Article Text

Patient and physician perspectives guiding intra-articular treatment choice in knee osteoarthritis: stakeholders are aligned on treatment priorities but have different assessments of treatment effect
  1. Vandana Menon1,
  2. Caroline Huber2,
  3. Alexandria Portelli2,
  4. Marissa Baker-Wagner2,
  5. Scott Kelley1,
  6. Kathy Lang2
  1. 1 Flexion Therapeutics Inc, Burlington, Massachusetts, USA
  2. 2 Precision Medicine Group Inc, Bethesda, Maryland, USA
  1. Correspondence to Dr Kathy Lang, Precision Medicine Group Inc, Bethesda, Maryland 20810, USA; kathy.lang{at}


Objectives Knee osteoarthritis (OA) is a leading cause of health-related disability. In the absence of curative non-operative therapies, treatment goals are limited to symptom relief. Data are limited on how patients and physicians prioritise available treatment options. We assessed patients’ preferences for and physicians’ attitudes towards intra-articular treatments including corticosteroids (IACS), an extended-release corticosteroid (TA-ER) and hyaluronic acids (IAHA).

Methods We conducted a prospective, IRB-exempt, double-blind survey of patients with and providers who treat knee OA. Respondents were required to have received or prescribed TA-ER in a non-trial setting. We evaluated patients’ OA history, impact of knee OA and treatment preferences, and physicians’ decision-making and prescribing experiences.

Results Of the 97 patient participants, mean age was 56 years, 70.0% were women, 75.0% had bilateral knee OA and 46.4% were diagnosed over 5 years ago. Of the 50 physician participants, 42.0% were orthopaedic surgeons, 34.0% were rheumatologists and 60.0%, on average, treat 50+ patients with knee OA per month. Treatment selection factors considered ‘very important’ to patients and physicians included disease severity (88.7%, 82.0%), impact on quality of life (88.7%, 72.0%), disease extent (84.5%, 54.0%) and activity level (80.4%, 64.0%). A majority (93.8%) of patients indicated moderate to severe difficulty with their knees. Fewer patients (76.3%) reported shared decision making compared with physicians (92.0%). Half (50.5%) of the patients reported that they experienced months of pain relief with TA-ER, 27.7% with IACS and 18.8% with IAHA. Physician assessments were consistent but estimated a greater duration of treatment effects than that reported by patients across all therapies.

Conclusion While knee OA has a tremendous impact on patients, there are significant unmet treatment needs. The increasing use of patient-reported outcomes will allow patients and physicians to track pain and functional status over time and across therapies, improving shared decision-making.

  • knee
  • osteoarthritis
  • treatment / technique

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information. Data were reported in aggregate and deidentified.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, an indication of whether changes were made, and the use is non-commercial. See:

Statistics from

What are the new findings

  • For both patients and physicians surveyed, disease severity (88.7% of patients; 82% of physicians) and impact on quality of life (88.7% of patients; 72.0% of physicians) were assessed as the most important factors in making knee osteoarthritis (OA) treatment decisions.

  • While both patients and physicians reported discussing different treatment options for their knee OA (86% of both groups reported thoroughly weighing treatment options), fewer patients than physicians reported that the decision to choose an intra-articular treatment therapy was a shared one (76.3% of patients vs 92% of physicians).

  • Patients reported longer-lasting pain relief and functional improvement with triamcinolone acetonide extended-release compared with intra-articular injections of corticosteroids and intra-articular hyaluronic acid. Overall, physicians reported a greater duration of effect than patients across all treatments, however.


