Sex-Based Differences in the Utilization of Shoulder, Hip, and Knee Arthroplasty

Abstract

Introduction:

Studies show that females have a higher prevalence of osteoarthritis, worse symptoms, but lower rates of joint replacement surgery (JRS). The reason for this remains unknown.

Methods:

A database of JRS candidates was created for patients seen in 2019 at an academic center. Demographics, Kellgren-Lawrence grades, symptom duration, visual analogue pain score, Charlson Comorbidity Index, and nonsurgical treatments were collected. Patients who were offered but declined surgery were invited to focus groups. Two independent sample t-tests, Mann-Whitney U tests, and chi-square tests were used for continuous, scored, and categorical variables, respectively, with two-tailed significance <0.05. Qualitative, code-based analysis was performed for the focus groups.

Results:

The cohort included 321 patients (81 shoulder, 59 hip, and 181 knee) including 199 females (62.0%). There were no differences in proportions of females versus males who underwent JRS or in nonsurgical treatments. Female shoulder arthritis patients were older, had a higher visual analogue pain score, and had a higher Charlson Comorbidity Index. In focus groups, males prioritized waiting for technology advancements to return to an active lifestyle, whereas females experienced negative provider interactions, self-advocated for treatment, concerned about pain, and believed that their sex affected their treatment.

Discussion:

We found equal utilization of JRS at our institution. However, female patients experienced unique barriers to surgery.

Sex bias is pervasive throughout medical care, and the treatment of osteoarthritis (OA) is no exception. A cohort-based study found that when adjusted for willingness, females had three times the need for arthroplasty when compared with males and were notably less likely to have undergone surgery.¹ Studies have also found that female patients are less likely to have discussed surgical treatment with a provider, be referred to an orthopaedic surgeon, and offered joint replacement surgery (JRS) even when compared with standardized male patients with identical clinical presentations.^1-7 The sex-based underutilization of JRS is well documented, but the reason for the disparate use remains unknown.

Another consideration is the role sex has in the development of OA and the degree of symptomatology. There is a higher prevalence of radiographic and symptomatic OA in females, which may be attributed to genetic and anatomic variables.⁸ Females demonstrate increased heritability of OA in those with a specific homozygous allele of the estrogen receptor, have increased inflammatory cytokines in synovial fluid, and have different hormone receptors on chondrocytes.^9,10 For example, in the knee, females have a narrower distal femur with less cartilage volume and a larger Q angle than males, increasing the risk for developing OA.^11-13

Female patients also present with more severe symptoms and greater disability.^14-16 A retrospective analysis found that female patients who underwent total knee arthroplasty (TKA) were, on average, 3 years older and had worse knee function than male patients, indicating that they present further in their disease course.¹⁴

Arthroplasty is the benchmark surgical treatment for moderate-to-severe OA once nonsurgical treatment has been exhausted and provides durable, reliable pain relief and improved functionality and quality of life. Thus, it is imperative to ensure equitable access and minimal barriers to JRS. This study sought to (1) generate a database of shoulder, hip, and knee joint replacement candidates at our institution and determine whether there is a difference between rates of males and females undergoing JRS, (2) understand whether differences exist in the degree of symptoms between female and male patients, and (3) qualitatively describe reasons for not undergoing JRS. We hypothesized that our university cohort would follow prior trends, and female patients would undergo JRS less than male patients and would experience different barriers to surgery than male patients.

Methods

This study received institutional review board (IRB) approval and was conducted in two phases, including (1) development and analysis of a database of JRS candidates and (2) focus groups designed to investigate reasons for declining JRS.

Database Analysis

First, a database of joint replacement candidates was developed by reviewing new patient visits at orthopaedic sports medicine and orthopaedic arthroplasty clinics at an academic tertiary referral center for the calendar year 2019. A cohort of patients who qualified for and were offered JRS but declined surgical treatment was identified.

Cohort inclusion criteria included patients with Kellgren-Lawrence (KL) grade 3 to 4 radiographic changes,¹⁷ primary total shoulder arthroplasty (TSA) or reverse TSA candidates, TKA candidates, total hip arthroplasty (THA) candidates, and those with Medicare insurance. Prior surgery, if not JRS, was acceptable (ie, rotator cuff repair surgery or knee arthroscopy). The KL grade was determined by the review of radiographs by a trained orthopaedic surgery resident. KL grading was chosen because it has been shown to have substantial intraobserver and interobserver reliability.¹⁸

Exclusion criteria included KL grade 1 to 2, patients with diagnoses other than primary OA (ie, posttraumatic OA, inflammatory arthritis, or osteonecrosis), hip and shoulder hemiarthroplasty, unicompartmental knee arthroplasty, nonelective JRS such as THA for femoral neck fracture, revision arthroplasty candidates, and patients who were not offered surgery due to severity of medical comorbidities or body mass index >40 for hip and knee arthroplasty in accordance with the existing protocol.

