- Top of page
This article provides evidence that may be helpful in eliminating racial disparity in the rate of total knee replacements (TKRs) performed in the US, the sixth arthritis-related Healthy People 2010 objective (1). Knee osteoarthritis (OA) is a leading cause of disability in the United States, affecting approximately 6% of the adult population (15.7 million adults) (2–4). Total joint replacement is considered a valuable treatment for end-stage OA; it is cost effective (5), can alleviate the majority of disability associated with these joints, and can greatly improve quality of life (QOL) for these patients (6–8). However, TKR remains underutilized among the US population, with wide variation by race/ethnicity and gender (2, 9–13). Reasons for this variation remain uncertain. This study examines differences in patients' concerns regarding TKR by race/ethnicity and gender.
Significant disparities exist in TKR utilization between African Americans and white Americans. Medicare enrollment and claims data show that whites received TKR at a rate of 18.2 per 10,000, more than twice that of African Americans (12). The lower rates persist even when adjusted for age, sex, and insurance coverage (9). These findings suggest many minorities are not receiving needed interventions to decrease pain and disability.
Data from the Framingham study indicate that knee OA is more common in women, rates increase with increasing age, and some groups of women have 1.7 times the incidence of knee OA compared with men (14). Women also have significantly worse preoperative functional status than men at the time of TKR for OA (13, 15–17). These differences persist even after adjusting for demographic and clinical characteristics. Recent findings demonstrate that decreased preoperative functional status leads to worse postoperative function (18). These data suggest that women choose TKR at a more advanced disease stage than men, possibly at the expense of QOL.
- Top of page
Thirty-seven people, all actively considering TKR, participated in the 6 focus groups between 1998 and 1999 (Table 1). Participants ranged in age from 39 to 76 years (mean age 60 years); 12 were men and 25 were women. Twenty identified themselves as white Americans and 17 as African Americans. Of the white Americans, 5 said their culture and upbringing was mainstream white American but their ethnicity included Latino (n = 4) and Native American (n = 1). Because results for those 5 individuals were similar to those of the other white Americans, the results we present are based on analysis comparing 20 white Americans with 17 African Americans.
Table 1. Patient characteristics (n = 37)
| ||White American men||White American women||African American men||African American women|
|Age, range, years||51–75||31–67||57–72||48–74|
|Household income ≤$30,000/year, %||12.5||8.3||0||69.2|
|Education ≥high school, %||87.5||91.7||25.0||61.5|
|One or more other people in the house, %||75.0||75.0||50.0||30.8|
Participants spontaneously generated 126 questions representing areas they wanted to understand better prior to making a decision regarding surgery. We divided these issues into 3 categories: 1) preoperative phase, defined as issues that clarify patient understanding of the procedure up to the surgery; 2) intraoperative phase, defined as the surgery itself plus the immediate postoperative recovery period in the anesthesia recovery room; and 3) postoperative phase, defined as the period after the patient leaves the anesthesia recovery room. Table 2 lists participants' concerns, tabulated by race/ethnicity and gender.
Table 2. Types of concerns (n = 60) from focus group respondents divided by race and gender of respondents (n = 37)*
| ||White American men||White American women||African American men||African American women|
|Preoperative concerns (n = 60)|| || || || |
| No. concerns||11||34||4||11|
| Types of concerns||Lifespan of prosthesis||Lifespan of prosthesis||Finance||Lifespan of prosthesis|
| ||Timing of TKR||Options/alternatives||Trust in physician||Options/alternatives|
| ||Addiction to medications||Device/technology|| ||Addiction to medications|
| ||Employment||Timing of TKR|| ||Employment|
| || ||Addiction to medications|| ||Trust in physician|
| || ||Employment|| ||Candidate for surgery|
| || ||Trust in physician|| || |
| || ||Anatomy|| || |
| || ||Drawbacks to surgery|| || |
|Intraoperative concerns (n = 22)|| || || || |
| No. concerns||12||8||1||1|
| Types of concerns||Technique||Technique||Technique||Technique|
| ||Anesthesia||Anesthesia|| || |
|Postoperative concerns (n = 44)|| || || || |
| No. concerns||9||23||0||12|
| Types of concerns||Benefits from surgery||Benefits from surgery|| ||Long-term outcome|
| ||Long-term outcome||Pain after surgery|| ||Recuperation process|
| ||Recuperation process||Recuperation process|| ||Function after surgery/limits|
| ||Function after surgery/limits||Function after surgery/limits|| || |
| ||Quality of life after TKR||Methods of pain relief|| ||Quality of life after TKR|
| || ||Quality of life after TKR|| ||Support after TKR|
Figure 1 shows the distribution of questions asked by all respondents about the preoperative phase (n = 60). Questions concerning alternatives to TKR (other options) were most frequent (18%), followed by concerns regarding pain medication and addiction possibilities (17%). White American women asked 57% of the preoperative questions (n = 34), covering the widest range of topics, including knee anatomy, devices and device technology, employment issues, physician trust, possibility of pain medication addiction, lifespan of the prosthesis, other options, and optimal timing of TKR. This was the only group that asked about potential drawbacks to surgery. African American women expressed a desire to know the criteria for TKR and reasons for their own candidacy for TKR. They also mentioned finances. White American men expressed interest in devices and device technology, whereas African American men expressed concern over financial issues, specifically their health insurance coverage of TKR.
