Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Outpatient physical therapy bundled payment models are feasible for total hip arthroplasty patients: an evaluation of utilization, cost and outcomes

Outpatient physical therapy bundled payment models are feasible for total hip arthroplasty... Background Various episode-of-care bundled payment models for patients undergoing total joint arthroplasty have been implemented. However, participation in bundled payment programs has dropped given the challenges of meeting continually lower target prices. The purpose of our study is to investigate the cost of outpatient physical therapy (PT ) and the potential for stand-alone outpatient PT bundled payments for patients undergoing total hip arthroplasty ( THA). Methods A retrospective review of 501 patients who underwent primary unilateral THA from November 2017 to February 2020 was performed. All patients included in this study received postoperative PT care at a single hospital- affiliated PT practice. Patients above the 75th percentile of therapy visits were then classified as high-PT utilizers and compared with the rest of the population using univariate statistics. Stepwise multivariate logistic regression was used to assess the predictors of high therapy utilization. Results Patients averaged 65 ± 10 years of age and a BMI of 29 ± 5 kg/m . Overall, 80% of patients were white and 53% were female. The average patient had 11 ± 8 total therapy sessions in 42 days: one initial evaluation, one re-eval- uation and 9 standard sessions. High-PT utilizers incurred estimated average costs of $1934 ± 431 per patient, com- pared to $783 ± 432 (P < 0.001) in the rest of the population. Further, no significant differences in 90-day outcomes including lower extremity functional scale scores, emergency department returns, readmissions, or returns to the operating room were observed between high utilizers and the rest of the population (all P > 0.08). In the multivariate analysis, women (OR = 1.68, P = 0.017) and those with sleep apnea (OR = 2.02, P = 0.012) were nearly twice as likely to be high utilizers, while white patients were 42% less likely to be high utilizers than patients of other races (OR = 0.58, P = 0.028). Conclusions Outpatient PT utilization is highly variable in patients undergoing THA. However, despite using more services and incurring increased cost, patients in the top quartile of utilization experienced similar outcomes to the rest of the population. If outpatient therapy bundles are to be developed, 16 visits appear to be a reasonable target for pricing, given this provides adequate coverage for 75% of THA patients. *Correspondence: Justin J. Turcotte jturcotte@luminishealth.org Full list of author information is available at the end of the article © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. Stock et al. Arthroplasty (2023) 5:26 Page 2 of 9 Keywords Total hip arthroplasty ( THA), Physical therapy (PT ), Bundled payment Background phase of care may be more attractive to providers. The Physical therapy (PT) is an important step in any treat- purpose of our study was to investigate the cost of out- ment protocol following total joint arthroplasty (TJA) to patient PT and the potential effectiveness of stand-alone improve a patient’s mobility, strength, and independence. outpatient PT payment bundles for patients undergoing In patients undergoing total hip arthroplasty (THA), THA. PT is typically prescribed for 2 to 3 days a week for 6 to 8  weeks [1]. Therapy often starts the day of or the day Methods after surgery and continues for two to three days a week This study was deemed institutional review board exempt along with home exercises until activity goals are met [2]. as a review of existing medical records by the institutional As the performance of TJA continues to shift toward the clinical research committee, and a waiver of informed ambulatory setting, more intensive early therapy pro- consent was granted. A retrospective chart review of all grams are being implemented to facilitate early mobi- patients undergoing THA by 7 board certified surgeons lization and same-day discharge [3–6]. Such therapy at a single institution was performed. protocols have been shown to decrease hospital length of stay (LOS), decrease hospital costs, increase patient Study population satisfaction, and improve functional status more rapidly All patients included in this study underwent primary [3–6]. Additionally, the costs and benefits of various PT unilateral THA from November 2017 to February 2020. models, including formal outpatient, home-based, and no Patients undergoing bilateral or revision THA were therapy, have been evaluated [7–10]. With some recent excluded from this study. All patients included in this studies showing that formal outpatient PT was not nec- study received postoperative PT care at a single hospital- essary for all THA patients [10, 11], additional research affiliated practice. A total of 501 patients met the inclu - into new models that maximize the value of traditional sion criteria. therapy is needed in order to maintain access to these services. Perioperative protocol Outpatient PT represents a significant portion of TJA All patients were cared for in a coordinated Joint episode cost [11]. In an effort to reduce cost, bundled Replacement Center and received education materials payment initiatives such as the Comprehensive Care for including written materials, preoperative medical evalu- Joint Replacement (CJR), Bundled Payments for Care ations, preoperative home exercise or outpatient physical Improvement (BPCI), and BPCI Advanced programs therapy, and an education class for patients and their car- have been put in place [12–15]. While the mechanics of egivers. All patients were treated utilizing a multimodal each program differ, they operate in a similar manner by pain management protocol which, depending on patient incentivizing hospitals and/or providers to deliver care factors, included acetaminophen, oral NSAIDs, pregaba- under an established target price while meeting quality lin, ketorolac, and oral opioid medications as needed. thresholds. Upon implementation, these programs effec - tively reduced the costs of TJA, primarily through reduc- PT protocol tions in LOS and discharge to skilled nursing facilities Standard PT protocols are used across all therapy sites. (SNFs) [13, 16, 17]. Despite these promising early results, However, therapists might modify treatment based on participation in bundled payment programs has dropped their clinical judgment of patient progression. During given the challenges of meeting continually lower tar- weeks 0–2, therapy focuses on range of motion (ROM), get prices [12, 18]. With target pricing decreasing by the flexibility, quadriceps strengthening exercises and gait CMS, the withdrawal rate of the BPCI Advanced pro- training. During this period, patients were expected grams has risen to over 85% [19]. As the financial savings to transition from walker to cane-assisted ambulation. and quality improvement gains from participation in epi- During weeks 3–6, scar mobilization is initiated and sode of care bundled payment programs slow, the ulti- assistive devices were discontinued as the patient’s gait mate fate of such programs remains uncertain [19, 20]. normalizes. In this phase, exercises focus on quadri- In light of the debate over the value of formal outpa- ceps, hamstring and core strengthening, hip abduction tient PT and the challenges of successfully managing the and adduction, and proprioception. During weeks 7–12, cost of an entire episode of care, alternative value-based therapy focused on continued strengthening, single leg payment models, such as bundled pricing for a distinct stance and uneven terrain exercises, and gait training S tock et al. Arthroplasty (2023) 5:26 Page 3 of 9 with the goal of mastering functional activities, improv- Table 1 Total population demographics and comorbidities ing strength, and normalizing gait patterns. Finally, in Patient demographics All patients (n = 501) weeks 13–16 intense lower extremity weight training and Age 65.37 ± 10.06 sport-specific training programs began with the goals of Sex approximating muscle strength and returning to sport- Female 267 (53.3) specific activities. Male 234 (46.7) White race 399 (79.6) Independent variables BMI 29.29 ± 5.39 Data were collected using an administrative database for ASA 3 + 167 (33.3) patient demographics, including age, sex, race, and body OA 482 (96.2) mass index (BMI). Seventeen comorbidities (presented Fracture 10 (2.0) in Table  1) were evaluated as defined by International AVN 9 (1.8) Classification of Disease 10th Edition (ICD-10) diag - Other diagnoses 0 (0) nosis codes. The definitions of each comorbidity used Obesity 210 (41.9) are presented in the Additional file  1. American Society Diabetes of Anesthesiologists (ASA) score was used to quantify Type 1 diabetes 1 (0.2) preoperative health status. The Centers for Medicare Type 2 diabetes 61 (12.2) and Medicaid Services (CMS) Hierarchical condition Type 1 or 2 diabetes 62 (12.4) category (HCC) score was also used to quantify levels Sleep apnea 72 (14.4) of comorbidity burden. HCC quantifies patient health COPD 20 (4.0) status by assigning risk scores to patients based on diag- Liver disease 10 (2.0) nosis codes and demographic factors and was calculated Asthma 46 (9.2) for risk stratification of all patients in a payer-agnostic AFIB 30 (6.0) fashion at our institution [21]. The HCC model sums CHF 3 (0.6) demographic factors and disease-based condition catego- CAD 48 (9.6) ries based on diagnoses in the past year and applies an ESRD CKD 29 (5.8) interaction factor to adjust for increased risk in patients GERD 144 (28.7) with multiple related comorbidities. Scores are normal- Anxiety/Depression 106 (21.2) ized to 1.0, with higher scores indicating a greater comor- HTN 260 (51.9) bidity burden and higher expected medical expenditures PVD 0 (0) [22, 23]. The primary reason for THA was evaluated via Neoplasm 11 (2.2) manual chart review and classified as osteoarthritis (OA), Anemia 12 (2.4) fracture, avascular necrosis (AVN), or others. HCC score 0.49 ± 0.27 ASA American Society of Anesthesiologisits, AVN Avascular necrosis, OA Outcome measures Osteoarthritis, COPD Chronic obstructive pulmonary disease, AFIB Atrial Outcomes of interest included the total number of fibrillation, CHF Congestive heart failure, CAD Coronary artery disease, ESRD CKD therapy sessions, more than 3  months of PT, number End-stage renal disease/chronic kidney disease, GERD Gastroesophageal reflux disease, HTN Hypertension, PVD Peripheral vascular disease, HCC Hierarchical of evaluations in 3  months, number of re-evaluations condition category in 3 months, the total number of sessions, total therapy charge, days in PT, last lower extremity functional score (LEFS) within 3  months postoperatively, days to LEFS, applied to all secondary and tertiary current procedure 90-day emergency department return, 90-day readmis- terminology (CPT) codes. CPT codes were selected sion and 90-day return to the operating room (OR). All based on the most commonly used treatment modali- outcome measures were captured by manual review of ties for the various visit types at our institution. Using the electronic medical record. Emergency department this approach, charges for the three types of therapy returns and readmissions included returns to outside visits were estimated as follows: evaluation (CPTs institutions participating in the Epic Care Everywhere 97,161 [low complexity evaluation] + 97,116 [gait train- program. Physical therapy charges were estimated ing] + 97,140 [manual therapy], $181.01), re-evaluation based on the CMS Medicare Multiple Procedure Pay- (CPTs 97,164 [re-evaluation est. plan of care] + 97,110 ment Reduction (MPPR) 2022 Rate File [24]. Allowable [therapeutic exercises] + 97,116 + 97,140, $147.26), amounts were calculated using the carrier and locality non-evaluation/re-evaluation treatment session (CPTs codes of our institution, and the 50% rate reduction was 97,110 + 97,116 + 97,140, $79.43). Stock et al. Arthroplasty (2023) 5:26 Page 4 of 9 Statistical analysis The most common comorbidities observed were hyper - Descriptive statistics were used to determine the preva- tension (HTN) in 52% of patients, obesity (42%), and lence of demographics, comorbidities, outcomes, and gastroesophageal reflux disease (GERD, 29%). The preva - utilization patterns of all patients. Patients were then lence of specific comorbidities is listed in Table 1. classified based on the number of total PT sessions. The Across the entire population, patients utilized an top quartile of high utilizers requiring over 16 PT ses- average of 11.45 ± 7.69 PT visits over the three-month sions was compared to the rest of the population. The top postoperative period, accounting for $1065.43 ± 657.48 quartile was selected as the threshold for ’high-utilizers’ in charges (Figs.  1 and 2). On average, these included in alignment with prior studies, as the top 25% of health- one evaluation session, one re-evaluation session, and care utilizers account for a disproportionate amount 9.50 standard therapy sessions. The average time in PT of aggregate healthcare spend [25–27]. As depicted in was 42.27 ± 31.95  days and 7.4% of patients required Tables  4 and 5, univariate analyses, including chi-square over 3  months of therapy. Over the 3-month postop- tests and independent samples t-tests, were used to erative period, the last average LEFS was 30.45 ± 18.79. determine demographic, comorbidity, utilization, and Unplanned resource utilization, including 90-day ED outcome differences between these groups. The Fisher’s returns, readmissions, and returns to the OR occurred in Exact test and Mann Whitney U test were performed 5.0%, 6.8%, and 4.4% of patients, respectively (Table 2). when the assumptions of chi-square and independ- In comparison to the rest of the population, high-PT ent samples t-testing were not met. Stepwise multivari- utilizers requiring 16 or more sessions were significantly ate logistic regression was used to assess the predictors more likely to be female (61.8% vs. 50.5%, P = 0.038) of having more than 16 total sessions. Subgroup analy- and have sleep apnea (20.3% vs. 12.4%, P = 0.043). No sis was then performed to compare outcomes across PT statistically significant differences in demographics or utilization quartiles. One-way ANOVA and chi-square comorbidities were observed between the high-PT uti- testing were performed, with non-parametric tests used lizer group and the rest of the population (Table  3). As when the assumptions of parametric testing were not expected, high-PT utilizers required more overall ses- met. All statistical analyses were performed using R Stu- sions on average (21.6 ± 5.03 vs. 8.15 ± 5.05, P < 0.001) dio (Version 1.4.1717© 2009–2021 RStudio, PBC). Statis- and incurred higher charges ($1933.94 ± 431.20 vs. tical significance was assessed at P < 0.05. 782.82 ± 431.70, P < 0.001) than the rest of the population. Further, a significantly higher proportion of high-utilizers Results required over 3  months of PT (28.5% vs. 0.5%). No sta- Patients included in the study averaged 65 ± 10  years of tistically significant differences in last LEFS scores were age and a BMI of 29 ± 5  kg/m . Overall, 80% of patients observed in high-PT utilizers (35.0 ± 21.4 vs. 30.5 ± 20.6, were white and 53% were female. The majority (96%) P = 0.199), despite the fact that the time to last LEFS of patients underwent THA for the treatment of OA. was significantly longer in this group than the rest of the Fig. 1 Histogram of total number of therapy sessions: The red line depicts the mean number of therapy session utilized by the population. The black line depicts the 75th percentile threshold of 16 sessions used to classify high-PT utilizers S tock et al. Arthroplasty (2023) 5:26 Page 5 of 9 Fig. 2 Total therapy charge: The red line depicts the mean therapy charges of the population. The black line depicts the 75th percentile of therapy charges incurred by high-PT utilizers Table 2 PT utilization and outcomes last LEFS score (P < 0.001) existed. Notably, the highest levels of postoperative function were observed in the Outcomes All patients (n = 501) third quartile of patients undergoing 11–15 total PT ses- Total number of therapy sessions (0–3 months) 11.45 ± 7.69 sions. Finally, no statistically significant differences in 3 + months in PT 37 (7.4) rates of 90-day ED returns, readmissions, or returns to # Evaluations (0–3 months) 1.00 ± 0 OR were observed across therapy utilization quartiles # Re-evaluations (0–3 months) 0.95 ± 0.96 (Table 5). Total number of PT sessions 9.50 ± 6.88 In the multivariate analysis, women (OR = 1.68, Total therapy charge ($) 1065.43 ± 657.48 P = 0.017) and those with sleep apnea (OR = 2.02, Days in PT 42.27 ± 31.95 P = 0.012) were nearly twice as likely to be high-PT Last 90-day LEFS 30.45 ± 18.79 utilizers requiring 16 or more sessions. However, the Days to LEFS 19.50 ± 23.69 white race was protective against high therapy utiliza- 90-day ED return 25 (5.0) tion (OR = 0.58, P = 0.028); white patients were 42% less 90-day readmission 34 (6.8) likely to have over 16 sessions than non-white patients 90-day return to OR 22 (4.4) (Table 6). Data are expressed as mean ± SD or n (%); : n = 203; PT Physical therapy, LEFS Lower extremity functional score, ED Emergency department, OR Operating room Discussion The results of this study demonstrated that utilization of PT services after primary THA was highly variable, with the top quartile of high-PT utilizers averaging nearly 22 population (34.7 ± 28.8  days vs. 14.3 ± 19.1  days). No sessions postoperatively. These high utilizers were more statistically significant differences in rates of 90-day ED likely to be female and have sleep apnea, and less likely to returns, readmissions, or returns to OR were observed be of white race. While high levels of PT utilization were between groups (Table 4). associated with increased costs, they did not translate In the subgroup analysis of the population strati- to significantly improved physical function or decreased fied in quartiles of therapy utilization, statistically sig - ED returns, readmissions, or returns to the OR during nificant differences in all cost and utilization measures the 90-day postoperative period. With 75% of patients were observed, with the exception of the number of re- requiring 16 or fewer PT sessions, we suggest this thresh- evaluations and re-evaluation charges (all P < 0.001). As old may be considered as a target for the development of expected, each increasing quartile of therapy utilization future PT bundled payment models. incurred higher charges and resulted in longer therapy Various studies have demonstrated the effectiveness of duration. Across the utilization quartiles, significant dif - bundled payment models in reducing the costs of TJA ferences in the last LEFS scores (P = 0.025) and time to [16, 17]. The CMS has reported their findings from the Stock et al. Arthroplasty (2023) 5:26 Page 6 of 9 Table 3 Patient demographics and comorbidities results, the authors standardized their care pathway to ensure the same supplies and techniques were consist- Demographics and Less than or 16 16 + Sessions P-value ently utilized [16]. Financial incentives for physicians comorbidities Sessions (n = 378) (n = 123) meeting targets on quality, operation time, length of Age 65.07 ± 10.02 66.31 ± 10.15 0.239 stay, and patient participation in preoperative education Sex 0.038 classes may have further led to the recorded successes Female 191 (50.5) 76 (61.8) of this pilot program [16]. Another study by Whitcomb Male 187 (49.5) 47 (38.2) et al. created a pilot program for THA bundled payments White race 309 (81.7) 90 (73.2) 0.055 including a clinical model that based the number of phys- BMI 29.18 ± 5.31 29.66 ± 5.62 0.410 ical therapy visits allocated on the discharge date [17]. For ASA 3 + 126 (33.3) 41 (33.3) 1 example, those discharged on postoperative day two were OA 366 (96.8) 116 (94.3) 0.274 given 8 home physical therapy visits, those discharged Fracture 5 (1.3) 5 (4.1) 0.071* three days after surgery were