Will patients with meniscal tear who continue to experience pain following a course of Physical Therapy (PT) enroll in a trial comparing further nonoperative therapy with surgery? a qualitative study

Abstract

Background

We aimed to conduct a qualitative prospective preference assessment (PPA) to assess participants’ willingness to enroll in a planned randomized controlled trial (RCT). The planned RCT would enroll participants with meniscal tear with persistent knee pain following a course of physical therapy (PT) and would compare outcomes of arthroscopic partial meniscectomy (APM) vs. enhanced conservative care.

Methods

We identified participants 45–85 years old with suspected meniscal tear who were referred to PT. After 10 weeks of PT, participants were sent a questionnaire that assessed knee pain. We asked participants reporting persistent knee pain to participate in a follow-up interview. We conducted semi-structured interviews and coded and analyzed transcripts to identify themes related to enrollment in our planned RCT.

Results

We analyzed transcripts from twenty participants (mean age (SD): 61.6 (6.16); 60% male, 100% White). 70% said they would participate in the trial. Willingness to enroll was associated with participants’ preferences for treatment modality and most were willing to continue PT, despite reporting persistent pain. Of those unwilling to enroll, all were averse to surgery or did not believe they needed surgery. Others had concerns about oral medications and injections, but these were not “dealbreakers” for willingness to enroll. Provider trust was a dominant decision-making factor in respondents’ willingness to enroll.

Conclusion

These findings suggest that investigators should consider additional education for participants regarding the benefits and risks of all treatment options at the time of enrollment invitation. Additionally, the trial should be introduced by a trusted provider.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12891-026-09521-6.

Background

Randomized Controlled Trials (RCTs) provide rigorous assessments of treatment efficacy and shape evidence-based care. However, trial findings should be interpreted cautiously due to a host of potential threats to a study’s internal and external validity. Such challenges include differential attrition, low enrollment rate, and selection bias [1, 2]. For example, a subject who is eligible for the trial may choose not to participate because they prefer one treatment over the other or do not wish for treatment to be assigned on a random basis.

A prospective preference assessment (PPA) may be utilized during the trial design period to identify these potential threats and develop strategies to mitigate them. A PPA introduces a hypothetical trial to potential participants and asks if they would participate in the RCT if it were currently available. The research team uses these data to estimate enrollment rates, to appraise the feasibility and acceptability of the study protocol, and to modify the study design where indicated [3]. Previously reported PPAs conducted to prepare for orthopedic trials have addressed non-operative and operative treatments for femoroacetabular impingement and meniscal tear [4–6]. These PPAs suggested that 46–90% of potential subjects were willing to enroll in the hypothetical trials – a wide range that may be explained by the heterogeneity of interventions, study population, and sample size involved in the PPAs [7].

Our research team is planning an RCT focused on subjects ≥ 45 years old with knee pain, meniscal tear, and knee osteoarthritis (OA) who continue to report moderate knee symptoms after completing 10 weeks of physical therapy, the preferred first line of therapy [8, 9]. The treatment arms would include: (1) enhanced non-operative treatment, including intensified physical therapy (PT), intra-articular injections, oral medications, and education/coaching; and (2) arthroscopic partial meniscectomy (APM).

The trial is in the planning phase, including a PPA comprised of quantitative and qualitative components. The quantitative study showed that 46% of eligible participants after their initial course of PT were willing to enroll in the RCT. Furthermore, 54% and 30% of the participants indicated a preference to avoid APM and oral medications, respectively. Additional details of the quantitative study are reported separately [10]. The study team designed a qualitative study to help explain the quantitative results and offer insights into reasons influencing potential subjects’ decisions to enroll or decline participation in the hypothetical trial [11]. We report the qualitative PPA component in this manuscript. The goals of the qualitative component were to: (1) learn from potential subjects what factors informed their decision to enroll or not enroll into the planned RCT; (2) examine factors that contribute to patient (un)willingness to enroll, such as experiences and attitudes toward specific treatments; and (3) solicit feedback about the feasibility and acceptability of specific aspects of the trial design such as recruitment methods.

