Urology at Livonia Center for Specialty Care Clinic

Urology at Livonia Center for Specialty Care Clinic Patient Flow Study Final Report Presentation R E BEKAH ANDREWS | K AYWEE LIAN | TEAM 6 DEC EMBER 13 T H , 2016 KRISTEN YDOATE

Introduction Client: Clinic At Livonia Center for Specialty Care Director: John Wei, MD Professor of Urology Manager: Karen Moore Ambulatory Care Manager Intermediate Coordinators: Process and Operations Analysis Office Mary Duck Industrial Engineer Expert Kyle Worley Lean Coach Industrial Engineer

Urology Clinic at Livonia Provides General Urologic Care 8 4 1 -5 Medical Assistants Registered Nurses Providers, depending on the schedule

General Patient Flow

Patient Stratification by Patient/Visit Type, and Patient Diagnosis Urology Clinic Patient/Visit Type Patient Diagnosis Kidney Stones New Patient Return Visit Benign Prostatic Hyperplasia Consultation Erectile Dysfunction Elevated Prostate. Specific Antigen Nurse Visit Urinary Tract Infection Hematuria Procedural Incontinence Others

Decision Tree for Patient Scheduling Time Clinic utilizes a pre-arrival scheduling approach

Current Process Flow Has Several Issues Lack of quantifiable data Unknown areas of waste Disparity between scheduled and actual time

Goals and Objectives Background: Clinic wants to understand the patient flow process better through collecting information on timing of each step, and where waste resides in the current process. Goals: Quantify Current Patient Flow Identify Wastes and Opportunities for Improvement

Methods Observations Combined Time Study Form Literature Review Interviews Mi. Chart Data Surveys

Data Analysis Value Stream Mapping Mi. Chart Analysis Pareto Chart of Wastes

Value Stream Mapping 4 Value Stream Maps 1. 2. 3. 4. New Patients Return Visit Consultations Return Visit Procedural Nurse Visits

Value Stream Mapping Return Procedural Patient Flow Most Inefficient Value Stream Map Summary Table Stratified by Patient and Visit Type Source: Time Studies Data from 11/2/15 - 10/31/16, N = 594

Value Stream Mapping Monday, Tuesday, and Wednesdays Experience Longer MA and Provider (Mondays and Tuesday) Wait Times Value stream map totals across day of the week Source: Time study data 10/21/16 - 11/15/16, N = 513

Value Stream Mapping Excess Wait Time for MA, Nurse, and Provider Value Stream Map Summary Table for Steps in the Patient Flow Process in Minutes Source: Time Studies Data from 11/2/15 - 10/31/16, N = 548

Mi. Chart Analysis Patients with elevated PSA spend the longest time in the clinic Interviews reveal a possible explanation is elevated PSA patients are often sensitive conversations and involve teaching Mi. Chart data shows that 6 is significantly higher than 5, 3 and 1 Source: Michart Data from 11/2/15 - 10/31/16, N = 7500; 1 = Stones, 2 = Benign Prostatic Hyperplasia, 3 = Urinary Tract Infection and Cysts. 4 = Incontinence, 5 = Erectile Dysfunction, 6 = Elevated PSA, 7 = Hematuria

Pareto Chart Forms of waste identified from analyzing time study and process diagnostic form data are: ● Wait times exceeding 5 minutes ● Actual times exceeding allotted times ● >2 provider interactions ● Added-on procedures

Pareto Chart Top 3 Forms of Waste 1. Patients spend >60 minutes in the clinic 2. >5 minute wait for MA 3. > 5 minute wait for provider Figure 6: Pareto chart of the frequency of waste within the clinic. Source: Time study data 10/21/16 - 11/15/16; N = 594

Waste #1: Patients Spend >60 Minutes in the Clinic Data Summary Median = 61 minutes Mean = 67. 04 minutes 54% of patients spend > 60 minutes at the clinic 60% patients who spend > 60 minutes, spend up to 90 minutes at the clinic

Waste #1: Patients Spend >60 Minutes in the Clinic Stratified by Provider – Large variation between providers 83. 9 90. 0 Time (minutes) 80. 0 Median = [VALUE] 70. 7 68. 9 69. 0 70. 0 60. 0 90. 5 89. 2 60. 9 58. 6 55. 8 64. 0 60. 9 51. 4 61. 8 55. 9 50. 0 40. 0 27. 4 30. 0 20. 0 80% 67% 70% Percentage > 60 Minutes 100. 0 60% 50% 41% 40% 41% 35% 33% 30% 20% Median = 41% 58% 56% 21% 17% 5% 10. 0 1 2 3 4 5 6 7 8 9 Provider 10 11 Average time in clinic by provider Source: Time study data 10/21/16 - 11/15/16; N = 240 12 13 14 15 0% 1 2 3 4 5 6 7 8 9 Provider 10 11 12 13 14 Percent of patient visits greater than 60 minutes by provider Source: Time study data 10/21/16 - 11/15/16; N = 240 15

