« Previous | Contents | Next »
Listen
8. Summary of principal findings
8.1 The principal findings arising from investigation of this incident both by staff at the BOC and by the Inspector are summarized in the following paragraphs.
8.2 In May 2005 an upgrade to the existing computer package to the new Varis 7 system was installed at the Beatson Oncology Centre. This upgrade overcame previous technical and operational obstacles to allow the data calculated by the Eclipse treatment planning module within Varis 7 to be transferred electronically (rather than manually) to the RTChart module which verifies these treatment parameters prior to treatment delivery and records those set at delivery. To optimise the benefit of the change to electronic data transfer, the previous practice at the BOC whereby all treatment plans were computed using a standardized dose fraction of 1 Gray (100 centiGrays) was changed so that they were then calculated using the actual prescribed radiation dose per fraction.
8.3 This new facility has a number of potential advantages for planning efficiency and for patient safety. For example, the Royal College of Radiologists' Clinical Oncology Information Network 'Guideline for External Beam Radiotherapy' [2] recognizes that ' Manual transfer of data either from planning to treatment units or between treatment units is associated with a high risk of transcription errors' and recommends, therefore that ' The transfer of treatment data sets should be by local area IT network as far as possible'.
8.4 The 'whole CNS' procedure involved in this incident, comprises separate treatment fields for the head, the upper spine and the lower and spine. Of these, the spine fields are the more complex and it was decided at the BOC that these were not amenable to electronic data transfer within Varis 7. For the head fields separate electronic transfer of data would have been possible but for various reasons this was not pursued. Therefore, manual data transfer was retained for all elements of the whole CNS procedure.
8.5 Despite the retention of manual data transfer, the change in practice referred to above was applied, whereby treatment parameters for the head fields were calculated using the prescribed radiation dose per fraction instead of a standardized dose fraction of 100 centiGrays.
8.6 Therefore, the form of the data contained in the paper Treatment Plan Report from the Eclipse treatment planning module had changed. Specifically, whereas previously a specific entry on the Treatment Plan Report was always in units of 'Monitor Units per 100 centiGrays' the data on the Report for Miss Norris was the number of Monitor Units per treatment fraction. (The prescribed treatment fractions for Miss Norris were each of 175 centiGrays.)
8.7 Had the potential implications of this change been fully assessed then the potential for this difference in the form of the data output to introduce a critical error into manual planning might have been identified. However, no evidence has been presented of such an assessment having taken place.
8.8 Any such assessment would have necessitated a proper review of related written procedures and appropriate training for all staff. However, the last update of the BOC's Work Instruction number WI.14.01.01 for 'Medulla Planning' was in August 1998, and no evidence of any subsequent review of this document has been presented. This is contrary to the requirement of BOC quality system document QS 03 'Document and Data Control' which states that 'All controlled documents are reviewed at least annually…'.
8.9 The available training records for Planner B who undertook the bulk of the treatment planning give no indication of any formal competence in planning for the whole CNS procedure in question. However, he had been involved in planning of one treatment of this type since his training records had last been updated and this was considered by Principle Planner A to be sufficient to allow him then to plan similar treatments under supervision (in accordance with the provisions of Section 11(3) of the IR(ME) Regulations).
8.10 For this single previous experience, the prescribed radiation dose per treatment fraction happened to be 100 centiGrays. Therefore, the number of Monitor Units per treatment fraction written in the Treatment Plan Report from Eclipse was already in Monitor Units per 100 centiGrays and therefore would not have changed even if the normalization procedure had been applied.
8.11 Planner B (in common with the other planners at the BOC) therefore had no experience of dealing with the conversion from MU per treatment fraction to MU per 100 centiGrays that was required for completion of the Medulla Planning Form. Further, he had no access to any appropriately revised written procedures and, by common agreement among those interviewed, had received no training or instruction on this conversion. Planner B was therefore unaware of the need for this critical normalization step and omitted it.