Methods

Eligibility

We recruited participants into the PPA study from two northeastern cities, comprised of ten orthopaedic outpatient clinic locations. Eligible participants were 45–85 years old with confirmed or suspected symptomatic meniscal tear who attended formal PT, met the required KOOS pain score [12] threshold (> 20 on a 0-100 scale with 100 = worst) after their course of PT, and indicated they would be interested in participating in an interview. (Note, we score KOOS Pain with 100 = worst to be consistent with most other pain scores, such as the numeric rating scale or visual analogue scale, which routinely score ‘no pain’ as zero.) Participants who agreed to participate in the PPA study [10] were invited to complete a questionnaire before and after their course of PT (“Baseline” and “End of PT” questionnaires) (see End of PT Questionnaire in Additional file 2). Participants who planned to begin or recently started a course of PT were sent the baseline questionnaire. They were sent the End of PT questionnaire after they completed their course of PT. Those that had recently completed or were in the middle of PT prior to the study were sent only the End of PT questionnaire. If the participant reported KOOS pain > 20 on their End of PT questionnaire, they were asked on the questionnaire: (1) if they would agree to participating in the hypothetical RCT if offered today and (2) if they were interested in a one-on-one interview. Research assistants only contacted participants who stated they were interested in the interview and who could be scheduled for their interview within a month of completing their End of PT questionnaire because we wanted to capture their most recent experiences with therapy.

Consent procedures

This study was approved by the Institutional Review Boards of the two affiliated institutions. We adhered to an informed consent process approved by the Mass General Brigham IRB, which allows for implied, verbal, and/or written informed consent processes based on the components of the study. We used implied informed consent methods for the questionnaire portion of this study followed by verbal informed consent for the interview. First, when potential participants were offered enrollment into the first phase of the PPA study they received a fact sheet that described the purpose of the study, what was being asked of them as study participants, and the voluntary nature of study participation. The fact sheet (see Fact Sheet in Additional file 3) also explained that participants might be invited for a Zoom interview as a follow-up to the primary study. If participants were agreeable to participate in the interview, they provided verbal informed consent.

Data collection

Data included transcripts from semi-structured interviews conducted by trained study personnel. With participant consent, we utilized the recording and transcription functions in Zoom. After describing the goals of the study, we asked questions related to the nature of their knee pain, treatment experiences with PT, injections, oral medications and/or surgery, and their willingness to participate in the RCT we described (see Interview Guide in Additional file 1). Interviews lasted between 30 and 60 min. Following each interview, members of the research team verified transcripts against audio recordings to ensure accuracy. All recordings and transcripts were proofread, deidentified, and stored in an encrypted database.

Investigators

Eleven researchers (three male, eight female), who are distinct from the clinicians who cared for the participants, contributed to this study. The investigators included a behavioral scientist, physician investigator, statistician, epidemiologist, and several research assistants. The first author, who has a non-clinical doctorate in public health and is trained in mixed methods and health services research, provided extensive training to four research assistants (one male, three female) who conducted interviews over an eight-month period from December 2023 to July 2024.

Data analysis

After proofreading the transcripts, at least two research team members coded each transcript in Dedoose Analytic Software [13] using the Framework Analysis approach [14, 15] (Supplementary Fig. 1). We developed a preliminary codebook based on predetermined/neutral themes from the structured interview guide and emergent themes that arose from the interviews. We then mapped the codes into domains in a matrix (respondent x code) using Microsoft Word. We participated in an iterative process of data analysis and reduction and identified major and minor themes. The analytic team met biweekly for approximately three months to create the codebook, discuss coding, identify domains, summarize themes, and identify patterns between respondents. We created field notes and analytic memos to guide our meetings. We conducted further data reduction and interpretation with all co-authors in the manuscript drafting phase. During several group meetings (e.g. after interviews # 16 and 20), we discussed the extent to which we met saturation, defined as the point at which no new ideas emerged from interview data. Once met, we made the decision to cease interviewing participants. We resolved any discrepancies through adjudication until we had 100% agreement.

Results

144 participants completed the End of PT questionnaire, and 62 were eligible and invited to participate in the qualitative component. Of the 62 participants, 23 were interviewed over Zoom and 20 were included in the analytic sample (32.3%). Of the 39 who were not interviewed, 3 (7.7%) were no longer interested, 26 (66.7%) were out of the one-month eligibility window, 6 (15.4%) declined participation in the interview, and 4 (10.3%) had other reasons (e.g., zoom difficulties, lost to follow up). We excluded data from 3 of the 23 interviewed participants (7.1%) because they either already received an APM (n = 1) or meniscal tear was not a primary reason for seeking care (n = 2). This exclusion led to a final analytic sample of n = 20.

