University Of Michigan, Ann Arbor Industrial and Operations Engineering April 28, 1999 Project Final Presentation: Adrienne Niner Jessica Guibord Jeff Bieske Program and Operations Analysis Team: Pilot Project CCMU/Pharmacy
Contents Letter of Transmittal 1 Executive Summary 2 Introduction and Background 5 Current Situation 5 Approach and Methodology 6 Analysis 7 Testing for Normality 7 Process Control Charting 7 Process Capability 8 Inspection of Individual Steps 9 Findings and Conclusions 9 Recommendations 10 Action Plan 12 Summary 13 Appendix Contact Information 15 Data Sheet 17 Description of Steps 19 Census 21 Moving Range and X Charts 23 CUSUM Charts 28 Formulas 32 Excel Data 34 Liaisons Notes 44 Flow Chart 55 Average Process Times for Step Intervals 58 Total Process Time Distribution 60 Bibliography 61
28 April 1999 University of Michigan And Operations Engineering, Department of Industrial Adrienne Niner Jessica Guibord Jeff Bieske Sincerely, We would be glad to answer any questions and discuss and suggestions that you may have regarding the project plan. Please feel free to contact us at anytime. (Appendix A is plan. a list of contact information for our team.) Thank you for your time in considering our implementing the plan. The plan details the tasks for a proportion of 1999: from 8 March 1999 to 31 March 1999. The main focus of this project was to analyze STAT orders written on the sixth floor in this time frame. What we have proposed is merely a suggestion. It is up to the Critical Care Medicine Unit to alter the plan to their liking before actually adapting and The attached report is a project pian for the Critical Care Medicine Unit (6D.) Dear Mrs. Wurster, Mrs. Diccion, Mrs. Grondin, Mrs. Kinsey: Kathy Kinsey, Interim Supervisor IPCP Ann Arbor, MI 48104 Heather Wurster, RN Director Patient Care Stephanie Diccion, RN Head Nurse of CCMU University of Michigan Hospital Louise Grondin, RN Clinical Nurse Specialist Ann Arbor, MI 48104 1316 Geddes Ave. #4
3/21/99) EXECUTIVE SUMMARY The 6D Critical Care Medicine Unit (CCMU) at The University of Michigan Medical Center identified a little over a year ago that improvements were needed with regards to the delivery and availability of patient medications. The primary focus of the CCMU is to decrease the cycle time of STAT orders (a STAT order is defined as a drug that must be administered within 30 minutes of the time of which it was originally written). It is the expectation of the CCMU, that by decreasing the cycle time of a STAT order, the staff would be able to provide superior patient care, increase the level of job satisfaction of the staff on the unit, and improve the cost effectiveness of the system by having a higher turnover rate of the unit s beds. On February 2, 1998, a pilot project was implemented where the role of a pharmacy technician was added to the CCMU in response to some of the nursing staffs concerns involving Pharmacy Services. The pharmacy technician was required to make rounds to 6D on the hour to retrieve any orders that had been written and to drop off any orders that had been filled to the ward. If a STAT drug was needed and the pharmacy technician was not present, the technician was to be paged, at which point was required to stop doing whatever they were working on and proceed directly to the unit to pickup order. To evaluate how well this system functioned with respect to STAT drugs and to find a basis of comparison, our Senior Design Project Team collected time data for two weeks (3/8/99 without the pharmacy technician picking up any STAT orders from the CCMU. When the two weeks ended, the pharmacy technician was put back into place and time data continued to be gathered on STAT orders (from 3/22/99 3/31/99). The data collection sheets that were used to obtain time data were broke down into seven discrete steps. The times were recorded on the data collection sheets every time a task was completed by whomever was responsible for executing that assignment. The data was then analyzed and what was found was that of the six intervals between the seven steps, three of these time intervals accounted for 81% of the total process time. The three intervals previously mentioned and their average times were as follows: Interval 1 time for clerk to notice flagged chart (21 minutes), Interval 3 time for order to travel from the CCMU to the 6th floor satellite pharmacy (16 minutes), Interval 6 time for the STAT drug to travel from the pharmacy back to the CCMU and be administered. Overall, the data that was collected yielded an average process time of 1 hour and 12 minutes, which was more than double the time that a STAT drug should take. It was the Senior Design Project Team s goal to find ways to reduce the average times of the three intervals that consume the vast majority of the total cycle time. After thorough analysis of the current situation, possible solutions were discussed amongst the project team and the following short and long term recommendations were made. The average time that it took for the nursing staff to administer a medication once it was placed in the omnicell by the pharmacy technician was 30 minutes. One of the solutions that could be implemented almost immediately to reduce that time is the placing of red 2
3 the order over to the satellite pharmacy located down the hall. It is understood that there the staff that there is something waiting for them. Furthermore, the procedures of all the process as a whole. Is the 30-minute time limit being met? What does staff think of An area should be assigned within the 6D CCMU where STAT orders can be placed. are certain difficulties in hiring new employees, especially in the case of creating a new the changes? These could be some relevant questions asked of the staff, along with a doctor writes the prescription until the time when the clerk notices the patients chart is A lighting mechanism should be placed by the clerk s counter. The lighting system a five-minute time limit on each and every step once the physician has written the is ordered, the unit runner would immediately stop whatever they were doing and deliver waiting to be given and green would signal that there is currently not a drug waiting to be would be used by the doctors to indicate to the clerk that a chart is flagged and a STAT order is ready to be processed. By carrying out this recommendation, it is hoped to reduce the current average cycle time of 21 minutes, which includes the time from when whether or not a drug needs to be picked up. Red would indicate that there is a drug administered. recommended that the 6D CCMU should hire a unit runner that is only responsible for delivering orders and filled prescriptions to and from the pharmacy. When a STAT drug and green signs to every room on the 6D CCMU. The intention of the signs is to indicate flagged. The third major cause of delay is the travel time to the pharmacy. Therefore, it is their unit. The runner would be accountable for making rounds on the unit, and position, however, the benefits in terms of amount of time saved in the process are outstanding. prescription. This would require that all staff involved in the process must keep their Due to the urgent nature of STAT orders, it is not terribly unreasonable to strictly enforce eyes and ears open. If the previous recommendations are to be implemented, this fiveminute time limit will be rather simple to adhere to, as there will be visual cues to alert drug orders must be clearly defined and strictly enforced throughout the unit and The satellite pharmacy should have STAT stickers to be placed on every STAT order. This will indicate to the person receiving the order that it is a STAT drug and that it basis while these changes are still new. The goal of the surveys would be to find out what differences, positive or negative, the changes that were implemented have had on pharmacy. Failure to comply will result in meetings to find the source of delay and to should not be placed in the omnicell, but handled in a different manner. Surveys should be distributed to the nursing staff and pharmacy technicians on a monthly asking them for their opinions and suggestions. This would hopefully eliminate most of the calls to the pharmacy inquiring where the order is. This area is the only area that a STAT order is placed until the nurse is available remedy the problem as soon as possible.
