Research Question: To what extent does gender influence length of hospital stay for MI patients? And the pair of articles that I’m considering is the following: 1- Gender-specific characteristics of individuals with depressive symptoms and coronary heart disease. By Doering, L. et al.

Final Submission of Final Project Part II: Statistical Report

Demonstrate your mastery of the following course outcomes:
· Perform basic, context-appropriate statistical calculations and hypothesis testing in accurately analyzing biostatistical data.
· Interpret key biostatistical metrics, methods, and data for addressing population-based health problems.
· Communicate biostatistical results, procedures, and analysis to other health professionals and the general public for informing their decisions related to population-based health problems.

Research Question: To what extent does gender influence length of hospital stay for MI patients? And the pair of articles that I’m considering is the following:

1. Gender-specific characteristics of individuals with depressive symptoms and coronary heart disease. ; By Doering, L. et al.
2. Gender Differences in Self-Reported Symptoms of Depression among Patients with Acute Coronary Syndrome. By Lorraine Frazier et al, 2012.

In relating this article to the course readings, this particular article was a cross-sectional observational study. ; It is a cross-sectional observational because the researchers did control the values of the variables. This study was purposely made to studying or examining the gender differences, gender specific characteristics of individuals with depressive symptoms of Coronary Heart Disease, Acute Coronary Syndrome, and the prevalence of self-reported depressive symptoms and somatic. ; This study is cross-sectional with a convenience sample of 789 adults including 248 females and 541 males hospitalized with ACS. ;
With regards to the CHD particularly, study shows that, women with CHD and depressive symptoms have fewer resources, greater anxiety, and lower perceived control than men. In women, targeting modifiable factors, such as anxiety and perceived control, is warranted. ; The purpose of this study was to identify socio demographic, clinical, and psycho behavioral characteristics that distinguish men from women with both condition. According to certain researchers, depression that follows acute MI is certainly related with these two events cardiac and non cardiac mortality.
According to our textbook, “biostatistics is the discipline concerned the treatment and analysis of numerical data derived from biological, biomedical, and health-related studies”. ; And these articles have shown all the inclusive about biostatics. The numerical data of the number of men and women in the study plays significant role in biostatistics.
The reason why I picked these articles over the other ones, because I want to learn more these acute disease, how they are relating to biostatistics, and what their differences are. ;
In reading these articles, they give a bright idea to how I should investigate about a disease and the conclusion I should bring fourth after the investigation. ; Because, for any epidemiological study about a disease, the investigator should know about where, what, and when the patient gets this particular disease. ; For example, if I should an investigation about food poisoning on my patient, I should ask what restaurant he/she ate, where, and when it was happening.

Week 3

;A- ; ; ; ; ;State the overall health question you have been asked to address in your own words. Be sure you capture the key elements of the question, using language that a non-technical audience can understand.

B- ;Describe the data set using summary statistics.

; – Identify any limitations you think the data has.

– Identify the variables you will use in your study.

A: ; ; ;Here Age and the Length of hospital are the two variables. There might be some relationship exists between these two variables or not. If the relationship exists between these variables, we can find the intensity of the relationship.

Here the Length of hospital stay of MI patients is one variable which assumes depends on the variable ‘age’. So Length of hospital stay is a dependent on the age of the patients.

So here you need to measure the intensity of the relationship, which is the extent the age influence the length of hospital stay.

First take the data set of Age of patients and Length of stay of MI patients. First find the means of the variables of the variables Age and Length of stay.

Next calculate the standard deviations of Age and Length of stay. Also find the combined variance between the variables Age and Length of stay.

The intensity of the correlation is defined as the ratio to combined variance and to the product of their individual standard deviations. That is correlation coefficient is

=covariance of age, length of stay/(sd of age * sd of length of stay)

; ; ; ; ; ; ; ; ; ; ; The limitations of the study: Here our study confines how extent the age influences on the length of stay. But in actual the length of stay for MI patients depends upon many factors like the type of treatment, the severity of the disease, age, sex, drug usage etc.

But we limited the study to find the relationship between length and age of patients.

The variables used in this study are Áge of MI patients’ and ‘Length of stay of MI patients’.

