ITECH1103: Data Analysis Report on Healthcare Dataset, Semester 2

This report, a group assignment for ITECH1103, utilizes SAS Visual Analytics to analyze a healthcare dataset, aiming to uncover trends, patterns, and insights. The analysis focuses on the "READMIT-HISTORICAL" dataset, exploring various factors such as ICU days, diagnosis groups, departmental patient numbers, patient demographics (race, gender), and admission trends. The report includes detailed visualizations like line graphs, horizontal graphs, and bar graphs to represent the data. Key findings reveal correlations between diagnoses and ICU stays, the popularity of certain diseases (e.g., congestive heart failure), and departmental patient distributions. The report also provides justifications for the visualization methods and discusses the implications of the findings, including recommendations for improving healthcare practices and resource allocation. The report concludes with an executive summary and a reflection on the project, highlighting the importance of big data analysis in healthcare decision-making.

ITECH 1103: BIG DATA
AND ANALYTICS
(REPORT)
Group Assignment
By
(Name of Student)
(Institutional Affiliation)
(Date of Submission)
Table of Contents

Task 1: Background Information.............................................................................................................2
Task 2: Reporting/Dashboards.................................................................................................................3
Task 3: Additional Visualization............................................................................................................12
Task 4: Justification................................................................................................................................14
Task 5: Discussion of Findings...............................................................................................................14
Task 6: Executive Summary...................................................................................................................15
Task 7: The Reflection............................................................................................................................16
References................................................................................................................................................16
List of Figures
Figure 1: ICU days and Diagnosis
Figure 2: Diagnosis and Region
Figure 3: CHF most and least popular disease
Figure 4: AMI most and least popular disease
Figure 5: COPD most and least popular disease
Figure 6:Departments and number of patients
Figure 7: Patient number and gender
Figure 8: Proportion of patients discharged to home
Figure 9: Race and Region on Patients number
Figure 10: Number Visist by Hopsital

Figure 11: Patient number and admission date
Figure 12: City by patient number
Figure 13: Admission date by gender
Figure 14: Trends of Patients admission date
Figure 15: Geo Map of hospital and patients number
Figure 16: Trend of patient numbers among CHF
Figure 17: Trend of diagnosis group by months
Figure 18: Patient number, operation counts, trend by months

Task 1: Background Information
Big dataset analysis in SAS visual analytics helps in ensuring that timely and accurate
results are made available to improve the quality of data demand use. That is why analysis is
important and is recommended for every health organization that has a given focus, this helps in
improving productivity which in turn brings efficiency and affectivity and also valuable health
insights are achieved. With this, the organizations reputation can be upheld and thus business
firms and health organizations can compete fairly and improve their performance and operations.
In the assignment, "READMIT-HISTORICAL" dataset from teradatauniversitynetwork.com is
taken into consideration.
In decision making the SAS visualization analytics has been used as it makes the process
much easier. Moreover, it is advantageous in so many ways among them helping in
communicating the general trends observed and also aids in the discovery of any patterns hidden.
The main reason why the health industry of America is the focus is because it has become
expensive and Common Diseases and availability of Physicians and also Development and
updating of strategic health plans on certain issues (Halpin, Paprocki & McDonough, 2019).
Therefore the key to evaluating the quality of services provided by hospitals and health industries
at large lies with the use of Big Data analysis. For instance, between June 2011 to July 2012, the
data is huge hence using the SAS visual analytics in visualizing such data makes it easier when it
comes to interpretation. In addition, it helps to predict the future of U.S health Services which
makes it easy when it comes to decision making and planning for healthcare services.
Because the health department deals with people’s lives in the country (U.S), there must
be a continuous operational improvement. Usually, there are a wide range of benefits that data

visualization comes with among them includes having a glance to changes exposed within a
timely manner. (Yang, Li, & Zhang, 2018). For instance, through data visualization one is able to
manipulate dashboards in order to discover hidden factors that may be affecting the performance
of an organization/ institution or company. Therefore, Data visualization in comparison to static
charts as it has more additional interactions in which case static charts do not have. With Static
charts you can only use that which has been presented. Thus many people prefer to use data
visualization to static charts because of its interactive nature that can change many analytics
general perceptions (Hepworth, & Canon, 2018).
Task 2: Reporting/Dashboards
ICU Days and Diagnosis Group
First, based on the analysis in regard to the average number of ICU days with respect to diagnose
group and gender it was found that the Acute Myocardial Infection had the highest ICU days by
diagnosis. On the other, Congestive Heart Failure was highly diagnosed as showed in the
histograms below.
Figure 1: ICU days by Diagnosis group

Figure 2: Frequency of Diagnosis Group
In regard to this, the number of ICU days among the male participants in the acute myocardial
infection should be lowered. In addition to this, medical training should be enhanced in which
medical practitioners who are specialized in congestive heart failure should respond to the
patients’ requirements.
There is need of benchmarking in other locations which has low levels of congestive heart failure
such as location 1 to come up with lessons learnt while adhering the cases in the locations with
high levels of congestive heart failure.
Popularity of the diseases and diagnosis groups
Under this, the analysis shows that heart congestive heart failure is the most popular with an
average percentage of 92.2% while the least popular ones was disorders of fluid electrolyte with
0.1 percent.
Figure 3: AMI most and least popular disease

According to further analysis of the acute myocardial infection, pneumonia organism unspecified
recorded the highest popularity of about 81 % while bronchopneumonia recorded the least
popularity of about 0.24 percent. The figure 4 below shows the display of the findings.
Figure 4: CHF most and least popular disease
Figure 5: COPD most and least popular disease

The analysis about chronic obstructive pulmonary disease showed that chronic bronchitis is the
leading group as far as popularity of disease is concerned. In addition to this, many doctors
specialized in pneumonia organism unspecified and the chronic bronchitis should adequately
adhere to the patients resulting illness conditions should undergo training.
Department and the Number of Patients
In regard to the departmental analysis it was found that the department of heart has numerous
clients amounting to 770 patients in total as compared to the other departments. The analysis
shows that neuroscience had the least number of patients consisting of about 10 individuals as
seen in the same period. Due to this, there is need to reduce the number of doctors from the
neuroscience department and addition of more medical doctors into the department of health.

Number of patients by race, hospital, gender and discharge
Based on the analysis in regard to patient race, hospital, discharge and gender, following findings
were obtained;
There are highest number of female in region 11 compared to region 3 which had the lowest
number of female patients.
Figure: Patient number by Gender (Female)

Secondly, it was found that an average of about 65% of patients get discharged to homes after
treatment.
Figure 6: Frequency of discharge to homes after treatment
The highest number of asthma patients came from Hospital 35 while the least number came from
hospital 28.
Figure: Number of Visits by Hospitals by Asthma Patients

Therefore in order to encounter these, the number of patients being discharged to home should be
increased through ensuring proper and better attention during the time of hospitalisation. In
addition to this, the number of doctors working in hospital 35 should be increased because they
attend to many asthmatic patients.
Figure 7: Frequency of Region and Race
Number of Patients, Admission and City
The analysis here mounted to the following
March was found to have the highest admission while October had the highest admission through
January 2012.
Because of this therefore, there is need to raise the number of medical doctors that are being
deployed at the city of Delray Beach since they attend to many patients as compared to other
locations and regions.
Figure: Patient number by date of admission

Based in the city, Delray Beach recorded the highest number of patient as compared to other
cities.
Figure: Frequency of City