Table 1.

Failure modes table for Gamma Knife radiosurgery

No.	Step	Potential failure modes	Potential causes of failure	Potential effects of failure	O	S	D	RPN	Notes and examples of causes and failures
1	Create paper chart/eChart	Incorrect patient ID data	Errors in manual entry, most likely causes: 1. Omission in entry 2. Human transcription error 3. Miscommunication	Very wrong dose	2.00	8.00	1.50	24.00	Patient name is typed into the hospital database incorrectly. Information is requested from another department for a different person who actually exists. Information for the wrong patient is sent back. Suboptimal dose prescribed.
2	Previous/adjuvant treatment record	Incorrect or incomplete previous treatment history	1. Personnel omission 2. Miscommunication	Very wrong dose	2.00	6.13	2.50	30.63	Historical information about patient treatment inside/outside the dept. is inaccurate or incomplete. The patient & family don't remember treatment details.
3	Physical examination	Incorrect or incomplete test results	1. Lack of standardized procedures 2. Miscommunication 3. Personnel omission	Very wrong dose	2.50	3.25	1.75	14.22	Necessary physical or sensory tests are not done prior to the treatment, i.e. the most recent hearing test result is not available for an acoustic patient. Other test results may include vision, taste, motor function etc.
4	Physical examination	Incorrect or incomplete patient implant info	1. Lack of standardized procedures 2. Miscommunication	Patient injury	2.13	5.38	2.13	24.27	Patient has pacemaker, dental implant etc. but the information about the implant is not available or inaccurate at the time of the treatment. MR images were acquired instead of CT. Patient injured.
5	Frame/fiducial box check	Cannot start	1. Miscommunication 2. Personnel omission	Inconvenience	3.00	3.38	2.25	22.78	Patient does not show up because of a severe weather condition or a miscommunication between patient and staff. Patient shows up but finds out he/she is not on the schedule.
6	Frame/fiducial box check	Use of distorted frame	1. Lack of standardized procedures 2. Personnel omission	Inaccurate volume Inaccurate dose distribution delayed treatment	1.63	7.00	4.13	46.92	A distorted frame is used. The accuracy of the image definition/dose calculation is compromised. If the distortion is too much, patient can not be docked and the treatment can not proceed.
7	Frame/fiducial box check	Bad fiducial box assembly	1. Lack of standardized procedures 2. Personnel omission 3. Inadequate training/orientation	Inaccurate volume Inaccurate dose distribution Imaging may need to be repeated	3.38	4.88	3.75	61.70	A fiducial box is not assembled correctly, i.e. screws are loose, frontal/back or left/right pieces are switched etc.
8	Frame/fiducial box check	Too much air bubble	1. Lack of standardized procedures 2. Personnel omission	Inaccurate volume Inaccurate dose distribution	3.25	4.00	2.88	37.38	The length of the air bubbles in a MR imaging box should be minimized. Air bubbles in the fiducial box may cause certain image slices to be excluded from the image definition process.
9	Frame positioning/fixation	Inappropriate local anesthesia injection	Personnel inadequately trained	Patient uncomfortable	3.38	4.38	3.13	46.14	Local anesthesia not given to where pins will be inserted into skull bone.
10	Frame positioning/fixation	Frame not fixed firmly	1. Personnel omission 2. Personnel inadequately trained	Very wrong dose distribution Inconvenience	1.50	5.00	2.13	15.94	A frame is not attached to the patient skull firmly and becomes loose during the course of a treatment. The procedure has to start over if someone finds out about the loose frame before the treatment. Otherwise, treatment with a loose frame could mean a wrong dose distribution.
11	Frame positioning/fixation	Unsterilized pin/post	1. Lack of standardized procedures 2. Personnel omission	Patient infection	1.50	3.13	1.63	7.62	The use of unsterilized pin/post may cause infection to patient and/or staff.
12	Frame positioning/fixation	Inappropriate frame placement	1. Lack of training/experience 2. Miscommunication	Inconvenience (staff and patient)	2.25	6.50	1.88	27.42	The stereotactic frame is not shifted correctly when treating a peripheral lesion. A collision with the helmets (on the 4C) or the source cap (on the Perfexion) is reported from the planning system. Frame and imaging need to be repeated.
13	Frame cap/adapter test	Distorted frame adapter	1. Lack of standardized procedures 2. Personnel omission	Inconvenience (staff and patient)	1.75	4.88	1.75	14.93	The frame adapter for Perfexion is bad but not found out before a treatment. Treatment can not proceed.