Week 5

This section should highlight the major findings of each article you selected for supervisor and peers. Specially address the following:

1. What are the findings of each articles and what implications do they have individually and collectively for solving the health question? Support your answer with specific example from your field.
2. Explain how key biostatistical calculations and methods support the conclusions in each article. Cite relevant information from the articles that support your answer.

Both studies were involved men and women from different stages of life. For the Acute Coronary Syndrome (ACS) the study is made of 789 adults’ samples including 248 women and 541 men patients from the age group ranged 28 to 96 years. ; For the Coronary Heart Disease (CHD) this study consists of 1951 hospitalized patients 691 women and 260 men. Among those 691 women studied 26.9% of them were diagnosed without depressive symptoms.
Collectively, these diseases are the remaining leading cause of mortality, morbidity in the US and around the world. But for the ACS alone study showed in 2008 that 616,000 deaths. ; Every year nearly 785,000 American suffer a first heart attack and approximately 470,000 will suffer additional MI. ; And women have less perceived control over their health then men. ; Also, a single study about CHD patients has showed that men and women had similar depression scores, but men reported higher personal control than women.

TABLE OF COMPARISON

BETWEEN ACUTE CORONARY SYNDROME 9ACS)AND CORONARY HEART SYMPTOMS (CHD).

VARIABLES ACS	MEN	WOMEN	VARIABLES CHD	MEN	WOMEN
N=789	541	248	N=1951	260	691
;	68.56%	31.43%	;	13.32%	35.41%

The key calculations and methods support the conclusions in each article by getting the p-value less than alpha value. So, we reject the null hypothesis that there is no any significant difference between the average age for the male and female patients without depressive symptoms.

This means we conclude that there is sufficient evidence that there is a significant difference between the average age for the male and female patients without depressive symptoms are addressed in women with CHD. ;

Week 7

What does your evaluation of strengths and weaknesses of the articles you selected suggest for future research in this field?

The research study in the first article is based on the large sample so results for this study would be more reliable for the future use. But it is essential to check whether the given sample is selected by using the proper random sampling method or not. Also, it is important to check whether the given random sample is homogeneous for entire population or not. For the first article the selected random sample is not homogeneous and this sample does not represent the entire population properly because this sample consists of most of the Caucasians with a high number of males. So, we evaluate that the given analysis is good for the prediction of the Caucasians with a high number of males but it would not be applicable for overall population under study. Also for the second research study the sample consist of most of Caucasian and thus we could not generalize the results of this study for other ethnic people. Also, from the overall study in the second article it is observed that all significant factors not included in the study and therefore results would be biased. But however all the results for this research study are very useful for the future research study only based on the Caucasian population. So with accordance to future use of the given research studies, it is important to use the proper techniques of random sampling and the selected sample would represent the entire population under study. Also, all significant factors responsible for the research variables would be include in the study. All types of measurement and instrumental errors would be avoided in future for getting more unbiased results.

From the given two research articles, it is observed that these two articles reached the different conclusions. The first article concludes that there is a significant difference between the male and females with the symptoms of the depression based on the responses from the selected patients, however the second article concludes that the significant difference between the male and females with socio-demographic, clinical, and psycho-behavioral characteristics. Also, from the given research study it is observed that the age of the patient is responsible for the length of the hospital stay for the MI patients. A positive considerable linear relationship or association exists between the age of the MI patients and length of hospital stay for MI patients. This means as the age of the patients’ increases, the length of hospital stay also increases.

Uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu

Created a table listing the tests you are going to complete to investigate your health question. In Milestone Four, you will actually complete these calculations. Specifically, you must address these critical elements:
A. Graphs: In this section, you will use graphical displays to examine the data and formulate an initial hypothesis. In particular, you should:
1. Create key graphical displays that give a sense of potential relationships between variables. Include the graphs and discuss why you selected these graphical displays as opposed to others.

