An Online Program for Primary Care Practitioners to Enhance Confidence in Ability to Care for Patients With or at Risk of Painful Knee Osteoarthritis

Abstract

Objective

Primary care practitioners (PCPs) care for the majority of patients with knee osteoarthritis (KOA). Despite the existing evidence‐based guidelines, PCPs often feel unequipped to evaluate and effectively treat patients with KOA. To address this need, we designed and implemented a free internet‐based program focusing on the diagnosis and treatment of KOA. We assessed whether the program led to improvements in participants’ confidence in their knowledge and skills related to effectively recognizing and caring for patients with or at risk of KOA.

Methods

We used Caffarella's integrative model to develop a program aligned with the American College of Rheumatology 2019 Guideline for the treatment of KOA. The program incorporated 18 case‐based questions to provide retrieval practice and mastery experiences. We assessed changes in participants’ confidence in their KOA knowledge and skills after program completion.

Results

Of the first 353 learners completing the program, 53.8% were women, 41.9% had a clinical focus in primary care, and 69.1% were nonphysicians. Overall confidence in KOA knowledge and skills improved after program completion (P < 0.001; effect size = 1.28, 95% confidence interval 1.12–1.45), with largest improvements among participants with lower pre‐program confidence. A total of 95.8% of participants indicated they would recommend the program to others.

Conclusion

A free online program focusing on the effective care for patients with KOA attracted a wide range of learners, even though it targeted PCPs. Participants overwhelmingly endorsed it as highly relevant and would recommend it to others. Whether improvements in confidence translate into better patient outcomes is an important area for future research.

INTRODUCTION

Osteoarthritis (OA) is a leading cause of disability in the United States. ¹ It is estimated that almost half of US adults will develop knee OA (KOA) by age 85. ² About a quarter of all adults have knee pain on most days, most of which is due to KOA. ³ These already high rates of painful KOA are likely to increase substantially over the coming decades because of skyrocketing obesity rates, a dramatically aging population, and increases in activity‐related knee injuries. ⁴ , ⁵

SIGNIFICANCE & INNOVATIONS.

• This online program to support primary care practitioners’ evidence‐based care for patients with knee osteoarthritis is perceived as relevant and recommended by a wide range of participants.

• Completion of this program resulted in an improvement in participants’ confidence in their knowledge and skills to educate and manage patients with knee osteoarthritis.

Although the demand for KOA care is likely to increase, there is a growing shortage of medical musculoskeletal specialists. The 2015 American College of Rheumatology (ACR) Workforce Study projects that by 2030, demand for adult rheumatologists in the United States will exceed supply by more than 4,100 clinical full‐time equivalent practitioners. ⁶ There are similar projections for national shortfalls for physiatrists. ⁷ As a result, primary care practitioners (PCPs), including nurse practitioners, physician assistants, and other health professionals, in addition to other physicians, will be increasingly tasked with diagnosing and managing KOA.

The ACR published an updated evidence‐based guideline for the treatment of OA in 2020, ⁴ and other countries and organizations have developed similar recommendations. ⁵ , ⁸ Unfortunately, a significant discordance persists between recommended guidelines and clinical practice. ⁹ , ¹⁰ Despite overwhelming evidence of benefit, many patients are never offered educational, behavioral, psychosocial, mind‐body, and physical approaches to treat their KOA by their PCP and are often provided with oral medication without an individualized risk assessment. ⁹ , ¹⁰ , ¹¹ Even when KOA is treated, most patients have significant residual pain. ¹ , ¹² Despite available guidelines, PCPs feel they lack knowledge to effectively treat OA and consider themselves underprepared to manage their patients’ OA. ¹³

There are multiple practitioner‐level barriers to implementing evidence‐based KOA management. The misperception that KOA pain is an inevitable part of aging can lead to the prioritization of other medical conditions, leaving KOA unaddressed. ¹⁴ , ¹⁵ Misperceptions of the risks and benefits of various therapies for KOA also interfere with the initiation of evidence‐based care. PCPs often find guidelines lack specifics on how recommendations can be easily translated into practice. ¹⁶ , ¹⁷ In addition, although there is an abundance of information for clinicians and patients online, finding high‐quality, useful resources can be overwhelming, and PCPs are often unaware of trusted educational and community resources. ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²²

To address these shortfalls, we created a free online program incorporating retrieval practice and mastery experiences to address barriers impeding PCPs’ recognition and treatment of KOA. We assessed whether completing the program led to improvements in participants’ confidence in their knowledge and ability to effectively manage KOA.

