Week 10-16 and 10-26 All Data Analysis

For years 2010~2017

1. How many unique hospital IDs? If any missing?
2. How many admissions per hospital from 2010 to 2017?
3. Identify unique patients/hospital from 2010 to 2017.

CORE2010 = read.csv("~/OneDrive/Johns Hopkins/Ali Sobhi Afshar - HCUP/Data/SIDC_MD_2010/rds/MD_SID_2010_CORE.csv");
CORE2011 = read.csv("~/OneDrive/Johns Hopkins/Ali Sobhi Afshar - HCUP/Data/SIDC_MD_2011/rds/MD_SID_2011_CORE.csv");
CORE2012 = read.csv("~/OneDrive/Johns Hopkins/Ali Sobhi Afshar - HCUP/Data/SIDC_MD_2012/rds/MD_SID_2012_CORE.csv");
CORE2013 = read.csv("~/OneDrive/Johns Hopkins/Ali Sobhi Afshar - HCUP/Data/SIDC_MD_2013/rds/MD_SID_2013_CORE.csv");
CORE2014 = read.csv("~/OneDrive/Johns Hopkins/Ali Sobhi Afshar - HCUP/Data/SIDC_MD_2014/rds/MD_SID_2014_CORE.csv");
CORE2015q4 = read.csv("~/OneDrive/Johns Hopkins/Ali Sobhi Afshar - HCUP/Data/SIDC_MD_2015/rds/MD_SID_2015q4_CORE.csv");
CORE2015q1q3 = read.csv("~/OneDrive/Johns Hopkins/Ali Sobhi Afshar - HCUP/Data/SIDC_MD_2015/rds/MD_SID_2015q1q3_CORE.csv");
colnames(CORE2015q1q3) = colnames(CORE2015q4);
CORE2015=rbind(CORE2015q1q3, CORE2015q4);
CORE2016 = read.csv("~/OneDrive/Johns Hopkins/Ali Sobhi Afshar - HCUP/Data/SIDC_MD_2016/rds/MD_SID_2016_CORE.csv");
CORE2017 = read.csv("~/OneDrive/Johns Hopkins/Ali Sobhi Afshar - HCUP/Data/SIDC_MD_2017/rds/MD_SID_2017_CORE.csv");

How many unique hospital IDs? If any missing?

uniqueHospitals <- function (yearData) {
    yearData$DSHOSPID = as.factor(yearData$DSHOSPID);
    numHospitals = nrow(as.data.frame(levels(yearData$DSHOSPID)));
    numMissingID = sum(which(yearData$DSHOSPID == ""));
    numHospitals;
}
missingHospitalIDs <- function (yearData) {
    numMissingID = sum(which(yearData$DSHOSPID == ""));
    numMissingID;
}

How many admissions per hospital from 2010 to 2017?

totalAdmissionsPerHospital <- function (yearData) {
    totalAdmissions <- as.data.frame(table(yearData$DSHOSPID));
    totalAdmissions <-totalAdmissions[order(-totalAdmissions$Freq), ];
    colnames(totalAdmissions) = c("DSHOSPID", "totalAdmissions");
    totalAdmissions;
}

Identify unique patients/hospital from 2010 to 2017.

uniquePatients <- function (yearData) {
    yearData$DSHOSPID = as.factor(yearData$DSHOSPID);
    patients <- select(yearData, DSHOSPID, VisitLink);
    colnames(patients) = cbind("DSHOSPID", "VisitLink");
    numPatients = data.frame(matrix(ncol = 2, nrow = 0));
    colnames(numPatients) = c("DSHOSPID", "numUnique");
    numPatients[1, 1] = 21001;
    numPatients[1, 2] = 1;

    for (i in c(1:uniqueHospitals(yearData))){
        hospitalPatients <- patients[which(patients$DSHOSPID == as.data.frame(levels(yearData$DSHOSPID))[i,1]), ]
        hospitalPatients$VisitLink = as.factor(hospitalPatients$VisitLink)
        numPatients = rbind(numPatients, c(as.data.frame(levels(yearData$DSHOSPID))[i,1], nrow(as.data.frame(unique(hospitalPatients$VisitLink)))))
    }
    numPatients = numPatients[-1, ]
}

Table of # of patients with # of admissions.

admissionNumPerPatient <- function (yearData) {
    admissionPerPatient <- as.data.frame(table(yearData$VisitLink));
    colnames(admissionPerPatient) = c("VisitLink", "Frequency");
    admissionPerPatient <- as.data.frame(table(admissionPerPatient$Frequency));
    colnames(admissionPerPatient) = c("numAdmission", "numPatients");
    admissionPerPatient;
}

210007: University of Maryland Saint Joseph Medical Center (now 210063)
210054: MedStar Southern Maryland Hospital (now 210062)
213300: Mount Washington Pediatric Hospital (gaped the year 2013)

  Hospitals in 2012 but not in 2013
1                            210007
2                            210054
3                            213300

  Hospitals in 2012 but not in 2014
1                            210007
2                            210054

How many admissions per hospital from 2010 to 2017?

   DSHOSPID totalAdmissionsAllYears
8    210009                  358963
4    210004                  289280
2    210002                  264834
19   210023                  253251
11   210012                  202131
13   210015                  196566

Identify unique patients/hospital from 2010 to 2017.

   DSHOSPID totalUniquePatientsPerHospital
8    210009                         188790
4    210004                         157173
19   210023                         133843
2    210002                         132401
13   210015                         105734
11   210012                         101270