Project Description

• This Networked Data Lab analysis by the NDL lab in Wales focusses on providing new insights into reablement care at a local authority level in Wales, including exploration of the demographics and health of those accessing reablement care.
• Please note that these research outputs have not yet been peer-reviewed and should be treated as preliminary.

Data Sources

• Annual District Death Extract (ADDE) (as at 01/04/2024) - used to determine individuals who meet the study criteria for inclusion in the cohort.
• Patient Episode Database Wales (PEDW) (as at 04/08/2024) - used to determine hospital referrals to reablement care services.
• Reablement care data for Bridgend Local Authority (as at 07/08/2024) - used to identify individuals in Bridgend who accessed reablement care during the study period.
• Reablement care data for Rhondda Cynon Taf Local Authority (as at 04/07/2024) - used to identify individuals in RCT who accessed reablement care during the study period.
• Welsh Demographic Service Dataset (WDSD) (as at 04/08/2024) - used to determine individuals who meet the study criteria for inclusion in the cohort and their demographic information.
• Wales Longitudinal General Practice (WLGP) (as at 04/08/2024) - used to identify individuals who meet the study criteria for inclusion in the cohort, and to determine long-term conditions.

Requirements

SQL was written to query a DB2 database in SAIL Project 1658. R scripts were written in R 4.3.3 and run in RStudio.

Getting started (Health Foundation Briefing)

• Under each section folder, the following folder structure is required:

  • data
  • outputs
  • scripts

• Section 1: Cohort creation

  • 1.01_sail_gp_registration_lookup.sql: Creates lookup table for continuous periods of registration at a SAIL GP.
  • 1.02_cohort_creation.sql: Creates reablement cohort using inclusion criteria
  • 1.03_matching_cohort.sql: Creates matching pool of potential 'non-reablement' matches
  • 1.04_matched_cohort_creation.R: creates matched cohort

• Section 2: Demographics

  • 2.01_demographics.R: Demographic analysis of reablement cohorts. Produces counts, percentages and plots by sex, age group, deprivation and rurality.
  • 2.02_ons_demographics.R: Demographic analysis of general adult popultaion in LAs of interest. Produces counts, percentages and plots by sex, age group, deprivation and rurality, using ONS mid-2022 population estimates.

• Section 3: LTC (long-term conditions)

  • 3.01_ltc_codelist.R: Creates Read code lookup for long-term conditions from Cambridge Primary Care Unit and Hanlon et. al. (2022). Adapted from code written by Dr Will Parry, sourced from HFAnalyticsLab GitHub. Original code: https://github.com/HFAnalyticsLab/CMD_multimorbidity/blob/master/R/02_Create_CamCodeList.R
  • 3.02_ltc.sql: Extracts long-term conditions from WLGP.
  • 3.03_ltc.R: Combines long-term conditions data extract with Bridgend cohort data. Adapted from code written by Dr Will Parry, sourced from HFAnalyticsLab GitHub. Original code: https://github.com/HFAnalyticsLab/CMD_multimorbidity/blob/master/R/03_Create_Cambridge_Score.R
  • 3.04_ltc_analysis.R: Analysis of long-term conditions (multimorbidity and most common conditions) in reablement/matched populations.

• Section 4: Referral Source

  • 4.01_referrals.R: Provide descriptive analysis and accompanying plots on the types of referral sources in each local authority.

• Section 5: Service Type

  • 5.01_types_of_services.R: Provide descriptive analysis and accompanying plots on the types of reablement services received by individuals in each Local Authority.