PATIENTS AND METHODS

Program development model

We applied Caffarella's integrative model for program development. ²³ This comprehensive, collaborative, and customizable model incorporates adult learning principles. It recognizes learners’ autonomy, the importance of context, the realities of institutional complexity, and the potential of educational programs to foster learner capacity. The process includes discerning the context, building a solid base of support, identifying, sorting, and prioritizing program ideas, developing program objectives, selecting formats, preparing marketing plans, designing instructional plans, forming evaluation plans, and communicating results.

Program content

The educational program prioritized evidence‐based content aligned with the 2019 ACR/Arthritis Foundation Guideline for the Management of Osteoarthritis of the Hand, Hip, and Knee. ⁴ The program objectives were for those who completed the program to be able to (1) recognize patients at high risk for painful KOA; (2) diagnose painful KOA during a patient encounter, including distinguishing it from other causes of knee pain; (3) individualize management of patients with KOA; (4) access resources to support patients’ self‐management of painful KOA, including long‐term behavior change to reduce knee pain; and (5) feel more confident in their ability to care for patients with KOA.

The program was organized into three sequential modules: Module 1: Recognizing and managing patients at risk for painful KOA; Module 2: Diagnosing painful KOA, including how to examine the knee during a telehealth visit; and Module 3: Managing painful KOA. Specific learning objectives for each module are listed in Table 1.

Table 1.

Program learning objectives

Module	Learning objectives
1. Recognizing and managing patients at risk for painful KOA	1. Recognize modifiable and nonmodifiable risk factors for KOA
	2. Address modifiable risk factors for KOA
2. Diagnosing painful KOA	1. Formulate a differential diagnosis for adults with knee pain
	2. Prepare patients for telemedicine visits that will facilitate assessment
	3. Ask questions to distinguish OA from other causes of knee pain
	4. Perform relevant aspects of the physical examination using a virtual platform
	5. Determine if and when to refer patients with knee pain for imaging or other tests
	6. Incorporate patient‐reported outcomes and physical performance measures to monitor the impact of KOA
3. Managing painful KOA	1. Teaching patients that pain associated with KOA is something treatable
	2. Engaging with patients in shared decision‐making regarding therapeutic options
	3. Developing a management plan based on best evidence for nonpharmacologic and nonsurgical care options
	4. Supporting positive health behavior change in patients with KOA
	5. Addressing psychosocial issues often associated with KOA
	6. Supporting engagement in physical activity, including facilitating access to physical therapy for KOA
	7. Engaging in discussions with overweight patients about nutritional/weight management programs, and referring patients to such programs
	8. Describing the role of surgical approaches to KOA
	9. Feeling more confident in one's ability to care for patients with KOA

KOA, knee osteoarthritis.

Program format

We designed the program to incorporate active learning with retrieval practice (the act of recalling information from memory), as this has been shown to effectively enhance future recall, retention, and transfer of knowledge. ²⁴ , ²⁵ Eighteen case‐based, multiple‐choice questions were embedded within the modules, after sections presenting informational content. For the 8 questions on KOA diagnosis, there were 11 answer options for each question; for the 10 questions on KOA management, there were 4 answer options. After responding to each question, participants were provided with feedback and an answer rationale, including up to date information and links to resources relevant to that question. Feedback that informs the learner of the correct answer and provides an explanation for why that answer is correct allows the learner to correct errors and maintain correct responses with confidence. ²⁶ , ²⁷ All referenced resources were curated by study investigators to ensure a repository of carefully vetted links to evidence‐based information. Participants were required to correctly answer at least 80% of the questions in each module to move to the next module and complete the program. This design provided participants with mastery experiences, which are expected to promote confidence in their abilities to correctly diagnose and care for patients with KOA. ²⁸ There was no restriction on the number of attempts to answer the embedded questions.