14	Bubble helmet measurement	Inaccurate bubble measurement	1. Inadequate training of personnel 2. Human error	Inaccurate dose distribution	2.63	5.00	2.63	34.45	Bubble measurement is not accurate enough and is not noticed at the final check stage. Dose delivery is based on an inaccurate patient skull geometry.
15	Post/pin measurement	Incorrect post/pin measurement	1. Inadequate training of personnel 2. Human error	Inconvenience Patient injury	1.88	5.00	1.75	16.41	Post or pin measurement is not correct. A collision is not detected by the planning system. A collision with a post or pin could stop the treatment or injure the patient on treatment.
16	Fiducial box attachment	Fiducial box not attached properly	1. Inadequate training of personnel 2. Personnel omission	Inaccurate volume	2.50	3.88	2.25	21.80	The imaging fiducial box should be completely engaged with the frame and locked securely. Failure to do so may cause distortion in acquired images.
17	Patient positioning/immobilization	Unsecured imaging adaptor	1. Personnel inadequately trained 2. Inadequate materials/tools	Inaccurate volume Inaccurate dose distribution	1.63	8.63	2.75	38.54	Artifacts in images. Re‐imaging may be needed sometimes.
18	IV contrast administration	Contrast not used when needed	1. Inadequate communication from MD to sim staff 2. General procedures not clearly documented	Target volume not detected	2.13	4.38	4.25	39.51	Imaging contrast is not used appropriately. May cause inaccurate definition of the treatment volume.
19	Imaging protocol selection	Not enough imaging series	1. General procedures (e.g. all patients of a particular type should have certain series) not clearly documented 2. Miscommunication	Wrong target and/or OAR volume	1.75	6.38	2.38	26.50	Whole head study is not acquired, makes it difficult to check patient skull definition. Volumetric study is not acquired when needed, cause difficulty in target and/or critical organ delineation. Angio study is not acquired when needed.
20	Imaging protocol selection	Wrong scan protocol used (e.g. wrong slice thickness)	1. General procedures not documented 2. Miscommunication	Inaccurate volume inaccurate dose	2.38	5.38	2.63	33.51	Compromised CTV/OAR delineation and inaccurate dose delivery.
21	Imaging protocol selection	Patient move too much during scan	1. Poor communication to the patient 2. Imaging process too long	Inaccurate dose distribution	3.00	3.25	2.25	21.94	Patient may need to be rescanned.
22	Images transferred to TPS	Acquired image set accidentally deleted	1. Inattention to details 2. Inadequate training 3. Inadequate backup procedures	Inconvenience (staff and patient)	2.50	3.63	1.75	15.86	Imaging needs to be repeated. Inefficient imaging process.
23	Images transferred to TPS	Incorrect image data set associated with patient	1. Pull up wrong patient's record 2. Inadequate training	Very wrong volume Very wrong dose distribution	1.88	4.88	3.00	27.42	Patient treated on the basis of another patient's volumes.
24	Images transferred to TPS	Dicom communicating failure	1. Inadequate commissioning and acceptance testing 2. Limitations of treatment planning systems or scanners	Very wrong dose distribution Very wrong volume	1.38	7.63	2.25	23.59	A dicom communication failure causes corrupted files imported to the treatment planning system. Patient treated based on wrong information.
25	Images transferred to TPS	Can not transfer image via network	Defective materials (software or hardware)	Inconvenience Delayed treatment	1.88	6.00	1.50	16.88	Network problem not identified and fixed before a treatment. Images can not be transferred from a scanner to a treatment planning system. CD ROM may need to be used in some cases.
26	Machine daily QA	Staff not available	Miscommunication	Inconvenience Patient uncomfort Delayed treatment	3.00	1.63	2.63	12.80	After imaging, it has become clear that no or not enough coverage from the radiation oncology team is available. A patient treatment has to be cancelled after frame placement. A patient has to wait excessively long for a treatment.
27	Machine daily QA	Undetected console computer malfunctioning	1. Lack of standardized procedures 2. Personnel omission	Min: Inconvenience and delayed treatment Max: Very wrong dose	1.88	4.13	2.38	18.37	The treatment console computer is not working properly (timer, clock, system configuration etc.) but not identified during the morning QA process. Wrong clock or system configuration would prevent the system to start a treatment. Wrong timer would cause a wrong dose delivery.
