Management of hip and knee osteoarthritis in general practice: a systematic review

Abstract

Purpose: Osteoarthritis (OA) management by general practitioners (GPs) is aimed at improving quality of life, self-management and preventing further progression of the disease. Understanding GP reported clinical practice is key when trying to improve adherence to the well-established management guidelines. The aim of this study was to summarise GP reported management of hip and knee OA, including pharmacological and non-pharmacological management, referrals, diagnostic tools and assessment patterns. Methods: Databases (Embase, MEDLINE, CINHAL and Cochrane Central Register for Controlled Trial) were searched using the keywords “general practice”, “treatment” and “osteoarthritis” from inception to June 2018. Published studies that comprised GP reported management for people with hip and knee OA were identified. Risk of bias was assessed based on predetermined measures of informativeness and internal validity. Results: Out of a total of 13,930 records identified through the databases search, 14 original articles were included in this review, all of which used a prospective cross-sectional design. Results were collected from ∼19,446 GPs; response rates ranged from 17% to 90%. Of the included studies, seven studies reported respondents’ male percentage (mean (SD) 72% (18)), six reported respondents’ practice location (urban 60% (18) versus rural 40% (14)) and five reported respondents’ time since graduation (20 years (8)). Overall, studies scored high in most internal validity measures including reproducibility of results and standardising GP management definitions across cohorts. However, eight of the included studies did not adjust analyses for confounding factors such as patients and GPs’ characteristics. Informativeness rated low as seven included studies lacked data on respondents’ baseline characteristics. Pharmacological management: Twelve of the 14 studies included data on GP reported management with non-steroidal anti-inflammatory drugs (NSAIDs) with prescription rates ranging from 15% to 84%. NSAID prescription rates differed by stages of disease severity (stage 1: 43%, stage 2 and 3 >70%) and before and after treatment (19% before, 9.3% after). Paracetamol prescription rates ranged from 7% to 91%, as reported in 10 out of the 14 studies. One Australian study showed that for every 100 patients treated for hip and knee OA, 23 were prescribed paracetamol. Prescription of NSAIDs combined with paracetamol was reported in one study, while prescription of NSAIDs combined with analgesics was reported in three out of the 14 included studies. The rates ranged from 1.4% to 42.8%. Non-pharmacological management: GP reported non-pharmacological management of OA using exercise prescription and/or weight reduction advice was reported in 10 of the 14 studies. The percentages of GPs recommending general exercise ranged from 15% to 87%. Four of the 10 studies which included data on exercise prescription showed that types of recommended exercise included walking, swimming, cycling and hydrotherapy. Weight reduction advice rates ranged from 32% to 68%, as reported in nine of the 14 studies. Five of the 14 included studies reported on management using intra-articular steroid injections and rates were low ranging from 4% to 25%. There was a low percentage of GPs recommending the use of walking sticks in early-stage OA (25% as reported in two studies) and this percentage increased to 45% in the case of acute exacerbation. Referrals: Referral rates to allied health professionals (physiotherapists, exercise physiologists and/or nutritionists) ranged from 0.1% to 62.7%, as reported in seven of the 14 included studies. Referral to orthopaedic surgeons ranged from 7% to 75%, as reported in three studies. GPs also referred to other health care practitioners (6% to 18%), secondary care (13%), rheumatologists (5%) and clinical specialists (46%). Diagnostic tools and assessment patterns Five of the 14 studies reported data on GPs’ use of diagnostic tools when assessing patients with knee or hip OA. Across these studies, 35% to 85% of GPs reported requesting plain radiography (X-ray). One study reported that 28% of GPs would request additional imaging even if patients had previously undergone radiography within the past six months. Two studies reported that >50% of GPs request blood tests to investigate erythrocyte sedimentation rate and blood count. Conclusions: Pharmacological management largely concentrated on NSAIDs and paracetamol, with similar prescription rates for both. Non-pharmacological management using exercise and/or weight reduction and referral to allied health professionals varies widely but appears to be underutilised. Diagnostic imaging use appears high, given OA can be diagnosed clinically, suggesting that GPs over investigate the diagnosis. More work needs to be done to support GPs’ implementation of evidence-based management of hip and knee OA.