Participant demographics

Interviewed participants were on average 61.6 years old (SD = 6.16), the majority male (60%), and all White. Participants reported an average 36.58 KOOS pain score (SD = 11.11) with most (85%) reporting “daily” or “always” knee pain. There were no differences between interviewed and excluded participants by age, race/ethnicity, BMI, education level, or frequency of knee pain (Table 1). Those interviewed were disproportionately male compared to those excluded (60.0% vs. 21.4%). Furthermore, the participants we interviewed disproportionately reported they would participate in the hypothetical trial compared to those excluded (65.0% vs. 35.7%).

Table 1.

Characteristics of participants eligible for interview, by interview status

	All (n = 62)	Interviewed and Analyzed (n = 20)	Excluded1 (n = 42)
Age - Mean (SD)	61.1 (7.5)	61.6 (6.2)	60.8 (8.2)
45–54	15 (24.2)	5 (25.0)	10 (23.8)
55–64	25 (40.3)	7 (35.0)	18 (42.9)
65+	22 (35.5)	8 (40.0)	14 (33.3)
Sex
Male	21 (33.9)	12 (60.0)	9 (21.4)
Female	41 (66.1)	8 (40.0)	33 (78.6)
Race
White	58 (93.6)	20 (100.0)	38 (90.5)
Other/multiple races	4 (6.5)	0	4 (9.5)
Ethnicity
Not Hispanic	58 (93.6)	19 (95.0)	39 (92.9)
Hispanic	4 (6.5)	1 (5.0)	3 (7.1)
BMI – Mean (SD)	29.29 (5.3)	29.46 (6.0)	29.21 (4.9)
Underweight (< 18.5)	0	0	0
Normal (18.5–24.9)	11 (17.7)	3 (15.0)	8 (19.1)
Overweight (25-29.9)	26 (41.9)	9 (45.0)	17 (40.5)
Obese ( > = 30)	25 (40.3)	8 (40.0)	17 (40.5)
Education
High school graduate	12 (19.4)	2 (10.0)	10 (23.8)
College/technical school graduate	50 (80.7)	18 (90.0)	32 (76.2)
KOOS Pain Score – Mean (SD)2	36.76 (12.7)	36.58 (11.1)	36.84 (13.5)
Frequency of knee pain
Less than Daily	11 (17.7)	3 (15.0)	8 (19.1)
Daily	43 (69.4)	15 (75.0)	28 (66.7)
Always	8 (12.9)	2 (10.0)	6 (14.3)
Agree to participate in RCT3
Yes, definitely / Yes, probably	28 (45.2)	13 (65.0)	15 (35.7)
Undecided	18 (29.0)	6 (30.0)	12 (28.6)
No, probably not / No, definitely not	16 (25.8)	1 (5.0)	15 (35.7)
How many physical therapy sessions did you complete for your knee pain?
1–6 sessions	11 (17.7)	0	11 (26.2)
7–9 sessions	13 (21.0)	6 (30.0)	7 (16.7)
10 + sessions	36 (58.1)	14 (70.0)	22 (52.4)
Missing	2 (3.2)	0	2 (4.8)

¹ Of the 42 participants excluded, 39 were not interviewed and 3 were interviewed but their interview transcripts were not included in our analytic sample because they either had already received an APM (n = 1) or meniscal tear was not a primary reason for seeking care (n = 2)

² KOOS scored on a 0-100 scale with 100 = worst

³ Now that you have completed your initial course of physical therapy (PT), if offered, would you agree to participate in a randomized control trial (RCT) that compares arthroscopic partial meniscectomy (surgery for meniscal tear) versus an enhanced nonoperative multimodal treatment program for patients with meniscal tear and osteoarthritis? The nonoperative treatment program would include some combination of personalized physical therapy, injections into the knee joint, education and coaching support, and medication for reliving knee pain (such as anti-inflammatories). In a randomized trial, you are assigned to either surgery or non-operative therapy by the flip of a coin

Willingness to enroll in the planned RCT

When asked if they would be willing to enroll in a hypothetical trial (Aim 1), 5 participants (25%) responded they would not. Of the remaining participants, 14 (70%) provided a conditional “yes” to trial participation under specific circumstances related to logistics or provider preference. One participant said they needed more information to decide. To explore patient thought processes and solicit feedback on trial logistics (Aims 2 and 3), we identified six major themes corresponding with our study aims. Four themes were related to participant experiences and attitudes toward specific treatment modalities (PT, APM, oral medications, and injections), one was related to provider trust, and the final theme was related to recruitment strategies. Labels and short descriptions of each theme, along with selected respondent quotes, are summarized in Table 2.