The objective data collected from each new patient clinic visit note included age, sex, Charlson Comorbidity Index (a validated measurement of the risk of 1-year mortality and disease burden),¹⁹ visual analogue pain score (VAS), time from symptom onset to presentation to orthopaedic arthroplasty or sports clinic, and nonsurgical treatments tried (NSAIDs, opiate medications, injections including steroid or viscosupplementation, physical therapy, and bracing treatment). All orthopaedic arthroplasty surgeons at this institution are male.

Focus Groups

Members of the cohort were invited through e-mail to participate in virtual, hour-long focus groups to better understand their current experiences with arthritis, experiences with care providers, goals and perspectives, and current opinions on JRS. All those who responded to the invitation participated in the focus groups. In total, there were two focus groups that included six participants of the same sex along with one study team moderator and one study observer. Both study team members had prior training and experience in conducting interviews in qualitative research methodologies. Each group was asked the same set of questions from a semistructured interview guide, included in Appendix A, http://links.lww.com/JG9/A293. The questions were based on the framework from the prior focus group and survey studies investigating preferences and concerns in patients considering JRS.^20-22 The focus groups were recorded and professionally transcribed using Trint software. All participants received a $20 gift card for their participation.

Focus groups were coded using Dedoose software. The initial codebook was generated following the focus groups with a multistaged mixed inductive and deductive approach (Table 1). The parent codes were based on each section of the focus group interview guide and included Current Experiences with Arthritis, Patient Experiences with Care Providers, Patient Goals and Perspectives, and Current Opinions. Salient themes were identified within each parent code through a mixed iterative process after multiple passes through the focus group transcripts.

Table 1.

Codebook From the Focus Group Analysis

Code	Description
Current Experiences with Arthritis
Impact on daily tasks and responsibilities	Mentions the impact of arthritis on activities that patients are obligated to do (not recreation)
Impact on hobbies and recreational activities	Discussions of how arthritis affects the patient's abilities to participate in sports and hobbies they enjoy
Preoperative pain	Patient's experience with pain and discomfort before treatment
Patient Experiences with Care Providers
Advised to delay treatment	Patient's being advised by healthcare providers to put off treatment or surgery; this can include patients being told that their condition is not severe enough yet or that surgery will not help
Age-based differences	Patient's experiences in receiving care that they explicitly attribute to be based on their age; this may include being told that they are too young for certain treatments
Lack of access to providers	Patient struggles trying to access appropriate healthcare providers to care for their arthritis
Negative provider-patient interactions and distrust	Discussions of experiences with orthopaedic healthcare providers which patients found unsatisfactory; this can also include distrust in providers
Positive provider-patient interactions and trust	Discussions of experiences with orthopaedic healthcare providers which patients found positive; this can include general trust in providers
Seeking information outside of the doctor's office	Discussions of seeking more information outside of doctors' visits to answer further questions or gain more clarity or insight
Self-advocacy	Patients' experiences needed to advocate for themselves when receiving care; this can include feeling that physicians do not take their experience seriously or feeling the need to demand certain care or treatments
Sex-based differences	Patients' experiences of differences which they explicitly attribute to be based on their sex; this can include mentions of unideal surgery conditions and an explicit mention in describing physician-patient dynamics
Patient Goals and Perspectives
Postoperative relief	Patients' desires to relieve their joint pain through surgery
Regain partial functionality	Discussions in which patients are not seeking complete return to baseline but at least want to regain some degree of functionality
Return to active life	Patients' desires to be able to participate in sports and activities again; these will be discussions of wishing to once again participate in the recreational activities they find fulfilling
Current Opinions: Why or Why Not?
Apprehension	Patients' general overall apprehension or nervousness about undergoing the operation
Doubts of degree of success	Concerns over how much functionality or pain relief can actually be gained from surgery
Negative accounts from friends and families	Stories of friends and families who have undergone JRS that were negative; this can include surgeries that were complicated or that did not adequately relieve symptoms
Positive accounts from friends and families	Stories of friends and families who have undergone JRS that were positive
Not ready for surgery	Feelings that patients' condition is not bad enough to warrant surgery at this moment
Postoperative pain	Patients' concerns over pain and pain management after surgery
Postoperative recovery and accommodations and support	Discussions of concerns or thoughts regarding the postoperative recovery period and adjustments that may be needed in this time and support they are seeking
Surgical complications	Concerns over complications during or after surgery
Technology advancements	Patients' thoughts on the state of technology advancements in JRS

Codes were generated for experiences with arthritis, experiences with care providers, patient goals and perspectives, and opinions on JRS.