Target learners

The target audience for this program was primary care clinicians who care for patients with KOA. Secondary audiences included other physicians, nurse practitioners, physician assistants, physical therapists, physiatrists, rheumatologists, and orthopedic surgeons; enrollment was not restricted to any specific specialty or profession.

Learner recruitment

The Hospital for Special Surgery (HSS) Education Institute Marketing and Communication Department posted promotions on HSS's LinkedIn, Facebook, and Twitter accounts, which included direct links to program registration. Email campaigns were also sent to both HSS email subscriber lists and to a rented list from the American Medical Association (AMA), which included all AMA members from New York, New Jersey, Connecticut, and Florida, who are in family practice, general practice, or internal medicine. The program was promoted on the American Academy of Nurse Practitioners website (aanp.org), the American Academy of Physician's Associates website (aapa.org), and myCME.com, a hub for Continuing Medical Education (CME) and Continuing Education courses. Completion of this activity enabled the participant to earn up to 4.00 Medical Knowledge Maintenance of Certification points in the American Board of Internal Medicine Maintenance of Certification program and 4.0 AMA Physician's Recognition Award Category 1 Credits^TM for CME. Participants were required to complete the program to receive these points. Learners self‐reported their race and ethnicity using fixed National Institutes of Health categories, including the option of not answering or writing in additional categories.

Assessments of study outcomes

We used a previously developed instrument to assess participants’ confidence in their knowledge and skills, including their ability to engage in discussion and provide education about OA evaluation and management, with phrasing of the questions adapted to focus specifically on KOA. ²⁹ Self‐rated confidence for each item was measured on a 5‐point Likert scale (1 = not at all confident; 5 = very confident) with two cumulative subscales capturing (1) Confidence in KOA Knowledge (11 items, score range 11–55) and (2) Confidence in KOA Skills (16 items, score range 16–80). A Composite Score, which sums all items, was also calculated (27 items, score range 27–135).

Participants were also asked about the relevance of 12 prespecified barriers they may have encountered to delivering best practice nonpharmacologic and nonsurgical care to people with KOA. ²⁹ Participants rated the extent to which each barrier was applicable to them on a 4‐point Likert scale (1 = not at all applicable to me; 4 = highly applicable to me). In addition, participants were asked to describe any other barriers they encountered. The assessment instrument was administered to each participant prior to and after completion of the three modules.

We used several strategies to assess participant engagement. Our learning management platform captured the time elapsed between when participants started and completed the program. The number of times web‐based links were accessed from the program was captured using the link management tool on bitly.com. The learning management platform also recorded the answers participants selected as their first responses to the embedded case‐based questions.

Upon completion of the three modules, participants were asked about the relevance of the content. In addition, they were asked to provide suggestions for improving the program and to describe anticipated changes in their practice attributed to their participation in the program.

Statistical methods

Characteristics of all participants who completed the three online modules during the study period were summarized with descriptive statistics. Our primary analysis evaluated changes in participants’ overall confidence for caring for patients with KOA after completion of the three online modules. Means with SD were reported for the Composite Score, the Confidence in KOA Knowledge and Confidence in KOA Skills subscales, as well as each single item in the instrument, which included specific questions regarding participants’ confidence in their ability to engage in discussion and provide education about KOA management. Confidence scores were compared pre‐ and post‐program completion with the Wilcoxon signed rank test. Cohen's d effect size, including 95% confidence intervals (CIs), was calculated. To explore the factors associated with changes in confidence, statistics from Wilcoxon sum rank test, Kruskal‐Wallis test, or Pearson's correlation coefficient were calculated, as appropriate. Predictors of the change in the Composite Score were modeled using multivariate linear regression. Certainty to implement relevant changes in practice after completion of the course, reported by participants with or without a clinical focus in primary care, were compared using the Wilcoxon sum rank test. All P values were two sided, and 95% CIs were calculated to assess the precision of the obtained multivariate estimates. The frequency of how often the correct answer was selected for each multiple‐choice question was recorded, and point biserial statistics calculated to measure the correlation between each participant's first answer and their overall correct answer rate. The frequency of perceived barriers pre‐ and post‐program were compared. All analyses were performed in SAS Version 9.4 (SAS Institute, Inc.). This study was granted an exemption by the HSS Institutional Review Board (IRB #2021‐0317).

