A Population-based Ankylosing Spondylitis [PAS] cohort

Studying long term conditions can be difficult as a result of the timescale involved and the difficulty in capturing associated non-medical costs such as disability or lack of earnings due to illness. A great deal of information is already regularly collected on all people using the NHS. However, this information is held in many different NHS systems, such as with the general practitioner, their specialist, the hospital and laboratories. By linking this information, people with specific conditions can be followed through the health system. In addition, these people can also contribute directly to the data collected by completing questionnaires about the effects of their disease, such as function or pain. This information can be fed back to their usual doctor, thereby contributing to their personal care, while also improving the quality and value of the data collected. In this study we intend to apply these strategies of linking data relating to people living in Wales with ankylosing spondylitis, the second most common inflammatory arthritis. This data will be collected and analysed in a secure and anonymised manner to help address important research questions about this condition. For example, this linked information will help us to better understand the variations in this lifelong condition; help to predict early in the disease which patients will develop the most disabling disease requiring more aggressive therapy; facilitate estimates of the true cost of this condition, thereby speeding up assessments of expensive new treatments as they become available; and help to better plan the pathway of these patients through the health system. This study will facilitate genetic and biological studies, as well as trials of new medications for ankylosing spondylitis, thereby further improving understanding of this condition and attracting high quality research to the NHS. The methods used here build on existing resources in the UK and can also be applied to many other conditions in the future.

Ankylosing spondylitis (AS) is the second most common form of inflammatory arthritis. However, the natural history of AS is poorly understood with considerable variability in presentation and progression. In particular, there is a paucity of research on the cost of AS which has resulted in delayed NICE approval of new effective therapies for AS. The cost-effectiveness issue is further limited by the current inability to identify patients at risk of developing aggressive disease (who would benefit most from early aggressive therapy) and inability to predict which patients will fail to respond to these therapies. Identifying and recruiting AS patients, in a prompt and cost-effective manner, for genetic validation studies and clinical trials of new therapies is becoming increasingly difficult in the current fragmented NHS system. This proposal is for a population based cohort of people with AS, utilising disease-specific data (rheumatology database, radiology imaging) and patient-centred data (quality of life, work limitations, disease activity, function) linked with existing electronic data from disparate clinical, laboratory and administrative systems held by HIRU (GP records, out-patient and in-patient data, laboratory, A&E, social services datasets etc). All patients with AS (diagnosed by a rheumatologist) living in Wales will be approached to participate in this cohort, resulting in a well characterised cohort with robust clinical data at its base. The unique link with existing datasets will allow the cohort to be followed both prospectively and retrospectively. Therefore the cohort can be used to provide data on the natural history of AS, give an objective estimate of the cost of AS at each stage of the disease, identify early risk factors for severe disease, help identify potential interventions to prevent surgery and disability, in addition to facilitating patient identification and recruitment for genetic studies and clinical trials of new interventions. This study builds on existing UK resources and enhances the NHS?s ability to attract research. The cohort can also contribute to registers and post-marketing surveillance of new therapies. The use of routine data will ensure that this study will not suffer from the ?cohort effect? as all patients nationally can be included in the linking of routine data, and the use of rheumatology and GP clinical records will ensure the cohort will not date with time as it will include all new patients every year. Therefore this cohort, which can act as a pilot for other conditions, builds on and enhances existing UK resources and infrastructure.