given 6 visits, and on the AVN 7 (1.9) 2 (1.6) 1* day of surgery discharge were given 4 visits included in Obesity 153 (40.5) 57 (46.3) 0.298 the bundled payment [17]. The 45 THA patients included Diabetes in the payment bundle clinical model were more likely to Type 1 diabetes 1 (0.3) 0 (0) 1* be discharged home, had a shorter LOS, and decreased Type 2 diabetes 46 (12.2) 15 (12.2) 1 overall surgical costs including post-acute care and post- Type 1 or 2 diabetes 47 (12.4) 15 (12.2) 1 hospital cost when compared to those in the pre-pilot Sleep apnea 47 (12.4) 25 (20.3) 0.043 period [17]. Both studies have noted decreased costs due COPD 14 (3.7) 6 (4.9) 0.597 to greater home discharge as opposed to SNF utilization Liver disease 7 (1.9) 3 (2.4) 0.713* [16, 17]. Our study was built upon these results, sug- Asthma 33 (8.7) 13 (10.6) 0.665 gesting that opportunities to standardize postoperative AFIB 23 (6.1) 7 (5.7) 1 therapy protocols exist, which may lead to cost savings CHF 2 (0.6) 1 (0.8) 0.571* without compromising postoperative outcomes. CAD 38 (10.1) 10 (8.1) 0.651 Physical therapy is often an essential component ESRD CKD 23 (6.1) 6 (4.9) 0.783 of the preoperative and postoperative care of THA GERD 109 (28.8) 35 (28.5) 1 patients. However, predicting which patients require Anxiety/depression 78 (20.6) 28 (22.8) 0.708 formal PT and utilization patterns is difficult given the HTN 192 (50.8) 68 (55.3) 0.446 multitude of factors influencing recovery. Zeng et  al. PVD 0 (0) 0 (0) 1 developed a predictive model to determine the physi- Neoplasm 7 (1.9) 4 (3.3) 0.476* cal therapy placement for total knee arthroplasty, home Anemia 8 (2.1) 4 (3.3) 0.500* health service or outpatient PT [28]. This model found HCC score 0.49 ± 0.29 0.48 ± 0.21 1 four predictors including increased age, female gender, P-values < 0.05 are in bold; * Denotes Fisher’s Exact Test; Data are expressed lack of access to transportation, and lack of motivation as mean ± SD or n (%); BMI Body mass index, ASA American Society of to participate in outpatient PT as predictive of home Anesthesiologists, AVN Avascular necrosis, OA Osteoarthritis, COPD Chronic health care rather than outpatient therapy utilization obstructive pulmonary disease, AFIB Atrial fibrillation, CHF Congestive heart failure, CAD Coronary artery disease, ESRD CKD End-stage renal disease/chronic [28]. Further, Klement et  al. found increased Charlson kidney disease, GERD Gastroesophageal reflux disease, HTN Hypertension, PVD Comorbidity Index (CCI) scores, increased body mass Peripheral vascular disease, HCC Hierarchical condition category index (BMI), and increased preoperative Short Form 12 (SF-12) mental scores to be independent predictors for BPCI Advanced Model for the first two years and found outpatient, rather than in-home PT [8]. In addition to surgical savings of 3.6% compared to cost without the the risk factors described by prior studies, we identified bundled model, driven mainly by orthopedic procedures female, non-white, sleep apnea patients to be increased [13]. Moreover, CMS reported improvements in the qual- utilizers of postoperative therapy. While we observed ity of care given reduced unplanned readmissions and no improvement in outcomes in these high utilizers, it reduced post-acute care utilization [13]. In examining is also important to note that this population did not approaches to successfully manage bundled payments, experience outcomes inferior to those utilizing fewer Froemke et al. reported savings of over $250,000 with an than 16 therapy sessions. It is therefore possible that estimated 62% of patients coming in or under the target increased therapy did have a protective effect in this price [16]. These savings were a result of shorter lengths population, enabling them to achieve early functional of hospital stay, more home discharges, and lower post- improvements and low complication rates similar to operative resource utilization [16]. To achieve these lower-risk patients. In the context of potential bundled S tock et al. Arthroplasty (2023) 5:26 Page 7 of 9 Table 4 Outcomes by PT utilization Outcomes Less than or 16 sessions 16 + Total sessions P-value (n = 378) (n = 123) Total number of sessions (0–3 months) 8.15 ± 5.05 21.6 ± 5.03 < 0.001 3 + months in PT 2 (0.5) 35 (28.5) < 0.001 # Evaluations (0–3 months) 1.00 ± 0 1.00 ± 0 1 # Re-evaluations (0–3 months) 0.58 ± 0.62 2.08 ± 0.94 < 0.001 Total # of PT sessions 6.57 ± 4.59 18.5 ± 4.49 < 0.001 Total PT charge ($) 782.82 ± 431.7 1933.94 ± 431.2 < 0.001 Evaluation PT charges 181.01 ± 0 181.01 ± 0 1 Re-evaluation PT Charges 80.3 ± 84.8 281.01 ± 114.5 < 0.001 Days in PT 29.6 ± 22.1 81.1 ± 25.9 < 0.001 b c Last 90-day LEFS 30.5 ± 20.6 35.0 ± 21.4 0.199 Days to LEFS 14.3 ± 19.1 34.7 ± 28.8 < 0.001 90-day ED return 23 (6.1) 2 (1.6) 0.083 90-day readmission 29 (7.7) 5 (4.1) 0.240 90-day return to OR 19 (5.0) 3 (2.4) 0.179 a b c P-value < 0.05 are in bold; Data are expressed as mean ± SD or n (%); Denotes Fisher’s Exact Test; n = 151, n = 52; PT Physical therapy, LEFS Lower extremity functional score, ED Emergency department, OR Operating room Table 5 Outcomes by PT utilization quartiles Outcomes Quartile 1: Quartile 2: Quartile 3: Quartile 4: P-value ≤ 25th percentile 26-50th percentile 51-74th percentile ≥ 75th percentile (< 5 total sessions, (5–10 total sessions, (11–15 total sessions, (16 + total sessions, n = 122) n = 122) n = 134) n = 123) Total number of sessions (0–3 months) 1.89 ± 0.95 8.46 ± 1.97 13.6 ± 1.55 21.62 ± 5.03 < 0.001 3 + months in PT 0 (0) 0 (0) 2 (1.5) 35 (28.5) < 0.001 # Evaluations (0–3 months) 1.00 ± 0 1.00 ± 0 1.00 ± 0 1.00 ± 0 1 # Re-evaluations 0–3 months 0.02 ± 0.13 0.52 ± 0.53 1.14 ± 0.45 2.08 ± 0.94 < 0.001 Total number of PT sessions 0.87 ± 0.93 6.93 ± 1.73 11.4 ± 1.58 18. 54 ± 4.49 < 0.001 Total PT charge ($) 251.88 ± 76.73 803.62 ± 178.24 1248.59 ± 431.21 1933.95 ± 430.85 < 0.001 Evaluation PT charges 181.01 ± 0 181.01 ± 0 181.01 ± 0 181.01 ± 0 1 Re-evaluation PT charges 1.86 ± 15.1 71.0 ± 72.6 160.31 ± 57.2 280.57 ± 114.1 < 0.001 Days in PT 4.20 ± 6.93 31.5 ± 11.1 51.1 ± 12.7 81.07 ± 25.89 < 0.001 c d e f b Last 90-day LEFS 22.9 ± 11.2 31.5 ± 23.5 36.5 ± 22.4 35.0 ± 21.4 0.025 Days to LEFS 7.56 ± 15.6 14.8 ± 18.8 19.9 ± 20.7 34.7 ± 28.8 < 0.001 90-day ED return 7 (5.7) 9 (7.4) 7 (5.2) 2 (1.6) 0.165 90-day readmission 10 (8.2) 8 (6.6) 11 (8.2) 5 (4.1) 0.519 90-day return to OR 5 (4.1) 6 (4.9) 8 (5.9) 3 (2.4) 0.548 a b c d e f P-value < 0.05 are in bold; Data are expressed as mean ± SD or n (%); Denotes Fisher’s Exact Test; Denotes Kruskal Wallis Test; n = 48, n = 51, n = 52, n = 52; PT Physical therapy, LEFS Lower extremity functional score, ED Emergency department, OR Operating room payment models, these factors should be considered for Recent studies have evaluated new models of physi- the risk adjustment of payments. It is critical that sex, cal therapy, including home therapy or remote therapy. race, and comorbidities be included in risk adjustment One study by Menon et  al. investigated a pilot program methodologies to adequately compensate providers and Outpatient Physical Therapy Home Visits (OPTHV) to avoid the adverse effects of cherry-picking only patients improve postoperative outcomes and resource utiliza- likely to utilize few sessions. We see cost savings oppor- tion [9]. The OPTHV program offers additional in-home tunities if therapy centers are aligned in a value-based services including preoperative physical and environ- bundle. mental assessments for fall risks and equipment usage, Stock et al. Arthroplasty (2023) 5:26 Page 8 of 9 Table 6 Stepwise multivariate logistic regression: predictors of therefore influence the different results observed. A final 16 + sessions limitation of our study is the lack of published literature on existing outpatient PT bundles, as this payment model Predictors Odd ratio 95% Confidence P-value interval has not been previously implemented, to our knowledge. Further research is warranted to develop the most cost- Female 1.68 1.10–2.60 0.017 effective PT bundled payment model, and additional White race 0.58 0.36–0.95 0.028 investigation into adequate risk adjustment methodolo- Fracture 3.17 0.86–11.65 0.074 gies is needed. Sleep apnea 2.02 1.15–3.47 0.012 P-value < 0.05 are in bold Conclusion In this study, THA patients exhibited significant variabil - ity in postoperative outpatient PT utilization. While high home exercises, and stair and transfer training [9]. Post- levels of PT utilization were associated with increased operatively, this program includes in-home PT assess- costs, they did not translate to significantly improved ments within a week after surgery and helps initiate the physical function or decreased ED returns, readmissions, transition to outpatient PT while also evaluating preven- or returns to the OR during the 90-day postoperative tion measures to avoid complications and readmissions period. With 75% of patients requiring 16 or fewer PT [9]. Menon et  al. found those enrolled in the Outpatient sessions, we suggest this threshold may be considered as Physical Therapy Home Visits program had shorter a target for the development of future PT bundled pay- LOS and were more likely to be discharged home [9]. ment models. Given the waning participation in the full Mahomed et  al. randomly assigned 234 TJA patients to episode of care bundles, the development of PT bundles either home-based or inpatient rehabilitation and found may be an opportunity to increase the adoption of value- no significant differences between groups in pain, func - based payment models in the THA population. tional outcomes, or patient satisfaction [29]. Another study conducted a randomized control trial of 120 THA patients for a self-directed home exercise program for Abbreviations THA Total hip arthroplasty 10  weeks or standard protocol that included 2  weeks of BMI Body mass index in-home physical therapy visits followed by 8  weeks of LEFS Lower extremity functional score formal outpatient physical therapy [10]. This study found ED Emergency department OR Operating room/odds ratio no statistically significant differences in