Learning Goal: I’m working on a data engineering multi-part question and need su

Learning Goal: I’m working on a data engineering multi-part question and need support to help me learn.Introduction San Diego Medical System (SDMS) is currently considering implementing a computerized physician order entry (CPOE) system for its inpatient acute nursing units. Presently, the physician (MD) writes the patient orders (could be several per one patient chart) by hand on a paper order form. This form is that placed in an “in-box” for clerks, who then enter it into the computer, from where the order is sent either to the Lab for various tests to be performed on the patient (e.g. blood tests), to Radiology for various X-rays to be performed, or to the Pharmacy for various drugs to be prescribed and administered. Presently, the orders are printed out at the other end and processed appropriately. Obviously there can be considerable delays in the above process. With CPOE, the physician would enter the orders directly onto a PC (or perhaps a smartphone, which then is connected to PC and the data uploaded, similar to what you will do in time study). By implementing CPOE, it is believed that there could be four major effects: 1) an increase in physician productivity (i.e. decrease the time the physician spends writing up patient orders) 2) decrease unit clerk staffing requirements 3) a decrease the delay times currently involved for the clerk to enter the hand-written data into the computer and for orders to be printed out 4) a decrease in medical errors (see how big the problem is in To Err is Human or the searchable full text) Your assignment is to analyze physician order times and lab/radiology/pharmacy unit wait times to determine the potential benefits of implementing CPOE for San Diego Medical Systems. An ISYE intern has collected such data (using time study) in the Excel file (CPOEdata). There are three columns: 1) Column 1 = MD Time (sec) is the time it took for the physician to process one patient chart which could have several orders 2) Column 2 = Number of orders per patient chart. There are a total of 116 patient charts or physician data points (but 507 total orders, see Col. 3) 3) Column 3 = Order Wait Time (min) is the wait time for each of 507 unit orders Procedure 1) Calculate the average physician time per chart and per order. Consider a distribution of the data and, perhaps, some outliers can be eliminated. 2) Calculate average wait time per order. Again consider a distribution of the data. Perhaps outliers (lost order!) can be eliminated. In your report: 1) Include your results in tables and/or graphs. 2) Address each of the three main expected goals: a) Can a change in physician productivity be expected, i.e., is it reasonable to expect the physician to enter the data more quickly using a keyboard, than by hand? Use your best judgment here. b) How much of a decrease in wait times can be expected? c) Perhaps most importantly (but perhaps most difficult), could there be reduction in medical errors? Consider all the possibilities for errors under the old paper and clerk data entry system vs. CPOE. Is there still potential for error in CPOE? You may need to check the Web for data on medical errors. It may be difficult to find, since health systems are probably reluctant to admit to such problems. 3) Finally consider costs issues. How much would such a CPOE system cost? (You may need to search the Web. A rough value is around $2,000,000 for both software and hardware) However, there would be a decrease in the need for data input clerks. Consider that SDMS had 24 clerks paid $15.00/hr, which eventually could be reduced by roughly 50% once the system is fully functioning and most physicians utilizing it. How long a payback period would there be?If you incorporate an interest rate, 8% will be fine.
Requirements: As long as needed in Excel

Leave a comment

Your email address will not be published. Required fields are marked *