SAIVMs | Scottish Audit of Intracranial Vascular Malformation Study

External Links

»NHS Scotland
»NHS National Audits

> SAIVMs > Protocol > Design

SAIVMs recruits a longitudinal, concurrent cohort of newly-diagnosed ('incident') participants, using multiple overlapping sources of case ascertainment, in order to give a representative picture of what happens to adults with IVMs across NHS Scotland (see appended flowchart).

Case ascertainment
Primary sources of case ascertainment are:
  • neuroradiologists at the four Scottish Neuroscience centres
  • radiologists working at other departments in Scotland, and hospitals in England occasionally serving the Scottish population, with CT and/or MR brain imaging
  • all neurologists, neurosurgeons, stroke physicians, neuropathologists, and relevant clinical nurse specialists in Scotland, and hospitals in England occasionally serving the Scottish population

Secondary source of case ascertainment: With the approval obtained from the Privacy Advisory Committee on 3rd August 2007, the Information Services Division (ISD) runs electronic searches of hospital discharge (SMR01) and death certificate data on a six-monthly basis, seeking adults meeting our inclusion criteria (we have evaluated the accuracy of the limited ICD-9 and ICD-10 codes available for IVMs, and found the only worthwhile codes to use are those for brain AVMs)

Notification of the diagnosis of a potentially eligible adult may be received by post, fax (to a designated NHS office, which is locked when unoccupied), email (to a secure email address), or via the SAIVMs website. The website uses Secure Sockets Layer (SSL) 128-bit encryption, which is the strongest encryption supported by the current web browsers, giving the highest level of protection for confidential transactions over the internet. The online notification form can only be accessed by collaborators, who receive a unique URL (incorporating their identity) in emails sent to them to remind them about recruitment to the register.

Opt out consent
Information for patients: Posters and information leaflets about SAIVMs are distributed to all brain imaging facilities and outpatient and inpatient areas served by neurologists, neurosurgeons, and stroke physicians throughout NHS Scotland. Therefore any patient (their family and carer) has access to information about SAIVMs at the point of diagnosis. The leaflet and poster are freely available on the SAIVMs website ( At the time of case validation (see 5.8 below), a SAIVMs leaflet is sent to every patientís consultant (who is usually a collaborator, participating in the national audit) for them to forward on to the patient in case the patient is not already aware of their inclusion in the register. Health Rights Information Scotland (HRIS) also provides leaflets to inform patients about the uses of their data for audit.

If an adult wishes to opt out of this audit, SAIVMs honours that request, and they are excluded from the register (see 4.3 above).

The Confidentiality and Security Advisory Group for Scotland (CSAGS) considered categories where implied (opt out) consent would be acceptable, and these include the objectives of SAIVMs, "planning, managing, funding and auditing, where identifiable data cannot be anonymised" and, "Multiple uses (if cannot be anonymised) e.g. registries, epidemiology, national data banks."1

Case validation
Contact with responsible clinicians: Four weeks after an adult's notification to SAIVMs, the register approaches the GP and hospital consultant by post to check the adult's demographic details and diagnosis. If these data are correct, confirming the adult's inclusion in SAIVMs, the register's administrator obtains and/or creates anonymised digital copies of their diagnostic CT scans, MR imaging, and catheter angiograms in DICOM format. At this stage, we also ask the GP and hospital consultant if the participant is still alive and aware of their diagnosis. We ask hospital consultants to inform patients of their diagnosis, if they have not been informed already, and to inform the patient of the passage of their details to this nationwide audit project.

Review of brain imaging: One or both of the register's neuroradiologists reviews each participant's anonymised diagnostic imaging via a secure web-based reviewing system, to further validate their IVM diagnosis. The digital images are stored on password-protected NHS computers and external hard drives (for back up), both in the Institute of Neurological Sciences in Glasgow and in the Department of Clinical Neurosciences in Edinburgh. Any hard copies of participants' brain imaging (from hospitals that have not provided imaging in digital format) are kept in a locked filing cabinet in Dr Jo Bhattacharya's office at the Institute of Neurological Sciences, Glasgow, or in a secure storage area in the Department of Clinical Neurosciences, Edinburgh.

Medical records review
Patient identifiable data are needed to obtain the correct sets of medical records.

SAIVMs staff with NHS contracts (or CRA authorisation) collect data on clinical presentation and outcome, as well as markers of the process of care, from GP and hospital medical records. Co-workers who are affiliated to SAIVMs (for example, medical students), are asked to sign assurances of confidentiality.

Stickers are put on these medical records, so that they are not destroyed and GPs/consultants are reminded to send SAIVMs copies of correspondence about the participant during follow-up.

Copies of medical records are stored in case notes that are only labelled with a patient's SAIVMS number, in locked filing cabinets, in a notes storage area secured by keypad access.

SAIVMs sends each participant's GP a questionnaire a few weeks before the anniversary of the participant's diagnosis to check the participant's demographic details and check they are still alive, to obtain information about any hospital admissions/investigations/appointments, and to obtain important outcome data (brain haemorrhage, epilepsy and the GP's rating of the participant's level of dependence on the Oxford Handicap Scale).

Every participant is followed-up via their GP and hospital case notes on an annual basis in order to collect data that are not reliably coded in ISD SMR01 hospital discharge data (e.g. epilepsy), or are not coded at all (e.g. outpatient attendances, prescriptions). These outcome data on process and outcome underpin SAIVMs as a high-quality clinical audit: survival and morbidity are two of the most meaningful outcomes for neurological disorders and their treatment, and dependence/disability are impossible to extract from the Scottish Morbidity Record (SMR01), which only codes the diagnoses listed on a discharge summary.

During the annual follow-up process, the GP becomes the principal point of contact about each participant because adults with IVMs may not be followed-up for long by their hospital consultant. GP follow-up is essential to document outcome for the whole cohort at uniform time points, especially because many patients are discharged from secondary care, and morbidity is a far more meaningful measure of outcome for neurological conditions than death.

The primary outcome in SAIVMs is a composite outcome of dependence/disability/death rated annually by the GP on the Oxford Handicap Scale. Other important outcomes are the occurrence and/or recurrence of intracranial haemorrhage, seizure(s) after presentation, and time to being 1 or 2 years seizure-free (for those patients with a history of seizure(s)).

All clinical and radiological data are stored in databases on secure servers in the Department of Clinical Neurosciences in Edinburgh; these databases are password protected with access controlled by allocating users restricted permissions. An anonymised extract of the radiological data is kept in a password-protected database in Dr Bhattacharya's office in the Institute of Neurological Sciences in Glasgow. Identifiable data are not shared with any external agency, but they are required for communication with medical records departments at hospitals where participants have been seen and with participants' GPs, in order to collect essential outcome data for the clinical audit.

1 Confidentiality and Security Advisory Group for Scotland (CSAGS). Protecting patient confidentiality. Scottish Executive Health Department, 2002.

Endorsed by: