You are on page 1of 14

Auriell Frederick

General Statistics 2315-2


Case Study: Hospital K
03/28/2013

General Statistics 2315-2 Spring 2013 Case Study Hospital K


Case Study Hospital K
To fulfill the educational requirements at Our Lady of The Lake College the student
enrolled in the Pre-Professional Human Medicine Undergraduate Program shall successfully
complete the mathematical course General Statistics 2315, Instructed by Barbara Napoli.
A component of this course will focus on the practical application of concepts derived
from the course. The student will demonstrate her knowledge acquired from General Statistics
2315 by effectively drawing logical conclusions based on reason and evidence. The student shall
use concepts acquired from the course to investigate a sample of data provided by the Instructor.
The student shall demonstrate his or her knowledge in a competent manner that illustrates her
comprehension via written narrative. The narrative shall describe a sample of data (n) procured
from Hospital K for one calendar year. The student will arrive at logical conclusions from
Hospital Ks information by generating a compilation of data including the synthesis of visual
representations of the information, including, but not limited to, charts, mathematical formulas,
graphs, tables, and verbal descriptions. This paper shall accurately analyze statistical trends of
the Hospitals data, drawing inferences from the Hospitals events for the year the data was
analyzed.
This exercise shall serve as an excellent opportunity to develop and improve the efficacy
of the students analytical, logical, general reasoning, and critical thinking skills and her ability
communicate them to a diverse audience in an articulate manner. In addition to the practical
development of described skills, it shall aid the student in her preparation to develop the ability
to make informed decisions based on scientific reason and their accompanying statistical trends.
The long-term learning objective of this assignment is to better prepare the student in her

General Statistics 2315-2 Spring 2013 Case Study Hospital K


professional and personal endeavors so that she can independently identify, analyze, and predict
a potential range of outcomes, prevent and mitigate possible damages, and develop business and
life strategies for optimal results.
Statistics studies the instruction of gathering and compiling data to organize and interpret
it numerically. Statistics is the science of uncertainty while utilizing the extraction of
mathematical data to and analyzing it. 1 Statistics also consolidates detailed data into simplified
format for easy interpretation. Descriptive and Inferential statistics, a branch of mathematics that
focuses on the organization, analysis, and interpretation of a group of numbers (Aron, Aron, and
Coups, 2006, p.2). Inferential statistics are used to draw educational conclusions and inferences
based on numbers from research, but psychologists take the information further than the numbers
(Aron, Aron, and Coups, 2006).
Statistical visual representations can include bar graphs. They are appropriate for
qualitative or quantitative data. Pareto graphs are used to illustrate the decreasing (from high to
low) frequency of a data set. Pie graphs, or circle graphs are used for small categories, typically
fewer than ten, visually displaying data in a circular fashion. Histograms are complex graphs
indicating the frequency of a data value which may fall into a particular class, the uniformity or
asymmetry of data.
The statistical sample data sets will represent three categories of patients visiting Hospital
Ks Emergency Room Department. Hospital Ks data can be divided into two broad categories:
sample data or population data. Population data (N) is from all items and or persons of interest,

P.4, Brase and Brase, 2012,Understading Basic Statistics

General Statistics 2315-2 Spring 2013 Case Study Hospital K


whereas sample data (n) is concerned with particular items or persons of interest.2 Data provided
in the case study represents n, as the information only provides a particular (unknown) year, a
portion of the Hospitals patient information, and describes patients seen in the Emergency
Room Department (ED). To provide N data the case study would be conclusive providing the
data in full with all persons and events of interest. n shall represent three categories of persons
seen in the ED: those admitted to Hospital K (AD), those seeing a physician at Hospital K(EP),
and those arriving by ambulatory services (AMB) to Hospital Ks ED.
When describing an aspect of a population numerically it can be defined as a population
parameter, whereas when describing an aspect of a sample it is sample statistic.3
There are two different ways to describe statistical numerical data. Qualitative data
describes an individual by categorically placing them into one group. 4If data was provided about
a patient such as their social security number, hospital record number, or name and it would be
qualitative data. Quantitative data provides a numerical measurement or value for which
arithmetic can be applied yielding a logical result. Because qualitative data is descriptive in
nature it cannot be summed even if the data piece represents a numeral. Examples of quantitative
data include patients height, weight, basal body temperature, etc. Note these values can be
summed and a statistical mean can be found. The mean of n (x-bar) is the sum () of an entire
data set of information, letting x represent any value in the set, divided by the number of values

and is visually represented as such:5

P.5, Brase and Brase, 2012,Understading Basic Statistics


P. 5, Brase and Brase, 2012,Understading Basic Statistics
4
P. 5, Brase and Brase, 2012,Understading Basic Statistics
5
P. 85, Brase and Brase, 2012,Understading Basic Statistics
3

General Statistics 2315-2 Spring 2013 Case Study Hospital K


Data can also be described to determine the level of measurement. Statistical levels of
measurement can be divided into four levels: nominal, ordinal, interval, and ratio. Nominal data
allows for its placing data into categories. This is the lowest form of data. This is ascribed to
individual names such as patient names, race, gender, language, and biological background.
Often this information serves as the primary patient identification which is used to screen
patients coming into EDs. The second is ordinal data which can range from the lowest level of
measurement to the most detailed and highest order of measurement.