Only 2 concerns were raised (22 questions) regarding the intraoperative phase: surgical technique (64%) and anesthesia (36%). White Americans asked almost all of them, with women being more concerned about anesthesia and men focusing more on technique (Table 2).
Of the 44 questions concerning the postoperative phase, white women again raised the largest proportion (n = 23; 52%), addressing recuperation period, functional recovery and limitations, and pain control. African American women asked 27% of the questions (n = 12), focusing on support after surgery, recuperation process, and long-term outcome. White American men were concerned about QOL after surgery. African American men asked no questions (Table 2).
White American men demonstrated the greatest amount of factual background information. Nearly all had researched TKR, had actively chosen their physician, and had researched other available options. They were generally more prepared to discuss TKR in terms of information already acquired. Additionally, because most were already actively involved in their own care, they tended to ignore many issues that preoccupied other focus group participants (i.e., trust of physician), leaving them more focussed on the procedure itself.
Analyses of concerns by age group, use of assistive device, income, and activity level did not reveal informative patterns.
Issues addressed by patient information materials
Table 3 shows a systematic review of information distributed by 3 academic joint replacement centers, the AF, and the AAOS. These documents were taken from the organizations' Internet sites and are meant to address patients' operative concerns preoperatively. They are used in addition to time spent speaking with the physician and members of the health care team and are not meant as a replacement for discussion with a patient's physician.
Table 3. Patient concerns by topics addressed in patient total knee replacement (TKR) literature
|Preoperative|| || || || || |
| Anatomy||X||X||X||X|| |
| Bilateral TKR#|| ||X|| || || |
| Candidate for surgery||X|| || ||X|| |
| Devices/technology|| || ||X||X|| |
| Diagnosis of arthritis/evaluation of knee#|| ||X|| ||X|| |
| Drawbacks to surgery||X||X||X||X||X|
| Finance||X|| ||X|| ||X|
| Jehovah's Witness#|| ||X||X|| || |
| Job|| || || || ||X|
| Physician trust**|| || || || || |
| Medications/addiction**|| || || || || |
| Number of surgeries/TKR life||X|| ||X||X|| |
| Options|| ||X||X||X|| |
| Pathology#|| ||X||X||X|| |
| Planning for surgery#||X|| ||X||X||X|
| Questions to ask physician#|| || || || ||X|
| Timing of TKR|| || || || ||X|
| TKR definition#||X||X||X||X||X|
|Intraoperative|| || || || || |
| Anesthesia||X|| ||X||X|| |
|Postoperative|| || || || || |
| Benefits from surgery||X||X||X||X||X|
| Function after surgery/limitations||X||X||X||X||X|
| Golfing and TKR#|| ||X|| ||X|| |
| Long-term outcome||X||X||X||X|| |
| Methods of pain relief||X||X|| || || |
| Pain after surgery||X|| ||X||X||X|
| Quality of life after surgery|| || ||X||X||X|
| Recuperation, recovery, healing, postop therapy||X||X||X||X||X|
| Support after surgery|| ||X||X||X||X|
Altogether, participants raised 21 types of concerns: 11 pertained to the preoperative phase, 2 to the intraoperative phase, and 8 to the postoperative phase. Five (24%) were addressed by all 5 documents: drawbacks to surgery; technique; benefits from surgery; function and limitations after surgery; and recuperation, recovery, healing, and postoperative therapy. Two (addiction to pain medications and physician trust) were not addressed by any of the documents. Issues regarding candidacy for surgery, devices and technology, finance, jobs, lifespan of the prosthesis, other options, timing of TKR, anesthesia, methods of pain relief, and QOL after surgery were each addressed by 3 or fewer documents. The AF document was the only one that addressed job issues and timing of TKR.
