Skip to main content

Posts

Showing posts from February, 2019

Half-whiskers plot of difference in treatment effect

Below an alternative graphical representation of difference in treatment effects.



data ests; missing z; infile datalines delimiter=',' MISSOVER; input behandling time VASe VASs lowere uppere lowers uppers; datalines; 1,0,100,.z,100,100,.z,.z 1,1,78.88888889,.z,78.88888889,120,.z,.z 1,2,53.23552736,.z,53.23552736,100,.z,.z 1,3,27.7394636,.z,27.7394636,72.34042553,.z,.z 1,4,13.67521368,.z,13.67521368,78.72340426,.z,.z 1,5,7.948717949,.z,7.948717949,76.59574468,.z,.z 1,6,7.948717949,.z,7.948717949,77.65957447,.z,.z 1,7,2.948717949,.z,2.948717949,82.9787234,.z,.z 1,8,2.948717949,.z,2.948717949,81.91489362,.z,.z 2,0,.z,100,.z,.z,100,100 2,1,.z,64.40092166,.z,.z,11.11111111,64.40092166 2,2,.z,39.42105263,.z,.z,0,39.42105263 2,3,.z,22,.z,.z,0,22 2,4,.z,17.44444444,.z,.z,0,17.44444444 2,5,.z,9.545454545,.z,.z,0,9.545454545 2,6,.z,6.666666667,.z,.z,0,6.666666667 2,7,.z,5.409356725,.z,.z,0,5.409356725 2,8,.z,2.631578947,.z,.z,0,2.631578947 ;; run; ods graphics on /noborder; title '…

CUSUM plots for operating time

CUSUM plots have seen application within time series and quality analysis.

For both numeric, binary, ordinal and polytomous outcomes we often use residuals based on regression models instead of empirical means.

Below I apply a standard unpublished technique for adjusted CUSUM plots for operating times using a series based on information from 174 operations performed by the same team.

Simulations are added to illustrate randomness of the phenomena, based on residual standard deviation from a regression analysis.



optime<-c(375,205,255,190,195,235,255,235,305,225,225,255,255,274,255,255,195,270,235,235,255,295,195,375,259,185,205,255,205,285,175,155,215,190,195,315,185,255,405,255,235,135,255,215,255,175,215,195,205,195,345,315,195,135,233,135,315,180,155,315,255,200,315,375,255,375,270,255,210,255,185,235,195,225,145,215,230,278,245,280,230,271,270,213,290,325,375,202,214,435,165,175,325,195,435,245,225,220,215,195,285,315,230,275,355,255,175,160,255,165,145,255,225,265,255,172,375,2…

Comorbidity indexes in SQL

Generating Elixhauser comorbidity index from Danish National Health Register as relational database. (ICD 10 Coding in SAS)

A lookup-table based version of Charlson comorbidity index I made in SQL. A similar approach can be applied to Elixhauser.

SELECT V_CPR,
MAX(EI1)+MAX(EI2)+MAX(EI3)+MAX(EI4)+MAX(EI5)+
MAX(EI6)+MAX(EI7)+MAX(EI8)+MAX(EI9)+MAX(EI10)+
MAX(EI11)+MAX(EI12)+MAX(EI13)+MAX(EI14)+MAX(EI15)+
MAX(EI16)+MAX(EI17)+MAX(EI18)+MAX(EI19)+MAX(EI20)+
MAX(EI21)+MAX(EI22)+MAX(EI23)+MAX(EI24)+MAX(EI25)+
MAX(EI26)+MAX(EI27)+MAX(EI28)+MAX(EI29)+MAX(EI30)+MAX(EI31) AS Elixhauser
FROM
(SELECT V_CPR,
-- Congestive Heart Failure
CASE WHEN DIAG LIKE 'DI099%'
OR DIAG LIKE 'DI110%'
OR DIAG LIKE 'DI130%'
OR DIAG LIKE 'DI132%'
OR DIAG LIKE 'DI255%'
OR DIAG LIKE 'DI420%'
OR DIAG LIKE 'DI425%'
OR DIAG LIKE 'DI426%'
OR DIAG LIKE 'DI427%'
OR DIAG LIKE 'DI428%'
OR DIAG LIKE 'DI429%'
OR DIAG LIKE 'D…