...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
Description
...
The key theme emerging from discussions about the future of the dashboard statistics is the necessity to
...
align core offerings with
...
most of rstp measures are covered on at least some level
...
some have very little coverage - CKD, AKI
...
RSTP have very well established numerators/denominators we should use these or when we deviate from them it should be a conscious decision.
Areas without large amounts of coverage:
Features we wish to include
...
region level comparison/stats
...
RSTP measures. These measures have been carefully considered and can be easily defended. Currently, the dashboard statistics library already covers a significant portion of these measures, specifically the numerators on a per unit basis. Utilizing the well-established numerators and denominators from RSTP should be the foundation of the new stats. However, their integration requires interpreting the numerators and denominators within the context of the UKRDC dataset.
Since the UKRDC dataset differs from the RSTP dataset, adaptations will be necessary in most cases. When data is unavailable in the UKRDC, conscious decisions should be made to identify suitable substitutes. For example, in the case of CKD metrics, where we lack the total number of patients registered at a GP surgery, we can use the number of patients from whom we have received tests at that surgery instead.
It is crucial to effectively integrate the ability to compare these metrics, which is addressed in a separate use case found here:
https://renalregistry.atlassian.net/l/cp/
...
features we wish to extend
longditudinal coverage
...
RSTP Measure
...
Type
...
This will include both peer group and longitudinal comparisons. The integration with the current UI layout will be planned at a later date. For example there could be an audit tab with a list of audit measures which are displayed in a similar way to this:
https://public.tableau.com/app/profile/ukkidney/viz/KRTauditstandards/FunnelPlot?:render=false
and actually the metrics presented here are probably worth including at some point.
RSTP Measure | Type | Numerator | UKRDC Numerator Interpretation | Denominator | UKRDC Denominator Interpretation | Current |
|---|
Dashboard Stats Calculator | Comments | |||
|---|---|---|---|---|
Adult CKD 1 low eGFR | CKD | Of the denominator those who have received a kidney function monitoring test for eGFR in the previous 12 months. | Of denominator patients with a eGFR recorded in the result items within the last 12 months. | Patients aged 18 and over, with GP recorded CKD categories G3a to G5 (previously stage 3 to 5) and no subsequent CKD resolved code. |
Living patients over |
18 with a primary renal diagnosis ( |
G3a- |
G5). I think patients with a KRT treatment should be excluded from this cohort. | No coverage - to be included in CKD calculator | UKRDC result items table/dataset contains creatinine (QBLA1 ) as well as egfr results (QBLAP, QBLAR). Should we use creatinine in the numerator? Maybe it doesn’t matter because a lab order will contain all of the above. | ||
Adult CKD 1 low eGFR | CKD | Patients aged 18 and over with two eGFRs <60ml/min/1.73m2 more than 3 months apart, but with no record of CKD categories G3a to G5 diagnosis in the GP record. | Of denominator patients with two eGFRs <60ml/min/1.73m2 more than 3 months apart | Total number of patients registered with a GP practice aged 18 and over |
not sure how much we can do with this one? maybe patients with low egfr and missing PRD/Transplant/Dialysis…some other area of the dataset…
UKRDC cohort as defined here https://renalregistry.atlassian.net/l/cp/PzKHxCsx but without counting patients without KRT treatments as recorded by UKRDC treatment table. Patients with a CKD diagnosis (PRD G3a to G5) should also be excluded. | No coverage - to be included in CKD calculator | |||
Adult CKD Prevalence | CKD | Patients with a GP recorded diagnosis of Chronic Kidney Disease G3a to G5 (previously stages 3-5) who had no record of resolution | Total number of patients registered with a GP practice aged 18 and over |
Maybe this could become something like fraction of patients with previous CKD diagnosis not appearing on KRT (and possible conservative care) | |||||
Adult CKD with hypertension and proteinurea medicated | CKD | Of the denominator those who have received a kidney function monitoring test for eGFR in the previous 12 months. | Of denominator patients with a eGFR recorded in the result items within the last 12 months. | Total number of patients aged 18 years and above with GP recorded CKD categories G3a to G5 (previously stage 3 to 5), and GP recorded hypertension and proteinuria/albuminuria (as defined in the NDA proteinuria code set). NICE guidance suggests that a conf | Living patients |
over |
18 with a primary renal diagnosis (G3a-G5) and hypertension. I think patients with a KRT treatment should be excluded from this cohort. | No coverage- to be included in CKD calculator | ||||||
Proportion of Dialysis Patients on Home therapies | Dialysis | Prevalent home therapies (home HD and PD) patients. | Prevalent patients whose treatment with the most recent start date in the ukrdc whose treatment is HHD or PD | prevalent adult dialysis patients | Patients meeting following conditions:
| RRT Calculator | I will need to double check but most of these things are basically already part of the dialysis calculator. As it currently stands most of the numerators here are represented in some sort of way so it would be a case of turning them into a single number. The definition of prevalence is something which needs to be carefully gone over again I think to make sure it is what is expected. |
Proportion of Dialysis Patients on PD | Dialysis | Prevalent PD patients. | Treatment type is PD | prevalent adult dialysis patients | ““ ““ | RRT Calculator | |
Proportion of Dialysis patients presenting late | Dialysis | Incident patients referred < 90 days before starting renal replacement therapy | Incident patients referred < 90 days before starting KRT. | RRT Calculator | Not sure if this one is doable yet. However in the land of the v5 dataset that will change I think with the CAP codes. | ||
Definitive Vascular Access | Dialysis | Patients starting on AVF/AVG/PD | Incident dialysis adult patients starting KRT for the first time | RRT Calculator | |||
Total Admitted Days for Dialysis Patients | Dialysis | Total admitted days (at any time) | Incident haemodialysis and peritoneal dialysis patients | RRT Calculator | |||
Median time from dialysis start to transplant, excluding pre-emptive transplants | Transplant | Median time from dialysis start to transplant excluding pre emptive transplants | Total days from start of dialysis to transplant excluding pre emptive | RRT Calculator | |||
Median time from dialysis start to transplant, including pre-emptive transplants | Transplant | Median time from dialysis start to transplant excluding pre emptive transplants | Total days from start of dialysis to transplant | RRT Calculator | |||
Proportion RRT patients transplanted or listed | Transplant | RRT Calculator | |||||
Transplant rate per 100 RRT patients | Transplant | RRT Calculator |
Target User(s)
Components/Dependencies
Libraries
(delete where appropriate)
dashboard stats
ukrdc fastapi
ukrdc UI
UKRDC stats dependencies
abc base class
Steps
- Create xy model
- Write core query to …
- Write abc function to filter patient cohort …
...
- Sprint where CKD calculator is written from scratch
- review of changes that need to be made
- adapt remaining calculators to include remaining RSTP measures