Six sigma project

134 LABMEDICINE j Volume 39 Number 3 j March 2008 labmedicine.com

Reports

Waiting is an inherent component in many health care processes, including phlebotomy. Come in early and wait, check in and wait, be called back and wait, call the office and wait… it has become an expected part of the process. Recently, however, health care organizations are beginning to realize that they are in a service industry—the patient is the customer, and the patient’s concerns should be addressed. Patients should not have to wait for extended periods of time.

Lean Six Sigma is providing an avenue to deal with issues such as patient wait times. Lean Six Sigma—the marriage of the speed and efficiency of Lean processes with the structured approach to problem solving provided by Six Sigma—facilitates open communi- cation for all parties involved in a process to come together as a team and work toward a common goal. The common goal for Trinity Clinic management was to address the ever-present issue of extended patient wait times in the phlebotomy department.

DMAIC To break a project down into manageable sections, the

acronym DMAIC is employed. First, Define the project. Sec- ond, Measure the baseline of the process in question. Next, Analyze the data collected during the measure phase to de- termine root causes for the variations in the process. Then, Improve the process with a pilot run using the ideas generated by brainstorming from the team. Finally, Control the process with charts and written procedures and policies to guarantee the changes will remain in effect.

Define The phlebotomy department at most clinics and hospitals

is a central hub of activity—stat draws, rolling veins, glucose tolerance tests, urine specimens on disabled patients and tod- dlers. The amount and scope of work to be accomplished can be complex and challenging.

The team in this project was composed of 2 phlebotomy personnel, 1 front desk person, 1 nursing representative, a laboratory manager, a facilitator, and additional ad hoc repre- sentatives during the life of the project. All the major players involved in the process of collecting a patient’s blood speci- men were represented on the team. The scope of the project would begin with the time the laboratory order was placed by nursing personnel and would end after patient specimen

collection. The goals were to reduce the time patients were waiting to have blood collected and to reduce—if not elimi- nate—non-value-added tasks discovered in the process.

Measure The phlebotomy department was open from 6:30 am

to 6:00 pm Monday through Friday. There were 9 phleboto- mists—1 lead and 8 staff. There were 7 draw stations available at all times to collect specimens from patients. On average, 230 patient specimens were collected per day, or about 25 patients per phlebotomist. Seventy-five percent of the workload was com- pleted by 1:00 pm each day. The average time for a phlebotomist to collect a patient’s specimen was 4.5 minutes.

A flow chart was developed to illustrate the patient’s pro- cess. The process originated at the doctor’s office. The patient would be sent by nursing personnel to the phlebotomy de- partment for specimen collection (within the same building). The patient would walk to the phlebotomy department and check in with the front desk. The patient would then wait in the lobby until his or her name was called by the phleboto- mist. The phlebotomist would at that point collect the speci- men and dismiss the patient.

The only true value-added step in the patient’s process was having their specimen collected. All the traveling and waiting were non-value-added steps. The process contained 85% waste!

Analyze The question petitioning analysis was, “Why do patients

have to wait so long?” The team identified several potential causes affecting a phlebotomist’s daily routine, which in turn influenced the number of patient samples collected per hour. Five causes were identified:

1. Walking stat specimens one-third of a mile to the main hospital laboratory.

2. Picking up miscellaneous specimens from more than 20 doctors’ offices in the building during the morning rush.

3. Incorrect orders submitted. 4. Inventory stocked or counted during the morning rush. 5. Not having the 7 available workstations filled by

phlebotomists at all times during the morning rush (Figure 1).

Using Lean Six Sigma to Reduce Patient Wait Times Jennifer Jackson, MT(ASCP), MBA,1 Lori A. Woeste, EdD, MT(ASCP)2

(1Six Sigma Black Belt for Trinity Clinic, Tyler, TX, 2Assistant Professor, Clinical Laboratory Science Program, Illinois State University, Normal, IL) DOI: 10.1309/7UW17NMUET6M4XFP

D ow

nloaded from https://academ

ic.oup.com /labm

ed/article-abstract/39/3/134/2504683 by V A

Jerry P ettis Library user on 04 June 2020

Reports

labmedicine.com March 2008 j Volume 39 Number 3 j LABMEDICINE 135

Improve Improvement involves the critical decision to implement

changes identified in the process. (1) Stat specimens were no longer walked to the hospital. The phlebotomists began to use a pneumatic tube system in the building next door that was already routed directly to the hospital. (2) Specimens collected in the doctors’ offices were not picked up during the morning rush; rather, this collection was moved to later in the day when phlebotomy staffing and patient demand were balanced. (3) To reduce the number of incorrect laboratory orders, nursing personnel were encouraged to call the phlebotomy department before placing the order to obtain clarification if needed. (4) In- ventory and other business-related tasks were performed later in the day so the phlebotomists could concentrate on their primary daily function: patient specimen collection.