Osteoarthritis (OA) of the knee, a highly prevalent degenerative joint disease, is a leading global cause of disability.1 An estimated one in two Americans will develop knee OA over their lifetime.2 This chronic progressive disease is characterised predominantly by pain leading to substantial reductions in functional status, mobility and quality of life.3 4

Currently, there are limited therapeutic options and no non-operative curative treatments for knee OA. The primary goals of non-surgical management are to reduce pain and maintain or improve function and health-related quality of life (HRQoL).5 Available therapeutic modalities include non-pharmacological options, such as lifestyle modification, physical therapy and assistive devices, and pharmacological options, such as acetaminophen, oral and topical non-steroidal anti-inflammatory drugs (NSAIDs), opioids and intra-articular injections.6 7 Intra-articular injections of corticosteroids (IACS) and hyaluronic acid (IAHA) are commonly used, especially in patients with more advanced knee OA.8 9 For example, a large commercial claims analysis found that over 50% of patients with OA received IACS or IAHA.10 Another study found that IACS was the most frequently selected OA treatment among hip and knee surgeons.11 Recent clinical guidelines, however, recommend IACS over other forms of IA therapies based on the benefit-risk profiles.6 12

Given the progressive nature of knee OA, patients may try several different therapies over the course of their lifetime. Treatment choice for knee OA is highly patient-specific and depends on disease characteristics, expectations, preferences and clinician recommendations.13 These patient-level factors also create heterogeneity in treatment response.14 15 In 2017, the FDA approved an extended-release IA corticosteroid (TA-ER) for OA knee pain.16 While a number of studies have assessed preferences for and experiences with knee OA treatments, to our knowledge, none has specifically assessed these factors across IA treatments, including TA-ER.17–23 This study aimed to describe treatment priorities, perceptions of shared decision-making and treatment experiences among patients and providers who have experience with IA therapies, including TA-ER.

Material and methods

Study design and sample

We conducted a double-blind cross-sectional survey of patients with knee OA and physicians who treat such patients. Separate survey instruments were developed for patients and physicians based on a targeted review of the literature evaluating treatment satisfaction, treatment decision-making, including facilitators of and barriers to certain treatments, and the patient–physician relationship in knee OA (see online supplemental file). A convenience sampling approach was used to survey up to 100 patients and 50 physicians.

Supplemental material

We recruited for the surveys using two separate panels of geographically diverse individuals who agree to participate in survey research: adults (aged 18+) in the general population in the USA and physicians licensed and practicing medicine in the USA. Email invitations containing links to the surveys were sent to the panel members. Interested patient respondents completed a screener to assess certain demographic and clinical characteristics. The screener also contained an information statement providing an overview of the study and indicating that participation in the study was voluntary. All participants were asked to review the information statement and provide consent. Patients who gave informed consent and met the following inclusion criteria based on their responses to the screener questions were invited to participate in the full survey: aged 18 years and older, told by a physician they have knee OA in one or both knees and received TA-ER as a treatment for their knee OA in a routine (non-trial) setting. Similarly, interested physician respondents were asked to complete a screener to assess consent and certain demographic and clinical experience characteristics. Physicians who met the following criteria based on their responses to the screener were invited to participate in the full survey: actively licensed and practicing medical doctor (MD or DO), specialises in orthopaedics, rheumatology, rehabilitative medicine or sports medicine and has prescribed TA-ER as a treatment for knee OA in a routine (non-trial) setting.

The internet-based surveys were programmed, fielded and hosted electronically by Jibunu, a market research firm, between September and December 2019. No personally identifying information was collected, and the study was deemed to be exempt from IRB oversight by Advarra IRB (Study ID: Pro00037350). Both patients and physicians received compensation in the form of a nominal gift card for participation in the survey.

Study measures

We collected data from patients on their knee OA history, HRQoL, treatment decision-making processes and preferences, treatment satisfaction and relationship with their knee OA care physician. We adapted the Knee Injury and Osteoarthritis Score (KOOS) to assess HRQoL. The KOOS assesses patients’ opinions and problems with their knee OA and includes five subscales; we adapted the quality of life subscale.24 To assess treatment satisfaction, we adapted the Treatment Satisfaction Questionnaire for Medication (TSQM). The TSQM was designed to assess patient satisfaction with specific medications and comprises effectiveness, side effects, convenience and global satisfaction.25 Patients who indicated they had experience with intra-articular injections were asked about their experiences and satisfaction with the therapies, specifically IACS, IAHA and TA-ER. We adapted the Shared Decision Making Questionnaire (SDM-Q-9) to assess the patient–physician relationship in knee OA care and treatment decisions. The SDM-Q-9 is the patient version of an instrument that evaluates the extent to which patients are involved in the treatment decision-making process.26