functional out - PT Physical therapy come measures between the two groups [10]. Although TJA Total joint arthroplasty some authors have proposed no formal outpatient physi- TKA Total knee arthroplasty LOS Length of stay cal therapy following THA, most surgeons use therapy CJR Comprehensive Care for Joint Replacement for their patients [7, 10, 11, 30]. Further, self-directed BPCI Bundled Payments for Care Improvement home exercise programs are not suitable for all patients. CMS Centers for Medicare & Medicaid Services SNF Skilled nursing facility Therefore, payment bundles should be considered for COPD Chronic obstructive pulmonary disease outpatient physical therapy programs to help mitigate CHF Congestive heart failure costs without jeopardizing the quality of care. NSAID Non-steroidal anti-inflammatory drugs ICD-10 I nternational Classification of Disease 10th Edition Our study does not come without limitations. One lim- ASA American Society of Anesthesiologist itation is within our outcome measures for readmission HCC Hierarchical Condition Category rates and emergency department visits. These measures OA Osteoarthritis AVN Avascular necrosis were only recorded for institutions participating in Epic MPPR Multiple Procedure Payment Reduction Care Everywhere, therefore patients presenting to non- CPT Current procedural terminology Epic hospitals were not included. Second, it is possible HTN Hypertension GERD Gastroesophageal reflux disease that selection bias existed, as we only included patients CCI Charlson Comorbidity Index who completed PT at our hospital-affiliated therapy prac - SF-12 Shor t form 12 tice, in order to collect the most accurate data regarding OPTHV Outpatient physical therapy home visits the number of visits, evaluations, and re-evaluations. Third, it is possible that unmeasured confounding fac - Supplementary Information tors, including surgical approach, hospital length of stay, The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s42836- 023- 00179-2. and variability in adherence to PT protocols, impacted our results. Fourth, when comparing our results to Additional file 1. Comorbidity definitions based on ICD-10 diagnosis those of previously published studies, the therapy pro- codes. tocols used across institutions were variable, and might S tock et al. Arthroplasty (2023) 5:26 Page 9 of 9 Acknowledgements 10. Austin MS, Urbani BT, Fleischman AN, Fernando ND, Purtill JJ, Hozack WJ, Not applicable. et al. Formal physical therapy after total hip arthroplasty is not required: a randomized controlled trial. J Bone Joint Surg Am. 2017;99(8):648–55. Authors contributions 11. Yayac M, Moltz R, Pivec R, Lonner JH, Courtney PM, Austin MS. Formal physi- L.A.S., A.H.J., and J.J.T. were major contributors to the writing of this manu- cal therapy following total hip and knee arthroplasty incurs additional cost script. J.C.B. performed the statistical analysis regarding this project. L.A.S. also without improving outcomes. J Arthroplasty. 2020;35(10):2779–85. contributed to data collection for this project. J.J.T., P.J.K., and J.H.M. provided 12. Moore JD. Unpacking payment bundles. Phys Ther. 2016;96(2):139–41. oversight of this project. All authors read and approved this final manuscript. 13. Services CfMaM. BPCI advanced 2022 updated 02/1/2022. Available from: https:// innov ation. cms. gov/ innov ation- models/ bpci- advan ced# prici Funding ng. Accessed 1 Aug 2022. This project did not receive any funding. 14. Services CfMaM. Comprehensive care for joint replacement model 2019. Available from: https:// innov ation. cms. gov/ initi atives/ cjr. Accessed 1 Aug Availability of data and materials 2022. The datasets used and/or analyzed during the current study are available from 15. CMS. Bundled Payments for Care Improvement (BPCI) Initiative: General the corresponding author upon reasonable request. Information 2020. Available from: https:// innov ation. cms. gov/ innov ation- models/ bundl ed- payme nts. Accessed 1 Aug 2022. 16. Froemke CC, Wang L, DeHart ML, Williamson RK, Ko LM, Duwelius PJ. Stand- Declarations ardizing care and improving quality under a bundled payment initiative for total joint arthroplasty. J Arthroplasty. 2015;30(10):1676–82. Ethics approval and consent to participate 17. Whitcomb WF, Lagu T, Krushell RJ, Lehman AP, Greenbaum J, McGirr J, et al. Institutional review board approval was obtained. Experience with designing and implementing a bundled payment program for total hip replacement. Jt Comm J Qual Patient Saf. 2015;41(9):406–13. Consent for publication 18. Meyer H. Joint replacement bundled payments losing their appeal in BPCI All authors certify that they’ve read and approved this manuscript. Advanced2020 07/13/2022]. Available from: https:// www. moder nheal thcare. com/ finan ce/ joint- repla cement- bundl ed- payme nts- losing- their- Competing interests appeal- bpci- advan ced. Accessed 1 Aug 2022. No relevant disclosures. 19. Springer BD, Mcinerney J. Medicare bundles for arthroplasty : a journey back to fee for service? Bone Joint J. 2021;103-B(6 Supple A):119–25. Author details 20. Navathe AS, Emanuel EJ, Venkataramani AS, Huang Q, Gupta A, Dinh Anne Arundel Medical Center, Annapolis, MD 21401, USA. CT, et al. Spending and quality after three years of medicare’s voluntary bundled payment for joint replacement surgery. Health Aff (Millwood). Received: 6 September 2022 Accepted: 7 March 2023 2020;39(1):58–66. 21. Turcotte J, Sanford Z, Broda A, Patton C. Centers for Medicare & Medicaid Services Hierarchical Condition Category score as a predictor of readmission and reoperation following elective inpatient spine surgery. J Neurosurg Spine. 2019;31(4):600–6. References 22. Services CfMM. Risk Adjustment: CMS.gov; 2021 [updated 12/01/2021. 1. Center CM. How long do you need physical therapy after a knee replace- Available from: https:// www. cms. gov/ Medic are/ Health- Plans/ Medic areAd ment?: Conway Medical Center. 2021. Available from: https:// www. conwa vtgSp ecRat eStats/ Risk- Adjus tors]. ymedi calce nter. com/ news/ how- long- need- physi cal- thera py- after- knee- 23. Services CfMM. Report to congress: risk adjustment in medicare advantage. repla cement. Accessed 1 Aug 2022. CMS.gov; 2018. 2. Savyasachi C Thakkar MD. Hip replacement recovery: Q&A with a hip spe- 24. Services CfMM. Therapy Services CMS.gov2021 updated 12/01/2021. Availa- cialist: Johns Hopkins Medicine. Available from: https:// www. hopki nsmed ble from: https:// www. cms. gov/ Medic are/ Billi ng/ Thera pySer vices. Accessed icine. org/ health/ treat ment- tests- and- thera pies/ hip- repla cement- recov ery- 1 Aug 2022. qa. Accessed 1 Aug 2022. 25. Shah N, Greenberg JA, Leverson G, Funk LM. Predictors of high cost after 3. Temporiti F, Draghici I, Fusi S, Traverso F, Ruggeri R, Grappiolo G, Gatti R. Does bariatric surgery: a single institution review. Surgery. 2016;160(4):877–84. walking the day of total hip arthroplasty speed up functional independ- 26. Kyle MA, McWilliams JM, Landrum MB, Landon BE, Trompke P, Nyweide ence? A non-randomized controlled study. Arch Physiother. 2020;10:1–7. DJ, et al. Spending variation among ACOs in the Medicare Shared Savings 4. Frassanito L, Vergari A, Nestorini R, Cerulli G, Placella G, Pace V, et al. Program. Am J Manag Care. 2020;26(4):170–5. Enhanced recovery after surgery (ERAS) in hip and knee replacement sur- 27. Reschovsky JD, Hadley J, Saiontz-Martinez CB, Boukus ER. Following the gery: description of a multidisciplinary program to improve management money: factors associated with the cost of treating high-cost Medicare of the patients undergoing major orthopedic surgery. Musculoskelet Surg. beneficiaries. Health Serv Res. 2011;46(4):997–1021. 2020;104(1):87–92. 28. Chan Zeng P, Mark W. Melberg, MD, Heather M. Tavel, MPH, Suzanne E. 5. Juliano K, Edwards D, Spinello D, Capizzano Y, Epelman E, Kalowitz J, et al. Argosino, BSMT, BSN, Denise A. Kiepe, BSN, Ella E. Lyons, MS, Morgan A. Ford, Initiating physical therapy on the day of surgery decreases length of stay MS, Claudia A. Steiner, MD, MPH. Development and validation of a model without compromising functional outcomes following total hip arthro- for predicting rehabilitation care location among patients discharged home plasty. HSS J. 2011;7(1):16–20. after total knee arthroplasty. J Arthroplasty. 2020;35(7):1840–6.e.2. 6. Specht K, Kjaersgaard-Andersen P, Kehlet H, Wedderkopp N, Pedersen BD. 29. Mahomed NN, Davis AM, Hawker G, Badley E, Davey JR, Syed KA, et al. Inpa- High patient satisfaction in 445 patients who underwent fast-track hip or tient compared with home-based rehabilitation following primary unilateral knee replacement. Acta Orthop. 2015;86(6):702–7. total hip or knee replacement: a randomized controlled trial. J Bone Joint 7. Rao BM, Cieslewicz TJ, Sochacki KR, Kohlrieser DA, Moylan DD, Ellis TJ. Worse Surg Am. 2008;90(8):1673–80. preoperative pain and higher activity levels predict patient choice of formal 30 Freburger JK. An analysis of the relationship between the utilization of physical therapy after primary anterior total hip arthroplasty. J Arthroplasty. physical therapy services and outcomes of care for patients after total hip 2021;36(8):2823-8 e2. arthroplasty. Phys Ther. 2000;80(5):448. 8. Klement MR, Rondon AJ, McEntee RM, Greenky MR, Austin MS. Web-based, self-directed physical therapy after total knee arthroplasty is safe and effec- Publisher’s Note tive for most, but not all patients. J Arthroplasty. 2019;34(7S):S178–82. Springer Nature remains neutral with regard to jurisdictional claims in pub- 9. Menon N, Turcotte JJ, Stone AH, Adkins AL, MacDonald JH, King PJ. Outpa- lished maps and institutional affiliations. tient, home-based physical therapy promotes decreased length of stay and post-acute resource utilization after total joint arthroplasty. J Arthroplasty. 2020;35(8):1968–72. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Arthroplasty Springer Journals

Outpatient physical therapy bundled payment models are feasible for total hip arthroplasty patients: an evaluation of utilization, cost and outcomes

Download PDF
9 pages

Loading next page...
 
/lp/springer-journals/outpatient-physical-therapy-bundled-payment-models-are-feasible-for-W66zNwZVoG

References (41)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2023
eISSN
2524-7948
DOI
10.1186/s42836-023-00179-2
Publisher site
See Article on Publisher Site

Abstract

Background Various episode-of-care bundled payment models for patients undergoing total joint arthroplasty have been implemented. However, participation in bundled payment programs has dropped given the challenges of meeting continually lower target prices. The purpose of our study is to investigate the cost of outpatient physical therapy (PT ) and the potential for stand-alone outpatient PT bundled payments for patients undergoing total hip arthroplasty ( THA). Methods A retrospective review of 501 patients who underwent primary unilateral THA from November 2017 to February 2020 was performed. All patients included in this study received postoperative PT care at a single hospital- affiliated PT practice. Patients above the 75th percentile of therapy visits were then classified as high-PT utilizers and compared with the rest of the population using univariate statistics. Stepwise multivariate logistic regression was used to assess the predictors of high therapy utilization. Results Patients averaged 65 ± 10 years of age and a BMI of 29 ± 5 kg/m . Overall, 80% of patients were white and 53% were female. The average patient had 11 ± 8 total therapy sessions in 42 days: one initial evaluation, one re-eval- uation and 9 standard sessions. High-PT utilizers incurred estimated average costs of $1934 ± 431 per patient, com- pared to $783 ± 432 (P < 0.001) in the rest of the population. Further, no significant differences in 90-day outcomes including lower extremity functional scale scores, emergency department returns, readmissions, or returns to the operating room were observed between high utilizers and the rest of the population (all P > 0.08). In the multivariate analysis, women (OR = 1.68, P = 0.017) and those with sleep apnea (OR = 2.02, P = 0.012) were nearly twice as likely to be high utilizers, while white patients were 42% less likely to be high utilizers than patients of other races (OR = 0.58, P = 0.028). Conclusions Outpatient PT utilization is highly variable in patients undergoing THA. However, despite using more services and incurring increased cost, patients in the top quartile of utilization experienced similar outcomes to the rest of the population. If outpatient therapy bundles are to be developed, 16 visits appear to be a reasonable target for pricing, given this provides adequate coverage for 75% of THA patients. *Correspondence: Justin J. Turcotte jturcotte@luminishealth.org Full list of author information is available at the end of the article © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. Stock et al. Arthroplasty (2023) 5:26 Page 2 of 9 Keywords Total hip arthroplasty ( THA), Physical therapy (PT ), Bundled payment Background phase of care may be more attractive to providers. The Physical therapy (PT) is an important step in any treat- purpose of our study was to investigate the cost of out- ment protocol following total joint arthroplasty (TJA) to patient PT and the potential effectiveness of stand-alone improve a patient’s mobility, strength, and independence. outpatient PT payment bundles for patients undergoing In patients undergoing total hip arthroplasty (THA), THA. PT is typically prescribed for 2 to 3 days a week for 6 to 8  weeks [1]. Therapy often starts the day of or the day Methods after surgery and continues for two to three days a week This study was deemed institutional review board exempt along with home exercises until activity goals are met [2]. as a review of existing medical records by the institutional As the performance of TJA continues to shift toward the clinical research committee, and a waiver of informed ambulatory setting, more intensive early therapy pro- consent was granted. A retrospective chart review of all grams are being implemented to facilitate early mobi- patients undergoing THA by 7 board certified surgeons lization and same-day discharge [3–6]. Such therapy at a single institution was performed. protocols have been shown to decrease hospital length of stay (LOS), decrease hospital costs, increase patient Study population satisfaction, and improve functional status more rapidly All patients included in this study underwent primary [3–6]. Additionally, the costs and benefits of various PT unilateral THA from November 2017 to February 2020. models, including formal outpatient, home-based, and no Patients undergoing bilateral or revision THA were therapy, have been evaluated [7–10]. With some recent excluded from this study. All patients included in this studies showing that formal outpatient PT was not nec- study received postoperative PT care at a single hospital- essary for all THA patients [10, 11], additional research affiliated practice. A total of 501 patients met the inclu - into new models that maximize the value of traditional sion criteria. therapy is needed in order to maintain access to these services. Perioperative protocol Outpatient PT represents a significant portion of TJA All patients were cared for in a coordinated Joint episode cost [11]. In an effort to reduce cost, bundled Replacement Center and received education materials payment initiatives such as the Comprehensive Care for including written materials, preoperative medical evalu- Joint Replacement (CJR), Bundled Payments for Care ations, preoperative home exercise or outpatient physical Improvement (BPCI), and BPCI Advanced programs therapy, and an education class for patients and their car- have been put in place [12–15]. While the mechanics of egivers. All patients were treated utilizing a multimodal each program differ, they operate in a similar manner by pain management protocol which, depending on patient incentivizing hospitals and/or providers to deliver care factors, included acetaminophen, oral NSAIDs, pregaba- under an established target price while meeting quality lin, ketorolac, and oral opioid medications as needed. thresholds. Upon implementation, these programs effec - tively reduced the costs of TJA, primarily through reduc- PT protocol tions in LOS and discharge to skilled nursing facilities Standard PT protocols are used across all therapy sites. (SNFs) [13, 16, 17]. Despite these promising early results, However, therapists might modify treatment based on participation in bundled payment programs has dropped their clinical judgment of patient progression. During given the challenges of meeting continually lower tar- weeks 0–2, therapy focuses on range of motion (ROM), get prices [12, 18]. With target pricing decreasing by the flexibility, quadriceps strengthening exercises and gait CMS, the withdrawal rate of the BPCI Advanced pro- training. During this period, patients were expected grams has risen to over 85% [19]. As the financial savings to transition from walker to cane-assisted ambulation. and quality improvement gains from participation in epi- During weeks 3–6, scar mobilization is initiated and sode of care bundled payment programs slow, the ulti- assistive devices were discontinued as the patient’s gait mate fate of such programs remains uncertain [19, 20]. normalizes. In this phase, exercises focus on quadri- In light of the debate over the value of formal outpa- ceps, hamstring and core strengthening, hip abduction tient PT and the challenges of successfully managing the and adduction, and proprioception. During weeks 7–12, cost of an entire episode of care, alternative value-based therapy focused on continued strengthening, single leg payment models, such as bundled pricing for a distinct stance and uneven terrain exercises, and gait training S tock et al. Arthroplasty (2023) 5:26 Page 3 of 9 with the goal of mastering functional activities, improv- Table 1 Total population demographics and comorbidities ing strength, and normalizing gait patterns. Finally, in Patient demographics All patients (n = 501) weeks 13–16 intense lower extremity weight training and Age 65.37 ± 10.06 sport-specific training programs began with the goals of Sex approximating muscle strength and returning to sport- Female 267 (53.3) specific activities. Male 234 (46.7) White race 399 (79.6) Independent variables BMI 29.29 ± 5.39 Data were collected using an administrative database for ASA 3 + 167 (33.3) patient demographics, including age, sex, race, and body OA 482 (96.2) mass index (BMI). Seventeen comorbidities (presented Fracture 10 (2.0) in Table  1) were evaluated as defined by International AVN 9 (1.8) Classification of Disease 10th Edition (ICD-10) diag - Other diagnoses 0 (0) nosis codes. The definitions of each comorbidity used Obesity 210 (41.9) are presented in the Additional file  1. American Society Diabetes of Anesthesiologists (ASA) score was used to quantify Type 1 diabetes 1 (0.2) preoperative health status. The Centers for Medicare Type 2 diabetes 61 (12.2) and Medicaid Services (CMS) Hierarchical condition Type 1 or 2 diabetes 62 (12.4) category (HCC) score was also used to quantify levels Sleep apnea 72 (14.4) of comorbidity burden. HCC quantifies patient health COPD 20 (4.0) status by assigning risk scores to patients based on diag- Liver disease 10 (2.0) nosis codes and demographic factors and was calculated Asthma 46 (9.2) for risk stratification of all patients in a payer-agnostic AFIB 30 (6.0) fashion at our institution [21]. The HCC model sums CHF 3 (0.6) demographic factors and disease-based condition catego- CAD 48 (9.6) ries based on diagnoses in the past year and applies an ESRD CKD 29 (5.8) interaction factor to adjust for increased risk in patients GERD 144 (28.7) with multiple related comorbidities. Scores are normal- Anxiety/Depression 106 (21.2) ized to 1.0, with higher scores indicating a greater comor- HTN 260 (51.9) bidity burden and higher expected medical expenditures PVD 0 (0) [22, 23]. The primary reason for THA was evaluated via Neoplasm 11 (2.2) manual chart review and classified as osteoarthritis (OA), Anemia 12 (2.4) fracture, avascular necrosis (AVN), or others. HCC score 0.49 ± 0.27 ASA American Society of Anesthesiologisits, AVN Avascular necrosis, OA Outcome measures Osteoarthritis, COPD Chronic obstructive pulmonary disease, AFIB Atrial Outcomes of interest included the total number of fibrillation, CHF Congestive heart failure, CAD Coronary artery disease, ESRD CKD therapy sessions, more than 3  months of PT, number End-stage renal disease/chronic kidney disease, GERD Gastroesophageal reflux disease, HTN Hypertension, PVD Peripheral vascular disease, HCC Hierarchical of evaluations in 3  months, number of re-evaluations condition category in 3 months, the total number of sessions, total therapy charge, days in PT, last lower extremity functional score (LEFS) within 3  months postoperatively, days to LEFS, applied to all secondary and tertiary current procedure 90-day emergency department return, 90-day readmis- terminology (CPT) codes. CPT codes were selected sion and 90-day return to the operating room (OR). All based on the most commonly used treatment modali- outcome measures were captured by manual review of ties for the various visit types at our institution. Using the electronic medical record. Emergency department this approach, charges for the three types of therapy returns and readmissions included returns to outside visits were estimated as follows: evaluation (CPTs institutions participating in the Epic Care Everywhere 97,161 [low complexity evaluation] + 97,116 [gait train- program. Physical therapy charges were estimated ing] + 97,140 [manual therapy], $181.01), re-evaluation based on the CMS Medicare Multiple Procedure Pay- (CPTs 97,164 [re-evaluation est. plan of care] + 97,110 ment Reduction (MPPR) 2022 Rate File [24]. Allowable [therapeutic exercises] + 97,116 + 97,140, $147.26), amounts were calculated using the carrier and locality non-evaluation/re-evaluation treatment session (CPTs codes of our institution, and the 50% rate reduction was 97,110 + 97,116 + 97,140, $79.43). Stock et al. Arthroplasty (2023) 5:26 Page 4 of 9 Statistical analysis The most common comorbidities observed were hyper - Descriptive statistics were used to determine the preva- tension (HTN) in 52% of patients, obesity (42%), and lence of demographics, comorbidities, outcomes, and gastroesophageal reflux disease (GERD, 29%). The preva - utilization patterns of all patients. Patients were then lence of specific comorbidities is listed in Table 1. classified based on the number of total PT sessions. The Across the entire population, patients utilized an top quartile of high utilizers requiring over 16 PT ses- average of 11.45 ± 7.69 PT visits over the three-month sions was compared to the rest of the population. The top postoperative period, accounting for $1065.43 ± 657.48 quartile was selected as the threshold for ’high-utilizers’ in charges (Figs.  1 and 2). On average, these included in alignment with prior studies, as the top 25% of health- one evaluation session, one re-evaluation session, and care utilizers account for a disproportionate amount 9.50 standard therapy sessions. The average time in PT of aggregate healthcare spend [25–27]. As depicted in was 42.27 ± 31.95  days and 7.4% of patients required Tables  4 and 5, univariate analyses, including chi-square over 3  months of therapy. Over the 3-month postop- tests and independent samples t-tests, were used to erative period, the last average LEFS was 30.45 ± 18.79. determine demographic, comorbidity, utilization, and Unplanned resource utilization, including 90-day ED outcome differences between these groups. The Fisher’s returns, readmissions, and returns to the OR occurred in Exact test and Mann Whitney U test were performed 5.0%, 6.8%, and 4.4% of patients, respectively (Table 2). when the assumptions of chi-square and independ- In comparison to the rest of the population, high-PT ent samples t-testing were not met. Stepwise multivari- utilizers requiring 16 or more sessions were significantly ate logistic regression was used to assess the predictors more likely to be female (61.8% vs. 50.5%, P = 0.038) of having more than 16 total sessions. Subgroup analy- and have sleep apnea (20.3% vs. 12.4%, P = 0.043). No sis was then performed to compare outcomes across PT statistically significant differences in demographics or utilization quartiles. One-way ANOVA and chi-square comorbidities were observed between the high-PT uti- testing were performed, with non-parametric tests used lizer group and the rest of the population (Table  3). As when the assumptions of parametric testing were not expected, high-PT utilizers required more overall ses- met. All statistical analyses were performed using R Stu- sions on average (21.6 ± 5.03 vs. 8.15 ± 5.05, P < 0.001) dio (Version 1.4.1717© 2009–2021 RStudio, PBC). Statis- and incurred higher charges ($1933.94 ± 431.20 vs. tical significance was assessed at P < 0.05. 782.82 ± 431.70, P < 0.001) than the rest of the population. Further, a significantly higher proportion of high-utilizers Results required over 3  months of PT (28.5% vs. 0.5%). No sta- Patients included in the study averaged 65 ± 10  years of tistically significant differences in last LEFS scores were age and a BMI of 29 ± 5  kg/m . Overall, 80% of patients observed in high-PT utilizers (35.0 ± 21.4 vs. 30.5 ± 20.6, were white and 53% were female. The majority (96%) P = 0.199), despite the fact that the time to last LEFS of patients underwent THA for the treatment of OA. was significantly longer in this group than the rest of the Fig. 1 Histogram of total number of therapy sessions: The red line depicts the mean number of therapy session utilized by the population. The black line depicts the 75th percentile threshold of 16 sessions used to classify high-PT utilizers S tock et al. Arthroplasty (2023) 5:26 Page 5 of 9 Fig. 2 Total therapy charge: The red line depicts the mean therapy charges of the population. The black line depicts the 75th percentile of therapy charges incurred by high-PT utilizers Table 2 PT utilization and outcomes last LEFS score (P < 0.001) existed. Notably, the highest levels of postoperative function were observed in the Outcomes All patients (n = 501) third quartile of patients undergoing 11–15 total PT ses- Total number of therapy sessions (0–3 months) 11.45 ± 7.69 sions. Finally, no statistically significant differences in 3 + months in PT 37 (7.4) rates of 90-day ED returns, readmissions, or returns to # Evaluations (0–3 months) 1.00 ± 0 OR were observed across therapy utilization quartiles # Re-evaluations (0–3 months) 0.95 ± 0.96 (Table 5). Total number of PT sessions 9.50 ± 6.88 In the multivariate analysis, women (OR = 1.68, Total therapy charge ($) 1065.43 ± 657.48 P = 0.017) and those with sleep apnea (OR = 2.02, Days in PT 42.27 ± 31.95 P = 0.012) were nearly twice as likely to be high-PT Last 90-day LEFS 30.45 ± 18.79 utilizers requiring 16 or more sessions. However, the Days to LEFS 19.50 ± 23.69 white race was protective against high therapy utiliza- 90-day ED return 25 (5.0) tion (OR = 0.58, P = 0.028); white patients were 42% less 90-day readmission 34 (6.8) likely to have over 16 sessions than non-white patients 90-day return to OR 22 (4.4) (Table 6). Data are expressed as mean ± SD or n (%); : n = 203; PT Physical therapy, LEFS Lower extremity functional score, ED Emergency department, OR Operating room Discussion The results of this study demonstrated that utilization of PT services after primary THA was highly variable, with the top quartile of high-PT utilizers averaging nearly 22 population (34.7 ± 28.8  days vs. 14.3 ± 19.1  days). No sessions postoperatively. These high utilizers were more statistically significant differences in rates of 90-day ED likely to be female and have sleep apnea, and less likely to returns, readmissions, or returns to OR were observed be of white race. While high levels of PT utilization were between groups (Table 4). associated with increased costs, they did not translate In the subgroup analysis of the population strati- to significantly improved physical function or decreased fied in quartiles of therapy utilization, statistically sig - ED returns, readmissions, or returns to the OR during nificant differences in all cost and utilization measures the 90-day postoperative period. With 75% of patients were observed, with the exception of the number of re- requiring 16 or fewer PT sessions, we suggest this thresh- evaluations and re-evaluation charges (all P < 0.001). As old may be considered as a target for the development of expected, each increasing quartile of therapy utilization future PT bundled payment models. incurred higher charges and resulted in longer therapy Various studies have demonstrated the effectiveness of duration. Across the utilization quartiles, significant dif - bundled payment models in reducing the costs of TJA ferences in the last LEFS scores (P = 0.025) and time to [16, 17]. The CMS has reported their findings from the Stock et al. Arthroplasty (2023) 5:26 Page 6 of 9 Table 3 Patient demographics and comorbidities results, the authors standardized their care pathway to ensure the same supplies and techniques were consist- Demographics and Less than or 16 16 + Sessions P-value ently utilized [16]. Financial incentives for physicians comorbidities Sessions (n = 378) (n = 123) meeting targets on quality, operation time, length of Age 65.07 ± 10.02 66.31 ± 10.15 0.239 stay, and patient participation in preoperative education Sex 0.038 classes may have further led to the recorded successes Female 191 (50.5) 76 (61.8) of this pilot program [16]. Another study by Whitcomb Male 187 (49.5) 47 (38.2) et al. created a pilot program for THA bundled payments White race 309 (81.7) 90 (73.2) 0.055 including a clinical model that based the number of phys- BMI 29.18 ± 5.31 29.66 ± 5.62 0.410 ical therapy visits allocated on the discharge date [17]. For ASA 3 + 126 (33.3) 41 (33.3) 1 example, those discharged on postoperative day two were OA 366 (96.8) 116 (94.3) 0.274 given 8 home physical therapy visits, those discharged Fracture 5 (1.3) 5 (4.1) 