Table 2.

Themes and selected quotes identified to address study aims

Label	Theme Description	Selected Quotes
Theme #1 Participants Open to PT	Most were familiar with and open to undergoing PT again, regardless of past experiences and continuing symptoms.	“Oh, I think [physical therapy] a wonderful tool that I’ve used when I’ve got other matters and just recently used it and if given the proper prescription -- direction from the medical folks, I would certainly do that.” Male, 65–74 years old. “If it’s pretty much the same exercises using the same equipment that I’ve used, I would probably say, well, ‘I’ve tried that.’ If it was a different modality that would be something I would be very happy to discuss. It would also depend on where the physical therapy was going to be delivered.” Male, 65–74 years old.
Theme #2 Mostly Negative “Dealbreaker” Attitudes about APM	Feelings about surgery were more complex, and surgery was a “dealbreaker” for those unwilling to enroll in the hypothetical trial; among those who were willing to enroll, their “yeses” were conditional.	“No, I just as I spoke to the doctor within your organization, you know, he said it seems the preference now is physical therapy. You know, once you start going in and scraping some of that cartilage around the meniscus, it never grows back. So if PT works, then, which it has in this case, then I’m all in.” Male, 65–74 years old. “I would very much avoid surgery as much as possible… I don’t think the knee is bad enough where it needs surgery… I always view surgery as the last resort, and I don’t think I’m… I think there’s other options I should do to kind of heal my knee before I go to surgery.” Male, 45–54 years old. “If the physician said, ‘Hey look, you probably need surgery for this knee,’ and then it was like, ‘Hey, but what we want to do is, we’re not sure if surgery is always the best option, so we’re gonna enter you into a study and based on the randomization, you’ll either get surgery or physical therapy,’ then that would be, then I would be open to because it’s like I was going down the path of surgery.” Male, 45–54 years old. “I understand the science behind [randomization] and the reasoning of it, but, frankly, I think I would drop out if I flipped to the surgery.” Female, 65–74 years old.
Theme #3 Oral Medications Did Not Pose a Barrier for Enrollment	Patients were familiar with oral medications, but had concerns about taking them on a long-term basis; they did not pose barriers for enrollment	“No, I, matter of fact, anytime I’ve been prescribed pain medication for surgeries or other issues. I take very little if any of that whatsoever and defer to aspirin to the extent I can. And even get off of that as soon as possible.” Male, 55–64 years old. “I mean, I would do that, the orthopedist that I saw had prescribed, I guess it was just—it was naproxen. So, I took 2 courses of that. And I guess it did help to pain some. I’m not averse to it, but don’t really like being on medication either. So I don’t want that to be a long term thing.” Female, 55–64 years old.
Theme #4 Mixed Feelings about Injections	Most respondents reported receiving a joint injection in the past and of those, some were ambivalent about future shots.	“I have also had trouble with my other knee, my left knee. And many years ago I had a Euflexxa injection for that, which was like amazingly beneficial. I didn’t have a tear in that meniscus though, so I’ve been told that that treatment is not necessarily appropriate for the tear that I have in my right meniscus.” Male, 55–64 years old. “When I had my torn rotator cuff, I had cortisone injections and it was wonderful… And the most recent cortisone injections, I just wish it lasted longer. It only lasted about 3 weeks and when I was talking to the physician, it was, ‘well, this will last me 4 to 6 months,’ and I was thrilled. But it just didn’t last that long. […] No [concerns], as long as they work.” Male, 65–74 years old.
Theme #5 Trust in Provider Is Paramount	Provider trust was a dominant decision-making factor in willingness to enroll [and treatment decision-making overall].	“I know, you know, you guys have all the research and everything, but I would go with, you know, with with the surgeon, what she says.” Female, 65–74 years old. “I’d probably see what his best option would be, or recommendation would be because it’s not really getting much better. It’s pretty much remaining the same, and I’ve been doing this for 3 months now, so. If he was to say we can try the surgery, then I might be prone to do that.” Male, 45–54 years old. “I’d be open to it provided it’s something that [provider name] would prescribe for me to do and feel that I would benefit from. […] I really value his professional opinion and diagnosis.” Male, 55–64 years old.
Theme #6 Study Personnel and Convenience Matters	Study personnel and convenience influenced willingness to enroll.	“I don’t have an issue with traveling or, you know, going in-person visits, but you know, at the convenience of having remote, whether it be zoom calls or any other mechanism, I think, you know, I’d be happy to support that as well.” Male, 55–64 years old.

