Data analysis has a place in the configuration of healthcare today. In this module we will explore the benchmarking process, a commonly used method for collecting and comparing both public and private data. Daily organizations, both internal and external, collect and report healthcare data. Actually, external organizations such as insurance companies, mandate the sharing of private healthcare information. Threats to personal privacy through data mining occur frequently. Thus, privacy of healthcare data is a commonly expressed concern.
Answer the following questions 1-15 utilizing Microsoft Excel to display your answers.
Use this CHF data for you in an Microsoft Excel file for this project.
Descriptive Statistics Using Excel – Central Tendency
1. Calculate the mean, median, and mode: LOS and Total Charge.
2. Sort the charge values in ascending order.
3. Calculate the mean, median, and mode of the Total Charge excluding the maximum value.
4. Did the mean, median, and mode change more? Why?
5. Repeat that exercise for LOS.
6. What do you think is the ‘best’ statistic to describe the ‘typical’:
LOS or Total Charge
Admission source
Descriptive Statistics Using Excel – Spread
7. Calculate the sample variance and range for LOS and Total Charge.
8. Note: Use the var.s function to calculate the variance. This function calculates the sample variance.
9. Sort the charge values in ascending order.
10. Calculate the variance and range charge excluding the maximum value.
11. Did the variance and range mean or median change more? Why?
12. Repeat that exercise for LOS.
One Sample T-Test (two-tailed)
13. Test the hypothesis that the sample mean LOS for MS-DRG 291 is different from the national standard. Obtain the national Medicare standard for LOS from the information provided below. **Be sure to state the null and alternative hypotheses and set the alpha prior to performing the test.
Total hospital days for HF are estimated at 6.5 million annually. Although the average length of stay has decreased over the last decade to 6.3 days, the 30-day readmission rate has increased to 20% and is roughly 50% at 6 months .The average MS-DRG (291, 292, and 293) reimbursement is $5,759 for the acute care facility.
Two Sample T-Test (two-tailed)
14. Test the hypothesis that charges for patients with MS-DRG 291 are different for men and women. Assume equal variances.
**Be sure to state the null and alternative hypotheses and set the alpha prior to performing the test.
ANOVA
The basic principle of the IPPS payment system is that the ‘levels’ (no CC/CC/MCC) of a MS-DRG family has different resource intensity and therefore should be paid differently. Use the CHF data to test this hypothesis. Use charge as a measure of resource intensity. Be sure to state the null and alternative hypotheses and set the alpha prior to performing the test.
15. Can you conclude which MS-DRGs have significantly different average charges?