0.071* three days after surgery were given 6 visits, and on the AVN 7 (1.9) 2 (1.6) 1* day of surgery discharge were given 4 visits included in Obesity 153 (40.5) 57 (46.3) 0.298 the bundled payment [17]. The 45 THA patients included Diabetes in the payment bundle clinical model were more likely to Type 1 diabetes 1 (0.3) 0 (0) 1* be discharged home, had a shorter LOS, and decreased Type 2 diabetes 46 (12.2) 15 (12.2) 1 overall surgical costs including post-acute care and post- Type 1 or 2 diabetes 47 (12.4) 15 (12.2) 1 hospital cost when compared to those in the pre-pilot Sleep apnea 47 (12.4) 25 (20.3) 0.043 period [17]. Both studies have noted decreased costs due COPD 14 (3.7) 6 (4.9) 0.597 to greater home discharge as opposed to SNF utilization Liver disease 7 (1.9) 3 (2.4) 0.713* [16, 17]. Our study was built upon these results, sug- Asthma 33 (8.7) 13 (10.6) 0.665 gesting that opportunities to standardize postoperative AFIB 23 (6.1) 7 (5.7) 1 therapy protocols exist, which may lead to cost savings CHF 2 (0.6) 1 (0.8) 0.571* without compromising postoperative outcomes. CAD 38 (10.1) 10 (8.1) 0.651 Physical therapy is often an essential component ESRD CKD 23 (6.1) 6 (4.9) 0.783 of the preoperative and postoperative care of THA GERD 109 (28.8) 35 (28.5) 1 patients. However, predicting which patients require Anxiety/depression 78 (20.6) 28 (22.8) 0.708 formal PT and utilization patterns is difficult given the HTN 192 (50.8) 68 (55.3) 0.446 multitude of factors influencing recovery. Zeng et  al. PVD 0 (0) 0 (0) 1 developed a predictive model to determine the physi- Neoplasm 7 (1.9) 4 (3.3) 0.476* cal therapy placement for total knee arthroplasty, home Anemia 8 (2.1) 4 (3.3) 0.500* health service or outpatient PT [28]. This model found HCC score 0.49 ± 0.29 0.48 ± 0.21 1 four predictors including increased age, female gender, P-values < 0.05 are in bold; * Denotes Fisher’s Exact Test; Data are expressed lack of access to transportation, and lack of motivation as mean ± SD or n (%); BMI Body mass index, ASA American Society of to participate in outpatient PT as predictive of home Anesthesiologists, AVN Avascular necrosis, OA Osteoarthritis, COPD Chronic health care rather than outpatient therapy utilization obstructive pulmonary disease, AFIB Atrial fibrillation, CHF Congestive heart failure, CAD Coronary artery disease, ESRD CKD End-stage renal disease/chronic [28]. Further, Klement et  al. found increased Charlson kidney disease, GERD Gastroesophageal reflux disease, HTN Hypertension, PVD Comorbidity Index (CCI) scores, increased body mass Peripheral vascular disease, HCC Hierarchical condition category index (BMI), and increased preoperative Short Form 12 (SF-12) mental scores to be independent predictors for BPCI Advanced Model for the first two years and found outpatient, rather than in-home PT [8]. In addition to surgical savings of 3.6% compared to cost without the the risk factors described by prior studies, we identified bundled model, driven mainly by orthopedic procedures female, non-white, sleep apnea patients to be increased [13]. Moreover, CMS reported improvements in the qual- utilizers of postoperative therapy. While we observed ity of care given reduced unplanned readmissions and no improvement in outcomes in these high utilizers, it reduced post-acute care utilization [13]. In examining is also important to note that this population did not approaches to successfully manage bundled payments, experience outcomes inferior to those utilizing fewer Froemke et al. reported savings of over $250,000 with an than 16 therapy sessions. It is therefore possible that estimated 62% of patients coming in or under the target increased therapy did have a protective effect in this price [16]. These savings were a result of shorter lengths population, enabling them to achieve early functional of hospital stay, more home discharges, and lower post- improvements and low complication rates similar to operative resource utilization [16]. To achieve these lower-risk patients. In the context of potential bundled S tock et al. Arthroplasty (2023) 5:26 Page 7 of 9 Table 4 Outcomes by PT utilization Outcomes Less than or 16 sessions 16 + Total sessions P-value (n = 378) (n = 123) Total number of sessions (0–3 months) 8.15 ± 5.05 21.6 ± 5.03 < 0.001 3 + months in PT 2 (0.5) 35 (28.5) < 0.001 # Evaluations (0–3 months) 1.00 ± 0 1.00 ± 0 1 # Re-evaluations (0–3 months) 0.58 ± 0.62 2.08 ± 0.94 < 0.001 Total # of PT sessions 6.57 ± 4.59 18.5 ± 4.49 < 0.001 Total PT charge ($) 782.82 ± 431.7 1933.94 ± 431.2 < 0.001 Evaluation PT charges 181.01 ± 0 181.01 ± 0 1 Re-evaluation PT Charges 80.3 ± 84.8 281.01 ± 114.5 < 0.001 Days in PT 29.6 ± 22.1 81.1 ± 25.9 < 0.001 b c Last 90-day LEFS 30.5 ± 20.6 35.0 ± 21.4 0.199 Days to LEFS 14.3 ± 19.1 34.7 ± 28.8 < 0.001 90-day ED return 23 (6.1) 2 (1.6) 0.083 90-day readmission 29 (7.7) 5 (4.1) 0.240 90-day return to OR 19 (5.0) 3 (2.4) 0.179 a b c P-value < 0.05 are in bold; Data are expressed as mean ± SD or n (%); Denotes Fisher’s Exact Test; n = 151, n = 52; PT Physical therapy, LEFS Lower extremity functional score, ED Emergency department, OR Operating room Table 5 Outcomes by PT utilization quartiles Outcomes Quartile 1: Quartile 2: Quartile 3: Quartile 4: P-value ≤ 25th percentile 26-50th percentile 51-74th percentile ≥ 75th percentile (< 5 total sessions, (5–10 total sessions, (11–15 total sessions, (16 + total sessions, n = 122) n = 122) n = 134) n = 123) Total number of sessions (0–3 months) 1.89 ± 0.95 8.46 ± 1.97 13.6 ± 1.55 21.62 ± 5.03 < 0.001 3 + months in PT 0 (0) 0 (0) 2 (1.5) 35 (28.5) < 0.001 # Evaluations (0–3 months) 1.00 ± 0 1.00 ± 0 1.00 ± 0 1.00 ± 0 1 # Re-evaluations 0–3 months 0.02 ± 0.13 0.52 ± 0.53 1.14 ± 0.45 2.08 ± 0.94 < 0.001 Total number of PT sessions 0.87 ± 0.93 6.93 ± 1.73 11.4 ± 1.58 18. 54 ± 4.49 < 0.001 Total PT charge ($) 251.88 ± 76.73 803.62 ± 178.24 1248.59 ± 431.21 1933.95 ± 430.85 < 0.001 Evaluation PT charges 181.01 ± 0 181.01 ± 0 181.01 ± 0 181.01 ± 0 1 Re-evaluation PT charges 1.86 ± 15.1 71.0 ± 72.6 160.31 ± 57.2 280.57 ± 114.1 < 0.001 Days in PT 4.20 ± 6.93 31.5 ± 11.1 51.1 ± 12.7 81.07 ± 25.89 < 0.001 c d e f b Last 90-day LEFS 22.9 ± 11.2 31.5 ± 23.5 36.5 ± 22.4 35.0 ± 21.4 0.025 Days to LEFS 7.56 ± 15.6 14.8 ± 18.8 19.9 ± 20.7 34.7 ± 28.8 < 0.001 90-day ED return 7 (5.7) 9 (7.4) 7 (5.2) 2 (1.6) 0.165 90-day readmission 10 (8.2) 8 (6.6) 11 (8.2) 5 (4.1) 0.519 90-day return to OR 5 (4.1) 6 (4.9) 8 (5.9) 3 (2.4) 0.548 a b c d e f P-value < 0.05 are in bold; Data are expressed as mean ± SD or n (%); Denotes Fisher’s Exact Test; Denotes Kruskal Wallis Test; n = 48, n = 51, n = 52, n = 52; PT Physical therapy, LEFS Lower extremity functional score, ED Emergency department, OR Operating room payment models, these factors should be considered for Recent studies have evaluated new models of physi- the risk adjustment of payments. It is critical that sex, cal therapy, including home therapy or remote therapy. race, and comorbidities be included in risk adjustment One study by Menon et  al. investigated a pilot program methodologies to adequately compensate providers and Outpatient Physical Therapy Home Visits (OPTHV) to avoid the adverse effects of cherry-picking only patients improve postoperative outcomes and resource utiliza- likely to utilize few sessions. We see cost savings oppor- tion [9]. The OPTHV program offers additional in-home tunities if therapy centers are aligned in a value-based services including preoperative physical and environ- bundle. mental assessments for fall risks and equipment usage, Stock et al. Arthroplasty (2023) 5:26 Page 8 of 9 Table 6 Stepwise multivariate logistic regression: predictors of therefore influence the different results observed. A final 16 + sessions limitation of our study is the lack of published literature on existing outpatient PT bundles, as this payment model Predictors Odd ratio 95% Confidence P-value interval has not been previously implemented, to our knowledge. Further research is warranted to develop the most cost- Female 1.68 1.10–2.60 0.017 effective PT bundled payment model, and additional White race 0.58 0.36–0.95 0.028 investigation into adequate risk adjustment methodolo- Fracture 3.17 0.86–11.65 0.074 gies is needed. Sleep apnea 2.02 1.15–3.47 0.012 P-value < 0.05 are in bold Conclusion In this study, THA patients exhibited significant variabil - ity in postoperative outpatient PT utilization. While high home exercises, and stair and transfer training [9]. Post- levels of PT utilization were associated with increased operatively, this program includes in-home PT assess- costs, they did not translate to significantly improved ments within a week after surgery and helps initiate the physical function or decreased ED returns, readmissions, transition to outpatient PT while also evaluating preven- or returns to the OR during the 90-day postoperative tion measures to avoid complications and readmissions period. With 75% of patients requiring 16 or fewer PT [9]. Menon et  al. found those enrolled in the Outpatient sessions, we suggest this threshold may be considered as Physical Therapy Home Visits program had shorter a target for the development of future PT bundled pay- LOS and were more likely to be discharged home [9]. ment models. Given the waning participation in the full Mahomed et  al. randomly assigned 234 TJA patients to episode of care bundles, the development of PT bundles either home-based or inpatient rehabilitation and found may be an opportunity to increase the adoption of value- no significant differences between groups in pain, func - based payment models in the THA population. tional outcomes, or patient satisfaction [29]. Another study conducted a randomized control trial of 120 THA patients for a self-directed home exercise program for Abbreviations THA Total hip arthroplasty 10  weeks or standard protocol that included 2  weeks of BMI Body mass index in-home physical therapy visits followed by 8  weeks of LEFS Lower extremity functional score formal outpatient physical therapy [10]. This study found ED Emergency department OR Operating