Reflexivity

The focus group moderator is a female second-year preclinical medical student. She is concurrently enrolled in a master's program in public health, which includes training in qualitative research methodology. The study team members conducting the focus groups had not viewed any patient records before conducting the focus groups. Her prior interactions with the patients included reaching out through e-mail and phone to recruit participants for the study focus groups. The focus group observer is a female fourth-year medical student applying for residency in physical medicine and rehabilitation. She has a BA in psychology and has experience with multiple qualitative research projects. She was also involved in the chart review for this project but has otherwise never interacted with the focus group participants. On reflection, their identities as female-bodied individuals could potentially influence female focus group participants' comfort in discussing perceptions of sex-based biases in their arthritis care. The moderator and observer met immediately after interviews to reflect on interviewer-interviewee dynamics throughout the focus groups.

Statistical Analysis

Descriptive statistics were used to characterize data and demographic variables.

Two independent sample t-tests, Mann-Whitney U tests, and chi-square tests were used for continuous, scored, and categorical variables, respectively.

All statistical comparisons were performed with a two-tailed level of significance <0.05. The focus group recordings were transcribed and coded using Dedoose (qualitative data analysis platform, Los Angeles, CA). Each code was weighted equally, and the coding process was repeated iteratively to ensure intercoder reliability.

Results

Database Analysis

There were 1038 JRS candidates seen at our institution in the year of 2019 (226 shoulder, 251 hip, and 561 knee). The cohort of patients who met the inclusion criteria and were offered but declined surgery included 321 total patients (81 shoulder, 59 hip, and 181 knee) with 199 females (62%). The mean age was 75 + 7 years.

The rates of undergoing JRS were equal in women and men for shoulder, hip, and knee osteoarthritis (Table 2). Male knee arthritis patients trended toward longer symptom duration before presentation (65 + 117 vs. 39 + 48 months, P = 0.05) (Table 3). Female shoulder arthritis patients were older (78 ± 8 vs. 74 ± 8 years, P = 0.02), had a higher VAS (7 ± 2 vs. 5 ± 2, P = 0.02), and had a higher Charlson Comorbidity Index (5 ± 2 vs. 4 ± 2, P = 0.03) (Table 4). There were no other notable sex-based differences (Tables 3–5). There were no differences between females and males who trialed nonsurgical treatments (NSAIDs: 72% F vs. 69% M, P = 0.68; opioids: 9% vs. 7%, P = 0.51; injections: 44% vs. 38%, P = 0.26; physical therapy: 44% vs. 37%, P = 0.25; bracing treatment: 15% vs. 19%, P = 0.30).

Table 2.

Proportions of Female and Male Patients Who Were Offered and Underwent Surgery

Joint	Female	Male	P
Shoulder	36/84 (42.8%)	22/55 (40.0%)	P = 0.74
Hip	73/112 (65.2%)	57/77 (74.0%)	P = 0.20
Knee	97/209 (46.4%)	60/129 (46.5%)	P = 0.98

Table 3.

Knee Cohort Analysis

Knee	Female	Male	P
Age (y)	75 ± 7	75 ± 6	0.59
KL 3	73	43	0.65
KL 4	37	26	0.65
VAS	5 ± 3	5 ± 3	0.65
Symptom duration (mo)	39 ± 48	65 ± 117	0.05
CCI	4 ± 1	4 ± 2	0.84

CCI = Charlson Comorbidity Index, KL = Kellgren-Lawrence, VAS = visual analogue pain score. The bolded entries met statistical significance p<0.05 or near statistical significance at p=0.05.

Table 4.

Shoulder Cohort Analysis

Shoulder	Female	Male	P
Age (y)	78 ± 8	74 ± 8	0.02
KL 3	24	22	0.14
KL 4	22	11	0.12
VAS	7 ± 2	5 ± 2	0.02
Symptom duration (mo)	53 ± 108	48 ± 94	0.82
CCI	5 ± 2	4 ± 2	0.03

CCI = Charlson Comorbidity Index, KL = Kellgren-Lawrence, VAS = visual analogue pain score. The bolded entries met statistical significance p<0.05 or near statistical significance at p=0.05.