to pick it up. The reason for recommending the STAT order not be placed in the omnicell is due to the fact that it is difficult to see into the omnicell to see if the medication is in there without actually opening it up. Since there are rules against leaving prescription drugs unsecured in the unit, it will be necessary to either keep this area in a secured location behind the clerk s desk, much like how IV s are currently delivered and held until the nurse retrieves them. In addition to the short-term recommendations, a few long-term recommendations were also made and were provided. The primary recommendation that should be implemented in the future is to install a computer system to both Courtside and Riverside that links the CCMU to the satellite pharmacy. If this suggestion were to be implemented, the physician would simply enter the necessary drug into the computer. This would eliminate travel time, which currently accounts for a considerable portion of time in the process. Furthermore, it would also eliminate the importance of the clerk to notice that a chart has been flagged, which again accounted for a substantial portion of the total cycle time from the time data measured. In addition, more incentives should be provided to all employees in an attempt to increase job satisfaction, which in turn will ultimately contribute to a higher level of job performance which translates into a lower cycle time for the administration of STAT drugs. The focus should be looked into, increasing the base pay for pharmacy technicians from the current rate of $9.52 per hour. It was the project team s understanding that there exists a high turnover rate among pharmacy technicians due to the stressful nature of the job combined with a salary scale that does not adequately compensate for the job-related stress. Finally, training seminars should be developed whereby the CCMU staff and pharmacy staff would be trained to use the new computer system, along with other items of interest, such as methods on how to keep the process time within the 30-minute limit. 4
5 Approximately one year ago, it became evident that improvements were needed in Medicine Unit (CCMU), and after midnight it is responsible for the entire hospital. communication between the unit and the pharmacy in an event that a STAT order arrived. through the liaison. The purpose of the liaison was to observe what levels of throughout the entire hospital was to be implemented. recommendations given will hopefully lead to better patient care, minimize delays in the pharmacy staff. Presently the CCMU Pilot Project is still in effect with the pharmacy technician making distinguish a STAT order from a non-stat prescription order. Prior to this, all medication orders were placed together in a pile. The pharmacist or technician would In response to this problem, a pilot project was devised where pharmacy technicians could be from back up within the pharmacy. There is only one pharmacy on the floor arrive to the unit later than the expected time limit. One cause of delay for a STAT order that is responsible for all STAT orders as well as non-stat orders at certain times of the The pharmacy technician would then go to the unit as soon as possible to pick up the resolve problems that could arise unexpectedly. As a result, the pharmacy technicians drug is a medication that needs to be administered to the patient within 30 minutes from regards to the delivery and availability of patient medications. Currently, the problem in would make rounds on the CCMU on an hourly basis to receive the orders. They would were instructed to carry pagers with them at all times. These pagers were used for avoided is that the pharmacy is overloaded. rounds on an hourly basis. One change was implemented; STAT orders are to be placed inside a red envelope located at the clerk s desk. The purpose of the red envelope is to CURRENT SITUATION The objective of this project is to critically evaluate the pilot program initiated in the CCMU and provide recommendations to improve the system in place. The analysis and response time of a STAT order, and increase the level of satisfaction of the nursing and and administered by the required time. Presently, another critical problem that can not be improvement the pilot could achieve. If there were noticeable improvements, expansion the Critical Care Medicine Unit (6D) of the University of Michigan Hospital. the system is that STAT drugs are not meeting the time frame of 30 minutes. A STAT the time the prescription was written by the physician. In most cases the medication will day. The pharmacy is responsible for four units on the including the Critical Care The purpose of this project is to evaluate the Pilot Program that has been implemented in also meet with clerks and nurses to communicate pharmaceutical needs and efficiently order. During the trial and testing period, the only method used to fill STAT orders was A problem that arose was that there was not enough lead-time to get a prescription filled INTRODUCTION AND BACKGROUND
6 There was not an efficient way of identifying a STAT order in the pile from routine which STAT orders occur. Taking this into consideration, it was necessary that the P & patient s name, and the drug ordered. The data sheets also collected times: logbook located on each side of the unit to get any information necessary from the people CCMU unit. Once an order was encountered by the liaison, they finished their rounds, checked by the pharmacist, and delivered the order back to the unit. 2) When the clerk processed the order 3) When and how it was sent to the sixth floor satellite pharmacy 1) Physician wrote the order times themselves. This turned out to be infeasible because of the unpredictability with delays in the process occurred. Our data sheets were set up to collect the date, the The P & OA Team observed the process and separated it into steps where the different OA Team use a method to collect the data without being present. APPROACH AND METHODOLOGY expedite their delivery. satellite pharmacy to the CCMU, the P & OA would shadow the process and collect the orders, which only served to extend the process time. With the red envelopes in place, Originally, the Program and Operations Analysis Team (P& OA Team) proposed that in order to analyze the process of getting the necessary medication from the sixth floor time intervals could be written in (See Appendix B for the data sheet.) The data sheet would circulate along with the prescriptions and medications. From these break downs These data collection sheets were used from March 8, 1999 to March 31, 1999. satellite pharmacy to the CCMU. The liaison would make stops at both the river side and returned to the pharmacy, entered the order into the computer, filled the order, had it court side of the CCMU, communicate with the people present in the unit, and check a This liaison was a pharmacy technician who would circulate between the 6th floor 7) When it was administered to the patient (See Appendix C for descriptions) 5) When the pharmacist checked that the prescription was filled correctly 6) When and how it was sent back to the CCMU the pharmacist or technician would fill the order immediately and send it back to the unit. 4) When the prescription was entered into the computer then enter the orders into the computer. When a STAT order was discovered in the pile, the pharmacist can immediately identify which orders need to be filled without delay and the P & OA Team expected to see which steps took the longest and where the largest they did not make verbal contact with. The liaison made rounds every hour on the hour when possible. The liaison carried a pager to be contacted when they are not in the
The P & OA Team s goal was to collect at least 25 samples before the liaison started and at least 25 samples from when the liaison was running. The data collection sheets were picked up every three days until it was noticed that some of the administered times were missing. It was decided to pick them up everyday in order to look up the administered times on the MAR (Medication Administration Record) and communicate with any staff member, as necessary. The P & OA Team were able to collect 38 complete data sheets without the liaison but were only able to collect 3 with the liaison. When data was collected with the liaison, a census was conducted (See Appendix D). The census revealed that the admissions to the CCMU was lower between March 22 31 then March 8-21. Furthermore, the unit shut down the entire Courtside ward. After talking to the head nurse, Stephanie Diccion, it was confirmed that the low admissions does not occur very often, but rather roughly once a year. This is one reason why there was not enough data to analyze the process with the liaison. With that in mind, the group was unable to look further into the process and give any strong recommendations backed up by data for that procedure. ANALYSIS Testing for Normality Before it could be analyzed whether the process was in control, it was first necessary to determine whether the data was normally disthbuted. The two key factors in determining whether a distribution is normal are the skew and kurtosis of the collected data. A 95% confidence interval was created for both the skew and kurtosis based on the number of observations collected, If the value for the skew fell within ± 0.587 and the value for kurtosis fell within 0.93 kurtosis 1.06, it could be concluded with 95% confidence that the data was normally distributed. The skew of the data was 0.82694 and the kurtosis was 0.27069. Since the value for skew fell just outside of the range, and both values must fall within their respective ranges, it was initially infeasible to conclude that the data was normally distributed. However, several experts in the field of statistical quality control have investigated the effect on departures from normality. Burr (1967) notes that the usual normal theory control limit constants are very robust to the normality assumption and can be employed unless the data is extremely non-normal. Schilling and Nelson (1976) have also studied the effects of non-normality on the control limits of control charts. Their studies concur with those of Burr. It is also important to understand that we had collected only 38 observations. Usually, at least 100 to 200 observations are necessary to accurately determine whether the population is normally distributed. Since the value for kurtosis did fall within the specified range, and the amount that the skew fell outside of its range was negligible, it can be assumed that the data was normally distributed. Proce.rs Control C harting 7
