The US national Medicare inpatient hospital database is referred to as the Medpar data,
which is prepared yearly from hospital filing records. Medpar files for each state are also
prepared. The full Medpar data consists of 115 variables. The national Medpar has some
14 million records, with one record for each hospilitiztion. The data in the medpar file comes
from 1991 Medicare files for the state of Arizona. The data are limited to only one diagnostic
group (DRG 112). Patient data have been randomly selected from the original data.
Usage
data(medpar)
Format
A data frame with 1495 observations on the following 10 variables.
los
length of hospital stay
hmo
Patient belongs to a Health Maintenance Organization, binary
white
Patient identifies themselves as Caucasian, binary
died
Patient died, binary
age
Patient age range, categorical
age80
Patient age 80 and over, binary
type
Type of admission, categorical
type1
Elective admission, binary
type2
Urgent admission,binary
type3
Elective admission, binary
provnum
Provider ID
Details
medpar is saved as a data frame.
Count models use los as response variable. 0 counts are structurally excluded. The
data is also used to predict death as well as to understand the predictors which
bear on the death of a patient while in the hospital following surgery.
Source
1991 National Medpar data, National Health Economics & Research Co.
References
Hilbe, Joseph M (2015), Practical Guide to Logistic Regression, Chapman & Hall/CRC.
Hilbe, Joseph M (2014), Modeling Count Data, Cambridge University Press.
Hilbe, Joseph M (2007, 2011), Negative Binomial Regression, Cambridge University Press.
Hilbe, Joseph M (2009), Logistic Regression Models, Chapman & Hall/CRC.
first used in Hardin, JW and JM Hilbe (2001, 2007), Generalized Linear Models and Extensions, Stata Press.
Examples
# library(MASS) if not automatically loaded
# medpar in both LOGIT and COUNT packages
library(LOGIT)
data(medpar)
glmb <- glm( died ~ los + hmo + white + factor(type), family=binomial, data=medpar)
summary(glmb)
toOR(glmb)
library(LOGIT)
data(medpar)
summary(glmpb <- glm( died ~ los + hmo + white + factor(type),
family=binomial(link=probit), data=medpar))
library(LOGIT) # or library(COUNT)
data(medpar)
medpar$los<-as.numeric(medpar$los)
glmpb <- glm(los ~ hmo + white + factor(type), family=poisson, data=medpar)
summary(glmpb)
exp(coef(glmpb))
toRR(glmpb)
#library(COUNT) # nbinomial in both COUNT and msme packages
#data(medpar)
#nb2 <- nbinomial(los ~ hmo + white + factor(type), data=medpar)
#summary(nb2)
#exp(coef(nb2))
#library(LOGIT) # or library(COUNT)
#data(medpar)
#glmnb <- glm.nb(los ~ hmo + white + factor(type), data=medpar)
#summary(glmnb)
#exp(coef(glmnb))
#toRR(glmnb)
Results
R version 3.3.1 (2016-06-21) -- "Bug in Your Hair"
Copyright (C) 2016 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)
R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.
R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.
Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.
> library(LOGIT)
> png(filename="/home/ddbj/snapshot/RGM3/R_CC/result/LOGIT/medpar.rd_%03d_medium.png", width=480, height=480)
> ### Name: medpar
> ### Title: medpar
> ### Aliases: medpar
> ### Keywords: datasets
>
> ### ** Examples
>
> # library(MASS) if not automatically loaded
>
> # medpar in both LOGIT and COUNT packages
> library(LOGIT)
> data(medpar)
> glmb <- glm( died ~ los + hmo + white + factor(type), family=binomial, data=medpar)
> summary(glmb)
Call:
glm(formula = died ~ los + hmo + white + factor(type), family = binomial,
data = medpar)
Deviance Residuals:
Min 1Q Median 3Q Max
-1.3869 -0.9307 -0.8356 1.3751 2.5346
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.720149 0.219073 -3.287 0.00101 **
los -0.037193 0.007799 -4.769 1.85e-06 ***
hmo 0.027204 0.151242 0.180 0.85725
white 0.303663 0.209120 1.452 0.14647
factor(type)2 0.417873 0.144318 2.896 0.00379 **
factor(type)3 0.933819 0.229412 4.070 4.69e-05 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 1922.9 on 1494 degrees of freedom
Residual deviance: 1881.2 on 1489 degrees of freedom
AIC: 1893.2
Number of Fisher Scoring iterations: 4
> toOR(glmb)
or delta zscore pvalue exp.loci. exp.upci.