28	Machine daily QA	Undetected mechanical failure	1. Lack of standardized procedures. 2. Human error	Inconvenience Delayed or partially completed treatment	2.50	5.38	3.63	48.71	There is a mechanical failure in the treatment unit (APS problem, helmet micro switch problem, helmet hoist malfunctioning, sector switch malfunctioning, couch problem etc.) but not identified during the morning QA process. Treatment can not start or may have to be stopped in the middle.
29	Create patient in TPS	Wrong category (new patient/new examination)	1. Personnel omission 2. Inadequate training	Min: Confusion and inconvenience Max: Very wrong dose	3.25	3.88	3.00	37.78	A patient underwent previous Gamma knife treatment is created as a new patient. The previous treatments could be ignored during the planning process.
30	Generate skull geometry	Wrong numbers (24 point bubble measurement)	Human error	Inaccurate dose distribution	2.13	4.00	2.75	23.38	Bubble measurement was done correctly but wrong numbers were put into the planning system. This is not noticed during the final check. Dose calculation is based on inaccurate patient skull geometry.
31	Generate skull geometry	Wrong numbers (Pin or post measurement)	Human error	Inconvenience Patient injury	3.25	4.38	2.13	30.21	Post or pin measurement is not put in correctly. A collision is not detected by the planning system. A collision with a post or pin could stop the treatment or injure the patient on treatment.
32	Import images	Images of another patient imported	Personnel omission	Very wrong dose	1.63	5.88	3.25	31.03	Images of a wrong patient were imported. The images of the two patients are similar, so the error was not discovered during the planning process. Treatment is based on wrong image sets.
33	Import images	Images of another patient imported	Personnel omission	Inconvenience	1.88	1.88	1.50	5.27	Images of a wrong patient were imported. The error was identified during the planning process. Treatment planning has to start over.
34	Import images	Incomplete image sets or image series	Personnel omission	Inconvenience	2.38	4.63	4.13	45.31	Miss an image set, i.e., forget to import T2 volumetric study. Start to import images before the image transfer is complete. Only part of the image set is imported. This can usually be found out during the planning process.
35	Image definition/levelling	Large definition error not noticed	1. Inadequate training 2. Personnel omission	Inaccurate volume Inaccurate dose distribution	3.38	4.00	3.25	43.88	The definition error for an image series is too big, either because of the use of bad equipment/protocol or excessive patient movement during the imaging process. Imaging needs to be repeated for optimal result if the large definition error is noticed.
36	Image definition/levelling	Mismatch between the bubble/image not noticed	1. Inadequate training 2. Personnel omission	Inaccurate dose distribution	3.13	3.75	2.00	23.44	The skull shape from the bubble measurement and the images don't match very well. Usually bubble measurement needs to be repeated.
37	Image definition/levelling	Wrong patient orientation information	1. Inadequate equipment commissioning 3. User Error	Wrong volume Suboptimal plan	1.75	4.63	2.88	23.27	This happens most often with the definition of Angio images. The frontal/back, left/right marks were not put in correctly.
38	Image definition/levelling	Inadequate leveling	Inadequate training	Wrong volumes	2.63	6.50	3.50	59.72	Images are not levelled appropriately. The boundary of a treatment area is blurred. Certain areas are overlooked because of poor image contrast.
39	Image fusion/registration	Can not get needed image sets	1. Inadequate equipment commissioning 2. Inadequate communication	Suboptimal plan	3.25	2.63	2.13	18.13	A pre‐op/post‐op image set is needed to evaluate the progression of a previous treated (radiosurgery or surgery) disease site but there are problems transferring the images from certain image server.
40	Import previous examinations	Wrong number of blue circles	1. Inadequate training 2. Personnel omission	Very wrong volumes Very wrong dose delivery	2.63	4.75	3.25	40.52	Forget to import one of the previous examinations. Forget to turn on some of the blue circles. Cause a previously treated area with good control to be considered as a new area and treated again.
41	Image skull definition	Skull not clean	1. Inadequate training 2. Personnel omission	Wrong dose distribution	2.88	3.88	2.50	27.85	The skull shape from an image skull definition is not manually edited properly. There are larger errors in certain areas of the skull contour.