Here, we want to check whether there is any significant difference in the average BDI II females and BDI II males or not. For checking this we use the descriptive statistics and bar charts for comparison purpose. The descriptive statistics is summarized as below:

Variable	%	Mean	SD
Caucasian males	64%	61.35	11.56
BDI II Females	7.66%	11.89	9.68
BDI II Males	2.22%	9	7.93

For comparing the average BDI II for females and males, the bar chart is given as below:

From this bar chart it is observed that the average BDI II for females is greater than the average BDI II for males. We have to check whether this difference is significant or not. For checking this significant difference we have to use the one sample t test for the population means. The t test is explained in the next topic.

B. Conduct appropriate hypothesis tests, simple regressions, and other tests to analyse the data set.

Here, we want to check the hypothesis whether the there is any statistically significant difference in the average BDI II females and males or not. For checking this significant difference we have to use the two sample t test for the population mean assuming equal population variances. The null and alternative hypothesis for this test is given as below:

Null hypothesis: H₀: There is no ay significant difference in the average BDI II females and males.

Alternative hypothesis: H_a: There is a significant difference in the average BDI II females and males.

We assume the 5% level of significance for checking this hypothesis. The test statistic formula for this test is given as below:

Test statistic = t = (X1bar – X2bar) / sqrt[(S1^2/N1)+(S2^2/N2)]

We are given, X1bar = 11.89, X2bar = 9, S1 = 9.68, S2 = 7.93

We are given N = 789, N2 = 789*0.64 = 505 approximately and N1 = 789 – 505 = 284,

df = 283+504 = 787

Test statistic = t = (11.89 – 9) / sqrt [(9.68^2/284) +(7.93^2/505)]

Test statistic = t = 2.89 / 0.6379 = 4.5305

Critical value = 1.9630, P-value = 0.00 < Alpha value = 0.05

So, we reject the null hypothesis that there is no ay significant difference in the average BDI II females and males. This means we conclude that there is sufficient evidence that there is a significant difference in the average BDI II females and males.

C. Explain why these tests are the best choice in this context and how they compare with established best practices?
; ; ; ; ; ; ; ; ; ; ; Here, we have to see why the tests used in the given articles are best choice and how they compare with the established best practices. As we see in the previous discussion that the first article is based on the cross sectional observations study with a convenience sample of 789 adults. Test retest reliability was studied for reducing the error terms in the depression inventory. Pearson chi square test is used for finding the independence between the two categorical variables. We know that we use chi square test for independence between the two categorical variables. Also, t tests for independent samples for continuous variables are used for finding or checking the significant difference between the demographic characteristic between the males and females. This t test is best in this context because the given two samples are independent and we do not know any information about the population parameters regarding the demographic characteristic. So, it is better to use the t test instead of z test in this context. The Wilcoxon rank sum test is used for the purpose of comparison of the depression ordinal scores and symptoms by gender. We use this test in this context because we are given the level of measurement of variables as the ordinal level and parameter is not defined. For the second article, the t test for interval data and chi square test for ordinal data is used for significant difference in depressive symptoms between men and women. Also logistic regression model with backward elimination is used for the prediction of the depressive symptoms. We use the logistic regression model because the variables included this model are binary in nature and we know that we use logistic regression model in case of binary variables.

References

· Gerstman, B. B. (2015). ;Basic biostatistics: Statistics for public health practice ;(2nd ed.). Burlington, MA: Jones & Bartlett Learning.

· Lock, R. H., Lock, P. F., Morgan, K. L, Lock, E. F., & Lock, D. F. (2012).

Statistics: Unlocking The Power of Data.

Hoboken, NJ: John Wiley & Sons, Inc.

· Dale, Maria. T.G.; Bakketeig, Levi.S, & Magnus, Per. (2016). Alcohol

consumption among first-time mothers and the risk of preterm birth: a cohort

study. Annals of Epidemiology , 26(4), 275-282.

http://www.sciencedirect.com.ezproxy.snhu.edu/science/article/pii/S10472797150041160

· Gender Differences in Self-Reported Symptoms of Depression among Patients with Acute Coronary Syndrome (Lorraine Frazier et al, 2012) .

Retrieved November 4, 2016 from: https://www.hindawi.com/journals/nrp/2012/109251/

·Gender-specific characteristics of individuals with depressive symptoms and coronary heart disease. (Doering, L. et al. 2011). Retrieved December 5, 2016 from: https://www.ncbi.nlm.nih.gov/pubmed/20561880