## Warning: package 'rcompanion' was built under R version 3.4.4
Load Data
dat <- read.csv("~/nqf_caregivers/data/nqf_dat.csv", header = T, stringsAsFactors = F)
## Edit language
dat$LANGUAGE <- ifelse(dat$LANGUAGE == "ENGL", "ENGL", "OTHER")
prov <- provider_caremeasure_check(attending_check(dat))
prov <- expire_check(prov)
team <- team_caremeasure_check(attending_check(dat))
team <- expire_check(team)
Patient Level Analysis (Provider-Patient Interaction Level)
temp <- aggregate(cbind(NQF,
AGE,
SOFA) ~
ETHNICITY +
GENDER +
MARITAL_STATUS +
FIRST_CAREUNIT +
HADM_ID +
SUBJECT_ID +
CGID,
data = prov,
FUN = mean)
cat("There are", nrow(temp), "Unique Provider-Patient Interactions.\n")
## There are 4950 Unique Provider-Patient Interactions.
Provider-Patient Interaction (Provider Level) Statistics
plotDat(temp, "NQF", "GENDER", F, "Gender", "Gender", "Frequency")
test <- table(temp$GENDER, temp$NQF)
test
##
## 0 1
## F 1093 1397
## M 1289 1171
chisq.test(test)
##
## Pearson's Chi-squared test with Yates' continuity correction
##
## data: test
## X-squared = 35.497, df = 1, p-value = 2.555e-09
pairwiseNominalIndependence(
as.matrix(test),
fisher = F, gtest = F, chisq = T, method = "fdr")
## Comparison p.Chisq p.adj.Chisq
## 1 F : M 2.55e-09 2.55e-09
plotDat(temp, "NQF","ETHNICITY", F, "Ethnicity", "Ethnicity", "Frequency")
test <- table(temp$ETHNICITY, temp$NQF)
test
##
## 0 1
## BLACK 195 252
## OTHER 234 268
## WHITE 1953 2048
chisq.test(test)
##
## Pearson's Chi-squared test
##
## data: test
## X-squared = 4.8447, df = 2, p-value = 0.08871
pairwiseNominalIndependence(
as.matrix(test),
fisher = F, gtest = F, chisq = T, method = "fdr")
## Comparison p.Chisq p.adj.Chisq
## 1 BLACK : OTHER 0.3910 0.391
## 2 BLACK : WHITE 0.0421 0.126
## 3 OTHER : WHITE 0.3780 0.391
plotDat(temp, "NQF", "MARITAL_STATUS", F, "Marital Status", "Marital Status", "Frequency")
test <- table(temp$MARITAL_STATUS, temp$NQF)
test
##
## 0 1
## MARRIED 1101 970
## SINGLE 422 570
## UNKNOWN 118 93
## WIDOWED 741 935
chisq.test(test)
##
## Pearson's Chi-squared test
##
## data: test
## X-squared = 48.865, df = 3, p-value = 1.394e-10
pairwiseNominalIndependence(
as.matrix(test),
fisher = F, gtest = F, chisq = T, method = "fdr")
## Comparison p.Chisq p.adj.Chisq
## 1 MARRIED : SINGLE 4.67e-08 1.83e-07
## 2 MARRIED : UNKNOWN 4.88e-01 4.88e-01
## 3 MARRIED : WIDOWED 6.09e-08 1.83e-07
## 4 SINGLE : UNKNOWN 5.14e-04 1.03e-03
## 5 SINGLE : WIDOWED 4.23e-01 4.88e-01
## 6 UNKNOWN : WIDOWED 1.65e-03 2.47e-03
plotDat(temp, "NQF", "FIRST_CAREUNIT", F, "First Careunit", "First Careunit", "Frequency")
test <- table(temp$FIRST_CAREUNIT, temp$NQF)
test
##
## 0 1
## CCU 337 375
## CSRU 144 52
## MICU 1307 1715
## SICU 371 262
## TSICU 223 164
chisq.test(test)
##
## Pearson's Chi-squared test
##
## data: test
## X-squared = 121.24, df = 4, p-value < 2.2e-16
pairwiseNominalIndependence(
as.matrix(test),
fisher = F, gtest = F, chisq = T, method = "fdr")
## Comparison p.Chisq p.adj.Chisq
## 1 CCU : CSRU 1.44e-10 4.80e-10
## 2 CCU : MICU 5.34e-02 5.93e-02
## 3 CCU : SICU 4.50e-05 9.00e-05
## 4 CCU : TSICU 1.39e-03 1.74e-03
## 5 CSRU : MICU 3.20e-16 3.20e-15
## 6 CSRU : SICU 2.49e-04 3.71e-04
## 7 CSRU : TSICU 2.60e-04 3.71e-04
## 8 MICU : SICU 2.42e-12 1.21e-11
## 9 MICU : TSICU 1.17e-07 2.92e-07
## 10 SICU : TSICU 8.07e-01 8.07e-01
boxplot(temp$AGE ~ temp$NQF,
main = "Caremeasure Implementation by Age",
xlab = "Implementation (1 == Yes)",
ylab = "Age (Years)")
t.test(temp$AGE ~ temp$NQF)
##
## Welch Two Sample t-test
##
## data: temp$AGE by temp$NQF
## t = -10.265, df = 4943.4, p-value < 2.2e-16
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1.781107 -1.209871
## sample estimates:
## mean in group 0 mean in group 1
## 83.43432 84.92981
boxplot(temp$SOFA ~ temp$NQF,
main = "Caremeasure Implementation by SOFA Score",
xlab = "Implementation (1 == Yes)",