RESULTS

Program launch

The program, Effectively Caring for Patients with Knee Osteoarthritis: An Online Program for Primary Care Clinicians, was launched on the HSS eAcademy platform on April 5, 2022 (https://www.eacademy.hss.edu/courses/effectively-caring-for-patients-with-knee-osteoarthritis-an-online-program-for-primary-care-clinicians). It was sent to 1,185 institutional subscribers and 26,014 subscribers on AMA rented lists. In May and June 2022, links promoting the educational program had 2,820 and 45,000 impressions (ie, total number of opportunities to view the campaign) on the American Academy of Nurse Practitioners and the American Academy of Physician's Associates websites, respectively. From April to October 2022, a total of 2,479 learners were sent to the educational program via a promotion on myCME.com. From May through June 2022, promotion on LinkedIn led to 41,645 impressions and 209 “clicks” (209/41,645 = a click‐through‐rate of 0.5%). From April through June 2022, a Google Ad campaign led to 53,880 impressions and 2,397 clicks (a click‐through‐rate of 4.5%).

Participants

During the first 9 months after launch, 765 individuals registered; the current analysis reports on the 353 participants who completed the program during this time. Participants had different durations of clinical experience: 28.6% had spent 0 to 5 years in practice, and 31.2% had spent more than 20 years in practice. Despite targeting PCPs, only 41.9% self‐reported a clinical focus in primary care. Participants with an MD, MBBS, or DO degree comprised 30.9% of the participants; the remaining were nonphysicians. Of all participants, 53.8% identified as women, 69.1% identified as White, and 72.5% as not Hispanic/LatinX. The program had a global reach with 7.4% of participants being from outside the United States (see Figure 1). Characteristics of participants are described in more detail in Table 2.

Figure 1.

Participant locations (n = 353). Source: Figure created using mapcustomizer.com.

Table 2.

Description of participants

Characteristics	n = 353 (%)
Sex
Woman	190 (53.8)
Man	145 (41.1)
Prefer not to answer	18 (5.2)
Additional category	0 (0)
Race
White	244 (69.1)
Asian	34 (9.6)
Black or African American	16 (4.5)
American Indian or Alaska Native	3 (0.9)
Native Hawaiian or Other Pacific Islander	0 (0)
Multiracial	5 (1.4)
Prefer not to answer	46 (13)
No response	5 (1.4)
Ethnicity
Not Hispanic/LatinX	256 (72.5)
Hispanic/LatinX	30 (8.5)
Prefer not to answer	67 (19)
Years in practice
0–5	101 (28.6)
6–10	54 (15.3)
11–15	45 (12.7)
16–20	43 (12.2)
20+	110 (31.2)
Main area of clinical practice a
Primary care	148 (41.9)
Orthopedics	92 (26.1)
Geriatrics	32 (9.1)
Sports medicine	29 (8.2)
Physiatry	16 (4.53)
Rheumatology	10 (2.8)
Physical Therapy	9 (2.6)
Other b	49 (13.6)
Degree
MD/MBBS/DO	108 (30.9)
Nurse practitioner or registered nurse	97 (27.5)
Physician assistant	65 (18.4)
Master's degree, PhD, or Pharm D	58 (16.4)
Physical therapy	6 (1.7)
None of the above	19 (5.4)

^a Participants could select more than one.

^b Pain management (7), internal medicine (5), emergency medicine (3), neurology (3), pediatrics (3), cardiology (2), dermatology (2), general medicine (2), psychiatrics/mental health (2), radiology (2), acupuncture (1), allergy (1), college health (1), endocrinology (1), intensive care unit (1), informatics (1), neurosurgery (1), nursing professor (1), palliative care (1), pharmacy (1), public health (1), pulmonary care (1), surgeon (1), urgent care (1), not specified (4).