Theme 1: participants open to PT

Participants reported a range of durations and intensities of in-clinic PT (70% attended at least 10 sessions) with two individuals not completing formal PT (they called it a home-based strengthening program using an app). Most (n = 12) had a positive overall experience with PT, while 7 expressed mixed opinions, and one described an exclusively negative experience due to worsening pain following PT. Among the participants who reported mixed opinions, they mostly said that while their symptoms did not improve as much as they would have liked, their interactions with their physical therapist and the PT facility were positive (“Yeah, I think that the experience was fine. You know, I certainly enjoyed the interactions with my physical therapist. I, you know, I don’t think it made a huge amount of difference, unfortunately, when all of it was said and done” – male, 55–64 years old).

Despite mixed reviews and 8 participants describing some dissatisfaction with their PT experience, most (n = 19) participants stated they would try PT again. One participant wanted to know what made the enhanced model in the planned trial different from the PT they already received. Four participants mentioned time commitment or location as barriers for participation. Some participants noted they “reached the end of the line,” and doubted that additional PT would be effective. One participant said: “Hm. If I thought it would work, I’d probably go for it, but at this point, I just don’t know if it would help anymore” – female, 65–74 years old. Another said: “I feel like I tried that and it didn’t work” – female, 55–64 years old. Despite these experiences and attitudes, most participants reported that they were willing to participate in a trial with PT as one of the treatment arms; in other words, no one identified aversion to the PT arm as a reason not to participate in the hypothetical trial.

Theme 2: mostly negative “Dealbreaker” attitudes about APM

Compared to participants’ views of PT, their feelings towards surgery were more complex. The majority of participants (n = 17, 85%) expressed hesitancy and/or negativity towards surgery as a treatment option. Seven participants (35%) were fearful of uncertainties of surgery including risk of failure to relieve pain, complications, and ultimately needing total knee replacement. Participants also felt hesitant about the long recovery process or the irreversible nature of surgery.

Seven participants (35%) viewed surgery as a “last resort.” They reported that to consider surgery, they would need to have pain for a prolonged period, a high level of pain, low quality of life, and/or exhausted all other treatment options. Six participants (30%) noted that the current state of their knee is not serious enough to necessitate surgery. Moreover, in almost half of the participants (n = 9, 45%), provider recommendation played a factor in their decision to consider surgery as a treatment option now or in the future. Three participants felt that they needed more information about the surgery or wanted control over logistical aspects like date of surgery and surgeon.

For participants who were “conditionally willing” to be randomized to APM, conditions included provider recommendation, exhausting all other treatments, and candidacy for surgery (“No one wants to have surgery if they don’t have to, but I think that if you’ve exhausted all of the other, you know, options and sort of gotten to a point where this is like as good as it’s gonna get without surgery, then I think you have to, you know, be all in.” – female, 45–54 years old). As mentioned previously, more than half the participants (n = 14, 70%) reported they would agree to participate in the trial under specific conditions such as provider recommendation, being a candidate for surgery, and having control over logistical factors such as location of surgery. In contrast, 5 of the 20 participants (25%) said they would decline to participate in the trial if it was offered today; four stated that surgery was the “dealbreaker” to not participate in the RCT.

Theme 3: oral medications did not pose a barrier for enrollment

The majority of the 19 participants who commented on oral medications (n = 14, 74%) had reservations about taking them. But, none stated that the inclusion of oral medications in the planned trial would influence their decision whether or not to enroll in the planned RCT. Participants described several reasons for their reluctance to use oral medication in general: they did not see oral medication as a long-term solution, worried about side effects, and were reluctant to take medication altogether. Some participants mentioned that they only use oral medications when their pain level surpasses a particular threshold (“I only take medication when I’m in such pain that I can’t stand it” – female, 65–74 years old), and one participant reported a contraindication that limits the medication they could take. However, five participants stated that they would not have any concerns about taking oral medications if they enrolled in a trial that included them, and six participants discussed the efficacy of oral medications, with all finding them to be effective (“It’s actually worked very well” – female, 65–74 years old).

