Background
Osteoarthritis is defined as the wearing down of cartilage that protects joints in the body resulting in pain and stiffness (American Academy of Orthopaedic Surgeons, 2017). Osteoarthritis affects roughly 30 million people, with the knee and hip being two of the most commonly affected joints (Centers for Disease Control and Prevention [CDC], 2018). One treatment option is a total hip or knee arthroplasty where the degenerative joint is replaced with a prosthesis. However, according to the CDC (2018), osteoarthritis is one of the most expensive conditions to treat when it requires joint replacement surgery.
The Joint Commission (2017) estimates that 700,000 people undergo a hip or knee arthroplasty in the United States each year. By the year 2030, this number is expected to increase more than 600% compared with 2005 for total knee arthroplasties and 200% for total hip arthroplasties for the same time period (So, 2017). This increase is expected to result in billions of dollars in costs for hip or knee arthroplasties each year.
In the last 20 years, the overall care for knee and hip arthroplasty has changed. The average length of stay (LOS) has substantially decreased due to a variety of factors including advancements in technology, procedures, and medications. To put this into perspective, LOS has fallen from around 10 days in 1990 to generally 3 days or less today (Accelero Health Partners, 2014). With a shortened LOS, patients undergoing hip or knee arthroplasty need to be adequately prepared preoperatively to ensure positive outcomes and recovery. This preparation should begin with a preoperative educational class that informs patients about the operative process and postoperative recovery (Edwards et al., 2017).
In 2013, the Centers for Medicare & Medicaid Services introduced the Bundled Payments for Care Improvement initiative with the intention of improving patient experience, quality of health, and reducing the cost of care (Edwards et al., 2017). The introduction of this bundled payment method has resulted in the use of an enhanced recovery or fast-track programs that involve evidence-based interventions carried out by a multidisciplinary team to ultimately improve outcomes and recovery time (Christelis et al., 2015). A preoperative educational class is usually incorporated into these programs with several other interventions such as early mobilization and multimodal analgesia (Khan et al., 2014). There is ample research demonstrating improved patient outcomes when using an enhanced recovery program (Castorina et al., 2018; Khan et al., 2014; Pamilo et al., 2018; Zhu et al., 2017). However, the impact of preoperative education alone is controversial.
Preoperative education is defined as any educational intervention provided before surgery that is intended to improve knowledge, health behaviors, and health outcomes (McDonald et al., 2014). Topics vary but typically include information on the surgical procedure, postoperative care, complications, pain management, and information about discharge. (Ibrahim et al., 2013). A 2014 Cochrane review examined preoperative education in 18 trials involving 1,463 patients. The researchers concluded that preoperative education had a minimal benefit over standard patient care regarding anxiety, pain, functions, and adverse events (McDonald et al., 2014). However, more recent studies reported improved outcomes after a preoperative joint replacement class. These studies examined pain, anxiety, preparation, and LOS in relation to preoperative education before a hip or knee arthroplasty (Chen et al., 2014; Edwards et al., 2017; Huang et al., 2012; Kearney et al., 2011; McDonald et al., 2014).
Kearney et al. (2011) studied 150 patients undergoing hip or knee arthroplasty and found that those who received preoperative education reported feeling better prepared (p = .002) and better able to control their pain (p = .001), but they found no significant differences in postoperative pain, ambulation distances, or LOS. This study was limited in that the education was not standardized, varying in type-in person, online, and written-and quality. Because of these limitations, it is difficult to determine which educational model had an effect on the outcomes (Kearney et al., 2011).
Similarly Moulton et al. (2015), examined 318 patients who received preoperative education in the form of a multidisciplinary educational session prior to hip arthroplasty with 233 patients receiving preoperative education and 85 receiving no education. The results indicated a significantly reduced LOS for those receiving perioperative education (p = .046) (Moulton et al., 2015).
