Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean...

28
Bruce Thomadsen University of Wisconsin -Madison Error Management and Patient Safety in Radiation Therapy: Fault Tree Analysis

Transcript of Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean...

Page 1: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

Bruce Thomadsen

University ofWisconsin -Madison

Error Management and Patient Safety in Radiation Therapy:

Fault Tree Analysis

Page 2: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

Administrative Stuff

Learning objectives:To understand fault trees.To understand how to use fault tree

analysis to design quality management.

Conflicts of Interest: The author has none.

Page 3: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

Proactive Steps to Address Hazards:Initial Design

Design the procedure with the system in mind:

To facilitate communications To minimize energy (e.g., minimize

distances to move, have ready access to materials)

Provide all environmental support (e.g., lighting, sound control)

Page 4: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

SEIPS Model Work system(Balance Theory; Smith and Carayon, 1989; Carayon and Smith, 2000)

Slide from Pascale Carayon

Technologyand Tools

Organization

EnvironmentTasks

Person

Page 5: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

Proactive Steps to Address Hazards:Analysis and Actions

After initial design of a procedure (or looking at a procedure in place), assess the risks.

For the high risks, try to redesign them out. When not possible, develop interventions.

Page 6: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

What to Do?

1. Understand the process2. Assess the hazards (e.g., FMEA)3. Establish the failure propagation patterns

(e.g., Fault tree analysis - FTA)4. Address the hazards5. Test and evaluate

Page 7: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

1. APBI with Contura Process Map

Do not try to read the labels; enjoy the organization

Successful treatment

Consultation and decision to treat

Imaging and diagnosis

Plan approval

Subsequenttreatments

Weekly chart checks

Chart filing

Decision of treatmenttechnique

Physics Check

Treatment review

Decision of protocol

Immobilization equipment fabricated

Immobilization equipment documented,

labeled, and stored

Immobilization forImaging study

Set up data documented

Check anatomy

Check prescription

Check dose calc method

Check beams, machines, etc.

Check normal tissue and target goals, dose distributions

Check that previous treatmentswere accounted for

Time out

Positioning

Imaging (port films, CBCT, etc) 27

Documentation

Treatment 3

Treatment 3

Documentation

DosimetryPhysics

MD

Immobilization equipment

documented,labeled, and stored

DQA

Chart Check

Equipment Check

Scheduling

Construct BlocksCompensators

Bolus, Etc

Scheduling forplanning process

RTP anatomycontouringPatient database

information entered

MD review

Plan review bydosimetrist

Data into electronic Database 21

Data into written chart 21

Review of patient medical history

Immobilization equipment fabricated

Import and fuse images 16

MD: delineateGTV/CTV 1,2,5, 53, 56

PTV construction

Edit density map for artifacts

Delineate ROIs and planning structures

Indicate motion/uncertainty Management 13

Specify registration goals 23, 37

Specify protocol for delineating target and structures 17

Specify images for target/structure delineation 11

Specify dose limits and goals 26

Suggest initial guidelines for treatment parameters

Enter prescription 19, 47

Setup fields

Setup dose calc parameters

Optimization/Dose calculation 12, 29, 31

Evaluate plan 10, 28

Initial treatmentplanning directive

Treatment planning

Immobilizationand positioning

Imaging (CT/PET/MR)