Preprogram confidence in KOA knowledge and skills and perception of barriers

Participants reported a wide range of preprogram confidence in KOA knowledge (mean 33.4, SD = 9.58, potential range 11–55; higher = more confidence) and confidence in KOA skills (mean 49.7, SD = 13.8, potential range 16–80; higher = more confidence). Participants identified patients’ lack of access to care because of financial and geographic circumstances as the most relevant barrier to delivering high‐quality care to people with KOA (32.6% indicating that this was “applicable” or “highly applicable”). Other barriers to delivering high‐quality care included a lack of participant awareness of current clinical guidelines and patients seeking care that does not align with current evidence (30.3% and 27.7%, indicated these barriers were applicable or highly applicable to them, respectively). The frequency at which participants indicated that each of the 12 barriers listed were applicable or highly applicable to them can be found in Supplementary Table 1.

Program engagement

The median time elapsed between participants starting and completing the program was 1 hour and 44 minutes (range: 12 minutes to 54.7 hours). Of the 89 resource‐links provided in the modules, 87 were clicked on at least once. The two links with the greatest number of clicks were the HSS “Best Bet Exercises for Osteoarthritis of the Knee” webpage (https://www.hss.edu/conditions_exercises-for-knee-osteoarthritis.asp), which had 109 clicks, followed by the HSS “Hip Arthritis” webpage (https://www.hss.edu/condition-list_hip-arthritis.asp) with 52 clicks.

Performance of case‐based questions

Psychometric performance of the 18 case‐based questions indicated that they were challenging. On their first attempt, 22.7% of participants answered at least 80% of the eight case‐based questions focused on diagnosing painful KOA correctly, and 52.9% answered at least 80% of 10 case‐based questions on managing painful KOA correctly. Point biserial correlations were excellent, ranging from 0.44 to 0.67.

Change in confidence in KOA knowledge and skills and perception of barriers

Confidence in KOA knowledge and skills improved after completing the course (see Table 3). Mean Composite Scores increased from 83.1 to 109.4 (P < 0.001, effect size = 1.28, 95% CI 1.12–1.45). The mean Confidence in KOA Knowledge Subscale and the mean Confidence in KOA Skills Subscale both improved (33.4 to 44.2, P < 0.001, effect size = 1.25, 95% CI 1.09–1.41 and 49.7 to 65.2, P < 0.001, effect size = 1.26, 95% CI 1.10–1.42, respectively).

Table 3.

Participants’ reported confidence in their KOA knowledge and skills prior to starting and after completing the online modules^* (n = 353)

