To Do

  1. $\checkmark$ Additional columns required in extract
  2. $\checkmark$ Hospital site
  3. $\checkmark$ Date of death in hospital
  4. $\checkmark$ Discharge destination
  5. $\checkmark$ Missing data in CSP... columns over final six months of dataset
  6. $\checkmark$ Discrepancies between number of heart failure admissions in my analysis vs those in the presentation
  7. $\Box$ Control chart analysis for admissions (absolute and proportions)
  8. $\Box$ Time zone tests - and passes!!
  9. Link in
  10. $\checkmark$ the care bundle data
  11. $\checkmark$ national audit data
  12. $\checkmark$ make national audit linkage site-sensitive
  13. $\checkmark$ refactor linkage tests to use bundle/nicor id rather than row number
  14. $\checkmark$ add pt 1087099 to nicor link test - to check, in conjunction with 1073855, that multi-site spells are linked to by any appropriate nicor records, but that cross-site linking doesn't happen.
  15. $\Box$ Examine for changes in demographics over the period
  16. $\Box$ Write research plan
  17. Analyse outcomes:
  library(xtable)
  tcn <- c('Before-After Raw', 'Before-After Adjusted', 'Bundle Intervention Raw', 'Bundle Intervention Adjusted')
  trn <- c('LOS','Bed Days','Mortality','Readmissions')
r1 <- c('$\\Box$','$\\Box$','$\\Box$','$\\Box$')
r2 <- c('$\\Box$','$\\Box$','$\\Box$','$\\Box$')
r3 <- c('$\\Box$','$\\Box$','$\\Box$','$\\Box$')
r4 <- c('$\\Box$','$\\Box$','$\\Box$','$\\Box$')
tab1 <- matrix(c(r1,r2,r3,r4),nrow=4, ncol=4, byrow=TRUE)
rownames(tab1) <- trn
colnames(tab1) <- tcn
print(xtable(tab1, align="lcccc"), type="html")


HorridTom/clahrcnwlhf documentation built on May 7, 2019, 4:02 a.m.