room/odds ratio no statistically significant differences in functional out - PT Physical therapy come measures between the two groups [10]. Although TJA Total joint arthroplasty some authors have proposed no formal outpatient physi- TKA Total knee arthroplasty LOS Length of stay cal therapy following THA, most surgeons use therapy CJR Comprehensive Care for Joint Replacement for their patients [7, 10, 11, 30]. Further, self-directed BPCI Bundled Payments for Care Improvement home exercise programs are not suitable for all patients. CMS Centers for Medicare & Medicaid Services SNF Skilled nursing facility Therefore, payment bundles should be considered for COPD Chronic obstructive pulmonary disease outpatient physical therapy programs to help mitigate CHF Congestive heart failure costs without jeopardizing the quality of care. NSAID Non-steroidal anti-inflammatory drugs ICD-10 I nternational Classification of Disease 10th Edition Our study does not come without limitations. One lim- ASA American Society of Anesthesiologist itation is within our outcome measures for readmission HCC Hierarchical Condition Category rates and emergency department visits. These measures OA Osteoarthritis AVN Avascular necrosis were only recorded for institutions participating in Epic MPPR Multiple Procedure Payment Reduction Care Everywhere, therefore patients presenting to non- CPT Current procedural terminology Epic hospitals were not included. Second, it is possible HTN Hypertension GERD Gastroesophageal reflux disease that selection bias existed, as we only included patients CCI Charlson Comorbidity Index who completed PT at our hospital-affiliated therapy prac - SF-12 Shor t form 12 tice, in order to collect the most accurate data regarding OPTHV Outpatient physical therapy home visits the number of visits, evaluations, and re-evaluations. Third, it is possible that unmeasured confounding fac - Supplementary Information tors, including surgical approach, hospital length of stay, The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s42836- 023- 00179-2. and variability in adherence to PT protocols, impacted our results. Fourth, when comparing our results to Additional file 1. Comorbidity definitions based on ICD-10 diagnosis those of previously published studies, the therapy pro- codes. tocols used across institutions were variable, and might S tock et al. Arthroplasty (2023) 5:26 Page 9 of 9 Acknowledgements 10. Austin MS, Urbani BT, Fleischman AN, Fernando ND, Purtill JJ, Hozack WJ, Not applicable. et al. Formal physical therapy after total hip arthroplasty is not required: a randomized controlled trial. J Bone Joint Surg Am. 2017;99(8):648–55. Authors contributions 11. Yayac M, Moltz R, Pivec R, Lonner JH, Courtney PM, Austin MS. Formal physi- L.A.S., A.H.J., and J.J.T. were major contributors to the writing of this manu- cal therapy following total hip and knee arthroplasty incurs additional cost script. J.C.B. performed the statistical analysis regarding this project. L.A.S. also without improving outcomes. J Arthroplasty. 2020;35(10):2779–85. contributed to data collection for this project. J.J.T., P.J.K., and J.H.M. provided 12. Moore JD. Unpacking payment bundles. Phys Ther. 2016;96(2):139–41. oversight of this project. All authors read and approved this final manuscript. 13. Services CfMaM. BPCI advanced 2022 updated 02/1/2022. Available from: https:// innov ation. cms. gov/ innov ation- models/ bpci- advan ced# prici Funding ng. Accessed 1 Aug 2022. This project did not receive any funding. 14. Services CfMaM. Comprehensive care for joint replacement model 2019. Available from: https:// innov ation. cms. gov/ initi atives/ cjr. Accessed 1 Aug Availability of data and materials 2022. The datasets used and/or analyzed during the current study are available from 15. CMS. Bundled Payments for Care Improvement (BPCI) Initiative: General the corresponding author upon reasonable request. Information 2020. Available from: https:// innov ation. cms. gov/ innov ation- models/ bundl ed- payme nts. Accessed 1 Aug 2022. 16. Froemke CC, Wang L, DeHart ML, Williamson RK, Ko LM, Duwelius PJ. Stand- Declarations ardizing care and improving quality under a bundled payment initiative for total joint arthroplasty. J Arthroplasty. 2015;30(10):1676–82. Ethics approval and consent to participate 17. Whitcomb WF, Lagu T, Krushell RJ, Lehman AP, Greenbaum J, McGirr J, et al. Institutional review board approval was obtained. Experience with designing and implementing a bundled payment program for total hip replacement. Jt Comm J Qual Patient Saf. 2015;41(9):406–13. Consent for publication 18. Meyer H. Joint replacement bundled payments losing their appeal in BPCI All authors certify that they’ve read and approved this manuscript. Advanced2020 07/13/2022]. Available from: https:// www. moder nheal thcare. com/ finan ce/ joint- repla cement- bundl ed- payme nts- losing- their- Competing interests appeal- bpci- advan ced. Accessed 1 Aug 2022. No relevant disclosures. 19. Springer BD, Mcinerney J. Medicare bundles for arthroplasty : a journey back to fee for service? Bone Joint J. 2021;103-B(6 Supple A):119–25. Author details 20. Navathe AS, Emanuel EJ, Venkataramani AS, Huang Q, Gupta A, Dinh Anne Arundel Medical Center, Annapolis, MD 21401, USA. CT, et al. Spending and quality after three years of medicare’s voluntary bundled payment for joint replacement surgery. Health Aff (Millwood). Received: 6 September 2022 Accepted: 7 March 2023 2020;39(1):58–66. 21. Turcotte J, Sanford Z, Broda A, Patton C. Centers for Medicare & Medicaid Services Hierarchical Condition Category score as a predictor of readmission and reoperation following elective inpatient spine surgery. J Neurosurg Spine. 2019;31(4):600–6. References 22. Services CfMM. Risk Adjustment: CMS.gov; 2021 [updated 12/01/2021. 1. Center CM. How long do you need physical therapy after a knee replace- Available from: https:// www. cms. gov/ Medic are/ Health- Plans/ Medic areAd ment?: Conway Medical Center. 2021. Available from: https:// www. conwa vtgSp ecRat eStats/ Risk- Adjus tors]. ymedi calce nter. com/ news/ how- long- need- physi cal- thera py- after- knee- 23. Services CfMM. Report to congress: risk adjustment in medicare advantage. repla cement. Accessed 1 Aug 2022. CMS.gov; 2018. 2. Savyasachi C Thakkar MD. Hip replacement recovery: Q&A with a hip spe- 24. Services CfMM. Therapy Services CMS.gov2021 updated 12/01/2021. Availa- cialist: Johns Hopkins Medicine. Available from: https:// www. hopki nsmed ble from: https:// www. cms. gov/ Medic are/ Billi ng/ Thera pySer vices. Accessed icine. org/ health/ treat ment- tests- and- thera pies/ hip- repla cement- recov ery- 1 Aug 2022. qa. Accessed 1 Aug 2022. 25. Shah N, Greenberg JA, Leverson G, Funk LM. Predictors of high cost after 3. Temporiti F, Draghici I, Fusi S, Traverso F, Ruggeri R, Grappiolo G, Gatti R. Does bariatric surgery: a single institution review. Surgery. 2016;160(4):877–84. walking the day of total hip arthroplasty speed up functional independ- 26. Kyle MA, McWilliams JM, Landrum MB, Landon BE, Trompke P, Nyweide ence? A non-randomized controlled study. Arch Physiother. 2020;10:1–7. DJ, et al. Spending variation among ACOs in the Medicare Shared Savings 4. Frassanito L, Vergari A, Nestorini R, Cerulli G, Placella G, Pace V, et al. Program. Am J Manag Care. 2020;26(4):170–5. Enhanced recovery after surgery (ERAS) in hip and knee replacement sur- 27. Reschovsky JD, Hadley J, Saiontz-Martinez CB, Boukus ER. Following the gery: description of a multidisciplinary program to improve management money: factors associated with the cost of treating high-cost Medicare of the patients undergoing major orthopedic surgery. Musculoskelet Surg. beneficiaries. Health Serv Res. 2011;46(4):997–1021. 2020;104(1):87–92. 28. Chan Zeng P, Mark W. Melberg, MD, Heather M. Tavel, MPH, Suzanne E. 5. Juliano K, Edwards D, Spinello D, Capizzano Y, Epelman E, Kalowitz J, et al. Argosino, BSMT, BSN, Denise A. Kiepe, BSN, Ella E. Lyons, MS, Morgan A. Ford, Initiating physical therapy on the day of surgery decreases length of stay MS, Claudia A. Steiner, MD, MPH. Development and validation of a model without compromising functional outcomes following total hip arthro- for predicting rehabilitation care location among patients discharged home plasty. HSS J. 2011;7(1):16–20. after total knee arthroplasty. J Arthroplasty. 2020;35(7):1840–6.e.2. 6. Specht K, Kjaersgaard-Andersen P, Kehlet H, Wedderkopp N, Pedersen BD. 29. Mahomed NN, Davis AM, Hawker G, Badley E, Davey JR, Syed KA, et al. Inpa- High patient satisfaction in 445 patients who underwent fast-track hip or tient compared with home-based rehabilitation following primary unilateral knee replacement. Acta Orthop. 2015;86(6):702–7. total hip or knee replacement: a randomized controlled trial. J Bone Joint 7. Rao BM, Cieslewicz TJ, Sochacki KR, Kohlrieser DA, Moylan DD, Ellis TJ. Worse Surg Am. 2008;90(8):1673–80. preoperative pain and higher activity levels predict patient choice of formal 30 Freburger JK. An analysis of the relationship between the utilization of physical therapy after primary anterior total hip arthroplasty. J Arthroplasty. physical therapy services and outcomes of care for patients after total hip 2021;36(8):2823-8 e2. arthroplasty. Phys Ther. 2000;80(5):448. 8. Klement MR, Rondon AJ, McEntee RM, Greenky MR, Austin MS. Web-based, self-directed physical therapy after total knee arthroplasty is safe and effec- Publisher’s Note tive for most, but not all patients. J Arthroplasty. 2019;34(7S):S178–82. Springer Nature remains neutral with regard to jurisdictional claims in pub- 9. Menon N, Turcotte JJ, Stone AH, Adkins AL, MacDonald JH, King PJ. Outpa- lished maps and institutional affiliations. tient, home-based physical therapy promotes decreased length of stay and post-acute resource utilization after total joint arthroplasty. J Arthroplasty. 2020;35(8):1968–72.

Journal

Arthroplasty – Springer Journals

Published: May 12, 2023

Keywords: Total hip arthroplasty (THA); Physical therapy (PT); Bundled payment

There are no references for this article.