Participants were asked to respond to each question below on 5‐point Likert scale from 1 = Not at All Confident to 5 = Very Confident	Pre‐course mean score (SD)	Post‐course mean score (SD)
How confident do you feel in your knowledge of:
The pathology (eg, involvement of articular and periarticular structures and neurobiology of persistent pain) and typical disease course of KOA	3.05 (0.99)	3.91 (0.75)
The relationship between pathology of KOA and pain and disability	3.15 (0.95)	3.99 (0.76)
Risk factors associated with the development of KOA	3.22 (0.94)	4.11 (0.73)
Current clinical diagnostic criteria for KOA	3.07 (0.95)	4.02 (0.76)
Current best practice nonpharmacologic and nonsurgical treatment strategies for KOA	3.10 (0.93)	4.08 (0.77)
Current best‐practice pharmacologic care for KOA	3.09 (0.94)	4.06 (0.76)
Principles of chronic disease self‐management for KOA	3.05 (0.96)	4.03 (0.76)
Principles for supporting self‐management and health behavior change	3.07 (0.94)	4.05 (0.71)
Physical performance measures to monitor the functional impact of KOA	2.86 (0.95)	3.96 (0.75)
Appropriate tools to capture patient‐reported outcomes to monitor the impact of KOA	2.79 (1.00)	3.93 (0.76)
Timing and suitability for surgical intervention for people with KOA	2.91 (0.97)	4.03 (0.74)
Confidence in KOA knowledge (subscale)	33.4 (9.58)	44.2 (7.61)
How confident do you feel in your ability to engage in a discussion and provide education about:
The disease of KOA (pathology, risk factors, typical disease course)	3.07 (0.96)	4.05 (0.72)
Role of exercise and physical activity	3.24 (0.97)	4.19 (0.72)
Role of nutrition management (incorporating weight loss, if appropriate)	3.16 (0.98)	4.12 (0.71)
Role of medicines	3.15 (0.92)	4.14 (0.72)
Role of surgery	3.04 (0.97)	4.08 (0.73)
How to practically manage pain, based on a contemporary understanding of pain neurobiology	2.99 (0.95)	3.98 (0.73)
Whether a diagnosis of KOA means that joint symptoms will inevitably worsen	3.00 (0.96)	4.08 (0.73)
How confident do you feel in your clinical skills to:
Clinically assess a person complaining of knee pain to determine the likelihood of KOA	3.20 (0.96)	4.03 (0.74)
Explain why a particular diagnostic test, imaging (eg, MRI), procedure, or surgery is not indicated to diagnose or manage KOA	3.13 (1.01)	4.08 (0.72)
Discuss with a patient their beliefs about KOA and their beliefs about therapeutic options	3.16 (0.95)	4.08 (0.70)
Engage in shared decision‐making regarding therapeutic options	3.21 (0.95)	4.11 (0.70)
Develop a management plan based on best evidence for nonpharmacologic and nonsurgical care options	3.11 (0.95)	4.09 (0.71)
Support positive health behavior change in a person with KOA	3.20 (0.93)	4.09 (0.70)
Develop a physical activity or exercise program that considers pain, disability, and beliefs about physical activity/exercise	3.03 (0.96)	4.04 (0.74)
Engage in a discussion with patients with overweight or obesity about a nutritional/weight management program and develop such a program	3.14 (0.95)	4.09 (0.73)
Measure the impact of KOA using standard outcome measures	2.89 (0.97)	3.96 (0.77)
Confidence in KOA skills (subscale)	49.7 (13.8)	65.2 (10.6)
Composite score	83.1 (22.8)	109.4 (17.8)

The Confidence in KOA Knowledge subscale, Confidence in KOA Skills subscale and the composite scores are bolded to distinguish them from the individual items that comprise them.

KOA, knee osteoarthritis; MRI, magnetic resonance imaging.

^* P < 0.001 for the difference between pre‐ and post‐course values for each item, subscale, and overall score.

In general, individual participants’ confidence in their KOA knowledge and skills increased (mean increase in Composite Score 26.3, SD = 19.8). Larger mean increases in confidence were seen in participants who described themselves as primary care clinicians (29.4 vs. 24.1, P = 0.011, effect size = 0.27, 95% CI 0.06–0.49), those in practice 15–20 years (32.5) or more than 20 years (30.1) (P for trend comparing changes between categories of years in practice; P = 0.002), those with medical as opposed to nonmedical degrees (29.4 vs. 25.0, P= 0.036, effect size = 0.22, 95% CI 0.00–0.45), and in those identifying as women (28.6 vs. 23.0, P = 0.043, effect size = 0.29, 95% CI 0.07–0.51). Participants with lower baseline confidence in their knowledge and skills related to KOA had greater increases in their self‐confidence after completion of the program, but those with higher baseline confidence still demonstrated increases (See Supplementary Table 2). In a linear regression model controlling for the participants’ baseline confidence in KOA knowledge and skills, identification as primary care clinicians, length in practice, medical versus nonmedical degree, and sex, only baseline confidence in KOA knowledge and skills remained significantly associated with changes in overall confidence in KOA knowledge and skills (B = −0.56, P < 0.0001).

After completing the three modules, participants described a significant decrease in their perception that lack of awareness of current clinical guidelines was a barrier to their care of patients with KOA (P< 0.001).

Anticipated behavior changes

Participants identified strategies that they planned to implement in their practice after completing the program's three modules, including encouraging weight management, exercise, and/or dietary changes (17.8% of participants), referring patients to physical therapy (9.9%), recommending cognitive behavioral therapy (8.5%), increasing their focus on patient education (8.2%), and relying more on therapies that did not involve surgery or taking medication (6.2%) (See Table 4 for more details).

Table 4.

