age groups.

Data period - the dashboards cover new diagnoses and their treatment, and also all surgical tumour resection activity, for the period 1 January 2018 onwards.

Cancer groups - these are defined by ICD10 codes as below:

[1] Patients have been excluded where a sex-specific diagnosis code does not match the person-stated gender. This may have excluded some transgender and non-binary patients; more information about this is available on the Inequalities in cancer page of our website.

[2] For more information on stage data availability, refer to the Comparison to National Cancer Registration Data section above.

Demographic factors – data is available with breakdowns by demographic factors, namely:

• Age – age group at diagnosis (<50, 50-59, 60-69, 70-79, 80+)
• Gender
• Socio-economic deprivation – deprivation quintile from the English indices of deprivation 2019, based on area of residence at diagnosis (from 1 – most deprived to 5 – least deprived)
• Ethnicity – broad ethnicity groups (Asian, Black, Mixed and Other, Unknown, White)
• Route to Diagnosis – Derived Routes to Diagnosis based on available data (Emergency presentation, GP referral, Screening, Urgent Suspected Cancer (USC) referral, Other outpatient, Inpatient elective, Death Certificate Only (DCO), Unknown; in the public-access version, the Inpatient elective, DCO and Unknown routes are combined into a single 'Other' route; this breakdown is not available for surgical tumour resection activity)
• Stage at diagnosis – TNM stage at diagnosis (only available for some cancer groups and/or for some of the cases within each cancer group, as detailed above; this breakdown is not available for surgical tumour resection activity)

Cancer alliance is based on patient’s postcode at diagnosis for incidence and treatment proportions data. For surgical tumour resection activity, cancer alliance is based on the provider recorded for the procedure.

Integrated care board (ICB) is based on patient's postcode at diagnosis for incidence and treatment proportions data. For surgical tumour resection activity, ICB is based on patient’s postcode at the time of treatment.

Provider, which is only available on the NHS-restricted-access version, is based on diagnosing trust (for most metrics) or trust of treatment (for the surgical tumour resection activity metric only). Choices of providers are restricted by previously selected ICBs or cancer alliances, displaying only providers with some activity for patients within the selected ICB(s) and/or cancer alliance(s).

Data in the dashboards is presented in two main ways:

1. Time-trend - showing number of diagnoses (incidence) or treatment proportion data by month of diagnosis.
2. Annual comparison - comparing data for a given month to data for the same month in a previous year (usually the preceding year is used for this comparison, but for March 2021 to February 2022, data is compared to the same month from March 2019 to February 2020, to avoid comparison to the months most affected by the pandemic in 2020 or 2021; 12-month rolling data is always compared to the preceding year).

The monthly incidence and surgical tumour resection activity data are mainly presented as working-day-adjusted counts. These are adjusted for the number of working days in each month and presented for an average month with 21 working days. This provides better comparability of monthly data to avoid variation resulting from differences in the number of weekend days or bank holidays. This is particularly important for the comparisons to the same month in a previous year. However, the adjustment for the number of working days each month means that the adjusted data presented will differ from record-level data and will not add-up to observed counts of the data. Unadjusted (observed) counts are also provided in some tables for completeness, to provide real counts of patient activity.

The most recent update, released on 9 January 2025, is based on data available by 7 December 2024. The next update is planned for 30 January 2025. If, for some reason, it is not possible to release on this date, it will be updated as soon as possible after this.

For the latest release, data is estimated to be near-complete for months up to:

It is therefore important to note that the data is not 100% complete, particularly for diagnoses and treatments in more recent months.

The dashboard update includes a full data refresh, updating data for months previously published. In addition, updates to the data published, by type of data, are shown in the following table, which summarises the latest month of data published in the previous and current release:

Other notable updates included in the latest release are:

• Surgery data (activity and proportions) has been updated to fully reflect the latest list of OPCS-4 procedure codes used for our treatments work (Linking treatment data for cancer diagnoses CAS SOP v 4.9), to include the procedure codes for skin cancers of external lip and genital sites which had previously been inadvertently missed. These changes have led to an increase of nearly 2% in the number of surgical tumour resection procedures reported overall, for all malignant cancers where surgery information is provided. This reflects increases for the gynaecological (12%), urological excluding prostate (9%) and head and neck (6%) cancer sites.

The dashboards are produced by the National Disease Registration Service (NDRS). Please send any feedback or queries to [email protected] 

Please do not include sensitive or patient identifiable information.

Latest NCRD incidence statistics - Link to the latest National Cancer Registration Data (NCRD) incidence statistics.

Staging data in England from NCRD - Based on NCRD, a dashboard presenting a range of information on staging data in England, including stage completeness, stage breakdowns, early-stage proportion and case-mix adjusted data.

Routes to Diagnosis data in England from NCRD - Based on NCRD, a dashboard presenting the latest data on the Routes to Diagnosis.

Treatment proportion data in England from NCRD - Based on NCRD, a dashboard presenting the latest data on the proportion of patients recorded to have received chemotherapy, radiotherapy and surgical tumour resections.

Further chemotherapy and radiotherapy dashboards - Several dashboards providing more in-depth information on chemotherapy and radiotherapy.