effective in detecting large shifts, but do not signal smaller shifts. Cumulative-sum charts shifts in our analysis of the CCMU s cycle time of STAT orders, both the Shewhart effective in detecting large shifts. Since it was desired to be able to detect both types of moving-range control charts and the standardized cumulative-sum control charts were performed (see Appendices E & F). It was chosen to use the moving range charts instead observations, therefore, x-bar and r charts were not applicable to the data. Instead, it was required to use control charts that were sensitive to the fact that samples of only size one average values. (See appendix G for equations) These are the standard limits used, because 99.7% of the population data would fall within these limits. If a point were to process is out of control and the point came from a shifted distribution. The other charts that are created from these calculations are the x-chart and the MR-chart. The x chart assesses the process stability in location, while the MR-chart assesses the process stability in variation. Both of these charts were inspected and signals were looked for of the more traditional x-bar and r charts because the data collected was individual possibility is that the process is in control and the point is just rare. The two control normally distributed process. No signals were found that would represent that the process was unstable, so it was concluded that there were not any large shifts in process location or variation. (See appendix E) observations. This means that on average, one would expect to see a point outside of the control limits of a process that is in control once in approximately every 370 observations. The value for h is similar to the upper and lower control limits on any control chart. Unless a CUSUM falls outside the ±h range, it can be assumed the process location and the process variation were calculated. These values were plotted and the charts were inspected for signals that the process had gone out of control. Since no signals were found that showed the process had become unstable, it was concluded that In constructing the CUSUM charts, values had to be chosen to use for the parameter h are designed to detect small shifts in both location and variation, but are not very both large and small shifts in location and variation. Shewhart charts are the most were collected. To create the MR-charts, a table of the observations was first constructed and the values for the moving range were calculated. From this table the upper and lower control limits fall outside of these limits, there are two possibilities to consider. The first is that the were determined, which are approximately ±3 standard deviations from the calculated that would indicate that the process is out of control, such as points beyond control limits, runs, and where the points are located relative to where it would be expected to see a variation. (See appendix F) variation, large or small. Since both the Shewhart MR-chart and the standardized CUSUM-chart were inspected, and all of the control charts indicated that the process had It was the intention to be certain the process did not experience any shift in location or the process had not experienced any small shifts in either process location or process is in control. Next, the positive and negative cumulative sums for both the process and parameter k. Values were selected that would give an average run length of 370 In monitoring the process of the cycle time of a STAT order, it was important to detect 8
remained in control, it was concluded that the process had remained stable throughout the entire time it was being monitored. Process Capability Unfortunately, just because it was confirmed that the process was in control, it did not mean that the process met specifications and is a capable process. Since the average total time to complete all stages in the cycle is higher than the upper specification limit of 30 minutes, it can be immediately seen that the majority of the time, the cycle time of a STAT order is above the upper specification limit. In order to make this assumption more concrete; the fallout of the process was calculated (See appendix G for calculations). The fallout was found to be approximately 8 1.327%. This means that over 80% of the time, CCMU is not meeting specifications on the cycle time of a STAT order. Inspection of Individual Steps The CCMU had not set target values or specification limits for the individual steps required to complete a STAT order. However, since the total time to complete a STAT order was frequently not meeting specifications, it was important to look at the areas that may be causing the problem. The CCMU felt that each step should, on average, take about the same amount of time. If the upper specification limit for total time to completion is exactly met, then each step should take no more than five minutes. For each observation, a breakdown of each step was made to see how much time was spent on each step (See Appendix H). For example, the average time to go from stage one to stage two takes about 21 minutes, which leaves a lot of room for improvement. On the other hand, going from stage four to stage five only requires approximately 4 minutes, which would indicate that the step is completed efficiently. FINDINGS AND CONCLUSIONS Overall, 38 completed samples were collected on the process times for STAT orders in the CCMU without the liaison. The average process time of the 38 samples was calculated to be one hour 12 minutes. The average time of the data collected was more than double the time allowed (30-minutes) for STAT drugs according to the unit s definition. Furthermore, of the 38 samples, only seven samples (18.4%) actually met the 30-minute requirement (See Appendix L). Therefore, the project team broke down the data collected into step intervals to see where the delays were occurring (See Appendix K). It was found that three of the six intervals accounted for 81% of the average process time. If any improvements are to be made on the process, these three areas must be targeted. Unfortunately, only three data points were collected when the liaison was in service from March 22 March 31. It was suggested by Louise Grondin to consult the census (See 9
low. Due to this the courtside was shut down in order to conserve resources. the problem. In the interview, she stated that she only received five STAT orders To investigate possible reasons for why there were so few STAT orders, the project team was unusually low. That was in fact the case that the number of patients admitted was Appendix D) of ward admissions on 6D to see if the number of patients in the CCMU interviewed pharmacy technician, Pat Starzec, to see if she could provide any insight on between 3/22/99 and showed the team the liaison s notes (See Appendix I). These notes showed the recording of a STAT order that they picked up. Furthermore, the liaison made note in the logbook that some members of the nursing staff were bypassing paging her. Interesting enough, the P & OA team found that the liaison did not receive paging system to see how long an actual page would take for the pharmacy technician to receive. The P & OA Team found that it takes 45 seconds to receive the page. This length of time does not justify why the liaison was not paged. Pat Starzec also mentioned assigned place orders are misplaced and nurses are not able to find them when they need collected that one of the steps that heavily prevented STAT drugs from meeting their 30- minute time limit was the time from when the doctor wrote the prescription to the time order for a STAT drug, the average time between steps I and 2 could be drastically reduced from the current average time of 18 minutes. receives the order is expected to and responsible for changing the sign to red and placing mechanism at the clerk s counter to indicate to the clerk that the doctor had written an Another area that prevented STAT drugs from meeting their 30-minute time limit was the patient they will then change the sign to green. RECOMMENDATIONS that the pharmacy receives calls daily in regards to nurses looking for their orders. She said that there is not an assigned area for STAT orders. Because of not having an The Senior Design Project Team concluded to offer the following recommendations for the Critical Care Medicine Unit for the short term. It was seen from the data that was when the clerk processed the order. Therefore, it was believed that by placing a lighting has arrived, green meaning that there is nothing waiting. The person (nurse or clerk) that In response to this, red and green signs should be placed on the outside of every room in the liaison altogether and tubing the STAT orders directly to the pharmacy, rather than any pages during the trial period. The P & OA team decided to do a trial test on the to be administered. time from when the order filled until the time it was finally administered to the patient. the unit to indicate when a prescription has been dropped off. Red meaning that the order the drug in the appropriate place. When the nurse is ready to administer the drug to the 3/31/99,
