Application of Analysis

FIND A SOLUTION AT Academic Writers Bay

By Elissa Torres
Application of Analysis
Essential Questions
What are the essential elements in evaluating prior research?
How does the analysis of quantitative versus qualitative studies differ?
How are results communicated from data collection and analysis?
Introduction
The use of statistics and statistical analysis is part of the clinical practitioner’s role. This may appear
in different ways from reviewing existing clinical research to participating in a study. There are
some critical questions when understanding statistics and the role of clinician in health care:
Why is it important to keep up-to-date on clinical research?
Why is it important for health care facilities to conduct ongoing studies?
What type of studies are important?
Previous chapters focused on understanding the elements of statistics and research, including how
to select and conduct hypothesis testing based upon the type of data collected. This chapter
focuses on the application of prior information to understand information written in prior research
studies and set up statistical tests and interpret results both statistically and clinically.
Academic Research Study Extraction
In the evaluation of research articles, it is important that key areas can be identified for
interpretation and understanding. In the review of both qualitative and quantitative research, it can
be daunting to extract the relevant information to determine the primary goals and outcomes of the
study. For clinical studies, this also means addressing the epidemiology.
The simplest way to extract relevant information is to first start with those key areas.
1. Topic: What is the broad topic research area/title?
2. Problem statement: What is the problem that the research is attempting to address? In many
studies, authors identify a lack of research in a specific area or population.
3. Purpose statement: Why did the author complete the study? In some studies, this often
appears in a sentence containing the phrase, “the focus of this study … ”
4. Research questions: What specific questions does the author need to address? In many
articles, this is not explicitly written but can be derived.
5. Hypothesis, variables, or phenomena: What are the variables the author has identified to
address the research goal (quantitative)? How is the phenomena described that the author
seeks to better understand (qualitative)?
6. Sample and location: What was the sample used, and where did the study take place?
7. Methodology: Was the research quantitative or qualitative? Did the author provide any more
details, such as quantitative correlational or qualitative case study?
8. Data collection: How did the author approach data collection? For example, did the author
use surveys, interviews, or clinical studies?
9. Data analysis: What approach did the author use to analyze the data? Did the author
mention statistical tests? What type of statistical data was provided? What type of
information is provided with qualitative studies?
10. Results: What were the results of the study? Did the author find anything significant? Did the
study address epidemiology?
These 10 questions for article evaluation are useful to perform a quick review of the study’s key
elements; however, it is important to start the process by first reading the full article. The format in
which information is displayed in Table 5.1 can be used as a template to organize information found
for each of these article elements. In some studies, information can be easily located in the abstract
and in clearly organized sections; however, this is not always the case.
Table 5.1
Quantitative Article Evaluation
Article Citation
Aljohani, A. H., Alrubyyi, M. A., Alharbi, A. B., Alomair, A. M.,
Alomair, A. A., Aldossari, N. A., & … Tallab, O. M. (2018).
The relation between diabetes type II and anemia. The
Egyptian Journal of Hospital Medicine, 70(4), 526.
doi:10.12816/0043795
Point
Description
Broad Topic Area/Title
The Relation Between Diabetes Type II and Anemia
Problem Statement
“There is consequently a need for more studies on the
incidence and prevalence of anemia among patients with
diabetes mainly those with renal malfunction” (p. 527).
Purpose Statement
“This study consequently purposed to determine the
pervasiveness of anemia due to renal insu[ciency among
patients with type 2 diabetes” (p. 526, 527).
Research Questions
Is there a relationship between patients with anemia and
patients with type II diabetes?
De^ne Variables/
Categorical variable: Gender
Hypotheses
Continuous variables: Age, Hb, Ferritin, MCV, TIBC, FBG,
Erythroietin, eGFR, Urea, Na, K, CA, and HbA1c
(found on pages 528 and 529)
Sample
50 participants
Case group: 25 participants with diabetes (8 male/17
female)
Control group: 25 participants without diabetes (7 male/18
female (p. 528)
Methodology
Quantitative, case-control study (p. 527)
How was Data Collected?
Medical records for the patients were examined from
physical examinations (p. 528)
How was Data Analyzed?
SPSS; descriptive statistics for categorical; summary
statistics, independent t-test; and ANOVA test; Pearson
correlation for Hb and HG for both male and female (p.
528)
What Were the Results?
The study indicated the following were statistically
signi^cant (low p-values) between the case group and
control group.
Hb Male and Hb Female
Ferritin Male and Ferritin Female
MCV
TIBC
Of the biochemical parameters, the following were
signi^cant:
FBG, Erthropoietin, eGFR, Urea, K, C1, Ca, HbA1c
Creatinine was not signi^cant
In the correlation test, HB and HG (female) was signi^cant,
but
HB and HG (male) was not signi^cant.
