Objectives About 25% of the patients who are fifty years and older suffer from degenerative knee pain and functional loss. According to Dutch orthopaedic guideline recommendations these patients are not required to be diagnosed and treated routinely with an MRI and/or an arthroscopy. The importance of these recommendations are underscored in the Choosing Wisely recommendation ‘No arthroscopic surgery and no MRI for patients of 50 years and older with degenerative knee complaints’ (in short: CW-recommendation) issued by the Dutch orthopaedic association. However, despite this recommendation, many patients with degenerative knee complaints still expect their orthopaedic surgeon to order an MRI and/or perform an arthroscopy for diagnosis and/or treatment. To improve the uptake of the CW-recommendation, this study aimed to investigate which barriers and facilitators were associated with the unnecessary referral for an MRI or an arthroscopy from a patient point of view. Method We performed an online questionnaire among 138 Dutch orthopaedic patients≥50 years with degenerative knee complaints, of whom 116 (84%) responded. This survey included questions about background characteristics, received care and factors influencing unnecessary referral for an MRI/arthroscopy from a patient perspective (4-point Likert scale). These factors were based on literature and interviews among patients with degenerative knee complaints, and classified according to the framework of Grol. Spearman rank correlations were used to investigate which factors were associated with unnecessary referral for an MRI/arthroscopy. Next, as individual factors may be related to each other, we included factors significantly associated with unnecessary referral for an MRI/arthroscopy into a multivariable logistic regression model (p<0.05). For this analysis, we dichotomized the answers on received care and individual factors because of few observations in some cells. Factors increasing unnecessary referral were classified as facilitators, factors decreasing unnecessary referral as barriers. Results 74 (64%) patients received an MRI and 58 (50%) underwent a knee arthroscopy. Two factors were independently associated with the unnecessary referral for an MRI: whether the orthopaedic surgeon discussed the added value of an MRI in the care trajectory with the patient (&bgr;=3.72 (CI 95% 0.88 till 6.56, barrier) and the patient’s own preference for an MRI (&bgr;=2.04 (CI 95% 0.19 till 3.90, barrier). Three factors independently increased the likelihood of referral for arthroscopic surgery and were thus barriers for the recommendation uptake: strong preference of orthopaedic surgeon for arthroscopy (&bgr;=3.64 (CI 95% 2.19 till 5.08), positive experiences of relatives and friends of the patient regarding an arthroscopy (&bgr;=1.68 (CI 95% 0.43 till 2.92) and preference of the patient for an arthroscopy (&bgr;=1.39 (CI 95% 0.07 till 2.71). Conclusions Barriers and facilitators among patients for unnecessary referral for an MRI or arthroscopy were found at the level of the individual patient (preference of patient), the social context of the patient (positive experiences of relatives and friends), and the individual professional (discussion of added value, preference of orthopaedic surgeon). Strategies to improve the uptake of the CW-recommendation among patients, by reducing the pressure of patients toward orthopaedic surgeons, should at least address these barriers and facilitators to be likely to be effective in reducing unnecessary MRI and arthroscopy.