Table 5.

Hip Cohort Analysis

Hip	Female	Male	P
Age (y)	75 ± 7	76 ± 9	0.53
KL 3	25	15	0.39
KL 4	14	5	0.39
VAS	6 ± 2	5 ± 2	0.57
Symptom duration (mo)	26 ± 21	25 ± 29	0.94
CCI	4 ± 1	4 ± 2	0.31

CCI = Charlson Comorbidity Index, KL = Kellgren-Lawrence, VAS = visual analogue pain score

Focus Group Analysis

Table 6 shows the salient themes that emerged from the focus group transcripts alongside exemplary quotes demonstrating each theme. We found that the theme of self-advocacy was unique to female patients, and negative interactions with care providers were much more prevalent among females interviewed. Females felt that physicians often did not engage them equally as partners in decision-making processes, leading to feelings of distrust.

“But I also know that there is underlying truth to the fact that if a physician doesn't buy into really thinking that you as the patient need to be treated with respect and needed to be listened to and need to be a partner in the process, that's where the most mistakes have been made in my history with the medical industry, so to speak. It is when I don't feel like people really engaged me as a partner in the process.” (F)

“I've learned that surgeons overestimate how successful they're going to be. You know that they, you know, in a way, it's positive. They've got egos. They're invested. But, you know, in a way, it's led me to sometimes underestimate the support that I would need afterwards.” (F)

Table 6.

Focus Group Themes With Exemplary Quotes

Theme	Exemplar Quotes
Similarities in current experiences with arthritis: pain	“Pain, I've had a lot of pain. I am sick of pain. And you know, that's the main thing.” (M) “I'm ready for the knee because it has gotten really painful while I had put it off for a long, long time. And so it's not going to get better over time. And when I was working as a librarian at college libraries and I like to walk a lot and that is difficult to do this way. And my husband and I had a little dog that likes to walk. We all do. I'm not an athlete.” (F)
Negative interactions with care providers among females	“I've learned that surgeons overestimate how successful they're going to be. You know that they, you know, in a way, it's positive. They've got egos. They're invested. But, you know, in a way, it's led me to sometimes underestimate the support that I would need afterwards.” (F) “Arthritis appears in different people, in different places, and it has a lot of effects, which I don't understand, and no doctor was ever very interested in my complaints.” (F)
Self-advocacy among females	“But I also know that there is underlying truth to the fact that if a physician doesn't buy into really thinking that you as the patient need to be treated with respect and needed to be listened to and need to be a partner in the process, that's where the most mistakes have been made in my history with the medical industry, so to speak. It is when I don't feel like people really engaged me as a partner in the process.” (F) “It takes a lot of willingness to like, strongly advocate for yourself to get out of the more automatic response ‘We don't give [opiates] or we don't give them past the second week’ or something like that. So if you're wanting to do that to really decide that you will advocate for yourself about that.” (F)
Differences in postoperative goals	“Also, I'd just like to go back to being 55 and going to the gym for two hours. That would satisfy me. I just, you know, the endorphins off football that and, you know, the activity. So that's my goal.” (M) “All I want to do is walk my dog and, you know, be able to go hiking and do some activities.” (F)
Positive experiences from friends and families	“I watched my father not do it and be crippled in his very last years in a lot of pain, and I am hopeful that I can get through this and get back to the activities that I want. And I've seen my sister do it earlier than I, and she's out there playing pickleball… The fact that she's done as well as she has, it's just been an inspiration for me to move forward with this.” (F)
Negative experiences from friends and families	“My father had both of them done in his late 70s, and … I really don't think that should have been done, even though he had a lifetime of bad knees and stuff. You know, it … should have been done earlier or less risky and, you know, just not both at the same time as that's a lot of pressure on the body…” (M)
Male-specific barriers: technological advancements	“I'm still waiting for, you know, maybe some sort of major breakthrough such that the replacement knee will be as strong as the existing knee.” (M)
Female-specific barriers: postoperative pain	“… I find it difficult to be very inspiring. And I don't want to just cower. So I guess I'm trying to deal with this realistically, and the pain does make me nervous and coming in the middle of that to argue in favor of getting more painkillers because the doctors aren't going to do that. That's not very positive, either.” (F)
Impact of sex on the quality of care received	“So I continue to worry about. The way in which people are either listened to or not listened to in the process. And I think when you talk about joints or joint replacement, it's interesting that we're a group of women, so I think sexism can come into play.” (F)

Males and females differed in their primary postoperative goals. Among males, returning to an active lifestyle was a prominent theme in discussions of postoperative goals that was not as strongly shared by females. For males, being able to play sports and exercise was a major driver of their desire to treat their arthritis. The themes around females' goals after JRS were more centered around maintaining their current functional status and being able to perform their activities of daily living.