We collected data from physicians on their knee OA patient population, treatment decision-making, treatment prescribing experiences including specific patient characteristics that would make them more or less likely to prescribe TA-ER and the patient–physician relationship in knee OA. Physicians who indicated they had experience with IA injections were asked about their experiences and satisfaction with TA-ER, IACS and IAHA. The patient–physician relationship was assessed using an adaptation of the SDM-Q-Doc,26 while the treatment decision-making and prescribing experience questions were developed based on the targeted literature review. The SDM-Q-Doc assesses the extent to which patients are involved in the treatment decision-making process from the physician perspective.

Data analysis

Data from all eligible study participants were included in the descriptive analysis. Participants were excluded if they had incomplete survey responses (abandoned or terminated the survey) or if they completed the survey in under 5 min. Demographic, clinical characteristics and treatment patterns were summarised using counts and percentages for categorical variables. Measures of central tendency (mean/median/SD/IQR) were used for continuous variables. Since participants were recruited using convenience sampling techniques, study data were not analysed to make inferences about the broader population of patients with knee OA and physicians who treat patients with knee OA. All analyses were conducted using SAS V.9.3.


Patient respondent characteristics

During the survey period, a total of 97 patients met the study inclusion criteria and completed the full survey, out of 4956 who responded to the survey invitation. Reasons for exclusion included ineligibility based on the screener and incomplete responses. The sample was predominantly women (72.2%) and Caucasian (82.5%) with a mean±SD age of 55.8±13 years (table 1). Three quarters of the sample reported bilateral knee OA and almost half reported long-standing symptoms (more than 5 years) (table 2). Nearly 60% of patients reporting seeing an orthopaedist or orthopaedic surgeon for their knee OA care, while 18.6% of patients see their primary care physician and 11.3% see a rheumatologist. The majority of patients reported having used topical creams or patches (72.2%), IACS (67.0%), acetaminophen (61.9%) and over-the-counter NSAIDs (57.7%), while only 16.5% reported IAHA use. Most patients indicated their knee OA significantly impacts their life. Nearly all patients (88.7%) reported being aware of their condition daily or constantly and indicated that they have moderate or severe difficulty with their knee(s) (93.8%), while 79.4% reported that they modified their lifestyle moderately or significantly because of their OA.

Table 1

Patient demographic characteristics

Table 2

Patient clinical characteristics

Physician respondent characteristics

Of 297 physicians who responded to the initial invitation, 50 physicians met the inclusion criteria and completed the survey. The sample was predominantly men (90.0%) and Caucasian (68.0%) with a mean±SD age of 57±15 years (table 3). More than half of physicians (60.0%) reported treating more than 50 patients with knee OA each month and nearly half (46%) reported treating more than 13 patients with TA-ER in the 6 months prior to the survey. The most commonly reported specialties were orthopaedic surgery (42.0%) and rheumatology (34.0%). All physician participants reported experience prescribing or recommending IACS (100.0%), while almost all reported experience with IAHA (98.0%), OTC NSAIDs (96.0%), prescription-only NSAIDs (94.0%) and topical creams or patches (94.0%).

Table 3

Physician characteristics

Factors considered in knee OA treatment selection

Patients and physicians reported similar attributes as highly important when making decisions about knee OA treatments (figure 1). For both groups, disease severity (mild vs debilitating) was assessed as the most significant factor when selecting a knee OA treatment, with 88.7% of patients and 82.0% of physicians classifying severity as ‘very important’. The impact of a treatment on a patient’s quality of life also was considered ‘very important’ for 88.7% of patients and 72.0% of physicians when making treatment-related decisions. The magnitude of a treatment’s effectiveness (81.4% of patients; 64.0% of physicians), patient activity level (80.4% of patients; 64.0% of physicians) and extent of knee OA (both vs one knee) (84.5% of patients; 54.0% of physicians) were also considered ‘very important’ factors in treatment determinations.