Waste #2: >5 Minute Wait for MA Data Summary Median = 8 minutes Mean = 11. 2 minutes 49% of patients experience > 5 minute waits for MA 41% of wait times are over 10 minutes

Waste #2: >5 Minute Wait for MA Excess Wait Times for MA at start of day and during lunch breaks 70% PERCENTAGE OF PATIENTS WAITING >5 MIN Stratified by Time of Day 64% 60% 52% Median = 52% 54% 57% 56% 48% 50% 42% 47% 40% 30% 20% 10% 0% 8 -9 9 -10 10 -11 11 -12 12 -13 Time of Day 13 -14 14 -15 15 -16 Percent of MA wait times greater than 5 minutes across time of day Source: Time study data 10/21/16 - 11/15/16; N = 296 16 -17

Waste #3: >5 Minute Wait for Provider Data Summary Median = 3 minutes Mean = 8. 1 minutes 69% waited over 10 minutes* *of patients who had to wait over 5 minutes 45% waited over 15 minutes*

Waste #3: >5 Minute Wait for Provider Stratified by Provider - Large variation between providers Average Wait Time for Provider (Minutes) 40. 00 37. 00 80% Mean = 9. 95 35. 00 69% 70% 30. 00 60% 25. 00 58% 57% 46% Percent 50% 20. 00 15. 00 8. 78 11. 17 10. 54 12. 00 7. 20 3. 43 4 5 6 7 27% 12 Provider Average wait time by provider Source: Time study data 10/21/16 - 11/15/16; N = 79 20% 10% 0. 00 2 40% 30% 10. 00 5. 00 42% 15 0% 0% 4 5 6 7 12 Provider (Sample Size > 5) 14 Percentage of time providers are late to appointment (Source: Time study data 10/21/16 - 11/15/16; N = 79) 15

Waste #3: >5 Minute Wait for Provider Scheduled 15 Minutes Appointment Insufficient for All Patient Care Tasks for a Single Patient Direct Care Indirect Care Direct Care 30. 0 40. 0 35. 0 10 20. 0 10 10. 0 5. 0 10. 9 14. 0 10 10 18. 5 10. 5 12. 2 10. 0 12. 3 11. 0 10. 9 8. 7 0. 0 2 4 5 6 7 9 Provider 10 12 13 Provider indirect and direct care time by provider, 15 minute appointment Source: Time study data 10/21/16 - 11/15/16; N = 79 14 15 TIME TAKEN (MINS) Time (Minutes) 25. 0 15. 0 Indirect Care 30. 0 10 25. 0 20. 0 10 10 10 15. 0 10. 0 5. 0 0. 0 13. 0 2 17. 1 17. 5 18. 3 19. 3 23. 6 10 22. 0 9. 0 4 5 6 PROVIDER 9 12 14 15 Provider indirect and direct care time by provider, 30 minute appointment Source: Time study data 10/21/16 - 11/15/16; N = 79

Waste #3: >5 Minute Wait for Provider Non-standardized Handling of Indirect Patient Care and Add-on Procedures another Source of Variability Providers complete required tasks: Before seeing a patient While seeing a patient During breaks in their schedule Providers accept add-on procedures: Perform immediately after consult Reschedule different appointment

Summary of Conclusions Value Stream Mapping § RV procedurals spend the longest time in clinic § Mondays, Tuesdays, and Wednesdays see a higher average wait time for MA’s § Longest wait time occur while waiting for Nurse, MA for vitals, and Providers, respectively Mi. Chart Data § RV patients with elevated PSA diagnosis spend the longest time in clinic Pareto Chart Top 3 Wastes: 1. 2. 3. Patients spend >60 minutes in the clinic >5 minute wait for MA > 5 minute wait for provider

Summary of Conclusions #1: Patients spend >60 minutes in the clinic § § § 54 % spend > 60 minutes Large variations by provider Big opportunity for improvement (goal <60) #2: >5 minute wait for MA § § § 49% wait > 5 minutes Large outliers High wait times at the start of the day and lunch hours #3: > 5 minute wait for provider § § Large outliers Variation between providers 15 minute allotted time is exceeded when indirect care is considered No standardization for indirect care or add-on procedures

