Final Milestone

3

Data Analysis Milestone Two

Deanna Buchanan

1

SNHU

Data Key Features

The evaluation is being carried out on data from the Worcester Heart Attack Study (WHAS100) conducted by the Department of Cardiology at the University of Massachusetts Medical School (Hosmer, Lemeshow & May, 2008). The appendix table shows that the dataset has 100 observations and 9 variables. The data was collected over a 9-year period starting in 1995 and focused on patients who suffered from myocardial infarction and were admitted to Worcester hospitals. The analysis will only use two variables - gender and length of stay (LOS). Gender is a categorical variable, with males coded as 0 and females coded as 1. LOS is a continuous variable with a mean of 6.32 and a standard deviation of 3.34. For LOS, the mean value is more indicative of the whole data set for men than for females since the variation and standard deviation are lower.

Limitations

The evident limitation of the data set is that the number of male observations (65) is approximately twice the number of female observations (35). This may introduce bias into the results, since it no longer accurately represents either male or female patients hospitalized due to myocardial infarction. Another drawback is that potential confounding factors such racial/ethnicity, treatment/medication disparities, insurance status, and preexisting/co-morbid illnesses were not included. These confounding variables might distort the analysis and findings of the statistical correlation between gender and hospital stay duration in patients with myocardial infarction (Gerstam, 2015).

References

Gerstman, B. B. (2015). Basic biostatistics: Statistics for public health practice. Jones & Bartlett Learning.

Hosmer, D. W., Lemeshow, S., & May, S. (2008). Applied survival analysis: Regression modeling of time to event data (2nd Ed.). New York, NY: John Wiley and Sons Inc.

Appendix

Table 1

Summary statistics for los:

Group by: gender

gender	Mean	Variance	Std. dev.	Median	Range	Min	Max	Q1	Q3	Std. err.	n
0	6.3230769	11.159615	3.3406011	5	16	1	17	4	7	0.41435056	65
1	7.8	79.517647	8.9172668	6	53	3	56	4	8	1.5072932	35