The importance of robust data to support clinical care was recognised in the Scottish Diabetes Framework(1) which included 'IM&T and Diabetes Registers' as one of the first stage priorities. In addition, the clinical standards for diabetes care in Scotland developed by NHS Quality Improvement Scotland (formerly the Clinical Standards Board for Scotland) included as the first standard the requirement for areas to have " an up to date electronic population clinical management system of all people with a recorded diagnosis of diabetes in the area". (2)
An agreed dataset provides a 'common currency' to ensure consistency of definitions and allow data to be shared between different IT systems. The significance of this has been acknowledged by the diabetes community in Scotland for a number of years and concerted efforts have been made to develop an appropriate dataset since the mid-1990s. In March 1996, the Scottish Intercollegiate Guidelines Network (SIGN) published "a recommended minimum data set for collection in diabetic patients" (SIGN 4). (3) This was updated in June 1998 (SIGN 25). (4) Recognising that the SIGN minimum dataset focused largely on outcomes of diabetes care, a CRAG Working Group on IT to Support Shared Care in Diabetes extended the dataset to include items for clinical management. This dataset was published in September 2000. (5) Building upon this work, the SCI-DC team, in conjunction with ISD, developed the Scottish Diabetes Core Dataset which was published at the start of 2003. (6)
The Diabetes Registers and Diabetes IT Systems Steering Group, (a CRAG working group which reported in 1999) recommended the establishment of a national diabetes register. The CRAG Working Group on IT to Support Shared Care in Diabetes, which followed in 2000, strongly supported this proposal both as a mechanism to monitor progress against the St Vincent targets (7) and also as a spur to the establishment of local diabetes registers. The work of these groups led, in September 2000, to the Scottish Executive issuing a Health Department Letter-HDL (2000)12 - Scottish Diabetes Survey. (8) This stated:-
"The Scottish Executive remains committed to improving the health of patients with diabetes. One of the key steps in achieving this objective is to improve the availability of data, particularly as many of the complications of diabetes can be prevented or delayed by effective monitoring of diabetic patients. This Circular outlines plans to compile a national picture of diabetes in Scotland through the central collation of information on diabetic patients and sets out the actions required by Health Boards, Trusts and individual clinicians."
In order to oversee the development and to evaluate the output of the national survey a Scottish Diabetes Survey Monitoring Group was established; (the membership is detailed in Annex A). The Group's remit is to:
(a) | Monitor and provide advice on the establishment and development of the Scottish Diabetes Survey. Comment on the quality of (i) the data; and (ii) the systems used by Health Boards and Trusts to provide the data. |
(b) | Ensure that data submitted for inclusion in the national diabetes register conforms to all data protection and data security requirements. |
(c) | Evaluate, on the basis of the Scottish diabetes survey, the progress of Health Boards in delivering diabetes services to their population and monitor Scotland's performance in meeting the St Vincent Declaration targets. |
(d) | Report to the Chief Medical Officer. |
The Scottish Executive established the Scottish Diabetes Group in March 2002 to support and monitor the implementation of the Scottish Diabetes Framework. At this time, in order to ensure integration of national diabetes initiatives, the Scottish Diabetes Survey Monitoring Group became a subgroup of the Scottish Diabetes Group.
A 'provisional' Survey was undertaken in 2001 and a report published in November 2001. (9) This 2002 Survey demonstrates some significant improvements on 2001, but again the Survey represents 'work in progress'. The intention to implement effective IT as part of the SCI-DC programme (Scottish Care Information Diabetes Collaboration) is widely welcomed and will significantly ease with burden upon those charged with collating data for the Diabetes Survey.
The Purpose of the Scottish Diabetes Survey
The purpose of the survey was set out in HDL(2000)12:-
(a) | To improve patient care by encouraging better monitoring of diabetic patients in order to provide a more effective response to the complications of diabetes. |
(b) | To allow Scotland's progress towards achieving the St. Vincent Declaration Targets to be monitored and provide evidence to show whether or not Scotland is achieving the targets. |
(c) | To enable the standards of care of patients with diabetes to be monitored between Health Boards and over time. |
(d) | To allow analysis of the Scottish diabetic population by, for example, age or deprivation category. |
(e) | To provide data to support and encourage the implementation of best practice, for example as set out in SIGN guidelines. |
(f) | To provide data to inform clinical governance. |
The intention remains to undertake the survey on an annual basis. However, as the HDL noted:
"IT developments may provide more effective solutions than an annual snapshot survey and service developments may demand other approaches. Evaluating the technology, methodology and the content of the survey will be a part of the Scottish Diabetes Survey Monitoring Group's remit."(8)
It is worth underlining that a key intention of the Survey is to promote the development of local diabetes systems and that one of the most important roles of the Survey is to monitor progress towards this objective.
A key driver for the Scottish Diabetes Survey is the wish to support and encourage the establishment of effective local diabetes systems which work to improve patient care and assist service delivery. To this end, the national survey is envisaged as a collation of returns from local area registers. However, it is acknowledged that some Health Board areas have more developed systems than others. The expectation is that the survey will fill out over time, i.e. it may take a number of years before the survey is comprehensively populated. (8)
The HDL also highlighted that the short-term goal of delivering data for the Survey should not detract from the longer term goal of implementing an effective clinical management system. This should ensure that NHS Boards do not feel compelled to divert resources from the implementation of clinical systems in order to capture data manually for the Survey. However, it also means that during this transition phase, (until comprehensive clinical systems are in place and populated with data), the Scottish Diabetes Survey will be a measure of the completeness of coverage of effective IT at least as much as a measure of the outcomes of diabetes care. The encouraging progress of SCI-DC in the development and roll out of SCI-DC Clinical (a hospital clinic system) and SCI-DC Network (a population register) will help significantly to ensure that effective diabetes clinical management systems are available throughout Scotland.
Data confidentiality and consent issues 2002
The 2002 Survey has not been hampered by concerns or doubts about the context of consent because, in accordance with a decision made in 2001, the data are, once again, anonymised and aggregated with no raw data being submitted for analysis.
Since the publication of the 2001 Diabetes Survey, the Confidentiality and Security Advisory Group Scotland (CSAGS) has produced guidance in relation to consent. (10) In essence, implied consent is sufficient to share data for purposes of direct clinical care but further consent would be required to utilise named data beyond this boundary. There is an intention to develop a system of acceptable anonymisation (psueudonymisation) so that linkage analysis will remain possible but without the need to transmit identifiers. However, this mechanism has not been established. Until this is in place - and is proven to work satisfactorily - the Scottish Diabetes Survey will not collect individual data.
Figure 1: Map of Scotland showing NHS Board populations (000s)