A few years later, Moulton et al. (2017) examined preoperative education prior to knee arthroplasty in 563 patients who were split into two groups: those who received preoperative education (n = 503) and those that did not (n = 60). Although the results indicated shorter LOS for those who received education, it was not statistically significant (Moulton et al., 2017). To date, there remains conflicting evidence regarding the role of preoperative education.
Purpose and Aims
The purpose of this study was to compare immediate postoperative outcomes following a total hip or knee arthroplasty based on educational session attendance. The primary aim of this study was to compare the immediate postoperative physical therapy (PT) performance in patients older than 50 years following a total hip or knee arthroplasty who attended a preoperative educational class with those who did not. A secondary aim was to compare the average LOS and discharge disposition between both groups and to investigate the effect of comorbidities and immediate postoperative performance. Finally, a tertiary aim was to perform a cost-benefit analysis.
Methods
This study took place on an orthopaedic floor at a community hospital in southwestern Pennsylvania. Institutional review board approval was obtained from the university and the hospital where this study took place. All patients who underwent a total hip or knee arthroplasty in the year 2018 and met inclusion criteria (listed later) were included in this study. The researcher retrospectively reviewed medical charts for immediate postoperative outcomes of both the patients who attended a preoperative educational class and those who did not. Classes were taught by a rotation of four experienced orthopaedic nurses with input from a physical therapist, occupational therapist, and orthopaedic navigator. Although the class was recommended by the surgeons, it was optional. Table 1 provides information regarding the educational class including structure and content.
This study used a descriptive, retrospective chart review design. The independent variable in this study was the preoperative educational class. The dependent variables were the PT performance, LOS, cost, and discharge disposition. Eligible patients were identified as the experimental or control group based on preoperative class attendance and were further grouped as hip or knee arthroplasty.
Following de-identification, patient electronic medical records (EMRs) were reviewed for demographic data, previous joint replacement surgical history, selected comorbidities, postoperative Day 1 PT notes, discharge disposition, and LOS. Convenience sampling was used to select patients who were 50 years of age or older and undergoing a total hip or knee arthroplasty. Patients were excluded if they were younger than 50 years, had prior joint replacement surgery in the past year, or had any mental or cognitive impairment. Patients were also excluded if they were undergoing any of the following orthopaedic surgeries: hip or knee revision, unicompartmental knee surgery, conversion of a unicompartmental knee to a total knee arthroplasty, or bilateral knee replacements.
All data were de-identified using a number system and Health Insurance Portability and Accountability Act (HIPAA) privacy was maintained throughout. Baseline demographic data collected included gender, age, ethnicity, type of joint being replaced, body mass index, class attendance, and previous joint replacement surgery. Data on prior joint replacement surgery included years since previous joint replacement surgery to examine the correlation between the number of years since their previous surgery and immediate postoperative performance. Data regarding comorbidities were also collected to determine whether there was a relationship between comorbidities, immediate postoperative performance, and attendance at a preoperative educational class. Data collected from the postoperative Day 1 physical therapist notes included ambulation distance, degrees of knee flexion and extension (in knee arthroplasty patients), and hip flexion and abduction (in hip arthroplasty patients).
Procedures
Operating room schedules were reviewed for all orthopaedic joint surgical procedures from January 2018 to December 2018, and all hip or knee arthroplasties were identified. Initially, duplicate surgical procedures due to rescheduling were included for a total of 1,092 entries.
Data were reviewed for exclusion criteria and duplicate entries. A total of 236 patients were excluded because of the type surgical procedure, more than one joint replacement in 2018, or duplicate entry (rescheduled). If patients had two joint replacements in the study year, the earliest instance was included in this study as the patient typically attended the educational joint replacement class prior to the first surgery and did not reattend for the second surgery. If a patient was a duplicate, only one instance was included. That left 856 patients. Class attendance was determined by reviewing the sign in sheets.