Three issues (definition of TKR, postoperative exercises, complications) were addressed by all 5 documents but not by subjects. Issues raised by at least 1 document but not by participants in our groups included: bilateral TKR (1 document), diagnosis and evaluation of the knee (2 documents), medical issues concerning Jehovah's witnesses (2 documents), pathology (3 documents), planning for TKR (4 documents), questions to ask the physician (1 document), and golfing postoperatively (2 documents).
- Top of page
Our recent knowledge of factors accounting for variation in TKR utilization has increased, largely due to work done by the TKR Patient Outcomes Research Team, which focused on understanding the reasons for regional variation in TKR utilization (20). These factors include likelihood of receiving TKR (11), orthopedic surgeons' and patients' perceptions of indications and outcomes for TKR (21, 22), patient predictors of physician referral for TKR (23), predictors of better physical function and other outcomes after TKR (24, 25), and outcomes of staged versus simultaneous TKRs (26). These studies, however, have not analyzed the mechanisms underlying the differences, either from the physician or patient perspective.
Our study has attempted to begin understanding these mechanisms by identifying themes underlying the decision-making process for TKR in patients with severe knee OA. Individual priorities and perceptions drive health care choice in any elective activity or procedure. For instance, Mancuso et al have recently shown that patients' expectations of benefits of knee surgery are linked to their requests for treatment (27). Because women and minorities underutilize costly health care interventions such as TKR (15, 28), it is important to explore the concerns that may be delaying their decisions. Research also suggests that both race/ethnicity and gender are important in patients' perceptions of costs and benefits in health care decisions (29–31). Evidence indicates these cultural differences are deeply ingrained (32). Understanding cultural differences may be critical in interpreting subjective health-related data, such as outcomes of elective procedures.
Our results raise some issues for consideration, both for counseling patients regarding TKR and the implications of not addressing these issues. With some overlap, different subgroups focussed on different phases of TKR and expressed different concerns. Women asked the most questions about both the preoperative and postoperative phases and broached financial issues most frequently. White women tended to raise process-oriented questions, addressing recuperation period, postoperative therapy, functional recovery and limitations, and pain control. African American women, on the other hand, asked more practical questions, focusing on support after surgery, benefits of surgery, and long-term outcome. They also expressed concern over the criteria used for TKR and their own candidacy for TKR.
Men raised fewer issues than women. In all phases, the issues concerned practical aspects of TKR, similar to previous findings (22). White men raised concerns that are easily answered by physicians and are the issues physicians classically tend to discuss (33, 34). For instance, white men asked questions such as, “If you have a good knee cap, then what is a replacement, does that [the knee cap] stay in there and they just work underneath it?” whereas white women asked questions such as, “Your friends who've had it [TKR], did you see them recover quickly or what?” This tendency of white American men to “speak the same language” as their physicians is an important communication and rapport concept long understood by medical anthropologists in their study of doctor-patient relationships (35–37). This could contribute to their proportionally increased use of TKR. The structural support of men and women in this study does not counter that argument. Previous investigators have suggested that women delay TKR because of lack of social support at home (22). Although this may still be relevant, 75% of both white American men and women had 1 or more others in the house; 50% of African American men had someone else in the house, compared with 30.8% African American women.
Although all focus group members were at the same stage in the TKR process, white American men's increased background knowledge concerning TKR may contribute to a greater proportion of this group choosing the procedure. For instance, 55% of their questions pertained to the intraoperative phase, focussing on technique.
African American men raised the fewest concerns (5 of 126 questions; 4%), the most common being finance. This topic was addressed in only 3 of the 5 patient information documents. The other issues they raised included distrust of physicians, other treatment options, and surgical technique. Although African Americans tend to have large potential supportive resources (38), it is well known that their medical informational networks are either less diverse or work differently than mainstream white Americans' (39–41). This may affect the number of nonsurgical options with which African Americans are familiar. Literature indicates African Americans have a much wider view of health and medical care than their white counterparts (42), which might cause them to consider a more diverse range of health options in treating knee OA. That both African American men and women raised issues pertaining to physician distrust, men more directly than the women, is particularly notable. As mentioned above, African American women expressed concern over the criteria for and their own candidacy for TKR. Although this may be an information-gathering technique, it also reflects distrust of the physician. Depending on the quality of their relationship with the physician, African American men may also be expressing distrust of their physicians when wondering whether they have been given all available options and alternatives to TKR. Expressed or implied, mistrust of the medical community is consistent with abundant medical literature describing mistrust among African Americans toward health care professionals in the aftermath of Tuskegee (43–45). Coupled with recent evidence that white Americans tend to know more people who have had TKR with better outcomes (46), this could contribute to the smaller proportion of TKRs performed in African Americans.