Strategies participants reported they planned to implement in their practice as a result of participating in the program

	n (%) a
Encourage weight management/exercise/diet	63 (17.8)
Recommend/refer to physical therapy	35 (9.9)
Implement improved diagnostic strategies, including physical examination skills	32 (9.1)
Recommend Cognitive Behavioral Therapy/psychological support	30 (8.5)
Increase focus on patient education/improve communication with patients	29 (8.2)
Apply strategies for medication management	24 (6.8)
Begin with therapies that do not involve surgery or taking medication	22 (6.2)
Recommend Tai Chi	16 (4.5)
Recommend topical treatments	15 (4.2)
Apply telehealth physical examination strategies	10 (2.8)
Increase screening for KOA/recognize and manage patients at risk for KOA	10 (2.8)
Apply current guidelines/use an evidence‐based approach	8 (2.3)
Apply/recommend pain management strategies	8 (2.3)
Develop individualized management plans	7 (2.0)
Recommend steroid injections	5 (1.4)
Implement team‐based approaches	5 (1.4)
Refer to surgical and nonsurgical specialties as needed	5 (1.4)
Share information with colleagues	5 (1.4)
Decrease use of MRI or x‐rays	3 (0.8)
Implement general strategies not otherwise specified	13 (3.7)

KOA, knee osteoarthritis; MRI, magnetic resonance imaging.

^a Participants could indicate more than one strategy.

Perceived relevance and feedback

A total of 93.4% of participants agreed or strongly agreed that the program content was useful and relevant to their scope practice, and 95.8% of participants indicated that they would recommend the program to others. A total of 99.4% of participants felt that the content of the program was balanced and unbiased.

In response to our request in the post‐program questionnaire for suggestions to make the program more effective, 17.2% of participants responded that no changes should be made. Submitted suggestions varied, including some recommending increased video (n = 11) and audio (n = 6) components.

DISCUSSION

Although KOA affects more than 250 million people worldwide and is one of the leading causes of disability in older adults, both patients with KOA and their treating clinicians often succumb to “therapeutic nihilism,” and assume that the pain and disability of KOA is an inevitable part of aging without effective therapies. ³⁰ , ³¹ This is a misconception, as comprehensive, multimodal management programs have shown success in decreasing pain and improving quality of life in patients with OA. ³²

For most patients with KOA, encounters with primary care clinicians are their initial interface with the medical system. ³³ If PCPs are able to recognize patients at risk of or with KOA, they have the opportunity to implement proven therapies, ideally before KOA pain becomes chronic or devolves to include pain sensitization. ³⁴ However, many patients with KOA feel that their knee symptoms are not taken seriously by their caregivers and that their caregivers do not provide adequate information and counseling on KOA. ¹⁷ This is perhaps not surprising, as many PCPs feel underprepared to treat KOA and lack confidence in their ability to provide effective treatment, despite the fact that the ACR and others have developed evidence‐based management guidelines. ¹³ Although these guidelines are well publicized, many studies have found that they are not being implemented and that clinicians find guidelines challenging to translate into practice. ¹³ , ¹⁶ , ²⁹

In response, the goal of this study was to assess whether an online program could improve participants’ confidence in their ability to recognize and manage patients with or at risk of KOA. We incorporated theory‐informed and evidence‐based strategies to maximize success. Physicians look for online educational experiences that are high quality, self‐paced or self‐directed, with opportunities for reflection. ³⁵ Interactive educational programming, as opposed to passive learning, has been shown to be effective in encouraging adherence to clinical guidelines. ³⁶ Our program was designed to promote durable and transferrable learning through retrieval practice and to enhance confidence in participants’ ability to recognize and care for patients with KOA through mastery experiences of correctly answering case‐based questions. ²⁸

Participation in our programs led to clear improvements in participants’ confidence in their knowledge and skills related to diagnosis and management of patients with KOA, with large effect sizes (>1), and received overwhelmingly positive endorsement from participants who found it useful and would recommend it to others. Online programs such as ours will allow the effective dissemination of high‐quality information without regard to geography or privileged, well‐resourced academic settings. Laws, such as the Accessible, Affordable Internet for All Act introduced in 2021 to make high‐speed broadband internet accessible and affordable nationwide, will further decrease barriers to online programming. ³⁷