42	Create treatment plan	Incorrect machine configuration	1. Inadequate commissioning 2. Personnel omission	Very wrong dose Very wrong dose distribution	1.63	3.13	3.50	17.77	Problems with the configuration of the treatment unit in the planning system. A wrong Gamma Knife unit is chosen (i.e., 4C vs Perfexion).
43	Create treatment plan	Incorrect algorithm configuration	1. Inadequate commissioning 2. Miscommunication 3. Personnel omission	Very wrong dose Very wrong dose distribution	2.75	3.00	2.00	16.50	Problem with the configuration of a dose calculation algorithm (wrong output factor, wrong CT density curve etc.).
44	Matrix definition/dose prescription	Wrong location of the targets	1. Inadequate training 2. Poor communication 3. Personnel omission	Very wrong volume Very wrong dose	2.50	3.75	2.38	22.27	Identification of the target area is not correct. Blood vessels, edemas are considered as tumors. Functional target left/right messed up, wrong location etc.
45	Matrix definition/dose prescription	Overlooked targets/target area	1. Inadequate training 2. Poor communication 3. Personnel omission	Ineffective treatment	1.75	4.50	2.63	20.67	Tumors or portions of tumors are not treated because of personnel omission or poor image quality.
46	Matrix definition/dose prescription	Inadequate prescription dose	1. Lack of standardized procedures. 2. Inadequate training 3. Poor communication	Very wrong dose	3.00	4.63	2.50	34.69	An inadequate prescription dose is used for a particular type of disease, depending on the experience of the planners, and/or the information gathered about the patient before the treatment.
47	Matrix definition/dose prescription	Inadequate matrix size/position	1. Lack of standardized procedures. 2. Inadequate training 3. Personnel omission	Inaccurate dose calculation	3.25	3.88	2.38	29.91	A dose calculation matrix is too sparse or not centered well. Dose calculation for this target is not accurate enough.
48	Shot placement & adjustment	Error in dose calc	Software error	Wrong dose distribution	1.38	6.50	1.63	14.52	Software cannot calculate dose correctly because of a computer problem. This can be checked by a secondary dose calculation program.
49	Contouring	Wrong organ, wrong site	1. Inadequate training 2. Personnel omission	Very wrong volumes	2.75	3.88	1.88	19.98	Wrong target volume or critical organ contours lead directly to very wrong dose distributions and volumes.
50	Contouring	Poorly drawn contours (spikes, sloppy, etc.)	Inattention	Suboptimal plan (worst case wrong dose distribution)	2.38	4.38	2.13	22.08	Drawings generally correct but have inappropriate spikes, sharp corners, etc. Plan evaluation is based on the imperfect contours. Wrong dose distribution in worst case.
51	Contouring/plan evaluation	Needed contour not drawn	1. Inadequate training 2. Inattention, lack of time 3. Failure to review own work	Suboptimal plan (worst case wrong dose distribution) Inconvenience	4.88	2.63	2.75	35.19	A critical structure is not drawn when needed. The structure is not included in plan evaluation. The plan is not optimal. Make it difficult to make clinical decisions for future treatments. Examples include skin dose is not evaluated, optical structure doses are not documented.
52	Plan evaluation	Inadequate evaluation	1. Not enough time/effort spent 2. Inadequate training 3. Poor evaluation strategy	Wrong dose Wrong dose distribution	1.63	6.88	2.38	26.53	One needs to look at DVH and dose distribution on a slice by slice basis. But because of poor training or not spending enough time to evaluate the DVH or isodose distribution, the result can be an inadequate evaluation of the plan.
53	Plan approval	Wrong patient	Inattention	Very wrong dose Very wrong dose distribution	1.75	6.50	4.13	46.92	Exported a treatment plan for a wrong patient. This happens rarely and can be found out in most of the cases.
54	Plan approval	Wrong plan approved	1. Miscommunication 2. Inattention 3. Inadequate procedure	Very wrong dose Very wrong dose distribution	2.25	5.88	3.63	47.92	Exported one of the trial plans rather than the final plan.
55	Plan approval	Matrix size not fine‐tuned	1. Inadequate procedure 2. Procedure not followed 3. Personnel omission	Inaccurate dose distribution	1.75	4.63	1.88	15.18	The sizes of the matrices are not adjusted appropriately before the plan is approved, resulting in inaccurate dose calculations.