Treatment preparation

Plan prep

Initial treatment

Patient Identified

Check that plan matches prescription

Special Instructions (pacemakers, allergies,

preps, etc.) 9

Account for previous treatmentsor chemotherapy 4

Check versionof the plan 20

Check plansatisfied objectives 7, 36

Motion management 8

Tx Unit operationand calibration 3

Information on previousor concomitant treatment 8

Protocol for delineationof targets 9, 17

Patient IDTreatment Site

Treatment settings

Imaging

Motion Management 8

Protocol for PTVMargin 6, 48

Specify PTV Margin

Select Images 25

4D imaging correct 14

Optimization ROI 18, 33, 44Optimization settings 22, 45, 51

Treatment accessories 24

Boolean operations 30, 46

Changes noted 32, 34

Pt prep 35

Monitor Pt/Tx 38, 42

Monitor Pt/Tx 38

Transfer patient data to treatment delivery 15

Manual data entry and plan modification 39

Specify treatment course

Delivery protocols

Scheduling

Automatic data entry and plan modification

Specify ROI for optimization 18

Check planidentity 36

Write finalprescription

Enter demographics, 49

Prepare DRR and other images 50

Treatment settings

Positioning

Pt prep 35, 52

Changes correct 40, 43

Transfer images and otherDICOM data: primary and

secondary data transferred 41

Run leaf sequencer 54

Pt changes noted 55

Imaging Studies

Patient prepped (contrast, tattoos,

BBs etc.)

Check version ofplan and patient ID 7

Treatment accessories 24

Page 8: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

2. Assess the Risk

Eric just discussed this.

Page 9: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

3. Establish the Failure Propagation Pattern

This is the fault tree analysis.For the fault treeBegin at the failureAsk what are all the possible causesRelate the causes through logical gatesFor each cause, ask what would be the causeRepeat as needed

