On this page:

National Health Demonstration Projects - Evaluation Task Group Review October - December 2003: Final Report

« Previous | Contents |

Listen

ANNEX A THE COLORECTAL CANCER SCREENING PILOT - the Fourth Health Demonstration Project
  1. Background
  2. Colorectal cancer is the second commonest cause of cancer death in the UK, with over 3000 new cases and 1700 deaths each year in Scotland.

    Three population based randomised trials 1,2,3 have shown conclusively that screening using faecal occult blood testing ( FOBt) can reduce death rates from colorectal cancer by about 20%. However, such trials are tightly controlled studies carried out by highly motivated individuals and there was some doubt as to whether similar results could be obtained in a routine setting within the NHS.

  3. Developing the Pilot
  4. The National Screening Committee ( NSC) held two workshops in 1997 and 1998 to canvas a wide range of expert opinion and critically review the available evidence. It was recommended that the feasibility of a national screening programme should be tested before committing the UK Health Departments to such a programme. To do this, a demonstration project ( DP) based on two populations of about one million each, in different parts of the country was proposed. An outline specification for the project was agreed - including the target age range (all 50-69 year olds) and the technique to be used (biennial FOB testing with colonoscopy as the follow up investigation for those testing positive).

    When this national specification had been approved, all UK Health Authorities were invited to submit expressions of interest in hosting a pilot site. 5 proposals were shortlisted for further consideration by an assessment panel appointed by NSC. Members of this panel visited these sites and, in early 1999, the final selection was made after presentations and interviews.

    The sites chosen for the pilot were North East Scotland (Grampian, Tayside and Fife) and Coventry and Warwick in England. An Executive group was set up to oversee the operational development and running of the pilot, comprising the National Screening Coordinators for Scotland and England, their central project managers, the two clinical directors and the two pilot managers. A UK steering group was also set up to formulate general policy.

    Working groups were set up to develop the specification for the computer system, to select the FOB test, to develop data sets, protocols, quality assurance systems and information materials.

    The first round of screening began in March 2000. FOBt kits were sent by post to the target population, with instructions to complete the test and to send it back to the screening centres in Dundee and Coventry for analysis.

  5. Evaluation Model and Methods
  6. 3.1 External evaluation

    Independent evaluation was an essential part of the project and a single evaluation covering both the Scottish and English arms of the pilot was agreed. This was put out to tender by the R&D Directorate of the Department of Health in London. On the basis of competitive bids, the evaluation was awarded to a group from the Universities of Edinburgh, Warwick and Essex.

    The evaluation covered many aspects of the pilot: performance against benchmarks, the impact on primary and secondary care services, uptake and the reasons for non - participation, physical and psychological morbidity associated with screening, and an economic evaluation.

    The evaluation was complex and required a multi-disciplinary approach, including epidemiology, health services management, psychology, health economics, primary and secondary care. The evaluation team were located at three sites, and needed to interact with the two pilots and with individuals in the Scottish Executive and the Department of Health.

    Developing a carefully managed relationship with the pilot sites was essential. The evaluation had to be independent and unbiased, yet good communication with the pilot sites was essential - e.g. to ensure agreement on definitions and on quality standards, and to coordinate the evaluation surveys with the screening programme. For these reasons, the evaluators developed a 'Terms of Engagement' document which aimed to make explicit the competing imperatives of independence and integration. This was found useful by all concerned.

    In February 2003, the Evaluation Group produced an independent report for the Department of Health R&D Directorate on the outcomes of the pilot, with the aim of informing a decision on a national programme.

    3.2 Internal monitoring/evaluation

    Initially, insufficient priority and resources were given to developing the internal monitoring capacity. There was an implicit assumption that the external evaluation would address all the issues. When this was identified as a priority, there were some problems with timely access to the data and ensuring the IT system could provide the necessary information.

  7. Results of the Evaluation
  8. A decision on whether to continue the pilot for a second round was needed during 2002, well before the results of the evaluation were available. A second round of screening began in Scotland in December 2002.

    The final report of the evaluation group was submitted to the Department of Health in February 2003. It showed that the UK pilot had demonstrated that key parameters of test and programme performance observed in randomised studies of FOBT screening can be repeated in population based pilot studies. The cost effectiveness also compared well with other forms of cancer screening.

    The independent external evaluation has not been continued to cover the second round of screening. The internal monitoring capacity has been enhanced to address this.

    In July 2003, the Scottish Executive gave a commitment in principle to roll out a national colorectal screening programme, while recognising that planning for its introduction could take up to 5 years.

  9. Key Issues
  • As a first stage in developing the demonstration project, national workshops were held to canvas expert opinion and critically review the literature. A detailed specification setting out core elements of the project, with supporting documentation, was then agreed nationally and subsequently put out to tender. Decisions on the successful proposals were made by an expert group, on the basis of detailed plans for implementation. This model differed significantly from that used for the other three demonstration projects.
  • There was a mismatch in timescales resulting in decisions about the future of the project being needed before the results of the independent evaluation were available. Similar difficulties were faced by the other three demonstration projects.
  • The external evaluation team recognised the need for a balance between independence from and engagement with the project team. This issue was shared by the other demonstration projects and the right balance will vary depending on the circumstances. The need for explicit 'rules of engagement' is important and recognised by all participants.
  • The importance of adequately resourced internal monitoring systems was not initially identified as a priority, but was subsequently addressed successfully. This was also an issue for the other demonstration projects.
  • The external evaluation was concluded with the final report of the team in February 2003. Evaluation of subsequent screening rounds of the project will be mainly through internal monitoring systems. Specific focussed research proposals are also being developed, seeking external funding through the usual channels.

« Previous | Contents |

Page updated: Thursday, April 7, 2005