recommended that the CCMU should hire a unit runner that is only responsible for their over to the satellite pharmacy located down the hail. It is understood that there are certain difficulties in hiring new employees, especially in the case of creating a new position, however, the benefits in terms of amount of time saved in the process are orders and filled prescriptions to and from the pharmacy. When a STAT drug is ordered, Due to the urgent nature of STAT orders, it is not unreasonable to strictly enforce a fiveminute time limit on each step once the physician has written the prescription. This The satellite pharmacy should have STAT stickers to be placed on every STAT order. This will indicate to the person receiving the order that it is a STAT drug and that it outstanding. would require that all staff involved in the process must keep their eyes and ears open. If rather simple to adhere to, as there will be visual cues to alert the staff that there is something waiting for them. Furthermore, the procedures of all drug orders must be clearly defined and strictly enforced throughout the unit and pharmacy. Surveys should be distributed to the nursing staff and pharmacy technicians on a monthly basis while these changes are still new. The goal of the surveys would be to find out what differences, positive or negative, the changes that were implemented have had on asking them for their opinions and suggestions. should not be placed in the omnicell, but handled in a different manner. necessary to either keep this area in a secured location behind the clerk s desk, much like CCMU to the satellite pharmacy. If this suggestion were to be implemented, the physician would simply enter the necessary drug into the computer. This would eliminate travel time, which currently accounts for a considerable portion of time in the the previous recommendations are to be implemented, this five-minute time limit will be the pharmacy inquiring where the order is. This area is the only area that a STAT order the omnicell to see if the medication is in there without actually opening it up. Since the future is to install a computer system to both courtside and riverside that links the is placed until the nurse is available to pick it up. The reason for recommending the In addition to the short-term recommendations, a few long-term recommendations were STAT order not be placed in the omnicell is due to the fact that it is difficult to see into An area should be assigned within the CCMU where STAT orders can be placed when the unit runner would immediately stop whatever they were doing and deliver the order unit. The runner would be accountable for making rounds on the unit, and delivering the process as a whole. Is the 30-minute time limit being met? What does staff think of the changes? These could be some relevant questions asked of the staff, along with the nurse is away or with a patient when the pharmacy technician or unit runner comes to the ward to deliver the STAT drug. This would hopefully eliminate most of the calls to how TV s are currently delivered and held until the nurse retrieves them. The third major cause of delay is the travel time to the pharmacy. Therefore, it is there are rules against leaving prescription drugs unsecured in the unit, it will be also made and provided. The primary recommendation that should be implemented in 11
12 performance which translates into a lower cycle time for the administration of STAT current rate of $9.52 per hour. It was the project team s understanding that there exists a combined with a salary scale that does not adequately compensate for the job-related job satisfaction, which in turn will ultimately contribute to a higher level of job In addition, more incentives should be provided to all employees in an attempt to increase process. Furthermore, it would also eliminate the importance of the clerk to notice that a chart has been flagged, which again accounted for a substantial portion of the total cycle time from the time data measured. high turnover rate among pharmacy technicians due to the stressful nature of the job time is about six weeks. staff would be trained to use the new computer system, along with other items of interest, Finally, training seminars should be developed whereby the CCMU staff and pharmacy details go wrong. uniform five-minute time limit on each step in the process, surveys to all staff for feedback on implementations, and the assignment of a designated area to place STAT weeks to implement. Actions that can be implemented in the first week include a enforce the procedures better. The lighting mechanism and a unit runner recommended It is estimated that none of the short-term recommendations should take longer than six ACTION PLAN drugs. The focus should be increasing the base pay for pharmacy technicians from the stress. they were initially undertaken. Finally, the computer system should be implemented and their competitors, analyze the budget to find where the funds will be allocated from, and to perform a study that will estimate the time, money, and lives saved through an introducing training seminars on how to adhere to the 30 minute time limit for STAT implemented with the goal in mind to have in place within six months from the time that within the target of two years. The two years will be used to research relevant systems such as methods on how to keep the process time within the 30-minute limit when minor orders upon arrival. Two weeks should be enough time for STAT stickers to be used by both pharmacy and the unit, for the red and green signs to be in use, and to define and The long-term recommendations include raising the base wage of pharmacy technicians, orders and how to avoid errors, and providing small incentives for the staff complying with the official procedure. The long-term recommendations suggested should be overall improvement in the quality of service that would result from the system.
process time for a STAT order was 1 hour 12 minutes. After collecting 38 samples without the liaison, we were able to conclude that the current process was not meeting their upper specification time limit of 30 minutes. This was due to a number of things. The P & QA Team found that within the Critical Care Medicine Unit, the total average 13 meet their STAT expectation. Collecting data at another point in time with the liaison, data. reason for the unit not meeting their expectations. Only 18.4% of the samples met the Inconsistency among the doctors, nurses, clerks, and pharmacy personnel is a primary 30-minute process time within the unit. In conclusion, changes that will drastically reduce some of the average process times must be implemented in order for the CCMU to when admissions is high, would be able to give the unit a more concrete finding with SUMMARY
Appendix A
734-665-3835 734-769-6349 734-662-1018 Jessica Guibord jessicaa@engin.umich.edu Adrienne Niner bieskej @ engin.umich.edu Contact Information: Jeff Bieske aniner @ engin.umich.edu
=.
CCMU Pharmacy/Nursing Pilot Stat Order Data Collection Sheet 1 Patient Name: Date: Drug: 2 3 Physician Writes Rx: Clerk Processes Rx: Method I] 4to El Pharmacy: El El 5 Rx Entered on C 6 Pharm. Checks Method El 7 to El Unit: El El 8 Rx Admin. To 9 Comments: C = Unit Clerk s Responsibility N = Nurse s Responsibility P = Pharmacy s Responsibility RETURN DATA SHEETS TO ENVELOPES LOCATED AT RIVER I COURT CLERK DESKS.
Appendix C
Description of Steps Step] to 2 Includes: time from when physician writes the prescription order, to when the doctor flags the chart, and the clerk notices that the chart is flagged. Step 2 to 3 Includes: time the clerk separates the RX, puts the white slip in the chart, and sends the slip to the pharmacy by one of three ways (tube, walking it over, or liaison.) Step 3 to 4 Includes: time to travel to the pharmacy, and time it was waiting to be entered into the computer. Step 4 to 5 Includes: time entered into the computer by either a pharmacist or a technician, and time to fill the RX order. Step 5 to 6 Includes: time it takes the pharmacist to check the order that has been filled. Step 6 to 7 Includes: travel time back to the unit, time waiting in the unit to be noticed by a nurse or clerk, and the time it is administered to the patient.