56	Plan approval	Collision check not performed	1. Inattention 2. Lack of procedure 3. Procedure not followed	Inconvenience Patient injury	1.88	4.88	1.75	16.00	There is a collision between the helmet/cap and the patient skull, frame post/pin. Treatment may be interrupted and replan maybe needed. A collision with a pin or skull could injury the patient as well.
57	Plan approval	Setup not checked	1. Inattention 2. Lack of procedure 3. Procedure not followed	Inconvenience Suboptimal plan	2.63	5.00	2.50	32.81	Setup/shot summary is not checked. The exported plan contains an unwanted Gamma angle or helmet change, a shot with very short treatment time etc.
58	Chart finish‐up	Not signed appropriately	1. Inattention 2. Miscommunication	Inconvenience	3.25	2.25	1.75	12.80	The hard copy of the treatment plan is not signed appropriately. It could also mean that the final plan has not been checked and agreed by all planners in some cases.
59	Chart finish‐up	Unnoticed plugged pattern	1. Miscommunication 2. Lack of standard procedures	Very wrong dose Very wrong dose distribution Very wrong volume	1.88	5.13	2.13	20.42	A plug pattern is used on 4C but the hard copy does not show it, either because of an incorrect printout selection or a breakdown of the printer. Treatment proceeded without the plug pattern.
60	Secondary dosimetry check	Undetected planning computer failure	1. Lack of procedure 2. Inadequate procedure 3. Procedure not followed	Very wrong dose Very wrong dose distribution Very wrong volume	1.50	7.38	3.13	34.57	A failure in the treatment planning system gives an erroneous delivery time but this not caught by the second check. Or the secondary check is simply not done.
61	Plan export for treatment	Wrong plan exported	Personnel omission	Very wrong dose; Very wrong dose distribution Very wrong volume	1.13	9.13	1.75	17.96	This is highly unlikely for the recent versions of the planning system because only an approved plan can be exported. There are not many reasons to have several outstanding approved plans at a time in the Gamma Knife planning system.
62	Load plan in treatment console	Communication failure	1. Cable pulled out accidentally 2. Bad wire 3. Network problem	Inconvenience	2.13	1.88	2.63	10.46	The exported plan can not get to the treatment console computer. Treatment can not start. The communication between the planning computers and the treatment console computer is usually not checked during the morning QA.
63	Load plan in treatment console	Wrong plan loaded	Personnel omission	Very wrong everything	1.13	7.00	3.50	27.56	Load a plan for a different patient. This is also highly unlikely because there are not many reasons to keep more than one plan in the treatment queue.
64	Patient ID check	Incorrect patient in the room	1. Lack of standard procedures 2. Poorly trained personnel	Very wrong everything	1.88	3.75	2.50	17.58	Inadequate check.
65	Frame adapter attachment	Frame adapter not attached properly	1. Inadequate training/orientation 2. Personnel omission	Inconvenience Inaccurate dose delivery	2.75	7.75	5.75	122.55	On the Perfexion, the frame adaptor should be attached properly. Failure to do so may result in an interruption in the treatment process or an imperfect dose delivery. Elekta has a field notice about this.
66	Patient positioning	Couch vertical position not ideal	1. Lack of standard procedures 2. Inattention to detail. 3. Poorly trained personnel	Patient uncomfortable	2.25	3.75	3.50	29.53	The couch vertical position should be optimized so that the patient neck does not get stretched much. This is important for long treatments. The couch vertical position may need to be adjusted before each run. A different run means a different Gamma angle or tumors in a different area are being treated.
67	Lock patient	Incorrect patient setup	Personnel omission	Very wrong dose distribution	1.88	4.13	2.00	15.47	Patient setup with wrong stereotactic coordinates in the trunnion mode. Patient setup at a wrong Gamma angle. On the Perfexion treatment can not start with a wrong Gamma angle.
68	Lock patient	Patient head position not firmly fixed	1. Bad equipment 2. Personnel omission	Wrong dose distribution	2.63	3.63	2.50	23.79	It is good practice to check if the patient head is fixed properly before a treatment. Equipment failure, untighten or loose screws may cause unsecured patient head position. This is more important for the trunnion mode on the 4C.
69	Clearance test	Undetected collision	1. Personnel omission 2. Procedure not followed	Interrupted treatment Suboptimal plan Inconvenience Patient injury	1.50	6.50	1.75	17.06	A necessary clearance test is not performed or not performed properly. A treatment is interrupted because of a collision. Replan is needed. Patient may get injured.