Theme 4: mixed feelings about injections

Participants spoke about intra-articular injections mostly from prior experience (n = 14, 70%). None of the participants identified injections as a reason why they would not enroll in the planned RCT. Many were open to receiving injections for their knee pain, with only a few (n = 2, 10%) who were unwilling. However, respondents also reported mixed opinions about injections, with a portion (n = 6, 30%) expressing ambivalent feelings related to efficacy. Reasons for the ambivalence included varied past experiences with injections and a fear of needles. Others (n = 3, 15%) mentioned that injections were medically inappropriate for them.

Of the 14 with a history of injections, most (n = 11, 79%) reported a positive experience with symptom relief. A couple participants discussed that injections helped accelerate healing or reduced the need for additional treatments, such as PT. In contrast, a few participants (n = 3, 21%) described dissatisfaction with past injections, stating injections did not provide adequate level or duration of pain relief.

Of the participants without a prior history of injections (n = 6, 30%), a few were open to it as a treatment option. Some had heard negative reports from friends or family and were not open to receiving injections. These participants expressed concerns that injections may mask symptoms, potentially impeding symptom improvement (“I’ve heard bad things about injections, and I think it just kind of masks what’s truly going on.” – male, 45–54 years old), or that injections simply delay surgery (“I have a couple friends that have gone through knee injections and it just seems like it either works or doesn’t work. […] And, you know, basically it was to delay surgery for them because they didn’t want surgery.” – female, 65–74 years old).

Theme 5: trust in provider is paramount

The importance of provider trust was an emerging theme in our analysis with most participants (n = 14, 70%) mentioning confidence in their provider. Some participants mentioned their orthopedic surgeon by name (“I’d be open to it provided it’s something that [provider name] would prescribe for me to do and feel that I would benefit from. […] I really value his professional opinion and diagnosis” – male, 45–54 years old). These participants stated that the opinion of their provider would strongly influence their willingness to participate in our planned trial. Almost half (n = 9, 45%) of participants reported that they would defer to their provider on the next steps for treatment, randomization between treatment options, and/or deciding whether to enroll in the planned trial.

Provider recommendation was an especially important condition for participants to be open to surgical treatment (“If I’m informed by my physician that I need surgery, then it, then it becomes a no-brainer for me.” – male, 45–54 years old). For those unsure about APM, they said they would rely on information from their provider before considering surgery (“Cause the length of the procedure, the length, the recuperation time. […] What’s the, what’s the chances of it reoccurring? That’s my question, for the doctor.” – male, 45–54 years old). However, a few participants (n = 2, 10%) did not have full provider confidence and expressed mixed opinions or hesitancies with the patient-provider relationship, (“Cause I feel people are afraid of doctors and they don’t ask as many questions as they should…I think they have to be very straightforward and easy to talk to.” - female, 65–74 years old).

Theme 6: study personnel and convenience matters

Most participants (n = 14, 70%) preferred physicians over research assistants to deliver the invitation to enroll in the trial. A considerable number of participants (n = 9, 45%) mentioned convenience in location or time of appointments as a factor in their decision to enroll in the trial (“I do work full time and I travel some, so scheduling I need to kind of look at that like if I was doing some type of PT and, you know, I have to work around my work schedule.” – female, 55–64 years old).

The majority (n = 13, 65%) stated they do not have a preference between in-person or phone for the mode of research invitation. Of the seven who had a preference, three participants preferred an invitation over the phone, and four participants preferred an in-person invitation. Three participants (15%) mentioned that cost would be a deciding factor in their decision to participate in the trial and said they would not participate if visits were not covered. Other participants preferred costs to be covered, but they indicated that cost would not be a deciding factor in their study participation. Seven participants explicitly mentioned the cost of prior PT appointments during the interview. Five noted that insurance made these costs manageable; one stated that the cost was higher than they preferred, but they believed the treatment was necessary and would choose to pay anyway.

Discussion

Summary

In this qualitative PPA study of participants with physician-suspected or confirmed symptomatic meniscal tear and persistent knee pain following a course of PT, we sought to examine the willingness of seemingly eligible subjects to enroll in a planned RCT comparing enhanced conservative care to APM. Most participants (70%) said they would enroll in the planned trial. That said, many of their “yeses” were conditional, based on the specifics of the trial, provider approval, and needing to learn more information before making a definitive decision.

We identified six major themes from the interviews, four related to treatment modality, one related to provider trust, and one related to study logistics. For those who said “no” to the hypothetical trial, most stated an aversion to surgery as their reason. Despite all having persistent pain after completing a full course of therapy, most participants were open to having additional PT, including those who doubted the efficacy of PT and those with negative experiences. Participants noted being open to education and coaching, enhancements we plan to include in the future trial; most assumed these components were already included in their current PT care.

