Background/Purpose: Intra-articular injections are an effective treatment for knee osteoarthritis (OA).  Despite being a simple and ambulatory procedure, most of the internists do not perform it regularly due to lack of confidence or inadequate training. Given the shortage of rheumatologists and long waiting times for referrals, we aimed to evaluate whether training during residency can increase residents’ confidence in performing knee injections during primary care visits. As a secondary measure, we chose to center this project at our public hospital’s clinic to increase access to care for this primarily underserved population.  

Methods: We included internal medicine residents from the University of Miami. Training was divided into 2 phases. Phase one consisted of a workshop led by a rheumatologist, which included a lecture followed by simulation with mannequins. A questionnaire was performed before this session to assess the residents’ referral practices for injections (1=never; 5=always refers), experience and confidence with the procedure (1=none; 5=high confidence).  In the second phase, residents on their primary care rotation at Jackson Memorial Hospital performed knee injections on patients with OA, supervised by a rheumatologist. A post-procedure survey was collected to assess the residents’ confidence level and perceived utility. Additionally, surveys were administered to the patients to evaluate their perceived benefit (5-point scale), pain before and after the procedure (scale of 1-10), and functionality.  

Results: 84 residents took the pre-workshop survey. On a 1-5 scale, the residents reported they refer OA patients for knee injections at a 2.3. Reasons for not referring included long waiting times to see a specialist (26%), resident being unsure about the benefits (31%) and patient refusing injections (29%). 9 residents had done prior knee injections, of which 2 residents can perform it independently. The average confidence pre-workshop was 1.4. 7 residents completed the survey post-real world injections. The mean confidence post-mannequin training was 2.7±1.4 and 4.4±0.8 after real world injections (mean difference 1.7; 95% CI 0.4- 3.0). The mean number of injections performed per resident was 4. All of them reported that this was useful and 5 (71%) that they would incorporate this into their primary care practice. 15 patients received knee injections. 12 (80%) reported that the procedure was helpful, with a mean effect of 3.3.  The mean pain score before the procedure was 8.9±1.6 and after 4.5±2.9 (mean difference -4.4; 95% CI -6.0 to -2.8). 6 patients (40%) decreased their use of NSAIDs/analgesics and 13 (87%) felt more functional. No complications were reported.

Conclusion: This is an ongoing study, but our results suggest it is easy and effective to train internal medicine physicians in knee injection. Implementation of this in real world primary care has the potential to increase access to health of patients in need and more timely treatment.  

« Back to ACR Convergence 2024

ACR Meeting Abstracts - https://acrabstracts.org/abstract/knee-injection-training-for-internal-medicine-residents-in-real-world-primary-care-practice/