Figure 1

OA health and treatment-related factors perceived most important by patient and physician respondents. Reported in proportion of responses. Participants could choose more than one option. OA, osteoarthritis.

Fewer patients and physicians reported route of administration (46.4% of patients; 30.0% of physicians), availability of other treatments to manage side effects (51.5% of patients; 36.0% of physicians) and frequency of treatment administration (55.7% of patients; 26.0% of physicians) as important factors. More patients assessed treatment guidelines (53.6% of patients; 32.0% of physicians) and a treatment’s impact on productivity (72.2% of patients; 56.0% of physicians) as well as other individuals such as spouse or family (46.4% of patients; 18.0% of physicians) as ‘very important’ factors compared with physicians.

Shared treatment decision-making

Almost all patients (86.6%) and physicians (92.0%) concurred that they discussed different treatment options for knee OA. Patients (86.6%) and physicians (86.0%) reported discussing the disadvantages and advantages of the knee OA treatment options. Fewer patients (76.2%), compared with physicians (90.0%), reported that treatment options were thoroughly weighed and that the decision to choose an IA therapy was a shared one (76.3% patients, 92.0% of physicians).

Experience with and perspectives on intra-articular injections

Across the three evaluated intra-articular knee OA treatment options, 50.5% of patients reported that they experienced pain relief lasting months with TA-ER, 27.7% reported the same with IACS and 18.8% with IAHA; 43.3% reported ‘significant’ improvement in pain with TA-ER, 20.0% with IACS and 18.8% with IAHA (figure 2). For functional improvement, 44.3% of patients reported months of improvement with TA-ER, 27.7% with IACS and 25.0% with IAHA and 43.3% patients report ‘significant’ functional improvement with TA-ER, 20.0% with IACS and 12.5% with IAHA.

Figure 2

Patient assessment of intra-articular knee OA treatments. IACS, Intra-articular injections of corticosteroids (n=65); IAHA, intra-articular hyaluronic acid (n=16); OA, osteoarthritis; TA-ER, triamcinolone acetonide extended-release (n=97).

Overall, physicians reported a greater duration of effect than patients for all therapeutic modalities. Sixty-eight per cent of physicians reported that their patients experience months of pain relief with TA-ER, 46.0% with IACS and 53% with IAHA. Similarly, 70% of physicians responded their patients experience functional improvements lasting months with TA-ER, 40.0% for IACS and 59.0% for IAHA.


Knee OA is a highly prevalent and debilitating disease with limited effective treatment options. Patients often cycle through various knee OA therapies in an attempt to achieve symptom relief and improve or maintain functional status. Given the widespread use of IA therapies in OA and lack of current evidence specifically focused on these therapeutic options, we sought to understand and describe patient and physician decision-making processes when choosing treatments for knee OA, as well as characterise their experiences with current IA therapies.

Our results indicate that patients and physicians are generally aligned on the factors that are considered important in knee OA treatment choice. In keeping with other studies, the respondents in our survey identified disease severity and extent, quality of life and activity level as very important attributes when selecting a treatment.5 13 23 27 Productivity and a treatment’s impact on other individuals including family members, however, were considered more important by patients than physicians. This finding represents an opportunity to better incorporate the non-medical consequences of knee OA such as work loss and caregiver burden in shared decision-making.28–30 To our knowledge, there have been limited studies that have focused on experiences and preferences with IA therapy. One recent qualitative study among US patients who had received IA therapies, however, found that disease severity and activity limitations were important factors in treatment selection.31 Further, participants in this study felt there was a lack of consensus among providers (especially across different specialties) around the effectiveness of different IA therapies. This finding aligns with our study results that found patients and providers perceive different magnitudes of effect among the IA therapies assessed.31