8.12 The choice of Planner B as the main planner for a procedure of this complexity was contrary to the (albeit flexible) staffing structure for treatment planning that was in place at the BOC. This structure (Table 6.1 in this report) allocates planning duties for complex (Category E) plans to senior planning staff with appropriate levels of experience. The underlying reason given by the Principal Planner A for allocating this plan to Planner B was non-availability of experienced staff.
8.13 The supervision received by Planner B in planning the treatment for Miss Norris was indirect in the sense that, following initial instruction, he was left to complete the planning process but with the facility to raise any issues of difficulty with a senior colleague. The use of this form of supervision appears to have been a consequence of the relationship between staffing levels and workload in treatment planning. Had direct supervision been possible, there is a greater likelihood that the more experienced supervisor would have identified the error.
8.14 In addition to the critical error for the head fields, Planner B also made other errors for the spine fields which were identified and corrected on checking by senior colleagues.
8.15 Checking of the treatment plan by senior colleagues was not carried out according to procedure, in the sense that the principal checker (Principal Planner A) was also involved in supervision and as such should not have undertaken checking. The plan should have been independently checked, as required by BOC Procedure QS.14.13. Again the deficiency appears to have been a consequence of the relationship between staffing levels and workload planning. However, the fact that the checking process did identify a number of other deficiencies in the original plan, suggests that the procedure was correct and that checking of the spine fields had been carried out diligently. The presence of these errors reinforces the need for more direct supervision of plans being prepared by staff in training.
8.16 The error was repeated by Planner B subsequently in planning a similar treatment for a new patient and, again, the checker (in this case Senior Planner C) failed to identify the error. However, prior to any treatment being delivered, as a result of a separate enquiry by a treatment radiographer the error for this patient was identified during further checking by Senior Planner D. It was this discovery that led to investigation of previous treatment plans using the same technique and to identification of the error in the plan for Miss Norris. None of the treatment plans for other patients was affected.
8.17 This investigation has identified a number of concerns regarding the proper allocation of staff responsibilities and lack of a common understanding about who was responsible for what (Section 6.3 of this report).
8.18 There were deficiencies in the BOC's compliance with the IR(ME) Regulations 2000. The decision to use a planning mechanism which did not minimise the risk of error (Schedule 1k), and failure to show particular regard to the maintenance of written procedures and training records contributed to the risk of occurrence for this incident.
8.19 The parallels identified in Chapter 7 of this report suggest that had the lessons learned and recommendations made following the previous incident at North Staffordshire Royal Infirmary been properly addressed at the BOC then the risk of occurrence for this incident would have been reduced.
8.20 Failure to ensure that the introduction of new technologies and the implementation of related changes in working practices were supported by proper assessment and provision of full written procedures and training have impacted significantly on the causes of this incident.
8.21 The main factors that contributed to this error were therefore;
(i) delays in achieving full compliance with the IR(ME) Regulations,
(ii) failure to assess fully and address both the risks and the potential improvements to patient safety following the introduction of new technologies including CT imaging, computer planning and the Varis 7 upgrade,
(iii) failure to keep written procedures and training records properly up to date,
(iv) inappropriate over-reliance on the limited experience of Planner B,
(v) failure to provide direct supervision in treatment planning,
(vi) failure to provide fully independent checking of treatment plans,
(vii) lack of written statements and of common understandings about individual responsibilities,
(viii) failure to address the lessons and recommendations from previous incidents at other radiotherapy centres.
8.22 An underlying cause of most these deficiencies is an evident insufficiency in staff resources which might be attributed to a number of factors including;
(i) continuing expansion of the service,
(ii) difficulties in recruiting and retaining experienced staff,
(iii) consequent implications of internal promotion of staff.
8.23 The result was that an overriding culture existed such that changes in practice could be made without proper assessment, evaluation and documentation. This culture was founded on reliance on a few well respected, experienced staff who had been employed at BOC for a number of years.
8.24 The BOC, both as an immediate consequence of the incident and in response to the recommendation made in the incident report by Dr Martin, has introduced a number of procedural changes aimed at minimizing the possibility of recurrence of any similar occurrences. However, further attention to quality system working is required to ensure that systems of work at the BOC are conducive with the appropriate level of patient safety (Section 9.4 of this report).
« Previous | Contents | Next »