Background Symptomatic slow-acting drugs for osteoarthritis (SySADOA) are medications used to alleviate long-term manifestations of osteoarthritis (OA). They differ pharmacologically from NSAIDs with little or no effect on the enzyme cyclooxygenase.

It is widely believed that continuous long-term NSAID therapy is undesirable given their inherent risk of gastrointestinal and cardiovascular side effects. The possibility that SySADOAs may exert a NSAID-sparing effect has generated considerable interest. However, while scientific literature lacks well-designed studies in real-world clinical practice evaluating the benefit-risk balance for SySADOAs, more data are needed to estimate the public health impact of these therapies.

Objectives The PEGASE study aims at 1) being so far the largest cohort conducted in primary care to evaluate the potential sparing effect of SySADOAs on NSAIDs use; and 2) providing large-scale and rather unique epidemiological data on OA and its management in primary care.

Methods The PEGASE study is an ongoing observational survey of a representative sample of general practitioners (GPs), rheumatologists, and corresponding patients nationwide, conducted in France since March 2010. It assembles patients with knee and hip-OA who are followed during a 16-month period.

Physicians are requested to comprehensively record OA clinical features, medical history and prescriptions at baseline and at the end of follow-up.

Patients have to complete a standard telephone interview at inclusion and then every 4 months up to 16 months, including: information on pain evaluation; Lequesne’s index; consumption of NSAIDs, SySADOAs and other medications, including switch of treatments, withdrawal, side effects, duration and doses. Events like hospitalizations, complications or recourse to health services are also recorded.

Analyses of exposure per time unit are performed. Since past exposure to drugs within the last 4 months is documented for every patient at each interview, a 2-month period interval unit of exposure is considered for both SySADOAs and NSAIDs. Latency and remanent effects of both SySADOAs have also been taken into account.

Odds ratios will be calculated to assess the relationship between SySADOAs exposure and the risk for NSAIDs use using logistic regression adjusted for age, gender, educational attainment, medical history of OA, pain degree, Lequesne’s index, body mass index, physical activity, recourse to kinesitherapy, other OA-pharmacological treatments as well as risk factors for NSAIDs contra indications. Autocorrelation between time-units is accounted for by generalised estimating equations (GEE). Further consideration on NSAIDs use at risk as outcome is also investigated.

Results To the best of our knowledge, PEGASE is the largest prospective cohort study conducted so far within general practice to evaluate the sparing effect of SySADOAs on NSAIDs consumption. The Pegase cohort assembles in January 2012 3,750 representative patients with OA recruited by 606 physicians from across France. It has been estimated that 2952 cumulated person-month of follow-up for each SYSADOAs will allow detection of a significant 20% decrease of overall NSAIDs use; such reduction is likely to have relevant implications for clinicians but also health policy makers.

Disclosure of Interest None Declared