Participants described mixed feelings toward injections and oral medication use (especially long-term use), but the inclusion of either of those treatments was not described as “dealbreakers” for participation in the trial. Provider opinion about the trial (including whether they qualified and if they should enroll) and about next steps for treatment was important to most participants. Many expressed the desire to be invited to the planned trial by a provider, rather than non-clinical research staff and several stated that convenience in timing and/or location was an important factor when deciding to enroll. Cost of treatments, in this study context at least, did not appear to be a major factor in participant decision-making.

In the End of PT questionnaire, 65% of the interviewed participants (13 of 20) said that they would “yes, definitely” or “yes, probably” participate in the hypothetical trial (Table 3). On the other hand, in their interviews, 70% said “yes” to participating in the hypothetical trial, but most of those yeses came with the conditions described above. Notably, less than half (45%) were consistent between their questionnaire and interview answers (i.e., said “yes” on the questionnaire and “yes” when interviewed); the remainder changed their minds (e.g., of the 6 participants who selected “undecided” on the questionnaire, all said “yes” during the interview; 4 who answered “yes” on the questionnaire answered “no” in the interview). Even though all interviews were conducted within a month of the completion of questionnaires, participants’ preferences may have changed during this time. Additionally, after being allowed to discuss more details about the planned trials in the interviews, some participants may have developed more nuanced perspectives. Ultimately, these differences between closed-ended responses on the questionnaire and more open-ended responses during an interview highlight the importance of qualitative work to capture a more nuanced and perhaps more accurate perspective regarding willingness to enroll in a planned study than a quantitative survey might.

Table 3.

Interview and questionnaire responses of interviewed participants

	Interview Responses2
Questionnaire Responses1	Yes	Yes (conditional)	Unsure	No	Total
Yes, definitely	3	1	0	2	6
Yes, Probably	1	3	1	2	7
Undecided	2	4	0	0	6
No, definitely not	0	0	0	1	1
Total	6	8	1	5	20

¹ Questionnaire responses based on the “End of Physical Therapy” questionnaire, self-reported by the participant and reported prior to the interview. The question was: “Now that you have completed your initial course of physical therapy (PT), if offered, would you agree to participate in a randomized control trial (RCT) that compares arthroscopic partial meniscectomy (surgery for meniscal tear) versus an enhanced nonoperative multimodal treatment program for patients with meniscal tear and osteoarthritis? The nonoperative treatment program would include some combination of personalized physical therapy, injections into the knee joint, education and coaching support, and medication for reliving knee pain (such as anti-inflammatories). In a randomized trial, you are assigned to either surgery or non-operative therapy by the flip of a coin.”

² Answer elicited by the interview question that occurred 8–36 days after the questionnaire was completed. The interview prompt was: “If you were asked to participate in this trial right now, what would you say?”

Alignment with prior literature

Previous qualitative studies report similar themes of treatment preferences in patients with meniscal tear, knee OA, and anterior cruciate ligament (ACL) injury. Like many of our participants who viewed surgery as a last resort, Ahmed et al. reports that participants with meniscal tear want to exhaust non-surgical treatments before undergoing surgery [16]. Several participants with knee OA described how their willingness to receive additional injections was in part motivated by a desire to avoid surgery [17]. Moreover, in a qualitative study where patients with an ACL injury were asked about participating in a hypothetical RCT involving metformin, 44% stated time and distance to the hospital are important factors to consider before enrolling in the trial, and 58% preferred to have a healthcare team member to introduce the study to them [18].

Other PPA studies have noted differences between quantitative and qualitative assessments of patients’ willingness to enroll in trials [11, 19]. In orthopedic RCTs with surgical and non-surgical study arms, researchers have observed that enrolling patients can be difficult due to treatment preferences [20, 21]. Of the interviewed participants, 65% indicated in the End of PT questionnaire that they would enroll as compared with 70% in the interviews. This is consistent with the range found in the literature of 46%-90% [4–6].

Limitations

While we enrolled participants across two cities and 10 orthopaedic clinics, the study population was relatively homogenous - all participants identified as White, and most were college graduates – and our findings may not capture the perspectives of those from other backgrounds. Future studies could mitigate these limitations by partnering with additional sites, including non-academic environments, and using purposive sampling to increase participation of different racial and ethnic groups. Furthermore, as the interviews were conducted on Zoom, this option could have excluded participants with limited digital access, leading to a biased sampling of interviewed participants.