“Also, I'd just like to go back to being 55 and going to the gym for two hours. That would satisfy me. I just, you know, the endorphins off football that and, you know, the activity. So that's my goal.” (M)

“All I want to do is walk my dog and, you know, be able to go hiking and do some activities.” (F)

Males and females also cited unique barriers to JRS. The desire to wait for technology advancements in joint replacements was a common theme among males, whereas for females, fear of postoperative pain was a primary concern that posed a significant barrier. Females expressed concerns that their pain would not be properly managed by their providers after surgery.

“I'm still waiting for, you know, maybe some sort of major breakthrough such that the replacement knee will be as strong as the existing knee.” (M)

“… I find it difficult to be very inspiring. And I don't want to just cower. So, I guess I'm trying to deal with this realistically, and the pain does make me nervous and coming in the middle of that to argue in favor of getting more painkillers because the doctors aren't going to do that. That's not very positive, either.” (F)

An additional theme unique to the females interviewed was the feeling that their sex affected the quality of care they received from providers. They expressed concerns that their identities as women posed a barrier to being engaged and listened to in the decision-making processes around treating their arthritis. This theme was absent from male focus groups but arose prominently among females.

“So I continue to worry about the way in which people are either listened to or not listened to in the process. And I think when you talk about joints or joint replacement, it's interesting that we're a group of women, so I think sexism can come into play.” (F)

Discussion

The objectives of this study were to determine whether the rate of joint replacement and symptoms differed between males and females and to identify whether there were sex-based barriers to undergoing JRS. We found that males and females underwent JRS at similar rates for hip, shoulder, and knee osteoarthritis at our institution. This study identified certain differences in joint replacement candidates. Ultimately, these data highlight potential sex-based differences in care that may influence patients' decisions to undergo surgical treatment or not.

Our study showed there was no sex-based difference in proportions of patients who were offered and underwent JRS. Hawker et al²³ conducted a prospective cohort study to evaluate the probability of undergoing JRS and found that willingness to consider JRS was the key determinant to receiving JRS, but, when willingness was removed, the education level became the prominent determinant. They found an increasing probability of undergoing JRS with increasing education level. This may be a factor that contributed to our results because adults in our city have higher levels of education and income compared with the national average in the United States. In the study city, 59% of adults have a bachelor's degree versus 38% in the United States, and 24% have an advanced degree, compared with 14%.^24,25 The higher levels of education in the study's city may have mitigated any sex-based differences in rates of JRS. In addition, the progressive culture in the study's city may have contributed to providers being more cognizant of providing equitable care.

The discussion from the focus groups highlights different barriers to JRS between the sexes. The barriers that male patients discussed in the focus groups were focused on physical outcomes. Male patients stated that they wanted to wait for technological advancements to be as good as a native knee with the postoperative goal to return to an active life. Male patients also spoke more frequently about negative JRS experiences from friends and families and how they inhibited participation in active hobbies. Karlson et al²² conducted focus groups of patients who were scheduled for or already had JRS and also found that male patients were focused on returning to high activity levels such as exercise, whereas female patients were not considering JRS until worse functional decline or being unable to conduct activities of daily living. Our focus group indicates that male patients undergo JRS to return to an active lifestyle and are willing to wait for technological advancements.

In the focus groups, female-reported barriers to care were largely interpersonal. Among female patients, themes surrounding negative experiences with care providers and the excessive need to advocate for their treatment emerged. These concerns were particularly prevalent in discussions around pain management. By contrast, the theme of self-advocacy as a barrier to JRS never emerged in the male focus group. These findings may be a result of female patient past experiences with healthcare providers, as female patients are treated differently for their pain. A recent literature review found that female patients with chronic pain received inferior care from their physicians, leading to higher levels of distrust in their providers.²⁶ In addition, female patients were less likely to receive pain medication and more likely to be prescribed antidepressants and mental health referrals.²⁶ This may at least be partially due to bias because sex has been found to introduce bias in pain evaluation, with female patients' pain often being underestimated.²⁷ The fear of postoperative pain may also be related to the fact that female patients experience more severe pain from OA preoperatively. Pain is multifactorial, but biology may play a role in this difference. Increased c-reactive protein (CRP) has recently been implicated in increased joint pain with OA in female patients but not male patients, and serum IL-17F and IL-23 were associated with increased bone marrow lesions in female patients with OA, suggesting that there may be a sex difference in acute phase reactants and the development of painful OA.²⁸