C I I I I I Appendix D I I I I I I I I I
a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) 0 fli Ct 0 0 çt C) 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 t -..... - 0 0 0 0 0 0 0 0 a) (ii )- U) U) U) U) U) U) U) U) U) U) U) U) U) U) U) U) U) U) U) U) U) U) U) U) U) U) U) U) U) - -..--..-.- S. t I i U) U) F ) F ) F ) IN-) F ) F ) F ) F ) F ) F ) f i I i I I s I s ê I- i i C) C) C) (fl C) co CX) I U) I CL) S.) Xn U) F ) C) CD I Ui U) F ) C) CL) CX) (SI U) F ) -S- --S S.5 -S-S -S-S S.. CD CD CO CO CO CO CL) CO CO CD CO CD CO CD CD CO CD CD CD CO CO CD CD CD CO CO CD CO CD CO CD CO CD CD CO CO CO CO CO CO CO CO CO CD CC) CO CO CO CD CD CD CD CO CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CO CO CD CD CD CD CD CO CD,-.- S. S. -S -S S. 0...-.- -.-. a) a) -. a) -. - - -. - - -. 0 a) 5 I- I- 0 0 - a) 0 0 0 CD CO a) J a) Ui a) Ui a) a) I- ) F- Ui J CL) I U) F ) IN-.) F ) CD 01 I- I- F- F a) F ) I- I- a) i i F-.1 F ) f i I- I I i F ) Ui F- a) I s U) I s F ) F ) U) I- I- I- I- - a) 1 (SI
Appendix E
CV 66.15809 2.0 2.5 3.0 median 50.0% 1.0000 N 38.00000 r> 0 0.890416 0.0011 W Prob.cW I Shapiro-Wilk W Test Quantiles) [Momen) cit for NorrnalitJ --3.25 50-0 :90 95 0 a, a, z minimum 0.0% 0.1000 Kurtosis -0.27069 0.5% 0.1000 Skewness 0.82694 quartile 25.0% 0.6450 Sum Weights 38.00000 2.5% 0.1000 Variance 0.63943 10.0% 0.2420 Sum 45.93000 quartde 75.0% 1.5425 Lower 95% Mean 0.94585 99.5% 2.8300 Std Dev 0.79964 97.5% 2.8300 Std Error Mean 0.12972 90.0% 2.7500 Upper 95% Mean 1.47152 maximum 100.0% 2.8300 Mean 1.20868 IF tc*.01.05.10 --1
R W b R I R. I II I I I 00ro0o000-0o)CyIa-.,1 C) I- C i- C 000 000 i1r iri ro P2o OCDO) ooo cxooi ooro oazcn 00)0) -1 (no D) CD1 41: zi 0 z G) xz CD 1 CD CD I! LcDO:.,I 0) 00)0)040)Cfl.P.ø -0 Cu F )
X-Chart 4.00 3.50 UCL = 3.6255 3.00 +2 sigma = 2.8199.2 2.50 +1 sigma =2 1 : E 1 50 CL=1.208 1.00 0.50 1 sigma = 0.4032 0.00 1 CL=0 I I 8 I I I I I I I I I 2 3 4 5 6 7 8 9 1011 12131415161718192021 22232425262728293031 32333435363738 Observation #
MR-Chart UCL = 2.9683 2.5 2 sigma = 2.2817 C 0 +1 sigma= 1.5 2 o 1.5 E I 1 CL=O.9086 0.5-1 sig a 0.2220 LCL = 0.0000 0 I I I 1 2 3 4 5 6 7 8 9 1011121314151617181920212223242526272829303132333435363738 Observation I
J X!pUddV J1
s smnss IR. (0 C3 C C) + z + C) z S. + S a + 2 a C 3, (0 C II Ii II II II U C I 0 -I r -1 20 - m C-) C (1) C
a a a a a a r a a a a a ai 1 Standardized CUSUM for Process Mean 6 U, D C-) N I.. C U) Observation #
W R... 1 1 1 1 K Standardized CUSUM for Process Standard Deviation 6 4 2 Cl) C) a) N 0 b. a) C a) 4-. Cl) -2-4 -6 Observation #
fj X!pUddV
Formulas Upper & Lower Control Limits for MR-Charts: X-Chart: UCL = (x-bar) + (E2 * RbWEm) =(1.21)+(2.66*0.91) = 3.6255 CL LCL =x-bar = 1.21 = (x-bar) - (E2 * R-barm) = (1.21) -(2.66 * 0.91) = -1.2080 =0 MR-Chart: UCL = D4 * R-barm =3.267*0.91 = 2.9683 CL LCL =R-barm = 0.91 = D3 * R-barm =0*0.91 =0 Process Fallout: 1 -ørusl-mul L sigma ] 1-4 (-0.8875) 1 (1 0.81327) =81.327%
Appendix H
_ a Data Collection Sheet Times (broken down into Steps) Without liaison completed sheets Step 1 2 3 4 5 6 7 Physician Writes Rx Clerk Processes Rx Method to Pharmacy Rx Entered on Computer Pharm. Checks FHled Rx Method to Unit Rx Admin. To Patient 1 17:20 17:20 17:25 17:26 17:26 17:26 17:30 2 12:10 13:05 13:05 13:14 13:15 13:15 13:15 3 10:45 10:50 10:55 11:25 11:30 11:43 12:00 4 10:00 10:15 10:20 10:32 10:33 10:33 10:50 5 22:00 23:00 23:00 23:10 23:10 23:10 23:20 6 23:55 2:00 2:00 2:01 2:02 2:02 2:45 7 13:30 13:45 13:48 13:55 13:58 13:58 14:10 8 1:00 1:30 1:30 1:48 1:50 1:50 2:00 9 18:50 19:05 19:05 19:23 19:24 19:24 19:40 10 7:40 7:40 7:43 7:45 7:45 7:45 7:48 11 19:00 19:30 19:33 19:45 19:47 19:47 19:55 12 12:10 12:13 12:15 12:20 12:25 12:40 12:50 13 12:00 12:10 12:10 12:41 12:42 12:42 13:30 14 11:56 11:56 11:58 11:59 12:00 12:00 12:02 15 13:10 13:20 13:20 13:40 13:44 13:45 14:15 16 16:00 16:00 16:00 16:10 16:20 16:21 18:50 17 9:00 9:10 9:10 9:18 9:18 9:18 9:28 18 11:00 11:25 11:25 11:40 11:41 11:41 11:50 19 16:30 16:43 16:45 18:05 18:07 18:08 18:10 20 16:40 16:47 16:49 16:58 17:02 17:03 17:10 21 16:30 17:15 17:15 17:25 17:27 17:28 17:35 22 16:15 16:19 16:19 16:30 16:30 16:30 16:48 23 5:00 5:10 5:22 5:28 5:32 6:15 6:15 24 5:00 5:10 5:22 5:28 5:32 6:15 6:15 25 23:30 23:30 23:40 23:41 23:41 23:41 24:00:00 26 7:45 7:45 7:45 7:50 7:50 7:50 8:00 27 8:20 8:25 8:30 8:37 8:38 8:38 9:05 28 8:00 9:40 9:45 9:50 9:51 9:51 10:07 29 6:45 6:52 6:55 7:24 7:24 7:25 9:30 30 10:45 10:55 11:00 11:22 11:27 11:27 11:40 31 19:05 19:10 19:10 19:25 19:26 19:27 20:30 32 7:00 7:25 7:25 9:13 9:18 9:18 9:25 33 7:55 8:00 8:00 9:14 9:15 9:16 10:00 34 0:05 0:12 0:19 0:35 0:35 0:35 1:05 35 10:30 10:35 10:43 10:55 10:59 11:00 11:04 36 9:30 9:45 9:50 10:00 10:04 10:07 12:15 37 9:00 9:00 9:00 9:30 9:45 9:50 10:00 38 8:00 8:40 8:45 8:55 9:00 10:00 10:35