Recommendations RV Procedural Patients Integration of Tasks § Currently there are 6 staff interactions § Cut down number of interactions by integrating tasks § Ex. MA’s are trained to obtain consent for procedures Parallelization § Currently clinic is conducted serially § Conduct tasks concurrently to reduce wait times § Ex. Nurse obtains consent while MA finishes vitals and interview

Recommendations Longer MA Wait Times on Mondays, Tuesdays and Wednesdays Revise Staffing Levels § Cope with greater amount of patients Reassign Providers § Move providers to less busy days § Smoothen out demand for MA’s

Recommendations Longer Nurse Wait Times Assign Nurses to Roles Versus Providers § One role to assist any providers (obtain consent and nurse teaching) § Other role is nurse visits

Recommendations PSA Patient Spend Longer Times at the Clinic Conduct a Follow-up Study § Clearly identify why these patients spend more time in clinic § May need to change scheduling of PSA patients

Recommendations Excess Wait Times for MA’s Begin Appointments at 8: 30 AM Versus 8: 00 AM § Allow more time for MA’s to prepare exams rooms Increase Staffing at the Start of the Day § Assign some MA’s to clinic set-up and others to attend to patients Examine Assignment of MA to Patients for Vitals § Determine if there any inefficiencies in the current MA assignment process

Recommendations Excess Wait Times for Providers Increase 15 Minute Appointment Times § Increase by increments of 5 minutes § Evaluate wait-times as consultation time increases Standardize Indirect Patient Care Tasks § Complete immediately after each patient is seen § Reduce variability between providers Standardize Add-on Procedures § Schedule patients for add-on procedures during breaks in a provider’s schedule – not immediately after consult

Project Goals: Quantify Current Patient Flow Identify Wastes and Opportunities for Improvement

Expected Impact Describe the current patient flow process and where waste resides Provide clear visualization of the overall patient process Guide the clinic in future process improvement efforts

Thank you! Questions?

Appendix §Value Stream Maps §Staff Survey Results §References

Value Stream Map: New Patients

Value Stream Map: Return Visit Consultation

Value Stream Map: Return Visit Procedural

Value Stream Map: Nurse Visit

Waste #2: MA Wait Times 350 300 Count 250 200 150 126 100 50 50 24 6 10 1 0 0 <= x <10 10 <= x < 20 20 <= x < 30 30 <= x < 40 40 <= x < 50 TIME RANGE (MIN) Frequency chart of MA wait times Source: Time study data 10/21/16 - 11/15/16; N = 519 50 <= x < 60 70 <= x < 80

Waste #3: Provider Wait Times 30 26 25 19 Frequency 20 14 15 10 9 8 6 5 0 5 < x < 10 10 <=x <15 15 <=x <20 20 <=x <25 time range (min) 25 <=x <30 >= 30 Wait time for provider when providers were late to the appointment Source: Time Study Data 10/21/16 - 11/15/16, N = 79

Waste #5: Late Patients 20 19 18 16 16 FREQUENCY 14 12 10 10 7 8 6 6 3 4 2 2 0 5 < x < 10 10 <= x < 15 <= x < 20 20 <= x < 25 25 <= x < 30 TIME RANGE (MIN) 30 <= x < 35 >= 35 Histogram of Patient Late Times for Patients That Arrived Past Pre-Arrival Time Source: Time Study Data 10/21/16 - 11/15/16; N = 63

Staff Surveys

Staff Surveys

References [1] Matt Bovberg et al. , “Analyzing Patient Flow and Process Waste in the General Thoracic Surgery Clinic”, IOE 481 Design Projects, Winter 2014, April 2014 [2] Senior Altarium Institute, “Applying Lean to Improve the Patient Visit Process at Three Federally Qualified Health Centers”, July 2011 [3] Lori Rutman et al. , “Improving Patient Flow Using Lean Methodology: an Emergency Medicine Experience”, Springer International Publishing, October 28 th 2015. [4] B. T. Denton and D. T. Brian, Handbook of healthcare operations management: Methods and applications. New NY: Springer New York, 2013, ch. 3, sec. 2. [5] L. Jiang and R. E. Giachetti, "A queueing network model to analyze the impact of parallelization of care on patient time, " Health Care Management Science, vol. 11, no. 3, pp. 248– 261, Dec. 2007. [6] A. M. Association, "How to handle patients who are always late, " 2009. [Online]. Available: http: //www. amednews. com/article/20090413/business/304139998/5/. Accessed: Dec. 6, 2016. [7] R. R. Lummus, R. J. Vokurka, and B. Rodeghiero, "Improving quality through value stream mapping: A case study of a physician’s clinic, " Total Quality Management & Business Excellence, vol. 17, no. 8, pp. 1063– 1075, Oct. 2006. York, cycle