Ordinal level data

differs from nominal data because there is an ordering scheme present for ordinal data.
(Sequois). An example of ordinal level data is certain financial information from patients at
Hospital K. This data is helpful by effectively allowing the Hospitals database to place patients
salary ranges into broad ranges for purposes such as billing. Data of patients earned salaries can
range from below poverty to, low, middle, or high providing a platform to determine medically
indigent individuals. These ranges do not provide the dollar or amounts earned. Poverty level
earners wages are lower than middle level earners but an exact amount is incalculable when
comparing only these ranges.
Interval data level is meaningful and its differences can be measured. (Sequois). An example of
interval level data is the initial temperature of a patient triaged at Hospital Ks ED. It can be
measured and compared to the time he or she was admitted. There is, however, no beginning
point or zero from which the numerical value will initially proceed; therefore the ratios
calculated between those two temperatures are meaningless.
Below is an example of interval data as represented by a bar graph.

P. 8, Brase and Brase, 2012,Understading Basic Statistics

General Statistics 2315-2 Spring 2013 Case Study Hospital K

Total Number of Patients Seen At


Hospital K's ED By Month
3500
3000
2500
2000
1500
1000
500
0

December

November

October

September

August

July

June

May

April

March

February

Janurary

Total of Patients being


seen in the ED By Month

Source: Hospital K
Year: Unspecified

Number of Patients Visiting the ED Seen By an EP or AD to Hospital


Month

Seen By An
EP

AD to
Hospital

January
February
March
April
May
June
July
August
September
October
November
December

2323
2012
2285
2328
2460
2551
2729
2735
2541
2586
2341
2332

372
304
402
338
343
370
372
370
328
403
391
396

Total Number of Patients Yearly

29,233

4,389

Source: Hospital K
Year: Unspecified

General Statistics 2315-2 Spring 2013 Case Study Hospital K


Meaningful comparisons of data values are represented in ratio level data. Ratio level
data is the highest order of data. It provides information based on the meaningful and measurable
differences in the data pieces. Examples of ratio level data are the Hospitals employees ages,
the lab work results of a patients hemoglobin sedimentation rate, the exact amount of salaries
that patients and employees earn, as well as their heights and body weights. In July 549 patients
were seen in the ED arriving by AMB whereas in April there were just 417. The difference is a
calculable 132 occasions of individuals arriving by AMB.

Central measures of tendency include the mean of quantitative data that can be described
by analyzing the three categories of patients visiting the ED. The mean of those seeing an EP is
29,223.00. The mean of those AD is 4,389.00 Finally those arriving via AMB is 5,0668.00. The
mean of patients seen in the ED for this year is 2695.75.
The mode is the number that occurs with the most frequency in a data set.7 The only
category in which a mode appeared was in AD. The mode is present twice in the AD category.
372.00 patients visited the ED and were admitted, which occurred in both January and July. The
mode reappeared in June and August at 370.00 patients visiting the ED . August has the highest
total of of individuals seeing an EP out of all the twelve months in the year. Perhaps its because
August was a busy travel month that particular year and the citys population increased, thereby
increasing the number of visits to the ED. One might speculate that perhaps there is a correlation
between more school-aged children home for summer vacation playing at home and outdoors
thus an increased risk of accidental injury may present which could be reflected in the month of
August which contains the highest monthly totals of patients visiting the ED that year.

P. 82, Brase and Brase, 2012,Understading Basic Statistics

General Statistics 2315-2 Spring 2013 Case Study Hospital K


The median of patients visiting the ED in the years sample represents the ordering of
each categorys values proceeding from smallest to largest, finding the middle two values sum,
and finally dividing the sum by two. EP=2,400.00, AD=371.00, AMB=480, with the total
median equaling 2,693.00. 8
The statistical range provides a method for cross reference of all ranges of data. The range is
defined as The difference between the largest and the smallest values of a data distribution. It
is a helpful measurement to begin to understand the distribution of information but does not
allow us to measure the values of the difference. i9 The hospital data ranges (by category) are EP
= 723.00, AD = 99.00, AMB = 200.00, equaling a total from all three categories of 687.00
patients seen at the Hospital ED for the year.
Standard deviation is a measure that is used to measure the varying risk. s is the measure of
deviation or risk. The sample standard deviation is used to describe a range of data about the
mean (x-bar). 10 Each of the three categories data was measured and a standard deviation
determined. This was performed by finding the value of a data piece (x) and finding the mean
(X-bar) and then squaring the sum of x minus x-bar. The sample standard deviation of each
category visiting Hospital Ks for the duration of one calendar year, as calculated by monthly
data are the following: patients seeing EP was 206.58, those patients arriving via AMB had a
deviation of 62.98, and the patients AD was 31.47. The deviation of patients seeing an
Emergency Physician was the highest of all three categories. Emergency Department physicians
are limited. Typically they are supported by Nurse Practitioners (NP) or Physician Assistants

