Variation reduction strategies for IP patient flow through operating rooms
Operational Performance Improvement
Research Engineer/Management Engineering
William Beaumont Hospitals
William Beaumont Hospitals
Managing beds in hospitals with high in-patient occupancy rates is aided by a steady flow of patients through the operating rooms. This presentation discusses techniques to determine variation of flow of patients in operating rooms. Day-to-day and hour-to-hour flow variation measurement and reduction techniques are discussed in a case study.
This presentation is a follow-up to a paper presented by Dr. Jayant Trewn at the 2004 SHS conference titled "A Treatise on Process Variation that Affects Patient Flow in Hospitals". Various methods to measure the effect of patient flow variation discussed above were:
Conceptual Throughput Model (velocity), Hamilton, K (2003) Schedule variation - Variation Partitioning using Sum of Squares, Yang, K. and J.Trewn (2003) Resource availability variation - Aggregate Hospital variation based on resource census, Resar, R. (2003)
High in-patient occupancy hospitals create a situation of bed shortage when throughput of in-patients is not steady. Patient flow variation is affected by a number of factors. These sources of sources of variation are identified as:
1 Provider practices variation between providers Methods, techniques, experience, etc.
2 Patient characteristics variation between patients Recovery, care choices, patient demographics, patients' reception to treatments, etc.
3 System variances process variation common cause vs. special cause variation
Equipment availability, resource availability (procedure rooms, beds, doctors, nurses, etc.), discharge procedure, etc. 4 Patient mix/severity/diagnosis variation between patients Variation in patient condition, patient recovery (backward acuity moves), etc.
5 Input variations
Patient appointments, Service unit schedule variations, Unit resource availability
6 Output variations
Availability of recovery space and resources, availability of admitting unit space, etc.
Surgeon scheduling practices are referenced as a source of artificial variation that causes a variation in day-to-day flow of in-patients through surgical operation rooms. This presentation measures artificial variation of in-patient flow through Gynecology Surgery operating rooms.
In this presentation, the researchers demonstrate how provider practices, in particular case scheduling, affects the throughput of patients through the operating rooms. Artificial variation introduced by surgeon office schedulers is a source of variation of patient departing operating rooms that post-operatively need in-patient beds. A case is made for using block scheduling as a technique to control artificial variation. The study is conducted in a major hospital system in Southeast Michigan, William Beaumont Health Systems in Royal Oak, MI, with data available from 2004 as a baseline and is compared to Q2 2005 after implementation of solutions. Preliminary analysis of data shows a marked reduction in variation from 1 patient per day to 0.6 patients per day on comparable day of week and time of day. ANOVA is used to partition variation and statistically show the reduction in variation. "Hills" charts are used to visualize the variation reduction. Control charts are used to study the trends in variation reduction.