Second, the participants we interviewed were more likely to say that they would participate in the hypothetical trial, compared to the excluded participants (65.0% vs. 35.7%). Given that our participants agreed to participate in the PPA study and be interviewed, they are more likely to agree to participate in the hypothetical trial, and this may have resulted in an overestimation of the proportion of otherwise eligible participants who are likely to agree to enroll. That said, we deliberately only interviewed participants who completed “End of PT” questionnaires and reported persistent pain; zeroing in on this patient population is a strength of our study. Additionally, there are likely unexplored factors that influence willingness to enroll that we did not capture. For example, we did not ask about participants’ expectations of treatments. We did not require MRI confirmation of meniscal tear in this pilot study and note that 84% of participants with similar entry and exclusion criteria in our recent TeMPO trial had meniscal tear confirmed by MRI [22]. Finally, it is possible that participant responses to our interview questions could differ from how they would actually answer if invited to participate in a trial. Nevertheless, PPAs are helpful tools for estimating enrollment sample sizes and can provide an estimated range of trial enrollment using participants’ responses [7].

Implications for research

These data indicate that despite an “unsuccessful” course of PT (all completed at least 7 sessions of PT yet 85% continued to report knee pain “daily” or “always” and reported a KOOS score of at least 20), participants are willing to continue with PT. This group of participants, for the most part, reported being reluctant to opt for an APM, at least at this stage in their disease progression. Some viewed surgery as an inevitability but wanted to postpone it for as long as possible. This finding will help to inform us how we design eligibility criteria for the trial. In addition, considering the various treatment-specific concerns participants expressed in the interviews, the trial may benefit from providing detailed education materials that help participants grasp the short and long-term benefits and potential harms of each treatment.

A major implication of this research is the strong influence of provider input on patient decisions. The high level of trust participants expressed for their provider’s recommendations underscore the importance of clinician equipoise – comfort recommending the trial. As many participants expressed nuanced concerns for various treatments in their interviews, allowing participants to discuss their thoughts with their providers may be an important recruitment step. Furthermore, it would be beneficial to work with enrolling providers on ways to present this trial to fully inform patients of their options and encourage enrollment. This might include providing potential participants with success rates of all available treatment options. Ensuring that the location and schedule (time/day) of treatment are convenient for participants in the future trial will be important.

Conclusions

This qualitative study suggests that our planned trial should include additional patient education regarding the value of all treatment options at the time of enrollment invitation (APM in particular) and that this should be delivered by a trusted provider. More generally, the findings underscore that potential RCT participants may have distinct preferences that should be queried to better understand whether they would enroll in a trial. Methodologically, this study suggests that qualitative studies may be an informative complement to quantitative studies to examine nuanced perspectives of patients.

Abbreviations

PPA

Prospective Preference Assessment

RCT

Randomized Controlled Trial

PT

Physical Therapy

APM

Arthroscopic Partial Meniscectomy

OA

Osteoarthritis

KOOS

Knee Injury and Osteoarthritis Outcome Score

SD

Standard Deviation

ACL

Anterior Cruciate Ligament

Authors’ contributions

*Conception and design: * JK, FS, NWL*Acquisition of data: * AC, JD, KF, KK, TT*Analysis and interpretation of the data: * All authors*Drafting of the article* : AC, JD, KF, JK, PO, TT, NWL*Substantive revision of the article: * All authors*Final approval of the article: * All authors.

Funding

The study was supported by NIH/NIAMS Grant 5R34AR082021-02.

Data availability

The data from the study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The institutional review boards (IRB) of Mass General Brigham and University at Buffalo approved the study protocol (Mass General Brigham: Protocol 2022P003110 and University at Buffalo: Protocol STUDY00007422). The research study was conducted in accordance with the Declaration of Helsinki and the Belmont Report. Written consent was not obtained, but implied and verbal consents (with a Fact Sheet for the PPA and later verbally at the start of the interviews) were provided by all the participants based on their specific participation. The Fact Sheet fully informed the participants of the potential risks, benefits, purpose, and nature of the study and that while some identifiable health information would be stored in our secure data capture system, all responses would be deidentified and stored securely. Agreement in both instances was recorded in the study’s data capture system, REDCap. Each of these procedures was approved by the governing IRB.

Consent for publication

Not Applicable.

Competing interests

JNK declares receiving support from Biosplice for a cohort study of OA patients.