The 856 patients were identified through the EMR and data were collected using premade data collection worksheets created by the researcher (see Figure 1). Based on the EMR review, an additional 149 patients were excluded from the study because LOS was greater than 3 days due to medical complications or insurance issues; cognitive impairment; surgery was rescheduled after the study period; not found in the EMR; previous joint replacement surgery in the last year; age less than 50 years; or joint surgical procedures other than total hip or knee arthroplasties. Following the second round of exclusions, 707 patients remained. The final data set was entered into IBM SPSS for data analysis.
Statistical Analysis
Descriptive statistics were calculated for all variables of interest. Continuous variables were summarized using means and standard deviations. Categorical variables were summarized using frequency tables.
Independent t tests were conducted with the preoperative educational class attendance as the independent variable and ambulation distance, degrees of knee flexion, and degrees of knee extension as the outcome variables for knee arthroplasty patients. Independent t tests were also conducted with preoperative educational class attendance as the independent variable and ambulation distance, degrees of hip flexion, and degrees of hip extension as the outcome variables for hip arthroplasty patients. Independent t tests were conducted with the preoperative educational class attendance as the independent variable and hospital LOS as the outcome for both knee and hip arthroplasty patients. Independent t tests were conducted with gender as the group and ambulation and LOS as the outcomes.
The data were split by joint replacement (knee vs. hip), and a series of independent t tests were conducted with LOS as the outcome and seven comorbidities (hypertension, depression, Type 2 diabetes, anxiety, gastroesophageal reflux disease, hyperlipidemia, and hypothyroidism). With the split data, a series of one-way analysis of variance tests were also conducted with categorical demographics as the predictors and postoperative performances as the outcomes.
Finally, a series of [chi]2 tests of independence was conducted to evaluate the relationship between discharge disposition and seven comorbidities. Another [chi]2 test of independence was conducted to evaluate the relationship between discharge disposition and postoperative educational class attendance for both knee and hip arthroplasty patients. All statistical analyses were conducted using SPSS version 26.0 and used a p value of .05 or less to attain statistical significance.
Results
The coded data were screened for out-of-range values resulting in 707 patients, 498 with knee arthroplasties and 209 with hip arthroplasties. Patient ages ranged from 50 to 89 years, with an average of 67.23 years, with 43% male and 57% female. A majority of patients were Caucasian (94.5%). Overall, 52.3% attended the preoperative educational class. Of the 498 knee arthroplasty patients, 264 (53%) attended the preoperative educational class and 234 (47%) did not. Of the 209 hip arthroplasty patients, 106 (50.7%) attended the preoperative educational class and 103 (49.3%) did not. Gender and discharge disposition between the education and control groups were statistically significant. Males were more likely to attend the educational class (56.6%) than were females (49.1%; p = .029). Patients discharged to home were more likely to attend the educational class (54.9%) than were patients discharged to a skilled nursing facility (SNF; 39.3%; p = .001). Demographic and descriptive statistics are shown in Table 2.
Knee Arthroplasty
Ambulation distance was statistically significant (p < .001) with preoperative class participants having greater ambulation (M = 131.34 ft) than nonparticipants (M = 97.22 ft). In addition, degree of knee flexion was also statistically significant (p < .001) with preoperative class participants having greater knee flexion (M = 77.18[degrees]) than nonparticipants (M = 65.61[degrees]). Knee arthroplasty PT performance is shown in Table 3.
Length of stay was statistically significant (p < .001) with preoperative class participants having a shorter LOS (M = 1.97 days) than nonparticipants (M = 2.10 days). The computed effect size for the mean difference, d = 0.526, is considered "medium" (Murphy et al., 2014). Thus, the results are clinically meaningful. Knee arthroplasty LOS shown in Table 3.
Hip Arthroplasty
Ambulation distance was statistically significant (p < .001) with preoperative class participants having greater ambulation (M = 146.13 ft) than nonparticipants (M = 111.38 ft). In addition, degree of hip flexion was also statistically significant (p = .012) with preoperative class participants achieving greater hip flexion (M = 63.54[degrees]) than nonparticipants (M = 60.31[degrees]). Hip arthroplasty PT performance is shown in Table 4.