Average Time / Step: Avg. Process Time 0:18 0:02 0:17 0:02 0:05 0:25 1:12 Time durations between each step (without liaison, completed sheets) Step 1 to Step 2 Step 2 to Step 3 Step 3 to Step 4 Step 4 to Step 5 Step 5 to Step 6 Step 6 to Step 7 1 0:00 0:05 0:01 0:00 0:00 0:04 2 0:55 0:00 0:09 0:01 0:00 0:00 3 0:05 0:05 0:30 0:05 0:13 0:17 4 0:15 0:05 0:12 0:01 0:00 0:17 5 1:00 0:00 0:10 0:00 0:00 0:10 6 2:05 0:00 0:01 0:01 0:00 0:43 7 0:15 0:03 0:07 0:03 0:00 0:12 8 0:30 0:00 0:18 0:02 0:00 0:10 9 0:15 0:00 0:18 0:01 0:00 0:16 10 0:00 0:03 0:02 0:00 0:00 0:03 11 0:30 0:03 0:12 0:02 0:00 0:08 12 0:03 0:02 0:05 0:05 0:15 0:10 13 0:10 0:00 0:31 0:01 0:00 0:48 14 0:00 0:02 0:01 0:01 0:00 0:02 15 0:10 0:00 0:20 0:04 0:01 0:30 16 0:00 0:00 0:10 0:10 0:01 2:29 17 0:10 0:00 0:08 0:00 0:00 0:10 18 0:25 0:00 0:15 0:01 0:00 0:09 19 0:13 0:02 1:20 0:02 0:01 0:02 20 0:07 0:02 0:09 0:04 0:01 0:07 21 0:45 0:00 0:10 0:02 0:01 0:07 22 0:04 0:00 0:11 0:00 0:00 0:18 23 0:10 0:12 0:06 0:04 0:43 0:00 24 0:10 0:12 0:06 0:04 0:43 0:00 25 0:00 0:10 0:01 0:00 0:00 0:19 26 0:00 0:00 0:05 0:00 0:00 0:10 27 0:05 0:05 0:07 0:01 0:00 0:27 28 1:40 0:05 0:05 0:01 0:00 0:16 29 0:07 0:03 0:29 0:00 0:01 2:05 30 0:10 0:05 0:22 0:05 0:00 0:13 31 0:05 0:00 0:15 0:01 0:01 1:03 32 0:25 0:00 1:48 0:05 0:00 0:07 33 0:05 0:00 1:14 0:01 0:01 0:44 34 0:07 0:07 0:16 0:00 0:00 0:30 35 0:05 0:08 0:12 0:04 0:01 0:04 36 0:15 0:05 0:10 0:04 0:03 2:08 37 0:00 0:00 0:30 0:15 0:05 0:10 38 0:40 0:05 0:10 0:05 1:00 0:35
Physician Writes Rx Clerk Processes Rx Method to Pharmacy Rx Entered on Comp Pharm. Checks Rx Method to Unit Rx Admin. To Patient 1 2:50 3:05 3:30 4:30 2 2:50 3:10 3:10 3:30 3:40 3:50 3 9:25 9:40 9:58 10:05 11:00 11:00 4 8:00 8:05 8:07 8:16 8:20 w * Without Iiai icomplete sheets Step 1 2 3 4 5 6 7 5 14:20 14:20 15:24 15:28 15:28 15:40 6 1:30 1:49 2:55 3:00 7 9:00 9:15 9:30 9:35 8 9 12:35 12:50 13:00 13:05 13:25 13:30 10 20:45 21:00 21:14 21:16 21:18 21:21 11 3:48 3:49 3:51 3:55 3:56 3:56 12 14:40 14:55 15:00 15:05 15:15 13 10:30 10:31 10:31 10:50 14 10:30 10:31 10:31 10:50 10:50 15 7:30 7:31 7:32 7:35 7:36 7:36 16 9:15 9:20 9:49 17 19:25 19:26 19:27 19:28 19:32 19:33 18 18:15 18:15 18:20 18:30 18:35 18:38 19 5:00 6:25 6:30 6:35 6:38 6:38 20 5:00 6:25 6:30 6:35 6:38 6:38 21 14:30 15:15 15:15 15:23 15:26 15:30 22 19:25 19:26 19:27 19:28 19:29 19:31 23 17:10 17:10 17:15 17:15 17:20 17:25 24 13:00 13:10 13:13 13:46 13:46 15:00 25 26 11:00 14:00 14:00 27 17:45 17:45 17:54 18:13 18:20 28 13:17 13:20 14:07 14:08 14:20 29 30 31 32 33 34 35 8:00 8:35 8:42 9:10 9:10 9:11 36 2:45 2:48 3:35 3:37 3:37 3:37 37 12:15 12:17 12:17 12:50 12:55 12:55 38 7:00 7:35 8:05 8:13 8:14 8:14 39 8:28 8:28 8:28 8:30 8:30 8:30 40 19:15 19:28 19:29 19:44 19:45 19:46 41 18:00 18:25 18:33 20:30 20:42 20:42 42 43 1:00 1:20 1:20 1:40 1:44 1:45
A A A A A A K K I I MJ 44 1:00 1:30 1:49 1:50 1:50 45 46 0:00 0:30 0:31 0:32 0:35 47 10:00 11:00 11:05 11:25 11:30 12:30 48 10:00 11:00 11:05 11:25 11:30 12:30 49 10:00 11:00 11:05 11:25 11:30 12:00 50 4:25 4:45 5:00 5:20 5:30 8:00 51 4:25 4:45 5:00 5:20 5:30 7:00 52 4:25 4:45 5:00 5:20 5:30 9:00 53 2:45 2:58 3:00 3:00 3:15 54 1:05 1:10 1:40 2:10
Time durations between each step (without liaison, incompleted sheets) Step 1 to Step 2 Step 2 to Step 3 Step 3 to Step 4 Step 4 to Step 5 Step 5 to Step 6 Step 6 to Step 7 1 1/0/00 0:25 2 1/0/00 0:20 1/0/00 0:00 1/0/00 0:20 110/00 0:10 1/0/00 0:10 3 1/0/00 0:15 1/0/00 0:18 1/0/00 0:07 1/0/00 0:55 1/0/00 0:00 4 1/0/00 0:05 1/0/00 0:02 1/0/00 0:04 5 1/0/00 0:00 1/0/00 1:04 1/0/00 0:04 110/00 0:00 1/01000:12 6 1/0/00 0:05 7 1/0/00 0:15 1/0/00 0:05 8 9 1/0/00 0:15 1/0/00 0:10 1/0/00 0:05 1/0/00 0:20 1/0/00 0:05 10 1/0/00 0:15 1/0/00 0:14 1/0/00 0:02 1/0/00 0:02 1/0/00 0:03 11 1/0/00 0:01 1/0/00 0:02 1/0/00 0:04 1/0/00 0:01 1/0/00 0:00 12 1/0/00 0:15 1/0/00 0:05 1/0/00 0:10 13 1/0/000:01 1/0/000:00 1/0/000:19 14 1/0/00 0:01 1/0/00 0:00 1/0/00 0:19 1/0/00 0:00 15 1/0/00 0:01 1/0/00 0:01 1/0/00 0:03 1/0/00 0:01 1/0/00 0:00 16 1/0/00 0:05 17 1/0/00 0:01 1/0/00 0:01 1/0/00 0:01 1/0/00 0:04 