Download presentation
Presentation is loading. Please wait.
1
Comparison of Rates Question 1
Tiffany Woods George Mason University Instructor: Farrokh Alemi, Ph.D. HAP 725
2
The Problem We are asked to construct a control chart showing the rate of prophylactic antibiotic overuse over time at Southeast Alabama Medical Center. Our data will come from the 2015 & 2016 datasets which contain the most recent recordings of antibiotic overuse.
3
Step 1: Download the data
Go to Under the “2016 Annual Files” section, open the .zip file Hospital_Revised_FlatFiles_
4
You will see a .csv file called Timely and Effective Care – Hospital.
*This file contains the data on antibiotic overuse. Save this file as an Excel Workbook (.xlsx) & rename it according to its date ( ) Save as “ ”
5
Complete these steps for the following 9 .zip files:
2016 Annual Files Hospital_Revised_FlatFiles_ ( ) HOSArchive_Revised_FlatFiles_ zip ( ) HOSArchive_Revised_FlatFiles_ zip ( ) 2015 Annual Files HOSArchive_Revised_FlatFiles_ zip ( ) HOSArchive_Revised_FlatFiles_ zip ( ) HOSArchive_Revised_FlatFiles_ zip ( ) HOSArchive_Revised_Flatfiles_ zip ( ) HOSArchive_Revised_Flatfiles_ zip ( ) HOSArchive_Revised_Flatfiles_ zip ( )
6
End the end, you should have 9 .xlsx files saved:
7
Step 2: Import the 9 .xlsx files into Microsoft SQL Server
Use ‘Microsoft Excel’ as your data source In the end, you should have something similar to this: *You may want to rename each dbo according to its date, as such.
8
Step 3: Query the 9 tables This code selects the antibiotic overuse data (Measure ID = SCIP_INF_3) for Southeast Alabama Medical Center (Provider ID = 10001) from each table. Then, it compiles this data into a singular output using UNION. SQL Code for Comparison of Rates, Question 1 SELECT [Provider ID], [Measure ID], CAST([Measure Start Date] as DATE) as [Measure Start Date], CAST([Measure End Date] as DATE) as [Measure End Date], [Score], [Sample] FROM dbo.[ ] WHERE [Provider ID] ='10001' AND [Measure ID] = 'SCIP_INF_3' UNION FROM dbo.[ ] FROM dbo.[ ] FROM dbo.[ ] FROM dbo.[ ] FROM dbo.[ ] FROM dbo.[ ] FROM dbo.[ ] FROM dbo.[ ]
9
Query Output - Provider ID: specifies Southeast Alabama Medical Center
- Measure ID: refers to prophylactic antibiotic use - Measure Start Date/End Date: time frame from which data was collected - Score: the # of patients with antibiotic overuse - Sample: all surgical patients with no evidence of prior infection
10
Step 4: Transfer query results to Excel
Highlight the entire output, right click & select “Copy with Headers” Open Excel & paste into a new spreadsheet
11
Step 5: Calculate rates & control limits
To calculate: Midway Point: ( [Measure Start Date] + [Measure End Data] ) / 2 *Remember to convert this result into a date format Rate of Overuse: [Score] / [Sample] Grand Rate: SUM (Score) / SUM (Sample) Upper Limit: Grand Rate *SQRT((Grand Rate*(1-Grand Rate))/Sample) Lower Limit: Grand Rate *SQRT((Grand Rate*(1-Grand Rate))/Sample)
12
Step 6: Plot the control chart
Limits spread wider due to a smaller sample size Analysis: There was a statistically significant increase in the rate of antibiotic overuse for both the 4/1/15 and the 5/17/15 time frames. All other time frames had rates that were within our control limits.
13
Feel free to contact me if you have any questions!
Thank You Feel free to contact me if you have any questions!
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.