(pp. 528-529)
Clinical implications:
The study did ^nd a higher occurrence of anemia in
patients with diabetes (87.5% males, 82.3% female). The
study also concluded that the presence of anemia may
increase the likelihood of poorly controlled diabetes (p.
529).
Check for Understanding
1. Would there be any additional evaluation of the article?
2. Did the researchers appear to follow ethical guidelines?
3. What were the assumptions and limitations of the study?
Table 5.2
Qualitative Article Evaluation
Article Citation
Jangland, E., Nyberg, B., & Yngman-Uhlin, P. (2017). It’s a
matter of patient safety: Understanding challenges in
everyday clinical practice for achieving good care on the
surgical ward – a qualitative study. Scandinavian Journal
of Caring Sciences, 31(2), 323-331.
doi:10.1111/scs.12350
Point
Description
Broad Topic Area/Title
Identify the challenges and barriers linked to quality care
and patient safety in the surgical ward.
Problem Statement
“Identify the challenges and barriers linked to quality of
care and patient safety in the surgical ward” (p. 324). Study
addresses gap where there were only a few studies that
looked at both the nurses’ and leaders’ perspective.
Purpose Statement
“The aim of this study was to explore, from the
perspectives of care leaders, the situations and processes
that support or hinder good and safe care on the surgical
ward” (p. 324).
Research Questions
What are the perspectives of leaders on the processes that
support good quality care in the surgical ward?
What are the perspectives of leaders on processes that
hinder good quality care in the surgical ward?
How do leaders’ experiences inform improvement in
clinical practice?
Describe Phenomena
Categorical variable: Gender
The two evaluations above provide a roadmap for reviewing prior research. Much of the research
completed in the clinical setting may not be as comprehensive; however, it is important to
understand the process. In a clinical setting, there may be opportunities to reduce cycle time,
increase quality, or participate in studies that influence health outcomes. Understanding the
process, knowing how to evaluate the data, and communicating the results enables contribution to
the organization.
Application of Statistics to Scenario
A medical office has noticed an increase in patient dissatisfaction and as well as an increase in
usage of urgent care facility services rather than seeing their primary care physicians (PCPs). To
increase understanding of the patient perception, the office surveyed the patients and received 81
Continuous variables: Age, Hb, Ferritin, MCV, TIBC, FBG,
Erythroietin, eGFR, Urea, Na, K, CA, and HbA1c
(found on pages 528 and 529)
Sample
“10 leaders in surgery departments (four department
leaders and six nursing managers) from 1 university
hospital and 2 county hospitals in different regions in
Sweden” (pp. 324-325).
Methodology
Qualitative-descriptive design
How was Data Collected?
Repeated renective interviews using semistructured
interviews
How was Data Analyzed?
Systematic text condensation
What Were the Results?
Study identi^ed four major themes (pp. 326-328):
1. Constant demands for increased e[ciency and
production
2. Continual nursing turnover and loss of competence
3. A traditional hierarchical culture
4. Vague goals and responsibilities in the development
of nursing care
Clinical implications:
Based upon the study, which has limitations as it was
performed in one country (Sweden), organizational
changes are required to ensure higher levels of
competence of staff and resources available to surgical
ward nurses to ensure higher quality care (p. 330).
responses. The survey includes a total of eight questions. The first five questions capture
satisfaction and urgent care utilization responses, and the last three questions capture data on
education, gender, and age group.
Q1: You meet with your Primary Care Physician greater than one time per year. Responses
Strongly Disagree to Strongly Agree.
Q2: You spend more than 10 minutes with your Primary Care Physician discussing health
concerns. Responses Strongly Disagree to Strongly Agree.
Q3: You are more likely to go to urgent care versus your Primary Care Physician. Responses
Strongly Disagree to Strongly Agree.
Q4: What is the number of times you went to urgent care in the past 12 months? Numerical
response requested.
Q5: Rate your overall satisfaction with the medical office. Responses Strongly Disagree to
Strongly Agree.
Q6: What is the highest level of education you completed?
Q7: What is your gender?
Q8: What is your age?
To review the responses from the data collected in the scenario, click on the button below.
Table 5.3
Patient Dissatisfaction Application Scenario
Point
Description
Broad Topic Area/Title
Understand the relationship between patient satisfaction
and usage of services at urgent care facilities.
Problem Statement
Recent indicator identi^ed lower patient satisfaction and
higher incidence of using services at urgent care facilities.
There is a need to understand the perception of patient
satisfaction for the XYZ medical o[ce and decrease
usage of urgent care.
