© 2010 Brighton Collaboration
The general practice research database (GPRD) collates the computerised medical records of general practitioners (GPs) in the UK who act as the gatekeepers of healthcare and maintain patients’ electronic health records. As such they are responsible for primary healthcare and specialist referrals, and they also store information stemming from the specialist referrals, laboratory tests, and hospitalizations. The data recorded in the GPRD include demographic information, prescription details, clinical events, preventive care, specialist referrals, hospital admissions, and major outcomes, including death. The majority of the data are coded in Read Codes; however, additional ‘free text’ data is also available at an extra cost, which can improve the sensitivity and specificity of data. Validation of data with paper records (including specialist letters and laboratory test results) is also available.
Importantly, GPRD operates a continuous quality control procedure that highlights which practices are considered to be “up-to-standard” for the purposes of research, (UTS), are included in the research dataset. The dataset is comparable with the UK population with respect to its age and sex distribution as well as its distribution of socioeconomic class and national geographic coverage. There are currently over 550 peer reviewed publications using GPRD data (www.gprd.com)
There are 479 currently active practices, of which 426 are currently reported to be contributing data that are considered to be UTS, which represents slightly over 5% of UK practices. There are 3.48 million currently active patients in the 426 practices, representing 29.8 million person-years (this excludes patients who are registered only as temporary patients). The research-usable population for a 10-year retrospective study is 6.17 million persons, representing 40.3 million person-years.
Data include demographics, all GP/healthcare professional consultations (phone, letter, email, in surgery, at home) that lead to a new diagnosis, diagnoses and symptoms that lead to a first prescription of any medicine or to a hospital referral, laboratory test results, treatments, including virtually all prescriptions issued in the GP practice, referrals to other care, hospital discharge summary (date and from READ code), hospital clinic summary, preventive treatment and immunizations, death (date and cause).
For a study to be carried out using GPRD data, protocols need to have been approved after scientific review by ISAC, which is usually conducted within 21 days. Ethics approval is in place and covers all observational research. Approval to communicate to 3rd parties: 21 days. Fast track: couple of days.