1/0/00 0:01 18 1/0/00 0:00 1/0/00 0:05 1/0/00 0:10 1/0/00 0:05 1/0/00 0:03 19 1/0/00 1:25 1/0/000:05 1/0/00 0:05 1/0/000:03 1/0/000:00 20 1/0/00 1:25 1/0/000:05 1/0/000:05 1/0/000:03 1/0/000:00 21 1/0/00 0:45 1/0/00 0:00 1/0/00 0:08 110/00 0:03 1/0/00 0:04 22 1/0/00 0:01 1/0/00 0:01 1/0/00 0:01 1/0/00 0:01 1/0/00 0:02 23 1/0/00 0:00 1/0/00 0:05 1/0/00 0:00 1/0/00 0:05 1/0/00 0:05 24 1/0/00 0:10 1/0/00 0:03 1/0/000:33 1/0/000:00 1/0/00 1:14 25 26 1/0/00 0:00 27 1/0/00 0:00 1/0/00 0:09 1/0/00 0:19 1/0/00 0:07 28 1/0/000:03 1/0/000:01 1/0/000:12 29 30 31 32 33 34 35 1/0/00 0:35 1/0/00 0:07 1/0/00 0:28 1/0/00 0:00 1/0/00 0:01 36 1/0/00 0:03 1/0/00 0:47 1/0/00 0:02 1/0/00 0:00 1/0/00 0:00 37 1/0/00 0:02 1/0/00 0:00 1/0/00 0:33 1/0/00 0:05 1/0/00 0:00 38 1/0/00 0:35 1/0/00 0:30 1/0/00 0:08 1/0/00 0:01 1/0/00 0:00 39 1/0/00 0:00 1/0/00 0:00 1/0/00 0:02 1/0/00 0:00 1/0/00 0:00 40 1/0/000:13 1/0/000:01 1/0/000:15 1/0/000:01 1/0/000:01 41 1/0/00 0:25 1/0/00 0:08 1/0/00 1:57 1/0/00 0:12 1/0/00 0:00 42
Avg. times Avg. total process incomplete: 0:23 0:09 0:15 0:05 0:07 0:35 N/A complete: 0:18 0:02 0:17 0:02 0:05 0:25 1:12 together: 0:21 0:06 0:16 0:04 0:06 0:30 N/A 44 1/0/00 0:30 1/0/00 0:19 1/0/00 0:01 1/0/00 0:00 45 46 1/0/00 0:01 1/0/00 0:01 47 1/0/00 1:00 1/0/000:05 1/0/000:20 1/0/000:05 48 1/0/00 1:00 1/0/00 0:05 1/0/00 0:20 1/0/00 0:05 49 1/0/00 1:00 1/0/00 0:05 1/0/000:20 1/0/000:05 50 1/0/00 0:20 1/0/00 0:15 1/0/00 0:20 1/0/00 0:10 1/0/00 2:30 51 1/0/00 0:20 1/0/000:15 1/0/00 0:20 1/0/000:10 1/0/00 1:30 52 1/0/00 0:20 1/0/00 0:15 1/0/00 0:20 1/0/00 0:10 1/0/003:30 53 1/0/000:13 1/0/000:02 1/0/000:15 54 1/0/00 0:05 1/0/00 0:30 25 39 38 34 28 15 K
- :hout liaison cc ated data sheets (broken down into shifts) )tal 1 2 3 4 5 6 7 Process Physician Writes Rx Clerk Processes Rx Method to Pharmacy Rx Entered on Comp Pharm. Checks Rx Method to Unit Rx Admin. To Patient Times ift 3am-7am 5:00 5:10 5:22 5:28 5:32 6:15 6:15 1:15 2 5:00 5:10 5:22 5:28 5:32 6:15 6:15 1:15 3 6:45 6:52 6:55 7:24 7:24 7:25 9:30 Average= 2:45 1:45 ft 7am-llam 1 7:00 7:25 7:25 9:13 9:18 9:18 9:25 2:25 2 7:40 7:40 7:43 7:45 7:45 7:45 7:48 0:08 3 7:45 7:45 7:45 7:50 7:50 7:50 8:00 0:15 4 7:55 8:00 8:00 9:14 9:15 9:16 10:00 2:05 5 8:00 9:40 9:45 9:50 9:51 9:51 10:07 2:07 6 8:00 8:40 8:45 8:55 9:00 10:00 10:35 2:35 7 8:20 8:25 8:30 8:37 8:38 8:38 9:05 0:45 8 9:00 9:10 9:10 9:18 9:18 9:18 9:28 0:28 9 9:00 9:00 9:00 9:30 9:45 9:50 10:00 1:00 9:30 9:45 9:50 10:00 10:04 10:07 12:15 2:45 Ii 10:00 10:15 10:20 10:32 10:33 10:33 10:50 0:50 2 10:30 10:35 10:43 10:55 10:59 11:00 11:04 0:34 13 10:45 10:50 10:55 11:25 11:30 11:43 12:00 1:15 14 10:45 10:55 11:00 11:22 11:27 11:27 11:40 Average= 0:55 1:17 ift llam-3pm 11:00 11:25 11:25 11:40 11:41 11:41 11:50 0:50 2 11:56 11:56 11:58 11:59 12:00 12:00 12:02 0:06 3 12:00 12:10 12:10 12:41 12:42 12:42 13:30 1:30 12:10 13:05 13:05 13:14 13:15 13:15 13:15 1:05 5 12:10 12:13 12:15 12:20 12:25 12:40 12:50 0:40 6 13:10 13:20 13:20 13:40 13:44 13:45 14:15 1:05 7 13:30 13:45 13:48 13:55 13:58 13:58 14:10 Average= 0:40 0:50 ift 3pm-7pm 16:00 16:00 16:00 16:10 16:20 16:21 18:50 2:50 2 16:15 16:19 16:19 16:30 16:30 16:30 16:48 0:33 3 16:30 16:43 16:45 18:05 18:07 18:08 18:10 1:40 16:30 17:15 17:15 17:25 17:27 17:28 17:35 1:05 16:40 16:47 16:49 16:58 17:02 17:03 17:10 0:30 6 17:20 17:20 17:25 17:26 17:26 17:26 17:30 0:10 7 18:50 19:05 19:05 19:23 19:24 19:24 19:40 0:50 Average= 1:05
Shift 7pm-llpiT 1 Average= 1:13 Average 1:20 19:00 19:30 19:33 19:45 19:47 19:47 19:55 0:55 2 19:05 19:10 19:10 19:25 19:26 19:27 20:30 1:25 3 22:00 23:00 23:00 23:10 23:10 23:10 23:20 1:20 Shift llpm-3am 23:30 23:30 23:40 23:41 23:41 23:41 24:00:00 0:30 23:55 2:00 2:00 2:01 2:02 2:02 2:45 2:50 0:05 0:12 0:19 0:35 0:35 0:35 1:05 1:00 1:00 1:30 1:30 1:48 1:50 1:50 2:00 1:00 Number of Stats Avg. Total Process Time Shift 3am-7am 3 1:45 Shift7am-llam 14 1:17 Shift llam-3pm 7 0:50 Shift3pm-7pm 7 1:05 Shift7pm-llpm 3 1:13 Shift 11 pm-3am 4 1:20