We identified effective strategies to publicize the program, which yielded click‐through rates on par with industry standards. The global average click‐through‐rate for LinkedIn advertising around the time of our campaign was 0.44% to 0.65% ³⁸ ; our rate of 0.5% was within that range. Average click‐through‐rates across all industries for Google Search ads around the time of our campaign were estimated to be anywhere from 2% to 5%. ³⁹ , ⁴⁰ Therefore, our campaign on Google Search ads of 4.5% performed well. However, these types of publicity campaigns require substantial resources and are not sustainable without additional funds; moving forward, there may be more efficient approaches to target relevant audiences. For example, we are planning to administer this program to medical trainees and will evaluate whether similar benefits are seen in this population.

We showed clear statistical improvements in all measured areas of confidence related to KOA knowledge and skills, and participants described high levels of certainty that they would implement relevant changes in their practice. This certainty to implement change was not limited to those whose practice was focused on primary care (See Table 5). Participants described plans to implement evidence‐based practices, including recommending weight management and exercise for their patients with KOA, which are two core tenets in the treatment of KOA. Although it is beyond the scope of this study to evaluate change in practice, perceived self‐efficacy and outcome expectancies (the anticipated consequence of one's actions) are strongly associated with behavior change. ⁴¹

Table 5.

Certainty to implement relevant changes in practice after completion of the course, reported by participants with or without a clinical focus in primary care^*

	Primary care (n = 148) mean (SD)	Nonprimary care (n = 205) mean (SD)	P value
Recognize and prioritize clinically significant KOA during a patient encounter	4.42 (0.62)	4.35 (0.87)	0.90
Develop of individualized management plans for patients with painful KOA	4.44 (0.63)	4.36 (0.88)	0.97
Access resources to support patients’ with KOA self‐management and long‐term behavior change	4.36 (0.67)	4.32 (0.90)	0.57

^*Participants responded to each question on a 5‐point Likert scale (1 = not at all certain to 5 = very certain).

KOA, knee osteoarthritis.

Participants in this program represented a diversity of sex, race, ethnicity, and duration of time in clinical practice, suggesting a broad need for this type of programming. Interestingly, less than 50% reported a focus on primary care, indicating relevance to specialists beyond our original target audience. In addition, because 69.1% of participants were nonphysicians, there is clearly a demand for evidence‐based information on KOA by nurses, physical therapists, and practitioners in other disciplines.

This study has some limitations. It was not designed to measure the clinical impact of improved self‐efficacy or whether participants followed through on their intentions, and thus we are unable to evaluate the downstream effects of improvements in KOA knowledge and confidence. Similarly, although we showed clear statistical improvements using a standard instrument, we cannot quantify the clinical impact of these improvements. We explored a variety of different associations and did not adjust for multiple univariate comparisons. Although an education program cannot directly reform systems‐wide structural barriers to care, our program did successfully decrease participants’ lack of awareness of current clinical guidelines as a barrier to clinical care.

Strengths of this study include using an integrative theory‐based model for developing a program to target specific deficiencies in KOA knowledge, skills, and related confidence. We reached a broad group of learners, who were extremely positive about the program, would overwhelmingly recommend it to others, and endorsed plans to implement evidence‐based practices. We used a standard instrument to evaluate the impact of our program, aligning with a recent systematic review of continuing professional development in rheumatology for PCPs, which identified program evaluation as an area for improvement. ⁴²

In conclusion, we have developed a free online program that improves participants’ confidence in their ability to care for patients with or at risk of painful KOA. Whether the impact on improvements in confidence translate into better patient outcomes is an important area of future research.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Xiaoyue Ma had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design

Mandl, Aizer.

Acquisition of data

Mandl, Lin, Gottesman, Mich‐Gennari, Wall, Nathif, Aizer.

Analysis and interpretation of data

Mandl, Ma, Aizer.

Supporting information

Disclosure form:

Supplementary Table 1: Barriers endorsed by participants prior to participating in the program.

Supplementary Table 2: Change in confidence in KOA knowledge and skills after participation in the program, according to pre‐program confidence.