The last root cause (5) had data that indicated supply was not meeting demand. With only 1 phlebotomist collecting patient samples at 6:30 am, a bottleneck had already formed, considerably increasing patient wait times. To improve cycle time efficiency, the phlebotomists’ exit rate had to equal the amount of work coming into the process. More phlebotomists were needed at 6:30 am. By combining creative scheduling and personnel requests, schedules were rearranged and a pilot was tested (Table 1).

The previous schedule had 1 supervisor and 8 full-time phlebotomists working during hours that did not adequately meet patient demand. During the project, 1 phlebotomist requested to work part time from 7:00 am to 11:00 am, 1 full-time phlebotomist requested to change to part time from 8:00 am to 12:00 am, the department was subject to the attri- tion of 1 phlebotomist, and the team suggested that 3 of the phlebotomists who began at 7:00 am now arrive at 6:30 am to help with patient demand. The improvement schedule in- corporated the requests, the staff reduction, and the modifica- tions into the department schedule and was tested as a pilot.

The pilot was successful. Patient wait times were reduced, and the small amount of wait time became stable and pre- dictable. The phlebotomy department did not feel as over- whelmed as they had previously. Prepilot, the phlebotomists would arrive to work and see numerous patients waiting; now the hourly supply of phlebotomists was better matched to the hourly demand of patients. While the phlebotomists were aware that patients were being called from the waiting room to have blood collected at a much better pace, the pending outcome would be the satisfaction of the patients. Patient re- sponses were positive. One patient exclaimed, “I only waited 2 minutes!” Another patient commented that he had brought along the daily paper to catch up on a little reading and was irritated because he was unable to do so!

Control To ensure the improvements had long-term results,

“report cards” were generated to illustrate collections per hour per phlebotomist. In the beginning, report cards were produced on a weekly basis to provide team members im- mediate feedback on the process and as a comparison with preimprovement implementation (Figures 2 and 3). As time went on and the changes became routine, the weekly report cards changed to monthly reviews. Patient wait times were also collected pre- and post-implementation to measure the improvement project’s direct effect (Figure 4). Currently,

random samples are collected and charted during the month to demonstrate long-term effects (Figures 5).

Conclusion The perception of some may be that this project should

have been a managerial decision alone and that front-line

Table 1_Revised Phlebotomists’ Morning Schedule

Previous Improvement

1 supervisor 1 supervisor 6:30 am 1 FTE 6:30 am 4 FTE 7:00 am 4 FTE 7:00 am .5 FTE 8:00 am 1 FTE 8:00 am .5 FTE 9:00 am 3 FTE 9:00 am 3 FTE 10 FTE 9 FTE

Figure 1_Seven phlebotomy workstations were available at all times. At no time were the workstations filled to capacity.

Figure 2_Preimprovement implementation data indicated only 1 phlebotomist was collecting specimens at an optimal level of 5 to 6 patients per hour.

Figure 3_Post-improvement implementation data indicated 3 phlebotomists were collecting specimens at an optimal level of 5 to 6 patients per hour.

D ow

nloaded from https://academ

ic.oup.com /labm

ed/article-abstract/39/3/134/2504683 by V A

Jerry P ettis Library user on 04 June 2020

Reports

136 LABMEDICINE j Volume 39 Number 3 j March 2008 labmedicine.com

workers need not have been involved. Furthermore, the issue should have been a quick fix from the start. Those state- ments are not all together incorrect; however, the key to the entire improvement process is the team concept. Empower- ing employees to examine their daily functions from diverse perspectives, and giving them the ability and time to make a difference, enables buy-in. Improvement decisions were not mandated by a top-down management strategy; rather, the

ideas came from the team who lives and breathes the process each and every day. LM

Acknowledgments: Special thanks to Kara Claiborne, Polly Booth, Marvetta Barker, Luann Hannah, Sherri Epperson, Linda Newman, Sandy Riggle, Winnie Malley, Amy Parks, Dana Gilbert, Heidi Davault, Billie Epperson, and Trinity Clinic management. LM

Figure 4_Preimprovement am wait times lacked consistency, and patients were waiting 30 minutes or less to be collected. Post- improvement am wait times indicate consistency and patients waiting 15 minutes or less to be collected.

Figure 5_Current random monthly wait times indicate sustained positive process outcomes.

D ow

nloaded from https://academ

ic.oup.com /labm

ed/article-abstract/39/3/134/2504683 by V A

Jerry P ettis Library user on 04 June 2020