The Million Women Study (MWS) is a large UK study of women's health. We have collected information on factors such as smoking and obesity through postal questionnaires, and on deaths, cancers and hospital admissions through linkage to NHS databases. We use this information to study risk factors for diseases such as cancer and stroke. However, we have limited information on drug prescribing, and on diseases managed in outpatient and primary care settings. General practice records are the most comprehensive source of such information, as GPs hold the full health record for NHS patients.
For this proposal, we plan to use linked CPRD records for over 102,000 women in our study to help our investigations of fractures, arthritis, and carpal tunnel syndrome (a common condition affecting nerves in the wrist and hand). We will compare the CPRD information with that already held by MWS and use it to improve our understanding of how factors such as obesity and physical exercise affect the risk of these conditions.
Women in the Million Women Study gave written consent to follow-up through medical records. We have extensive experience of using CPRD data, and the proposed study is approved by our NHS ethics committee.
In order to inform risk factor analyses within the Million Women Study cohort, we wish to continue using the linked CPRD-Million Women Study dataset provided to us in 2014 under Protocol 12-070R3.
The aim is to use information on primary care and outpatient activity in CPRD to supplement available study data (self-reported, medical records, and linked coded hospital admissions, death and cancer follow-up data provided to us through HSCIC). Using basic tabulation and description of data for comparisons, we will assess the reliability and completeness of coded NHS data (such as HES) for diagnostic outcomes, and of self-reported prescribing data for exposures. This new proposal deals with musculoskeletal outcomes (fracture, arthritis, joint replacement, carpal tunnel syndrome).
Large UK cohorts benefit from the availability of long term, cost effective, unbiased follow-up through linkage to NHS records: it is important to assess the reliability of such records for large-scale epidemiology, and to make the results available to others. There is very limited existing information. Our policy is to publish our results as methodological papers, or with adequate key word notification as part of analysis papers, and to ensure that data providers are kept informed.
Bone fractures Carpal tunnel syndrome Hip and knee replacement Osteoarthritis of base of thumb
Jane Green - Chief Investigator - University of Oxford
Dominic Furniss - Collaborator - University of Oxford
Jennifer Lane - Collaborator - University of Oxford
Lucy Wright - Collaborator - University of Oxford
Md Shajedur Rahman Shawon - Collaborator - University of Oxford
Owen (Tien Yu) Yang - Collaborator - University of Oxford
Valerie Beral - Collaborator - University of Oxford