P. 83, Brase and Brase, 2012,Understading Basic Statistics


P.94 Brase and Brase, 2012,Understading Basic Statistics
10
P.95 Brase and Brase, 2012,Understading Basic Statistics
9

General Statistics 2315-2 Spring 2013 Case Study Hospital K


(PA). The significant increase in the sample standard deviation may likely be explained by the
theory that the patients at the ED did not see an EP but a PA or NP instead.
Correlation and regression are two statistical topics which are used to make predictions based
on relationship(s), which are the relationships between selected values of X and observed values
of Y. A scatter diagram is a graph that uses an ordered pair X and Y, which are plotted
horizontally and vertically on a grid.11 This diagram visually represents the relationship between
X and Y based on their axis, respectively. Below is a scatter diagram representing the total
number of patients seen in Hospital Ks ED and those that were admitted. Let X equal the
number of patients admitted and Y equal the total of patients seen in the ED.

11

P.132 Brase and Brase, 2012,Understading Basic Statistics

General Statistics 2315-2 Spring 2013 Case Study Hospital K

Total Number of Patients Seen By Calendar Month During One Year

Patients Admitted to Hospital K After Being Seen in


the ED
3500

3000

2500

2000

1500

1000

500

0
0

50

100

150

200

250

300

350

400

450

Patients Admitted to Hospital K

It can be reasonably inferred that with this scatter diagram a line fits reasonably well. Typically
as the amount of individuals seen in the ED increases, so does the rate of patient admissions, thus
a positive correlation is determined between X and Ys relationship. Below is a table configured
to find the least square line (best fitting line).

General Statistics 2315-2 Spring 2013 Case Study Hospital K

X=AD
Y=Total n seen
372
2636
304
2295
402
2618
338
2550
343
2750
370
2789
372
2956
370
2982
328
2793
403
2814
391
2594
396
2572
SUM: 4389
32349

XY
980592
697680
1052436
861900
943250
1031930
1099632
1103340
916104
1134042
1014254
1018512
11853672

(X)2
(Y)2
138384 6948496
92416 5267025
161604 6853924
114244 6502500
117649 7562500
136900 7778521
138384 8737936
136900 8892324
107584 7800849
162409 7918596
152881 6728836
156816 6615184
1616171 87606691

The formula for the best fitting line the is: Y-hat= a + bx.
a equals the Y intercept and bx the slope. B is found by using the equation SSxy divided by SSx.
a is found by setting a to equal Y-bar - bx.
X-bar =365.75 and Y-bar equals 2695.75.
SSxy= The sum of XY minus the sum of the Xs times the sum of the Ys. SSx equals the
sum of X squared minus the sum of the Xs squared divided by n. b equals 0.0012, whereas a
equals 2695.31. The least squares line formula is Y-hat= 2695.31 + (0.0012)(365.75).

General Statistics 2315-2 Spring 2013 Case Study Hospital K


Patients Seen In the ED

Y-hat = 2.0217x + 1956.3

Patients Seen in ED That Were AD

Source: Hospital K
Year: Unspecified

3500
3000
2500
2000
1500
1000
500
0
0

50

100

150

200

250

300

350

400

450

Patients Admitted

In conclusion, statistics helps to better categorize and understand data of Hospital K. It


can be used to predict and determine data. Inferences made from Hospital Ks sample raw data
allow opportunity to better understand the events of Hospital K, such as the ratio of patients
admitted to those being treated in the ED. This may be useful to help determine the amount of
supplies necessary for patients, the facilities and amount of services required to meet an
increasing need of patients being admitted, or provide a comparison or prediction of budget
based on its history of expenses. Statistics will not accurately predict all events or outcomes
possible, nor will it be conclusive, particularly in this instance, as this is a sample of data, not a
population.

General Statistics 2315-2 Spring 2013 Case Study Hospital K


Works Cited
Aron, A., Aron, E. N., & Coups, E. J. (2006), Statistics for psychology (4th ed. ).
Braise, C.H., and Braise, P.B. (2012), Understanding Basic Statistics (4th ed.).
Sequois, C. o. (n.d.). Levels Of Measurement. Retrieved March 20, 2013, from
http://infinity.cos.edu/faculty/woodbury/stats/tutorial/Data_Levels.htm

General Statistics 2315-2 Spring 2013 Case Study Hospital K