Historical data in JRS show that female patients endorse higher levels of distrust and lower levels of confidence in their surgeons.²² The discrepancy in interactions may also be related to sex discordance because every arthroplasty surgeon at the researched institution is male. A recent cohort-based study found that female patients treated by male surgeons had notably worse postoperative outcomes and an increased risk for death and complications when compared with female patients treated by female surgeons or male patients treated by male or female surgeons.²⁹ Our findings further support this observation because our female focus group believed that their sex affected their care, whereas the male group did not. Furthermore, female patients in our focus group cited age as a limitation to their treatment twice as often as males. This may be partially explained by the age discrepancy in the shoulder subgroup. The female-reported barriers to care may be a result of sex discordance with providers, prior negative experiences, and fear of postoperative pain as preoperative pain is already more severe.

Shared decision making may assist the provider-patient relationship by encouraging open communication and understanding of the patient's perspectives, values, and needs while providing them with comprehensive information about treatments, potential risks, benefits, and available alternatives. This process enables patients to make well-informed decisions and can build trust with the provider.³⁰ One study investigating the decision-making process of treatment for knee OA similarly found that female patients had less baseline knowledge about OA and TKA and were less informed than male patients after their consultation.³¹ Another study comparing the patient-provider communication in recommendation of TKA with standardized patients found that there were less elements of informed decision making with female patients, with only 15% of physicians discussing the patient's preferences.⁷ Female patients experience bias that negatively affects the patient-provider relationship and presents barriers to informed decision making and ultimately treatment. Standardized educational materials or unbiased tools may help alleviate this barrier. A recent randomized clinical trial of patients presenting with painful knee OA compared artificial intelligence (AI)-integrated decision aids that included education, an assessment of patient preferences, and individualized estimations of outcomes to a control group who were given educational information and usual care.³⁰ The group with AI expressed better decision quality, collaborative decision making, satisfaction, and improved functional outcomes. The AI-enabled group did not differ in the rate of TKA or visit time.³⁰ AI may offer a solution for patients seeking consultation for OA where sex-discordant care is the only option.

Female patients may have had prior poor experiences that compromised their trust in healthcare providers, leading them to delay seeking care until functional decline from OA inhibits activities of daily living. This may be reflected in our findings that female shoulder replacement patients had higher VAS scores than male patients, indicating that they may have waited to present for consultation until they experienced greater pain and disability. Our findings suggest that the increased education level allows female patients in our university cohort to overcome this barrier, potentially by self-advocation, and receive JRS at the same rate as male patients. It is important to create awareness of these unique, interpersonal barriers to female patients receiving JRS. These findings present opportunities for improved patient care, building of trust and relationships in the healthcare system, and enhanced communication on different topics depending on the patient's individual goals and concerns.

Limitations

The limitations of this study include those that are inherent to qualitative data analysis and responder bias. To ensure that our focus group questions were limited in bias, we used prior studies to formulate the questions and had skilled personnel with a background in health and medical sciences including qualitative methods to create and proctor the focus groups. Because of our small sample size, we are unable to say whether saturation of themes was met. In addition, because of responder bias and the need to conduct the focus groups virtually in light of COVID-19, we were more likely to receive interest from patients who had higher technological literacy. This may or may not be associated with health literacy. Because our study included only participants from one medical center, we are unable to predict whether similar themes would emerge with patients from other geographic regions. Finally, although our statistical significance was defined by the cutoff of P = 0.05, the magnitude of difference may not all be clinically important or relevant.

Conclusions

We found equal utilization of JRS in males and females at our institution. We suspect that the increased education level in our geographic region may have aided female patients' ability to successfully advocate for and obtain JRS because prior studies have indicated that female patients may receive less adequate education about surgery compared with their male counterparts.²⁰ The focus group discussions highlight important barriers to JRS that warrant further investigation. Female patients expressed sex-based bias in their interactions with healthcare providers, and male patients did not. Female patients also reported needing to advocate for the treatment of osteoarthritis. Given these findings, individualized counseling regarding JRS is recommended. Subsequent studies aim to expand our understanding of barriers to JRS by using the focus group data to develop a detailed survey to investigate topics including education on joint replacement, provider biases, and patient perceptions.