ESTRO 36 Abstract Book

S565

ESTRO 36

_______________________________________________________________________________________________

divided into four groups: thermoplastic head masks,

patient specific indexed whole body vacuum cushions,

doubly indexed non-patient-specific immobilization – i.e.

indexed knee/feet rests together with indexed head or

head and arm rests – and not indexed at all. For the

definition of the tolerance windows we required that at

most one in forty setups should provoke an interlock.

Furthermore movements between reference marks and

target points have to be large enough to violate the

tolerances with a high probability if they are not

performed. We required this probability to be 99%.

Results

For all four subgroups feasible interlock thresholds can be

defined. Especially for patient specific immobilization

devices they can be set very tightly. For thermoplastic

masks the limits are below plus or minus one centimeter

in all three directions and for the vacuum cushions the

largest tolerance value, which is in the longitudinal

direction, amounts to not more than +/- 2.5 cm. But even

ample tolerances, as we find for non-indexed

immobilizations, should be implemented since they help

to decrease the risk of irradiating the wrong patient or

isocenter significantly.

Also the obtained minimum shifts from the reference

marks are feasible and can easily be adopted in routine

setup.

Conclusion

Tolerances to table coordinates help to detect shifts which

are not applied at all or not in all directions. They also

prove to be efficient in discriminating between two

different isocenters or patients.

The values presented here are both dependent on the

immobilization devices used and on the patient collective.

Therefore each department has to examine the

applicability of the values in its setting.

Especially if other technical means, e.g. surface scanning

or RFID technology, are not available, indexed

immobilization devices and couch coordinate tolerances

can serve as a simple and effective method to reduce the

risk of RTEs in treatment delivery.

PO-1025 Development of a in-house KPI tool

A. Wallis

, D. Moretti

Liverpool Hospital, Radiation Oncology, Liverpool,

Australia

Purpose or Objective

Health informatics and data mining have enabled the

analysis of operational performance and assist managers

in making informed decisions in their departments (1). In

2010 the New South Wales Government in Australia

requested that all departments, both public and private,

were required to report on the percentage of patients

treated within 10 minutes of their scheduled appointment

time. At the time, the Liverpool and Macarthur Cancer

Therapy Centres (LMCTC) did not have a tool which could

measure the patient’s waiting time. This was the catalyst

for developing an in-house tool to measure the patient’s

waiting time as well as a number of other key performance

indicators (KPIs). The purpose of this abstract is to present

how an in-house tool can be developed and

established within a department to measure

departmental KPIs such as individual patient appointment

times, patient waiting time, machine utilisation and the

impact of changing techniques and technology.

Material and Methods

In 2010, Mosaiq 2.0X was installed in LMCTC. This version

allowed the extraction of time stamps into a reporting tool

(Crystal Reports V11). Definition of a patient's

appointment required the standardisation of the

treatment processes. This ensured improved robustness of

patient data and allowed accurate extraction of time

stamps in Mosaiq. The data from the reporting tool is

imported into Microsoft Excel 2013 on a weekly basis for

visual display and actioning on the KPIs.

Results

A weekly in-house KPI tool which compares machine

utilisation and performance, completion of QA tasks and

individual patient appointments has been utilised at

LMCTC since 2011. The tool has enabled staff to monitor

patient appointment duration on a daily basis and allows

direct comparison with the patient’s scheduled time. A

traffic light system has been developed to allow easy

visualisation of patient appointments requiring

adjustment (Fig 1). A buffer time which is -12% and + 8%

of the scheduled appointment time is applied to allow

easy visualisation of appointments requiring action. Based

on the results and traffic light display, each patient’s

appointments are adjusted for the following week,

resulting in a machine schedule made up of individualised

patient appointments. Queue times are compared with

scheduled patient appointments to review the timeliness

of patients attending their appointment. The tool was

designed and released in October of 2010 for a trial period

of two months and has been in use in the department since

its introduction.

Figure 1: Traffic light system which compared individual

appointment duration against the scheduled appointment

time.

Conclusion

The development of an in-house KPI tool has many

advantages for a radiation oncology department.

Individual appointment times can be recorded and

adjusted to ensure adequate time is allocated for an

individual’s needs. Ensuring adequate scheduling results in

reducing patient waiting times and stress for treatment

staff. It also displays machine utilisation and overall

performance.