Research Questions
What is the patient perception of satisfaction with the
medical o[ce?
Do patients use urgent care as an alternative to the
primary care physician (PCP)?
Is there a relationship between patient satisfaction and
usage of urgent care facilities?
Hypothesis
H10: There is no relationship between the perception for
number of visits and perception of time spent with PCP.
H1A: There is a relationship between the perception for
number of visits and perception of time spent with PCP.
H20: There is no relationship between the perception for
number of visits and the likelihood to go to urgent care.
H2A: There is a relationship between the perception for
number of visits and the likelihood to go to urgent care.
H30: There is no relationship between the perception for
number of visits and the overall satisfaction.
H3A: There is a relationship between the perception for
number of visits and the overall satisfaction.
H40: There is no relationship between the perception time
spent with PCP and likelihood to go to urgent care.
H4A: There is a relationship between the perception of
time spent with PCP and likelihood to go to urgent care.
H50: There is no relationship between the perception of
time spent with PCP and overall satisfaction.
H5A: There is a relationship between the perception of
time spent with PCP and overall satisfaction.
H50: There is no relationship between the number of visits
to urgent care in past 12 months and overall satisfaction.
H5A: There is no relationship between the number of times
went to urgent care in past 12 months and overall
satisfaction.
Describe Phenomena
(qualitative) or De^ne
Variables/ Hypotheses
(quantitative)
Nominal: education, gender, age group
Ordinal: Survey Questions 1-3 and 5
Continuous: Survey Question 4: Number of visits to urgent
care in last 12 months
Sample
80 patients from XYZ medical o[ce
How is Data Being
Collected?
Sent electronic survey to 300 patients, and received 80
responses.
How Will Data be Analyzed
Descriptive statistics
Correlation analysis
Communicating Results
The data can be sorted for communication based upon summary and descriptive statistics for some
of the variables prior to the hypothesis tests. As an example, to describe the sample respondents by
age group and gender, the data can be converted in Excel to percentages (see Table 5.4). These
percentages can be written out or included in a table.
Even though the responses to the survey questions were ordinal as they were translated from
Strongly Disagree (1) to Strongly Agree (5), with larger samples, responses can be treated as
continuous. Frequently, the three most common forms of descriptive statistics are displayed in a
chart. These include the mean, median, and standard deviation (see Table 5.5).
What Were the Results?
Statistical relationships were identi^ed. The null
hypothesis would be rejected and the alternative
hypothesis would be accepted in all cases.
From a practical perspective, while the results indicated
higher scores for the likelihood to go to urgent care versus
the PCP, the actual descriptive statistics for urgent care
visits do not support this.
Table 5.4
Converting Frequency to Percentage Example
Age
Group
Female
Percent Female
Male
Percent Male
Total
Percent Total by Age Group
< 20
9
18.0%
2
6.7%
11
13.8%
20-25
7
14.0%
4
13.3%
11
13.8%
23-31
10
20.0%
5
16.7%
15
18.8%
32-37
8
16.0%
4
13.3%
12
15.0%
38-43
6
12.0%
4
13.3%
10
12.5%
> 44
10
20.0%
11
36.7%
21
26.3%
Total
50
30
80
Table 5.5
Example of Descriptive Statistics
Beyond addressing some information on descriptive statistics, the hypothesis tests need to be
addressed. Prior to conducting statistical testing, the data needs to be assessed for normality.
When assessing for normality, a statistical program, such as SPSS, determines if the data meets
the conditions of a normal distribution. Often, when data is derived from survey data responses
with ranges from strongly disagree to strongly agree, the data is not normally distributed unless the
samples are very large. In this case, the sample received was only 80. Table 5.6 displays the
normality tests for the variables that will be tested. Because the sample size is lower, the ShapiroWilk results should be used. The Kolmogorov-Smirnov test is most applicable for samples of more
than 2,000 data points. Based upon a 0.05 level of significance, a researcher would reject the null
hypothesis, which stated that the data was normally distributed.
Question
n
M
Mdn
SD
Q1
80
1.93
2.00
1.11
Q2
80
2.15
2.00
1.29
Q3
80
3.31
4.00
1.41
Q4
80
1.41
1.00
1.37
Q5
80
3.13
3.00
1.31
Table 5.6
Test for Normality
Because the test results identified that the data was not normally distributed, a nonparametric test
would be used to conduct the hypothesis testing for correlation. The correlation test to use in this
scenario is the Spearman Rho test. If the data was normally distributed, the commonly used
Pearson Product Moment test would be used. Table 5.7 demonstrates the SPSS output for the
Spearman Rho correlation test between survey Questions 1 and 2. Correlation coefficients are
reviewed on a scale of -1 to +1. The relationship is stronger if the calculated coefficient is closer to
either -1 or +1. In this case, there is a strong relationship between meeting with the PCP more than
one time per year and spending more than 10 minutes with the PCP discussing health concerns.
