epiforecasts / covid19.nhs.data Goto Github PK

Other variables are present in the raw data that it would be sensible to return.

It would probably be sensible to normalise by population and offer rates per 100,000 as a secondary option to straight admissions. This would need:

• Support for populations at LTLA and UTLA level.
• Support for population passing in functions with defaults to package built in
• Support for outputting rates (i.e calculate and include)
• Support for mapping rates + summarising by rate.

New variables about Adult G&A beds is from 17 November 2020 (not 01 August), so current set up for dates is bringing in some mistakes. Need to check with structure of raw data (https://www.england.nhs.uk/statistics/statistical-work-areas/covid-19-hospital-activity/).

Hello, I noticed that get_admissions is taking first admissions (patients admitted with COVID for the first time), and not total admissions (all patients admitted with COVID, regardless of whether they have any previous COVID admission/s). Is this as intended?

• Offline unit tests of all functionality
• Unit test of data extraction from URL

Raw Trust-level data is split between two files from 29 April 2021 onwards, so need to update download_trust_data to download two files (but keeping functionality to download single old file, if required).

To do:

• Add functionality to download new file (which has a different URL) if release_date is on/after 29 April 2021.
• Check 06 May 2021 that raw data format will continue

Hello, just to note that the 2021 LAD boundary changes are not currently reflected in the list of LTLAs.

The changes are:

• North Northamptonshire formed by merging Corby, East Northamptonshire, Kettering, and Wellingborough
• West Northamptonshire formed by merging Daventry, Northampton, and South Northamptonshire

This probably also affects the list of UTLAs since these new LADs are both UAs.

trust_code name

1 RBS Alder Hey Children's NHS Foundation Trust
2 RBV The Christie NHS Foundation Trust
3 RGM Royal Papworth Hospital NHS Foundation Trust
4 RQ3 Birmingham Women's and Children's NHS Foundation Trust
5 RT3 Royal Brompton & Harefield NHS Foundation Trust

I notice a few hospitals in your mapping that are essentially specialist centres for specific conditions and would in theory take patients from all over the country. Some of them have a p_trust of 1 which I am surprised at as I'd have expected them to have a very broad geographical coverage - maybe a small numbers thing. I suspect the covid patients in these hospitals are incidental rather than admitted because of COVID, and I wonder if it would be better to exclude these 5 trusts before running the mapping.

Currently only the point estimate from the mapping between NHS trusts and LTLA/UTLAs is output/summarised. In reality, and especially for low numbers, these is uncertainty that is not being presented to end users. It makes sense to do this either via confidence intervals in get_admissions or potentially as samples (those this would likely need to be at a request from the user).

Hello,

I noticed that the hospital admissions data is now split between 4 files:

Weekly Admissions and Beds up to 6 April 2021 (XLSX, 4.0MB)
Weekly Admissions and Beds from 7 April up to 30 September 2021 (XLSX, 2.6MB)
Weekly Admissions and Beds from 1 October 2021 up to 31 March 2022 (XLSX, 2.6MB)
Weekly Admissions and Beds 11 August 2022 (XLSX, 1.9MB)

Could you please update download_trust_data to include this?

Thank you!

It looks like all CI is failing due to sf not being able to be installed. This looks like it is due to gdal missing which is an unstated non-R dependency. It would be a good idea to fix this as could help spot other issues with development using the CI.

West Devon LTLA (LTLA code E07000047) is missing from Trust-LTLA mapping, since there were very few (<10) admissions to any single NHS Trust prior to 30 September, so they have been filtered out of the raw data when the public mapping was made.
Not sure how to add West Devon back into the mapping in a sensible way and without distorting the proportions (p_trust and p_geo) currently in the mapping.

Hello,

Thank you for your work on this package, it is great!

I was looking to link the data available from the NHS API on NHS Trusts to the LTLAs mapping in trust_ltla_mapping; unfortunately, the NHS API catalogue includes numerous trusts that are not available in trust_ltla_mapping. For example, the Herefordshire and Worcestershire Health and Care NHS Trust (code: R1A) seems to be gobbled in the Worcestershire Acute Hospitals NHS Trust (RWP) despite having a separate entry in the NHS API. I would maybe expect this to happen with the Nightgale hospitals (codes: 'NR0A', 'NR1H', etc. - which are missing too) but the LTLA mapping is also missing existing NHS Trusts in major cities (e.g. in Liverpool, both the Liverpool Heart and Chest Hospital NHS Foundation Trust (code: RBQ) and the Liverpool Women's NHS Foundation Trust (code: REP) are missing while available to download from the NHS API).

I understand this mapping is a probabilistic estimate based on the Secondary Uses Service (SUS) healthcare data for England that uses the counts of COVID-19 hospital discharges between 01 January 2020 and 30 September 2020. That said, all the NHS trusts mentioned above have counts during this period. (In total I think they are about ~82 NHS Trust names in the NHS API that are not in the trust_ltla_mapping)
Am I misunderstanding something or is there some limitation on the SUS dataset that makes this linkage infeasible? Is it that the mapping includes only acute NHS Trusts? (in a way my question is the opposite of #16 - I suspect @sophiemeakin's comment: "we exclude Trust-LTLA pairs with fewer than 10" is a partial answer but I try to confirm I don't misunderstand something cause they are not just a handful of trusts missing)

Thanks again for your work on this, it is really helpful!

Appropriately reflect NHS authors contribution to mapping. As discussed in #6.