Patients and physicians largely agreed that the decision regarding choice of IA therapy was based on a collaborative process in which different treatment options were discussed and evaluated, including the associated advantages and disadvantages. A greater proportion of physicians reported agreement with the shared decision-making attributes than patients suggesting the need for further improvements in patient-centred knee OA care.23 There were differences between patient reports and physicians’ perceptions of the effectiveness of all IA therapies, however, in terms of duration of pain relief and functional improvements. Compared with patients, however, physicians reported a greater magnitude of effect across all IA treatments assessed in this study. Our results are in line with research that suggests patient-reported outcomes (PROs) are useful markers of disease activity and should be more comprehensively included in knee OA care.32 33 The use of PROs in clinical practice allows patients and physicians to track pain, functional status and HRQoL over time and across therapies and thus refine the shared decision-making process for the management of knee OA.32

Future research is needed to explore differences in patient perspectives and experiences with IA treatments according to clinical and demographic characteristics (eg, race, geographical location). For example, this study did not assess patients’ underlying causes of OA (ie, post-traumatic vs degenerative), their experiences based on their severity or their lifestyle including work and other activities, which may impact treatment preferences. It also did not examine how geographic location or race may influence treatment decision-making. Further, given the numerous treatment options for patients, additional studies are needed to explore patients’ experiences with concomitant IA therapy and non-pharmacological interventions, such as lifestyle modifications and physical therapy. This evidence can further support optimal treatment decision-making in OA.

As with any convenience survey, the respondents may not be representative of the broader population of patients with knee OA and physicians who care for them. While web-based surveys have the advantages of reducing data entry errors and facilitating timely completion, results reflect the responses of patients and physicians with access and comfort using the technology. Given that OA commonly occurs among older adults, the use of a web-based survey may have resulted in a younger sample population (mean 56.5) than is typical of the general OA population. Our findings are in line with previous research, however, that found the mean age of symptomatic OA diagnosis in the USA was 53.5 years.34

Respondents may have inaccurately remembered past events, including treatment decisions and clinical outcomes and responses to treatment. Further, in order to ensure that we would gather data on patient and physician experiences with TA-ER specifically—which is not as commonly used as IACS—we required survey respondents to have either received or administered TA-ER. This may have introduced further bias around perceptions of treatment effectiveness and benefit, in that patients and providers who have used TA-ER may not be representative of patients with typical knee OA and provider populations. Given the study’s focus on understanding perspectives on IA treatments, this study also did not capture all possible OA treatments, including non-pharmacological options. As such, perceived treatment experiences may have been biased if patients were using concomitant medications. Last, this study was descriptive in nature, so no statistical tests were used to assess differences across the included treatments.


Knee OA has a tremendous impact on quality of life, and patients have significant unmet treatment needs. Patients and physicians in this survey identified similar clinical attributes (eg, disease severity) as very important in making knee OA treatment decisions. Differences among the patients and providers in this study around the value of non-medical consequences in treatment decision-making as well as perceived duration of effect among IA therapies, highlight the need for greater use of PROs in clinical practice. Findings from this study, especially around the differences in perceived duration of effect of IA therapies, may directly support clinical practice through a greater awareness of the value of shared decision-making. Future studies are needed, however, to better characterise the patient populations, including demographic and clinical characteristics and prior treatment patterns, who benefit most from different IA therapies, especially as new treatment innovations emerge.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information. Data were reported in aggregate and deidentified.

Ethics statements

Patient consent for publication


Supplementary material

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Correction notice Since Online First publication, the authors have noticed that the numbers and percentages for the reported specialities of physician participants were reversed in Table 3. This has now been corrected accordingly to 'orthopaedic surgery (42.0%) and rheumatology (34.0%)'.

  • Contributors CH, KL, MB-W and AP contributed to the conceptualisation and design of the study. CH, KL, MB-W, AP, VM and SK contributed to the analysis and interpretation of results. CH, KL, MB-W, AP, VM and SK also supported in the drafting of this manuscript, provided final approval for publication and agree to be accountable for all aspects of the work.

  • Funding This study was funded by Flexion Therapeutics.

  • Competing interests CH, AP, MB-W and KL are employees of PRECISIONheor, which received financial support from Flexion to conduct the study described in this manuscript. VM is a former employee of and held stock in Flexion Therapeutics. SK is a current employee of and holds stock in Flexion Therapeutics.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.