5.1 NHSScotland has a justifiable reputation for high quality data but there have been three specific areas of challenge for the Review:
- The systems that exist within Boards are appropriate and effective for the service's purposes but they have not been set up to answer the questions posed by the Review;
- The new approach saw the adoption of new data definitions for end-of-life and ultra-orphan medicines which limits before-and-after analyses;
- There are inconsistencies between Boards in relation to the data reported for IPTRs. This is the result of varying interpretation or application of data definitions and has been further complicated by a series of changes in the dataset gathered from Boards annually once again making comparisons across years difficult.
5.2 Within Boards data collection systems related to medicines largely support medicines ordering and prescribing but do not record outcomes. The systems do not allow tracking of the use of medicines in relation to patients. There are a number of local Board-level systems which capture elements of this information and there is a single chemotherapy system for NHSScotland but it is not used in the same way across the three cancer regions. The data challenges experienced by this Review would be addressed by a national electronic prescribing system which includes collection of data on outcomes and side effects.
5.3 Using existing systems some collation of data in relation to the number of times a specific medicine is prescribed is possible but for the purposes of the Review it has not been possible, with sufficient confidence, to equate the number of episodes where a specific medicine is prescribed to the number of patients treated. Similarly, the lack of a national electronic patient record prevents interrogation of data at the patient level. While it is possible to acquire this data from some Boards, a definitive national position cannot be described.
5.4 As stated above, the introduction of the new approach involved adopting new definitions for end-of-life and ultra-orphan medicines. The EMA definition for orphan had been in place since the inception of SMC and was kept. It has not been possible to retrospectively categorise medicines assessed under the previous process according to the new definitions and this has limited the Review's ability to make before-and-after comparisons. A further complication has been that some medicines can be categorised under more than one definition.
5.5 There are similar challenges with IPTR data and although Boards have been required to report on IPTRs to Scottish Government for several years this has been for the number of IPTRs received, the medicines involved and the decision reached. Data has not been requested on the condition being treated. The required dataset for reporting has been modified several times over the years and there has not been a requirement to break IPTRs down according to the end-of-life, orphan and ultra-orphan definitions. This has prevented year-by year comparisons other than at the highest level and limited the analysis of data to IPTRs for all medicines and not specifically for the categories which are the subject of the Review.
5.6 Going forward there would be benefit in developing datasets and collecting data which take account of outcomes. Outcomes comprise a number of eventualities including benefit accrued but also the stopping of treatment because of side effects or lack of response. There will be a significant role for Public Health Intelligence (PHI), formerly the Information Services Division (ISD), at NHS National Services Scotland (NSS) working with clinical teams and networks to agree and develop appropriate national datasets.
5.7 During the engagement process of the Review there was much discussion about the measurement of outcomes and this is considered in Sections 6.11 & 6.12 and paragraph 8.10.
5.8 A further limitation of the data is that it does not identify the number of patients who might be considered eligible for treatments but who have accepted SMC decisions and chosen not to pursue requests via IPTR or PACS. While the success rate for IPTR and PACS appears high this only reflects the experience of patients who have submitted requests.
5.9 A review of datasets and definitions and processes for collection and analysis would better prepare NHSScotland to meet the challenges that lie ahead. As discussed in Section 6.11, changes in the way that medicines are assessed for regulatory purposes mean that there will be a greater requirement to collect data in relation to individual episodes of treatment which allows assessment of outcomes including side effects. This will require a wider range of data including qualitative measures.
1 Develop, agree and implement national datasets and data definitions for end-of-life, orphan and ultra-orphan medicines and for IPTR/PACS processes. This will ensure that data from Boards is consistent and can be collated. This in turn will allow Boards' data to be used to support functions such as planning and resource allocation.
2 Develop, agree and implement a national chemotherapy dataset and equivalent datasets for medicines used to treat rare conditions
3 Develop, agree and implement sets of outcome measures for classes of medicines or, in the case of very rare conditions, specific medicines
4 Ensure that national systems being developed for electronic prescribing and electronic patient records are prioritised and support the above requirements
5 Establish a multi-agency taskforce or equivalent to report on data requirements to support the assessment and introduction of new medicines going forwards