The impact of guidelines on the use of imaging in osteoarthritis: a time trend analysis using electronic health record data

Abstract

The National Institute for Health and Care Excellence (NICE) suggests there is no role for routine radiography in the diagnosis of osteoarthritis (OA). It is not known how consistent this recommendation is across guidelines, or the impact of guidelines on the use of radiography in OA.
A systematic review of current guidelines identified 18 OA guidelines published between 1998 and 2019. Most recommended a clinical diagnosis of OA irrespective of joint site, although only three explicitly discouraged the routine use of plain radiographs to confirm the diagnosis of OA.
An analysis conducted from 2000-2015, on primary care electronic health record data from nine practices on the North Staffordshire CiPCA database, identified 23,784 patients with at least one recorded OA consultation. The highest annual rate of X-ray requests in patients consulting for OA was 31.8% in 2015. Those receiving an X-ray referral tended to be aged 55-64 years (adjusted OR=1.15 Ref: 45-54), consult more frequently (=10 times from 2000-2015 adjusted OR=11.69 Ref: 1-3 times) and registered to certain GP practices (for example Practice 2 adjusted OR=1.58 Ref: Practice 1).
A time-trend joinpoint analysis, restricted to the period 2000-2012, prior to the introduction of electronic X-ray requesting and reporting, was undertaken to determine whether the rate of Xray requests changed over time and whether any such change coincided with the publication of four relevant UK guidelines. From 2000-2003, I Identified a slight increasing trend in X-ray request rates. From 2003 2012, I identified a slight decreasing trend in X-ray request rates. From 2000-2012, four prominent national guidelines discouraged routine radiography to confirm OA, however only one joinpoint detected a change in the underlying trend in X-ray request rates. Consequently, it was concluded that guidelines may have a limited impact on reducing X-ray use in primary care.