70	Final check	Stuff on the couch	1. Personnel omission 2. Lack of standard procedure 3. Procedure not followed	Interrupted treatment	2.50	5.38	2.88	38.63	Pillows, blankets, screw drivers etc. are left on the couch. Treatment interrupted.
71	Final check	Tubes/cables not long enough	1. Personnel omission 2. Lack of standard procedure	Interrupted treatment Patient uncomfortable	2.88	2.75	3.50	27.67	A tube/wire (oxygen, pulse monitor, anesthesia equipment etc.) is attached to a patient. The tube/wire gets pulled when the couch moves into the treatment position. Treatment interrupted.
72	Final check	Wrong plug pattern	1. Personnel omission 2. Lack of double check	Wrong dose distribution	1.75	9.00	2.88	45.28	The plug pattern is not put in correctly on 4C.
73	Final check	Inserts/Plugs not secured	1. Personnel omission 2. Lack of double check	Interrupted treatment	3.25	5.13	3.63	60.38	An insert or plug is not firmly attached to a helmet. The insert/plug drops during a treatment. Cause the treatment to stop.
74	Final check	Bolus not used when needed.	1. Inadequate training 2. Inadequate procedure 3. Procedure not followed	Wrong or very wrong dose distribution	3.00	3.88	2.50	29.06	Bolus should be used for the treatment of superficial lesions. The surface portion of the lesion may be underdose if a bolus is not applied.
75	Beam on	Can not start treatment	1. Lack of training 2. Personnel omission	Delayed treatment	2.25	2.88	2.38	15.36	All the interlocks (couch release handle, side pieces, emergency buttons) has to be cleared before the beam‐on button lightens up. A patient is on the table but there is an un‐cleared interlock.
76	Beam on	Patient arms collide with machine	1. Lack of communication with patient 2. Inattention	Interrupted treatment Patient uncomfortable or injury	1.50	6.25	3.38	31.64	This happens most often when a restless or confused patient is being moved in or out of the treatment unit.
77	Treatment log book	Forget to record start/end time	Inattention	No effect on patient Inconvenience	2.50	2.25	2.25	12.66	Incomplete documentation.
78	Treatment monitoring	Treated shots not recorded correctly	1. Lack of standard procedure 2. Procedure not followed 3. Personnel omission	Very wrong dose Very wrong dose distribution	2.00	5.50	2.50	27.50	Delivered shots are not followed closely and recorded correctly on the treatment copy. Cause confusion in the shots that have been delivered. This happens more likely on 4C with trunnion shots.
79	Treatment monitoring	Patient hand reaches up	1. Lack of communication with patient 2. Inattention	Inaccurate dose delivery Interrupted treatment	3.00	2.63	3.38	26.58	Patient hands in the beam pathways. Cause underdose from some beam angles. Patient hand touch the source cap. Cause the treatment to be interrupted.
80	Treatment monitoring	Inappropriate sedation	Lack of experience/training	Patient comfort	1.88	6.75	1.88	23.73	A patient is overly sedated. Cause the treatment to be interrupted.
81	Treatment monitoring	Anesthesia problem	Lack of experience/training	Interrupted treatment	2.75	2.75	1.75	13.23	A patient is treated under general anesthesia. But an anesthesia problem cause a treatment to be interrupted.
82	Treatment monitoring	Poor choice of treatment sequence	Poor user choice	Ineffective treatment process	2.00	4.88	2.00	19.50	On 4C, when multiple helmets, multiple Gamma angles, high/low docking are involved, it is necessary to figure out a treatment sequence that minimizes the treatment time.
83	Treatment monitoring	Manmade treatment interruption	Inattention	Interrupted treatment	2.00	1.63	3.13	10.16	During the course of a treatment, someone accidently press the power button, door open, emergency, pause button, or spill liquids on the treatment console computer, cause the treatment to be interrupted.
84	Treatment monitoring	Unfinished treatment	Personnel omission	Very wrong dose	2.38	4.63	2.63	28.83	A treatment run is completely ignored.
85	Follow up phone calls	Unaddressed patient concerns	Inadequate procedure Personnel omission	Patient comfort	1.88	1.75	1.75	5.74	Patient has question to ask after Gamma Knife treatment, but is not answered appropriately.
86	Follow up scan	Follow up scan not at the right time	Miscommunication to patient	Patient care	2.50	1.88	1.63	7.62	May cause delay in further treatment.