Length of stay was statistically significant (p < .001) with preoperative class participants achieving a shorter LOS (M = 1.93 days) than nonparticipants (M = 2.07 days). The computed effect size for the mean difference, d = 0.557, is considered "medium" (Murphy et al., 2014). Thus, the results are clinically meaningful. Hip arthroplasty LOS is shown in Table 4.
After splitting the data by joint arthroplasty (knee vs. hip), independent t tests were conducted with LOS as the outcome variable and seven comorbidities (hypertension, depression, Type 2 diabetes, anxiety, gastroesophageal reflux disease, hyperlipidemia, and hypothyroidism). Prevalence of depression was higher (the difference was statistically significant) for hip arthroplasty patients. Length of stay was statistically higher for hip arthroplasty patients with depression than for patients without depression (p = .05).
Prevalence of Type 2 diabetes was higher (the difference was statistically significant) for knee arthroplasty patients (p = .001). Length of stay was statistically longer for knee arthroplasty patients with Type 2 diabetes (M = 2.11 days) than for patients without Type 2 diabetes (M = 2.01 days).
In addition, Type 2 diabetes was statistically associated with discharge disposition. A greater proportion of Type 2 diabetes patients were discharged to an SNF for both knee (p = .049) and hip arthroplasty patients (p = .001).
The data were again split by joint arthroplasty (knee vs. hip). Analysis of variance tests were conducted with categorical demographics as the predictor variables and postoperative performances as the outcome variables. Ambulation distance was statistically significant for age in both knee (p < .001) and hip (p = .002) arthroplasty patients such that, the older the age, the shorter the ambulation distance and the younger the age, the longer the ambulation distance in both knee and hip arthroplasty patients.
The relationship between gender and ambulation distance plus LOS was examined. Statistical significance was found between gender and ambulation distance for both hip and knee arthroplasty patients. Male patients had greater ambulation distance than female patients (p < .01). Statistical significance was also found between gender and LOS but for knee arthroplasty patients only. Female knee arthroplasty patients had a longer LOS (M = 2.07 days) than male patients (M = 1.98 days; p < .01).
The relationship between discharge disposition and preoperative educational class attendance for both knee and hip arthroplasty patients was statistically significant for knee arthroplasty patients only (p = .002). A greater proportion of knee arthroplasty patients who did not attend the class were discharged to an SNF (21.37%) than those who did attend the class (10.98%). No statistical relationship emerged between the years since previous surgery and the postoperative performance measures (p > .10).
The cost of the preoperative educational class was calculated by taking the median hourly salaries of all healthcare providers who taught the class multiplied by the time that they spent teaching during the class sessions. This cost included the average hourly rate of the four rotating nurses who led the class, as well as the mid-range hourly rate for a physical therapist, an occupational therapist, and an orthopaedic navigator. On days when the classes were taught, the nurses worked a full 8-hour shift during which they taught a 2-hour class session twice. One class was taught in the morning and one in the afternoon. There were 2 days of classes per month leading to four total educational classes per month, two on each day. The rest of their shift was used to prepare for future classes, which included calling future patients to make them aware of the class and gauge their interest in attending.
The physical therapist and the occupational therapist spent approximately 15 minutes each presenting during each class session. Therefore, the mid-range hourly rate for a physical therapist and an occupational therapist was each multiplied by a quarter of an hour to determine cost. This number was then multiplied by 2, one for each class session that day, to calculate their portion of the daily class cost. They were already on the unit working for the day and set 15 minutes aside to present to each class session. There was no preparation required as it was the same information at every class that involved various physical and occupational therapy exercises to practice before surgery. The orthopaedic navigator spent approximately 10 minutes talking during the class. Therefore, the mid-range hourly rate for the orthopaedic navigator was each multiplied by a sixth of an hour to determine cost. Again, this number was then multiplied by 2, one for each class session that day, to calculate their portion of the daily class cost. They were already on the unit working for the day and set 10 minutes aside to present to each class session. There was no preparation required as it was the same information at every class that involved introducing themselves and providing contact information if they needed anything regarding their surgery.