ylab = "SOFA Score")
t.test(temp$SOFA ~ temp$NQF)
##
## Welch Two Sample t-test
##
## data: temp$SOFA by temp$NQF
## t = -0.3143, df = 4888.3, p-value = 0.7533
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.1909025 0.1381490
## sample estimates:
## mean in group 0 mean in group 1
## 4.845508 4.871885
Hospital Admission Level Analysis (Care Team Level)
temp <- aggregate(cbind(NQF,
AGE,
SOFA) ~
ETHNICITY +
GENDER +
MARITAL_STATUS +
FIRST_CAREUNIT +
HADM_ID,
data = team,
FUN = mean)
plotDat(temp, "NQF", "GENDER", F, "Gender", "Gender", "Frequency")
test <- table(temp$GENDER, temp$NQF)
test
##
## 0 1
## F 191 442
## M 241 381
chisq.test(test)
##
## Pearson's Chi-squared test with Yates' continuity correction
##
## data: test
## X-squared = 9.8365, df = 1, p-value = 0.001711
pairwiseNominalIndependence(
as.matrix(test),
fisher = F, gtest = F, chisq = T, method = "fdr")
## Comparison p.Chisq p.adj.Chisq
## 1 F : M 0.00171 0.00171
plotDat(temp, "NQF","ETHNICITY", F, "Ethnicity", "Ethnicity", "Frequency")
test <- table(temp$ETHNICITY, temp$NQF)
test
##
## 0 1
## BLACK 32 73
## OTHER 38 87
## WHITE 362 663
chisq.test(test)
##
## Pearson's Chi-squared test
##
## data: test
## X-squared = 1.9838, df = 2, p-value = 0.3709
pairwiseNominalIndependence(
as.matrix(test),
fisher = F, gtest = F, chisq = T, method = "fdr")
## Comparison p.Chisq p.adj.Chisq
## 1 BLACK : OTHER 1.000 1.000
## 2 BLACK : WHITE 0.377 0.566
## 3 OTHER : WHITE 0.322 0.566
plotDat(temp, "NQF", "MARITAL_STATUS", F, "Marital Status", "Marital Status", "Frequency")
test <- table(temp$MARITAL_STATUS, temp$NQF)
test
##
## 0 1
## MARRIED 211 314
## SINGLE 67 179
## UNKNOWN 23 33
## WIDOWED 131 297
chisq.test(test)
##
## Pearson's Chi-squared test
##
## data: test
## X-squared = 17.223, df = 3, p-value = 0.000636
pairwiseNominalIndependence(
as.matrix(test),
fisher = F, gtest = F, chisq = T, method = "fdr")
## Comparison p.Chisq p.adj.Chisq
## 1 MARRIED : SINGLE 0.000646 0.00388
## 2 MARRIED : UNKNOWN 1.000000 1.00000
## 3 MARRIED : WIDOWED 0.002700 0.00810
## 4 SINGLE : UNKNOWN 0.059900 0.12000
## 5 SINGLE : WIDOWED 0.402000 0.48200
## 6 UNKNOWN : WIDOWED 0.153000 0.22900
plotDat(temp, "NQF", "FIRST_CAREUNIT", F, "First Careunit", "First Careunit", "Frequency")
test <- table(temp$FIRST_CAREUNIT, temp$NQF)
test
##
## 0 1
## CCU 39 114
## CSRU 42 19
## MICU 183 542
## SICU 107 89
## TSICU 61 59
chisq.test(test)
##
## Pearson's Chi-squared test
##
## data: test
## X-squared = 114.15, df = 4, p-value < 2.2e-16
pairwiseNominalIndependence(
as.matrix(test),
fisher = F, gtest = F, chisq = T, method = "fdr")
## Comparison p.Chisq p.adj.Chisq
## 1 CCU : CSRU 9.03e-09 3.01e-08
## 2 CCU : MICU 1.00e+00 1.00e+00
## 3 CCU : SICU 8.35e-08 1.67e-07
## 4 CCU : TSICU 2.82e-05 4.70e-05
## 5 CSRU : MICU 1.34e-12 6.70e-12
## 6 CSRU : SICU 6.84e-02 8.55e-02
## 7 CSRU : TSICU 3.11e-02 4.44e-02
## 8 MICU : SICU 8.32e-15 8.32e-14
## 9 MICU : TSICU 1.89e-08 4.73e-08
## 10 SICU : TSICU 5.94e-01 6.60e-01
boxplot(temp$AGE ~ temp$NQF,
main = "Caremeasure Implementation by Age",
xlab = "Implementation (1 == Yes)",
ylab = "Age (Years)")
t.test(temp$AGE ~ temp$NQF)
##
## Welch Two Sample t-test
##
## data: temp$AGE by temp$NQF
## t = -4.7878, df = 930.19, p-value = 1.961e-06
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -2.0173473 -0.8443392
## sample estimates:
## mean in group 0 mean in group 1
## 83.24378 84.67462
boxplot(temp$SOFA ~ temp$NQF,
main = "Caremeasure Implementation by SOFA Score",
xlab = "Implementation (1 == Yes)",
ylab = "SOFA Score")
t.test(temp$SOFA ~ temp$NQF)
##
## Welch Two Sample t-test
##
## data: temp$SOFA by temp$NQF
## t = -2.6818, df = 928.65, p-value = 0.007453
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.7880552 -0.1220467
## sample estimates:
## mean in group 0 mean in group 1
## 4.451389 4.906440