Page 10: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

Example: Calculation Fault Tree

Error in Calculation

Error in Input Data

Error in Data Entry

Error in Calculation Algorithm

Error in Prescription

Page 11: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

Example: Calculation Fault

Tree with QM Error in Calculation

Error in Input Data

Error in Data Entry

Error in Calculation Algorithm

Error in PrescriptionError in QC Error in QC

Error in QCError in QC

Error in QA

Error in PT Calculation

Page 12: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

Process Tree for Prostate Implants with Loaded Needles

Successful treatment

2 Prepare patient

Identify patient

Anesthetize patient

RO holds styletand removes needle

RO inserts needle intoindicated hole in template

RO rotates probe and mountingbracket to match previous images

5 Reviewimplant

1 Gatheringmaterials

MP brings plan to procedure room identifying needle

location and depth

MP brings needles toProcedure room

RO checks images along whole superior-Inferior range

RO scans throughImages looking for gaps

RO adds extra sourcesif needed and available

3 Position USprobe

4 Insertion ofneedles and sources

MP finds needle inpackage or container

MP brings planning imagesto procedure room

RO pushes needle untiltip shows at desired depth

Prepare and insert US probe

MP hands needleto RO

RO determinesstarting needle

Bring patient into procedure room

Move patient to table

Scrub patientRO selects next needle

1, 4,6,7

2,5,21

3,9,19

8,16,17,18

10,11

12,13,24

14,25

15,2022,23

1,6,7,18,20

2,5

3,4

8

9,11,12,13

14,15,16,17

10,19

Page 13: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

One Step of the Process Tree

RO holds styletand removes needle

RO inserts needle intoindicated hole in template

4 Insertion ofneedles and sources

MP finds needle inpackage or container

RO pushes needle untiltip shows at desired depth

MP hands needleto RO

RO determinesstarting needle

RO selects next needle

2,5,21

10,11

14,25

15,20

10,19

Page 14: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

Fault Tree for Prostate

Implants with Loaded Needles

Or

Sources placed in wrong location

Patient misidentified

US probe misaligned

Incorrect dose, dose

distribution, location

US images inadequate

Human error: Omission – Time-out not performed

Training – patient identified incorrectly

Or

Training – Probe cover not correctly

filled

Human Failure: Inattention/Poor performance -Temp/ate not

seated properly

Human Failure: Inattention/Poor performance -

Probe cover not correctly filled

Or

RO fails to match the volume study

images

US images inadequate – QA

failure

Human Failure: Poor performance

Changes in prostate

Or

RO fails to align images along the

range

Training failure

Human Failure: Inattention/Poor

performance

US images inadequate – QA

failure

Or

Or

Training failure: US images

inadequate – Needle order

Human failure: US images inadequate

– Needle order

RO fails to image adequately Or

MP fails to hand the correct needle

to ROPoor demarcation

of needles

Human Failure: Inattention/Poor

performance

Confusion between packagesOr

MP drops needle OrPoor room layout

Human failure: MP slips

RO fails to insert needle properly

RO fails to insert into correct hole

RO fails to insert needle to correct

depth

Or

Human Failure: Inattention

Bad viewing conditions

Confusion between holes

Or

US images inadequate – QA

failure

Training Failure

Human Failure: Inattention/Poor

performanceOr

Confusion between planes

RO fails to hold stylet stable during

retraction

Human failure: RO fails to hold stylet

Human failure: RO pushes on stylet

Or

RO selects next needle that obscures

subsequent needles

Training Failure

Human Failure: Inattention/Poor

performance

Or

Or

RO erroneously modifies source

distribution

RO adds unnecessary

sources

Or

RO fails to add necessary

additional sources

US images inadequate – QA

failure

Training Failure

Or

US images inadequate – QA

failure

Training Failure

Or

Again, don’t worry about reading it, this is for scale.

Page 15: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

4. So What?

Start with the branches of the fault tree that corresponds to the branches of the process tree with the greatest hazard, either PRN or severity.

Might as well start with the top of the branch, even though that is completely arbitrary.

Consider the possible interventions.

Page 16: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

Possible Interventions

First correct any environmental problems – that usually is a relatively inexpensive but effective operation.

Then consider the key core components identified by TG 100 Training Communication Standardized policies and procedures

Make sure resources are allocated as needed (i.e., staffing and equipment.

Page 17: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

Fault Tree with

Interventions

Again, don’t worry about reading it, this is for scale.

Or

Sources placed in wrong location

Patient misidentified

US probe misaligned

Incorrect dose, dose

distribution, location

US images inadequate

Human error: Omission – Time-out not performed

Training – patient identified incorrectly

Or

Training – Probe cover not correctly

filled

Human Failure: Inattention/Poor performance -Temp/ate not

seated properly

Human Failure: Inattention/Poor performance -

Probe cover not correctly filled

Or

RO fails to match the volume study

images

Human Failure: Poor performance

Changes in prostate

Or

RO fails to align images along the

range

Training failure

Human Failure: Inattention/Poor

performanceOr

Or

Training failure: US images

inadequate – Needle order

Human failure: US images inadequate

– Needle order

RO fails to image adequately Or

MP fails to hand the correct needle

to ROPoor demarcation

of needles

Human Failure: Inattention/Poor

performance

Confusion between packagesOr

MP drops needle OrPoor room layout

Human failure: MP slips

RO fails to insert needle properly

RO fails to insert into correct hole

RO fails to insert needle to correct

depth

Or

Human Failure: Inattention

Bad viewing conditions

Confusion between holes

Or

Human Failure: Inattention/Poor

performanceOr

Confusion between planes

RO fails to hold stylet stable during

retraction

Human failure: RO fails to hold stylet

Human failure: RO pushes on stylet

Or

RO selects next needle that obscures

subsequent needles

Human Failure: Inattention/Poor

performance

Or

Or

RO erroneously modifies source

distribution

RO adds unnecessary

sources

Or

RO fails to add necessary

additional sources

US images inadequate

Or

Taken care of by the generally complete training, establishing clear communication

modalities (possibly forms) and establishing protocols, policies and

procedures

Systemic corrections

Training failure

Training failure

Training failure

AN

D

US images inadequate

OrTraining failure

AN

D

US images inadequate

AN

D

US QA failure

US images inadequate

AN

D

US QA failure

US images inadequate

AN

D

US QA failure

AN

D

QC failure: Time out formQuality assurance

Quality control

Managerial changes

AN

D

QC failure: Time out for image

evaluation

AN

D

QC failure: Time out form

AN

D

QC failure: Prostate image

time out

Or

Procedural changes

AN

D

Failure of call and response

AN

D

Failure of call and response

AN

D

Failure of call and response

US QA failure

US QA failure

AN

D

Post-proceduralCT

Page 18: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

KeyTaken care of by the generally complete

training, establishing clear communication modalities (possibly forms) and

establishing protocols, policies and procedures

Systemic corrections

Quality assurance

Quality control

Managerial changes

Procedural changes

Page 19: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

Quality Management

Quality Control – Activities that force specific quality on a process. Often operates on inputs.

Quality Assurance – Activities that demonstrate the level of quality of a process. Often checks outputs.