Also included in the cost was the packet of written educational information and the folder that it came in for each class attendee. This was calculated by taking the cost of the reams of paper used to print the educational material on per day of class sessions and added to the cost of the packs of folders used per day of class sessions. This financial information came from the unit manager who ordered these materials in bulk as needed. Both hip and knee arthroplasty patients attended one joint class where both procedures were discussed. After computation, the cost of each day of class, which included both morning and afternoon sessions of the class, was determined to be $364.58 per day of class. With 2 days of classes provided a month, this equates to approximately $8,749.92 a year.
The benefit of the class is included for knee arthroplasties only as hip arthroplasty financial data were unavailable. The LOS in knee arthroplasty patients who attended the class was 1.97 days compared with 2.10 days in those who did not for a difference of 0.13 days. For the year 2018, financial data obtained from the hospital indicated a direct expense for knee arthroplasties to be $7,089 per case. The direct expense multiplied by the difference in LOS equates to approximately $921, which is the amount saved per knee arthroplasty when one patient attended the preoperative educational class. In 2018, the number of knee arthroplasty patients attending the class was 264 totaling $243,294 in direct expense savings. Subtracting the cost of the class, the preoperative educational class saved the hospital approximately $234,814 in direct expenses from knee arthroplasties performed in 2018. If the class was mandatory and the additional 234 knee arthroplasty patients had attended the class, an additional cost savings of $215,647 in direct expenses could have been achieved. When adding both costs savings values together, this equates to a total potential savings of $458,941 if the class was mandatory and attended by all patients who underwent knee arthroplasty in 2018. Although the financial data from hip arthroplasties were unavailable, it is estimated that the overall cost savings would be similar to knee arthroplasties because hip arthroplasty outcome data were similar.
Interrater reliability was calculated by having the unit manager review 20 of the patient charts. After 20 charts were randomly chosen, the unit manager independently recorded the data using the researcher's data collection worksheets. Interrater reliability was 91%.
Discussion
This study found that attendance at a preoperative educational class significantly decreased LOS after both hip and knee arthroplasties. The result of a shortened LOS after knee arthroplasty in patients who attend a preoperative educational class supports the findings of Jones et al. (2011),McDonald et al. (2014), and Dehorney and Ashcarft (2018) while conflicting with the findings of Kearney et al. (2011) and Moulton et al. (2017). Kearney et al. (2011) included various forms of preoperative education, such as online and written education; this lack of standardization may explain why they did not find a significant difference. Moulton et al. (2017) had a large sample size that included every patient in the enhanced recovery program; however, like the present study, patients were not randomized to intervention and control groups. Unlike this study-in which the intervention and control groups were relatively similar in size and patent profile-the Moulton study had 503 patients in the group who attended the educational class but only 60 patients who did not attend the class.
This study's findings of a shortened LOS after hip arthroplasty in patients who attended a preoperative educational class support the findings of Moulton et al. (2015) and Dehorney and Ashcarft. (2018) while conflicting the findings of Vukomanovic et al. (2008),Kearney et al. (2011), and McDonald et al. (2014). Vukomanovic et al. (2008) used randomization; however, they had a very small sample size of 36, which may explain why they did not reach significance. As mentioned previously, Kearney et al. (2011) used varying forms of preoperative education, including online and written education; their lack of standardization may explain why they did not reach significance. McDonald et al. (2014) was a systematic review; however, dates of the studies reviewed ranged from 1994 to 2008, which was outdated and much has changed in hip and knee arthroplasties since that time.