(Intercept) 0.4867 0.1066 -3.2873 0.0010 0.3168 0.7477
los 0.9635 0.0075 -4.7693 0.0000 0.9489 0.9783
hmo 1.0276 0.1554 0.1799 0.8573 0.7640 1.3821
white 1.3548 0.2833 1.4521 0.1465 0.8992 2.0412
factor(type)2 1.5187 0.2192 2.8955 0.0038 1.1446 2.0152
factor(type)3 2.5442 0.5837 4.0705 0.0000 1.6228 3.9887
>
> library(LOGIT)
> data(medpar)
> summary(glmpb <- glm( died ~ los + hmo + white + factor(type),
+ family=binomial(link=probit), data=medpar))
Call:
glm(formula = died ~ los + hmo + white + factor(type), family = binomial(link = probit),
data = medpar)
Deviance Residuals:
Min 1Q Median 3Q Max
-1.3725 -0.9284 -0.8359 1.3867 2.5427
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.45667 0.13086 -3.490 0.000483 ***
los -0.02065 0.00448 -4.610 4.03e-06 ***
hmo 0.01756 0.09236 0.190 0.849203
white 0.17801 0.12499 1.424 0.154385
factor(type)2 0.25459 0.08873 2.869 0.004113 **
factor(type)3 0.57887 0.14144 4.093 4.26e-05 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 1922.9 on 1494 degrees of freedom
Residual deviance: 1882.6 on 1489 degrees of freedom
AIC: 1894.6
Number of Fisher Scoring iterations: 4
>
> library(LOGIT) # or library(COUNT)
> data(medpar)
> medpar$los<-as.numeric(medpar$los)
> glmpb <- glm(los ~ hmo + white + factor(type), family=poisson, data=medpar)
> summary(glmpb)
Call:
glm(formula = los ~ hmo + white + factor(type), family = poisson,
data = medpar)
Deviance Residuals:
Min 1Q Median 3Q Max
-5.3063 -1.8259 -0.6319 1.0081 15.3827
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) 2.33293 0.02721 85.744 < 2e-16 ***
hmo -0.07155 0.02394 -2.988 0.00281 **
white -0.15387 0.02741 -5.613 1.99e-08 ***
factor(type)2 0.22165 0.02105 10.529 < 2e-16 ***
factor(type)3 0.70948 0.02614 27.146 < 2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for poisson family taken to be 1)
Null deviance: 8901.1 on 1494 degrees of freedom
Residual deviance: 8142.7 on 1490 degrees of freedom
AIC: 13868
Number of Fisher Scoring iterations: 5
> exp(coef(glmpb))
(Intercept) hmo white factor(type)2 factor(type)3
10.3081316 0.9309504 0.8573826 1.2481366 2.0329271
> toRR(glmpb)
rr delta zscore pvalue exp.loci. exp.upci.
(Intercept) 10.3081 0.2805 85.7439 0.0000 9.7728 10.8728
hmo 0.9310 0.0223 -2.9882 0.0028 0.8883 0.9757
white 0.8574 0.0235 -5.6131 0.0000 0.8125 0.9047
factor(type)2 1.2481 0.0263 10.5288 0.0000 1.1977 1.3007
factor(type)3 2.0329 0.0531 27.1457 0.0000 1.9314 2.1398
>
> #library(COUNT) # nbinomial in both COUNT and msme packages
> #data(medpar)
> #nb2 <- nbinomial(los ~ hmo + white + factor(type), data=medpar)
> #summary(nb2)
> #exp(coef(nb2))
>
> #library(LOGIT) # or library(COUNT)
> #data(medpar)
> #glmnb <- glm.nb(los ~ hmo + white + factor(type), data=medpar)
> #summary(glmnb)
> #exp(coef(glmnb))
> #toRR(glmnb)
>
>
>
>
>
>
> dev.off()
null device
1
>