Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 4 Current »

Contact

Doug Gould <Doug.Gould@icnarc.org>

Summary of Study

This request is to support a research study funded by the NIHR Health Technology Assessment
Research Programme that aims to establish the clinical and health economic effects of moving from
systemic heparin anticoagulation to regional citrate anticoagulation for continuous renal
replacement therapy for patients treated for acute kidney injury on non-specialist ICUs in England
and Wales by linking routinely collected data sources.

Objectives

  1. Investigate the short-term benefits, risks, and costs of citrate anticoagulation.
  2. Provide information on the subsequent development of end-stage renal disease (ESRD).
  3. Trial the efficient research techniques, that if successful could be used to track the effects of any change in critical care practice occurring in ICUs in England and Wales over a reasonably
    short time scale.

UKRR data are to be used for objective 2

Running

Install sql_runner -https://bitbucket.renalregistry.nhs.uk/projects/RR/repos/sql_runner/browse

Run:

python scripts\sql_run.py batch jobs\regular\RRAM Q:\Output

Definition

All patients extracted are restricted to those who were successfully matched and inserted into the STUDY_PATIENTS table. The original cohort was "for all patients with ERF diagnosed at anytime up to 31 March 2017 and alive at 1 April 2009".

These were found using:

SELECT
	A.RR_NO,
	A.DATE_BIRTH,
	A.SEX,
	ISNULL(A.NEW_NHS_NO, A.TRACING_NHS_NO) AS NEW_NHS_NO,
	B.POST_CODE
FROM
	PATIENTS A
LEFT JOIN
	(
		SELECT
			RR_NO,
			POST_CODE,
			ROW_NUMBER() OVER (PARTITION BY RR_NO ORDER BY DATE_START DESC) AS ROWNUMBER
		FROM
			RESIDENCY
		WHERE
			DATE_END IS NULL
	) B ON A.RR_NO = B.RR_NO AND B.ROWNUMBER = 1
WHERE
	(A.DATE_DEATH IS NULL OR A.DATE_DEATH > '20090401')
	AND
	A.RR_NO IN
	(
	SELECT
		RR_NO
	FROM
		QUARTERLY_TREATMENT
	INNER JOIN MODALITY_CODES ON TREATMENT_MODALITY = REGISTRY_CODE
	INNER JOIN LOCATIONS ON HOSP_CENTRE = CENTRE_CODE
	WHERE
		DATE_START <= '20170101' AND
		RRT = 1 AND
		COUNTRY_CODE IN ('GB-ENG', 'GB-WLS')
	)

NB: There is a new field in the STUDY_PATIENTS for the match score that was returned by NHS Digital should it be required in futue.


"ESRF"

Fields:

  • EDTA_DISEASE_CODE

  • EDTA_NEW_DISEASE_CODE

  • SECONDARY_DISEASE_CODE

  • PRIMARY_DISEASE_TEXT

From ESRF block for all study patients. A patient may have multiple rows if they have attended multiple renal units.

"Haemodialysis Session"

Fields:

  • HD_SESSION_DATE
  • SESSION_TYPE

For all study patients. Note that we default the session type to "HD" if it is not submitted. This would include all records prior to 2015 where we weren't officially collecting Plasma Exchange sessions.

"Patient"

Fields:

  • FIRST_SEEN_DATE
  • FIRST_EVER_RRT_MODALITY
  • FIRST_EVER_RRT_DATE
  • DATE_DEATH
  • COD_EDTA1
  • COD_EDTA2
  • COD_TEXT

The First Ever RRT fields are taken from the earliest Treatment (TXT) record where the modality was one we group as RRT. This should only include dialysis for the purposes of chronic care provided that the site is using the correct codes.

"Transplants"

Fields:

  • TRANSPLANT_DATE
  • TRANSPLANT_ID
  • TRANSPLANT_TYPE

Where TRANSPLANT_DATE is not NULL (i.e. a transplant has happened and it's not just a waiting list entry). Because sites are not submitting the new TRA block this has had to be done from the NHSBT data to obtain the TRANSPLANT_ID. Were this not required it could also be done from the Treamtent (TXT) records albeit with less information about Transplant Type.


  • No labels