Physician Writes Rx Clerk Processes Rx Method to Pharmacy Rx Entered on Computer Pharm. Checks Rx Method to Unit Rx Admin. To Patient 1 15:00 15:45 16:05 16:08 16:10 16:12 18:32 2 15:00 15:45 16:05 16:08 16:11 16:12 18:00 3 12:30 12:30 12:40 12:47 12:50 12:51 13:15 Step 1 to Step 2 Step 2 to Step 3 Step 3 to Step 4 Step 4 to Step 5 Step 5 to Step 6 Step 6 to Step 7 Total Process Times 1 0:45 0:20 0:03 0:02 0:02 2:20 3:32 2 0:45 0:20 0:03 0:03 0:01 1:48 3:00 3 0:00 0:10 0:07 0:03 0:01 0:24 0:45 AVG 0:30 0:16 0:04 0:02 0:01 1:30 2:25 Data Collection Sheet Times (broken down into Steps) With liaison completed sheets Step 1 2 3 4 5 6 7 Time durations between each step (with liaison, completed sheets) Avg. Total Process Time= 2:25 With liaison incompleted sheets Step 1 2 3 4 5 6 7 Physician Writes Rx Clerk Processes Rx Method to Pharmacy Rx Entered on Computer Pharm. Checks Rx Method to Unit Rx Admin. To Patient 1 14:55 14:55 14:56 14:56 14:57 15:10 2 11:20 11:20 12:00 12:02 12:07 12:07 Time durations between each step (with liaison, incompleted sheets) Step 1 to Step 2 Step 2 to Step 3 Step 3 to Step 4 Step 4to Step 5 Step 5 to Step 6 Step 6 to Step 7 1 0:00 0:01 0:00 0:01 0:13 2 0:00 0:40 0:02 0:05 0:00 AVG 0:00 0:20 0:01 0:02 0:00 0:13 N=1 2 2 2 2 1
Appendix I
I L\ LçT13 ci ô.cq \Do; ) ccc\ -- - UcCxTh._ Pcç 2o
D -?D 24S R-c -3: -
. C-cu- I. ac(a -- I R C) a I I I I. C) l-- - I - -i (L,cc S4-S I-- i 1 t /)I1._ - -ii %ct oçc I_ i U 1 I :t. 4ô %&\ )cç - (ç3 J I I - ç) - OflkfQ&o 4o Pcc - i r. i - Nu. c - cl I..?S UQ E v:lc p-c 3 Ic I
I cwu:,bcec I
I I
7 -i; (3f4 & J i - d9-99 -J c0q Dj\o JCw. dc o (o,i1p L..\VWL IL 9\% c0w -t9% :;, 10 ( r 1 o 1
I :.. I I
1
I. 3 3 I 4to LLb So ;- (i ) (b) received resolved Date PtfRm # Nurse Message for Tech Order Order CCMU Pharmacy Tech pilot I1 ôo- 1930 jq C i (w C) ( ;L) 5 ---
CCMU Pharmacy Tech pilot \Voo - Date PtfRm : Nurse Message for Tech Order received Order resolved. ( _) w 3h.( 7: ivzrzl p4- -- pfr (z) i4&ofi4 - V V 3-a9. - \% ;ilcz 1o1s, atoi&tii &W(ec Peof7c V OOflr
Appendix J
Process Flow Chart: STAT Medication Turnaround 7/21/94 MD Writes (RN Takes Verbal) Order 2 Order Noted by Clerk STAT Medication Turnaround Time Team Diagramed by: Bruce Chaffee, Pharm.D., team leader STAT medication QIT 3 Is Order Considered V No 19 Does NurselClerk Take Order to Pharmacy to Obtain Medication? 21 Nurse Sends Medication Order to Pharmacy via Pneumatic Tube Yes Yes 5 RN Prepares Medication for Mministration 9 Pharmacy Technician Picks New Orders 20 Nurse or Unit Clerk Walks to Pharmacy and Hands Order(s) to Pharmacy Technician or Pharmacist RN Administers Medication 10 Pharmacy Technician or Pharmacist Enters Orders in Pharmacy Computer 22 Pharmacy Technician or Pharamcist Pulls Order from Pneumatic Tubes RN Chirts 11 Technician Notes Order Is Stat and Informs Appropriate Staff of Urgency (IV TectiIRPH)
the Order be Acted as Written (i.e. no problems with the order such as restriction, non-form status, etc.)? No Pharmacst Clarifies Order In Conjunction with MD and/or Nu Technician Obtains Medication from Another UH Pharmacy or Inventory Pharmacist or Technician Sends Medication to Floor via Pneumatic Tube or Delivers Medication to Nurse or Unit Clerk, or Places Medication In Medication Drop-off Site Technician Takes Medication to Floor on Routine Delivery Run
Appendix K
Sample number Tot& process time 1 0:10 0:06 Minutes #ofsamples 2 1:05 0:08 :00-:30 7 3 1:15 0:10 :31-1:00 13 4 0:50 0:15 1:01-1:30 9 5 1:20 0:28 1:31-2:00 6 2:50 0:30 2:01-2:30 3 7 0:40 0:30 2:31-3:00 5 8 1:00 0:33 3:01-3:30 0 9 0:50 0:34 38 10 0:08 0:40 11 0:55 0:40 12 0:40 0:45 13 1:30 0:50 14 0:06 0:50 15 1:05 0:50 16 2:50 0:55 17 0:28 0:55 18 0:50 1:00 19 1:40 1:00 20 0:30 1:00 21 1:05 1:05 22 0:33 1:05 23 1:15 1:05 24 1:15 1:15 25 0:30 1:15 26 0:15 1:15 27 0:45 1:20 28 2:07 1:25 29 2:45 1:30 30 0:55 1:40 31 1:25 2:05 32 2:25 2:07 33 2:05 2:25 34 1:00 2:35 35 0:34 2:45 36 2:45 2:45 37 1:00 2:50 38 2:35 2:50
Total process time distribution* for STAT orders between CCMU and 6th floor satellite pharmacy 0-2 z Cl) w E Cl) I I 4 20 8 6 4 20 I- :00 :31 1:01 1:31 2:01-2:31 3:01 :30 1:00 1:30 2:00 2:30 3:00 3:30 Time Intervals (hours:minutes) *completed sheets, without liaison N=38 I
I I I I I I I I I Appendix L
Average Process Times for Step Intervals (without liaison) Cl) G) E iz 0:36 0:28 0:21 0:14 0:07 0:00 0:21 0:06 0:30 0:16 0:04 lto2 2to3 3to4 4to 5to6 6to7 Step Intervals I 5 0:06
Bibliography 1. Burr, I.J. (1967). The Effect of Nonnormality on Constants for Control Charts, Industrial Quality Control, Vol. 23 2. Montgomery, Douglas C. (1997). Introduction to Statistical Quality Control, John Wiley & Sons, Inc. 3. Schilling, E.G., and P.R. Nelson (1976). The Effect of Nonnormality on Control Limits, Journal of Quality Technology, Vol.8 4. Coffey, Richard J. (1998-99). JOE 481 Special Projects In Hospital Systems, Cousepack for Winter 1999.