In the ever-changing and ever-challenging world of healthcare, the ultimate goal is to reduce cost while maintaining a high level of care. A decreased hospital LOS stay equates to cost savings. This study suggests that the LOS at the study institution after both hip and knee arthroplasties is approximately 2 days, often less. This finding is consistent with the overall downward trend of LOS over the last 10 years as seen in the literature. A 2-day LOS is a substantial decrease from previous studies (Moulton et al., 2015; Vukomanovic et al., 2008), and it appears that the LOS is still trending downward. The role of preoperative education is even more important with a decreased LOS as it mentally and physically prepares patients for a smooth transition into the postoperative phase of the procedure.
The study results suggest that attendance at a preoperative educational class increases ambulation distance and degrees of knee flexion in knee arthroplasty patients and significantly increases ambulation distance and degrees of hip flexion in hip arthroplasty patients. An increase in ambulation distance and degrees of knee flexion in those patients who attend a preoperative educational class conflicts with the findings of Kearney et al. (2011) and McDonald et al. (2014), who found that preoperative education had a minimal benefit. In addition, an increase in ambulation distance and degrees of hip flexion in those patients who attend a preoperative educational class conflicts with the findings of Kearney et al. (2011),Vukomanovic et al. (2008), and McDonald et al. (2014).
Postoperative PT performance is clinically relevant as it plays a role in determining discharge disposition. Physical therapy performance helps determine whether the patients can safely return home after surgery or whether they require an SNF for more surveillance. Range of motion (ROM) is especially relevant as certain benchmarks must be met to perform activities of daily living (ADLs). A knee flexion of 67[degrees] is needed for the swing phase of gait, 83[degrees] for ascending stairs, 90[degrees] for descending stairs, and 93[degrees] for rising from a chair (Li et al., 2007). Hip ROM is difficult to measure and is generally believed to be questionable in evaluating outcomes regarding hip arthroplasty (Davis et al., 2007). Nevertheless, a hip flexion of 67[degrees] is needed for ascending stairs, 36[degrees] for descending stairs, and 112[degrees] for rising from a chair (Johnston & Smidt, 1970). Hip flexion is further complicated in hip arthroplasty patients as there is a risk of dislocation. Because of this, hip precautions need to be followed after total hip arthroplasty. These precautions include avoiding flexion at the hip past 90[degrees], twisting the operative leg, and crossing the legs (Barnsley et al., 2015). These precautions further complicate the value of measuring ROM after hip arthroplasty and may be why it is believed to be questionable in evaluating outcomes. In general, the greater the ROM after hip or knee arthroplasty, the better. The greater the ROM, the more ADLs the patient can safely perform. Therefore, patients with a greater ROM on postoperative Day 1 had shorter LOSs, which led to a cost savings.
This study's finding of an association between depression and LOS is consistent with the literature. A systematic review found that depression and depressive symptoms are related to a longer LOS in hospitalized patients (Prina et al., 2015). Although there is limited research regarding LOS specifically after hip arthroplasty, there is literature supporting an increase in LOS in patients with depression after knee arthroplasty. In another systematic review, depression was linked to 1.05 times longer LOS after knee arthroplasty (Oh, Gold, & Slover, 2020). Although knee and hip arthroplasties differ, the hospital course is somewhat similar and clinically this could also apply to hip arthroplasty.
This study's finding of an association between Type 2 diabetes and LOS is also consistent with the literature. Martinez-Huedo et al. (2017) examined Type 2 diabetes and LOS after both hip and knee arthroplasties and found a significantly longer mean LOS in patients with Type 2 diabetes (hip p = .037; knee p = .042). Lenguerrand et al. (2018) found that the median LOS was 1 day longer in patients with diabetes than in patients with no diabetes after knee arthroplasty (p = .004).
More research is needed regarding patients with diabetes who are discharged to an SNF after knee arthroplasty in terms of what they may need to prevent a prolonged LOS or being discharged to an SNF. We know that patients with diabetes have prolonged healing and a different recovery process after surgery. Lenguerrand et al. (2018) support a longer hospital LOS in patients with diabetes after knee arthroplasty; therefore, discharge to an SNF is somewhat expected. Because patients diagnosed with Type 2 diabetes and depression are at risk for a longer LOS, it may be particularly important to actively encourage them to attend an educational class prior to hip or knee arthroplasty.