Page 20: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

One Example

Patient misidentified

Human error: Omission – Time-out not performed

Training – patient identified incorrectly

OrAN

D

QC failure: Time out form

Systemic corrections

Quality assurance

Quality control

Managerial changes

Procedural changes

Page 21: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

Another Example

MP fails to hand the correct needle

to ROPoor demarcation

of needles

Human Failure: Inattention/Poor

performance

Confusion between packagesOr

MP drops needle OrPoor room layout

Human failure: MP slips

AN

D

Failure of call and response

Systemic corrections

Quality assurance

Quality control

Managerial changes

Procedural changes

Page 22: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

Another Example

RO fails to align images along the

range

Training failure

Human Failure: Inattention/Poor

performanceOr

US images inadequate

AND

US QA failure

Systemic corrections

Quality assurance

Quality control

Managerial changes

Procedural changes

Page 23: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

No Preventing Human Error

RO fails to align images along the

range

Training failure

Human Failure: Inattention/Poor

performanceOr

US images inadequate

AND

US QA failure

AND

QC failure: Prostate image

time out

Or

Add some QC down stream

Page 24: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

CommissioningOr

Error in delineating

GTV/CTV (MD) and other

structures for planning and optimization

>3*sigma error contouring errors:

wrong organ, wrong site, wrong

expansions (1)

Inadequate or lack of procedures /

practices

Defective materials/tools/equipment (2.1,

2.2, 2.3)

User Error (2.5, 2.6)

Inadequate design specification (3.2)

Inadequate assessment of

operational capabilities (3.5)

Inadequate programming (3.6)

Inattention, Rushed process,

lack of time or staff, fatigue,

failure to review work

Or

Excessive delineation errors resulting in <3*

sigma segmentation

Errors (2)

Inadequate or lack of procedures /

practices

Availability of defective

materials/tools/equipment (2.1,

2.2)

Materials/tools/equipment used

incorrectly or inadequate

assessment of materials/tools/

equipment for task (2.5, 2.6)

Inadequate design specification (3.2)

Inadequate assessment of

operational capabilities (3.5)

Inadequate programming (3.6)

Inattention, Rushed process,

lack of time or staff, fatigue,

failure to review work

Inadequate training /

orientation (6.1)

OrPoorly drawn

contours (spikes, sloppy, etc) (3)

Inadequate or lack of procedures /

practices

Availability of defective

materials/tools/equipment (2.1,

2.2)

Materials/tools/equipment used

incorrectly or inadequate

assessment of materials/tools/

equipment for task (2.5, 2.6)

Inadequate design specification (3.2)

Inadequate assessment of

operational capabilities (3.5)

Inadequate programming (3.6)

Inattention, Rushed process,

lack of time or staff, fatigue,

failure to review work

Inadequate training /

orientation (6.1)

Or

250

255

260

265

270

366

326

212

Taken care of by the generally complete training, establishing clear communication

modalities (possibly forms) and establishing protocols, policies and

procedures

Key item for commissing

Key item for facility managerial changes

Taken from TG 100

Page 25: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

Levels of QM

Commissioning Periodic Testing Per patient

Page 26: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

Good Beginnings

Acceptance Testing is making sure the procedure works as specified. Usually applied to hardware and software. Also applies to walking through new procedures.

Commissioning gets a procedure started Gathers all the data needed. Gives confidence that systems work as expected. Finds when the system fails.

Essential to any procedure.

Page 27: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

What to Do?

1. Understand the process2. Assess the hazards (e.g., FMEA)3. Establish the failure propagation patterns

(e.g., FTA)4. Address the hazards5. Test and evaluate – Quality

Improvement: a different presentation

Page 28: Recent Advances in Brachytherapy Dosimetry · 2010-08-18 · Treatment accessories 24 Boolean operations 30, 46 Changes noted 32, 34 Pt prep 35 Monitor Pt/Tx 38, 42 Monitor Pt/Tx

Conclusions

Fault tree analysis facilitates quality planning:Redesigning to eliminate hazards.Adjusting resources.Ensuring the key core components.Performing comprehensive

commissioning.Establishing quality management.