Another statistic to review in the output is the p value. If the p-value is less than the level of
significance identified in the study, the null hypothesis would be rejected and the alternative
hypothesis would be accepted.
Tests of Normality
Kolmogoroz-Smirnova
Statistic df Sig.
Shapiro-Wilk
Statistic df Sig.
Q1
.247
80
.000
.771
80
.000
Q2
.250
80
.000
.810
80
.000
Q3
.237
80
.000
.866
80
.000
Q4
.256
80
.000
.801
80
.000
Q5
.211
80
.000
.895
80
.000
a. Lilliefors Signi^cance Correction
Table 5.7
Test for Correlation Q1&Q2
Correlation coefficients are reviewed on a scale of -1 to +1. The relationship is stronger if the
calculated coefficient is closer to either -1 or +1. If the correlation coefficient is positive, then the
two variables are moved upward in the same direction. If the statistic is negative, then one variable
increases as the other variable results decrease (Levine, Krehbiel, Berenson, 2013). In this case,
there is a strong relationship between meeting with PCP more than one time per year and spending
more than 10 minutes with the PCP discussing health concerns. Another statistic to review in the
output is the p-value. If the p-value is less than the level of significance identified in the study, the
null hypothesis would be rejected and the alternative hypothesis would be accepted. Table 5.8
displays the remaining correlation coefficients depicted in the table as r and the corresponding pvalues for the test.
Spearman’s rho Q1 Correlation Coe[cient 1.000
Sig. (2-tailed) .
N 80
.777**
.000
80
Q2 Correlation Coe[cient .777**
Sig. (2-tailed) .000
N 80
1.000
.
80
Table 5.8
Correlation tests from Example
Variable
n
r’s
p-value
Q1&Q2
80
.777
.000*
Q1&Q3
80
.566
.000*
Q1&Q5
80
-.313
.005*
Q2&Q3
80
.419
.000*
Q2&Q5
80
-.348
.002*
Q4&Q5
80
-.212
.060*
Table 5.8 demonstrates that there is a statistical correlation between all variables tested at a 0.05
level significance except Q4 (number of times visited urgent care in the last 12 months) and Q5
(overall satisfaction with the medical office). The data output requires analysis to the original
hypothesis questions in the study.
Reflective Summary
This chapter reviewed the application of statistics to research, how to identify data, select the
appropriate tests, and apply this to data sets. The chapter also explored how to review articles or
studies for the key elements for understanding. This understanding was further applied to a
practical scenario including analysis of data collected. The statistical and practical analysis of results
for communication are essential in the roles of a clinician and the tools learned in this course
provided the framework for increased understanding.
Key Terms
Hypothesis: A testable statement of a relationship; an epidemiologic hypothesis is the relationship
is between the exposure (person, time, and/or place) and the occurrence of a disease or condition.
M: Table notation for statistical mean of data array.
Mdn: Table notation for statistical median of data array.
N: Table notation representing the sample size.
P values: The probability that there is enough evidence to make conclusions resulting from the data
collected in the study.
r: Table notation representing the coefficient of correlation.
SD: Table notation representing the standard deviation of the data array.
Variable: A data item such as characteristics, numbers, properties, or quantities that can be
measured or counted. The value of the data item can vary or be manipulated from one entity to
another. There are three different types of variables—dependent, independent, and extraneous.
References
Aljohani, A. H., Alrubyyi, M. A., Alharbi, A. B., Alomair, A. M., Alomair, A. A., Aldossari, N. A., & …
Tallab, O. M. (2018). The relation between diabetes type II and anemia. The Egyptian Journal of
Hospital Medicine, 70(4), 526. doi:10.12816/0043795
Levine, D. M., Krehbiel, T. C., & Berenson, M. L. (2013). Business statistics: A first course (6th ed.).
Upper Saddle River, NJ: Pearson.
Jangland, E., Nyberg, B., & Yngman-Uhlin, P. (2017). It’s a matter of patient safety: Understanding
challenges in everyday clinical practice for achieving good care on the surgical ward – a
qualitative study. Scandinavian Journal of Caring Sciences, 31(2), 323-331.
doi:10.1111/scs.12350
Copyright © Grand Canyon University 2018
Powered by | Academic Web Services

READ ALSO...   Ozhouse Clean -Policy and Procedure
Order from Academic Writers Bay
Best Custom Essay Writing Services

QUALITY: 100% ORIGINAL PAPERNO PLAGIARISM – CUSTOM PAPER