In this study, age was statistically significant for ambulation distance with younger patients reaching a longer distance. Generally, younger patients have fewer comorbidities, are more active, and have a higher baseline for physical activity than older patients. Therefore, a longer ambulation distance is expected.
In both hip and knee arthroplasties, males had significantly greater ambulation distance than females. In addition, female knee arthroplasty patients had a significantly longer LOS than males. These outcomes may be due to anatomical differences, pain response, and willingness to undergo surgery (Novicoff & Saleh, 2011). These differences have not been studied because of the multiple factors that are involved. A systematic review of 61 studies found that women had a higher rate of arthritis with greater disability and worse symptoms than men (Novicoff & Saleh, 2011). Women had increased cartilage loss, greater hip degeneration, worse symptoms of knee arthritis, and different gait patterns secondary to arthritis compared with men. Women also had more postoperative pain, more intense pain, and reduced ADL due to pain compared with men (Novicoff & Saleh, 2011). The results of this study could explain why men had a greater ambulation distance and shorter LOS. Whitlock et al. (2016) also found that women had a longer LOS of stay than men after knee arthroplasty (p < .001). Therefore, our study results regarding gender and LOS are consistent with the literature. More research is needed exploring gender disparities after hip or knee arthroplasty. However, it appears to be prudent to encourage female patients to attend a preoperative educational class, as it may prevent a prolonged LOS.
This study found that a significantly higher proportion of knee arthroplasty patients who did not attend the class were discharged to an SNF (21.37%) compared with those who attended the class (10.98%). Discharge to an SNF is dependent on multiple factors but most importantly it is related to PT performance of ADLs and safety. Ramkumar et al. (2019) found that discharge to an SNF after knee arthroplasty was related to older age (p < .001), female gender (p < .001), and various comorbidities (p < .001). Type 2 diabetes was not a significant variable in this study. Thus, it appears that preoperative educational class attendance, for those patients who are undergoing a knee arthroplasty, may be beneficial in decreasing the LOS.
Finally, no statistical relationship emerged between the years since previous surgery and the postoperative performance measures. This finding suggests that even if patients have experienced a previous joint arthroplasty, they should still be encouraged to attend a preoperative educational class. More research is needed to determine how quickly patients forget what they learned in preoperative educational classes and whether this information supports recovery after successive joint arthroplasty surgical procedures.
Study Limitations
Because of the retrospective design of this study, participants could not be randomized and this may have led to biases in the study. However, patients were evenly distributed by chance with 264 (53%) knee arthroplasty patients attending the preoperative educational class and 234 (47%) not attending. Of the 209 hip arthroplasties, 106 (50.7%) attended the preoperative educational class and 103 (49.3%) did not. Another limitation is the variation in surgeons' encouragement to attend the preoperative class, such that certain outcomes may have been related to specific surgeon's technique and practices. This is theoretical as specific surgeon data were not collected. In addition, the patients who did not attend the class were a self-selected group who may have based their decision on recommendation, motivation, attitude, or geographic issues. To address this situation, an online or taped version of the preoperative educational class could be an attractive alternative.
Conclusion
This study found statistically significant differences in the LOS and postoperative PT outcomes in patients who attended a preoperative educational class compared with those who did not attend as measured by ambulation distance, degrees of knee flexion, and degrees of hip flexion. Preoperative educational class attendance prior to knee arthroplasty equated to a cost savings of $921 per knee arthroplasty. A total of $234,814 was saved in direct expense from knee arthroplasties in 2018. Based on these results, attendance of a preoperative educational class prior to hip or knee arthroplasty should be strongly encouraged if not mandatory, especially in certain vulnerable patients.
References