African American men spoke the least during the focus groups and also raised the fewest issues. The reason for this is unclear. Sherman James et al have written on the relationship of powerlessness and stress in African Americans and how the stress manifests as hypertension (47). It is possible that powerlessness is also associated with a certain sense of hopelessness that manifests itself in decreased verbalization. Combined with a distrust of the medical system, this scenario is certainly plausible. Second, this group of African American men was very concerned about finances. Perhaps they felt TKR was not accessible to them. If so, it would be very important to address finances early with this population subgroup. Third, it is possible that African American men do not want to 1) appear naïve in asking a question others may consider fundamental, or 2) to speak unless it is from a position of power. Given the age of the African American men in the sample, it is important to consider that they lived through a time in this country when it was difficult, even dangerous, for African American men to speak up. Finally, it is possible that our sample was not representative and included a group of African American men who were not inclined to speak.
Because white American men have the most surgery proportionally, we had expected that the material handed out by TKR centers would address mostly “white male concerns.” To the credit of those developing the patient information materials, this was not found to be true in these 5 documents.
White American men or women asked the great majority of questions pertaining to the intraoperative phase; women were concerned about anesthesia and men focused on technique. Altogether, 64% (n = 22) of the questions addressed surgical technique, and 36% were concerned with anesthesia. White men tended to be more persistent about their concerns; although men comprised only 33% of the white Americans, they asked 64% of the questions regarding the intraoperative phase.
In this office, patients generally meet with an orthopedic nurse to address concerns after deciding to undergo TKR. In addition, they have the opportunity to discuss issues with the nurse at their appointment. According to focus group participants, their encounter with the orthopedic surgeon was brief, and many had no chance to think of questions during the visit (having just been told to consider TKR), so many specific concerns had not been addressed at the time of the focus group. However, all participants reported having a good understanding of why TKR was being recommended.
Analysis of thematic content of patient information materials from 3 large TKR centers, the AAOS, and the AF reveals <50% overlap between expressed patient concerns in our groups and material intended as reference information for the same population, although material varied in their rates of correlation (Table 3).
Qualitative studies such as this are limited in scope to the population being studied, in this case a university setting in an urban area. These data were obtained in the first phase of a larger study on race/ethnicity and gender variation in elective procedure utilization. Our patient population was a convenience sample; results are not generalizable to those patients with disabling knee OA who have never seen an orthopedic surgeon. Although that population is also important, locating and studying those patients will require a larger study. In addition, general limitations of focus group studies must be considered in relation to these results: 1) data are difficult to analyze, 2) groups are difficult to assemble, 3) researchers have less control over a group interview than an individual one, and 4) groups can vary considerably (19). The first 2 items are logistic issues, which were successfully managed in this study through application of analytic methods for qualitative data and recruitment efforts. Our use of a professional focus group moderator minimized the potential interviewer bias from participants' responses and also addressed the third issue. The last issue is a potential asset for the current study where the objective is to bring out a spectrum of issues. Focus groups are the method of choice when the goal is to investigate complex behaviors and motivations, as in this case (19). Our statistics show that patients in this population tend to make a decision regarding TKR within 1–2 weeks after discussion with the surgeon (the minimum time necessary to recruit enough patients for a single focus group). Although focus groups with specific race/ethnicity and gender composition may have lead to more open discussion, it was not possible to structure them that way given this characteristic. Finally, in qualitative work, any exceptions found to current understanding are important. In this case, that patients asked questions not covered in their literature and not discussed with their physicians or not understood when discussed is a relevant and noteworthy finding.
The importance of resolving health care disparities has recently been emphasized by the Healthy People 2010 objectives. TKR in OA is an important arena to investigate reasons for these differences because of the predominance of OA in aging populations, the high prevalence of OA among women and minorities, the potential for increased QOL among TKR recipients, and the relative cost effectiveness of TKR. Our research reveals that different population subgroups have different concerns when making decisions regarding TKR. Moreover, materials currently distributed to patients considering TKR do not cover some issues of concern to women and minorities in this study, such as candidacy for surgery, job issues, and physician trust. The physician's failure to discuss key issues may contribute to gender and race/ethnicity disparity in TKR utilization.
Although it is likely that physicians and staff at the different centers addressed additional issues and concerns with their patients outside those covered by the education materials, it is possible that the absence of opportunities to address their concerns or the lack of satisfactory answers decreases utilization of TKR by women and minorities. This research suggests that health care professionals should learn to recognize specific concerns relevant to women and minorities and target them in discussions. Overall, we believe the implications of our results indicate the value of pursuing this information further, as the differences we have described may contribute to the gender and race/ethnicity disparity seen in TKR utilization.