Comparative Analysis of Preadmission Testing Center and Primary Care Physician Clearance Before Total Joint Arthroplasty Authors: Jake Laverdiere, Swaroopa Vaidya, Gregory Panza, Thomas Giannasca, Jenna Bernstein Published: 2026 ABSTRACT BackgroundEffective preoperative evaluation is crucial for mitigating complications and cancellations in total joint arthroplasty. Limited access to a primary care physician (PCP) and fragmented coordination can delay clearance and increase perioperative risk. To address these challenges, preadmission testing centers (PATCs) have been implemented to streamline workflows. This study compares outcomes and cancellation rates between PATC and PCP-led pathways.MethodsRetrospective study of patients undergoing total joint arthroplasty between June 2022 and November 2024 at 2 community hospitals comparing demographics, outcomes, and cancellation rates between PATC vs PCP. Statistical analyses included sample t-tests, Mann-Whitney U tests, chi-square, or Fisher’s exact tests. Clinical outcomes including length of stay, discharge disposition, complications, and cancellations were adjusted for age, body mass index, and Charlson Comorbidity Index.ResultsAmong 8784 patients, 5641 (64.2%) were evaluated at PATC and 3143 (35.8%) by PCP. Patients evaluated at PATC were younger (68.98 ± 9.35 vs 70.38 ± 9.01 years, P < .001) and had a higher Charlson Comorbidity Index (median 1.0 [2.0] vs 1.0 [1.0], P < .001). Overall cancellation rates (P = .637) and 90-day postoperative complications (P = .212) were similar between groups. However, the PATC group demonstrated a slightly longer length of stay (median 27.9 [5.3] vs 27.5 [5.0] hours, P = .001), higher discharge to skilled nursing facilities (4.0% vs 2.3%, P < .05 after Bonferroni adjustment), and greater postoperative emergency department utilization (6.6% vs 4.0%, P < .001).ConclusionsDespite managing a higher-risk population, PATC evaluation demonstrated comparable cancellation and complication rates to PCP clearance, supporting centralized, protocol-driven preoperative evaluation as a viable alternative in real-world settings where access to primary care may be limited. Introduction Comprehensive assessment of a patient’s overall health, with particular attention to modifiable risk factors, is a critical component of preoperative medical clearance [1]. Without robust screening, patients incur a heightened risk of intraoperative and postoperative complications [2]. The financial burden of these complications is substantial. In 2019, readmissions following primary total hip arthroplasty (THA) and total knee arthroplasty (TKA) were estimated to cost $51,769 and $45,223, respectively [3]. Patients typically receive preoperative clearance within 30 days of total joint arthroplasty (TJA). Historically, this clearance has been managed by the patient’s primary care physician (PCPs); however, more than 100 million Americans are estimated to face barriers to accessing primary care [4]. Even for those with a PCP, scheduling difficulties and long wait times could result in preoperative evaluations falling outside of this optimal window [5]. Clearance through a PCP may also limit coordination among surgical, anesthesia, nursing, and laboratory teams, potentially resulting in inadequate patient education and higher rates of surgical cancellations due to insufficient preparedness [6]. Moreover, many preoperative laboratory tests ordered by PCPs are unnecessary, increasing the risk of iatrogenic harm and delays [7]. PCP-led evaluations may also involve multidisciplinary referrals for high-risk patients, further straining available resources [6]. In response, some institutions have developed preoperative clearance programs that offer integrated, multidisciplinary assessments in coordination with the anesthesia team. These internal preoperative evaluation clinics have been shown to provide a safe and accessible alternative to traditional PCP assessment [[8], [9], [10]]. At our institution, a preadmission testing center (PATC) is used to provide preoperative surgical clearance as an alternative to a PCP. This model may facilitate timely surgical clearance for patients and help to avoid unnecessary delays, regardless of PCP access. While one prior study has examined outcomes between institutional preoperative clinics and PCP evaluations in the TJA setting, findings are limited to a single institution and lack adjustment for key risk factors associated with surgical cancellations. Given the variability in patient populations, institutional protocols, and resource availability, further investigation is needed to assess the generalizability of previous findings. This study aimed to address this gap by evaluating differences in surgical cancellation rates and postoperative outcomes between patients who received preoperative assessments at our institutional PATC and those cleared by their PCP. We hypothesized that PATC evaluations would yield lower surgical cancellation and complication rates compared to PCP clearance, owing to standardized risk stratification protocols, direct coordination with anesthesia, and targeted management of modifiable risk factors. Material and methods This was a retrospective cohort study of patients who underwent primary THA and TKA using an institutional database from 2 high-volume community hospitals with standardized protocols through 1 large orthopaedic institute. The study was approved by the institutional review board prior to data collection. Eligible participants were adults aged 18-89 who underwent primary elective TKA or THA between June 1, 2022, and November 30, 2024, with preoperative assessments conducted at either the PATC or PCP. The PATC was staffed by advanced practice providers who followed institutional protocols. Although utilization of the PATC for patients was strongly encouraged at the institution, assignment to either pathway was non-randomized and likely reflective of various factors. This process was based on practical considerations such as surgeon discretion, PCP availability, proximity to the hospital or PCP office, access to transportation, and patient preference, with PATC serving as a streamlined option for patients without PCP access or requiring complex optimization. The PATC evaluation pathway became standard in June 2022. Data collection and outcome measures Patient demographics included age, gender, body mass index (BMI), Charlson Comorbidity Index (CCI), American Society of Anesthesiologists class, race, ethnicity, and insurance type. We evaluated surgical cancellation rates, defined as any cancellations occurring within 1 week of the scheduled procedure, and associated reasons. Key postoperative outcomes included hospital length of stay (LOS), discharge disposition, 30-day complications, 90-day complications, emergency department (ED) visits after surgery, and hospital readmissions. Postoperative complications of interest included medical complications such as acute myocardial infarction, blood transfusion, deep vein thrombosis, pulmonary embolism, pneumonia, sepsis, stroke, and in-hospital mortality, as well as surgical complications including implant failure, wound dehiscence, deep and superficial surgical site infection, and pressure injury. Preoperative patient-reported outcomes measures (PROMs) data included Patient-Reported Outcomes Measurement Information System-10, Knee Injury and Osteoarthritis Outcome Score for Joint Replacement, and Hip Osteoarthritis Outcome Score for Joint Replacement. PATC protocols The PATC followed a standardized, risk-based preoperative protocol. Preoperative optimization thresholds were established by a multidisciplinary council in accordance with clinical practice guidelines from the American Association of Orthopaedic Surgeons [11,12], American College of Cardiology and American Heart Association perioperative guidelines [13], and institutional testing protocols, which defined cutoff criteria for elective arthroplasty and indications for targeted testing. Hemoglobin A1c was obtained for patients with diabetes, prediabetes, hyperglycemia, or chronic steroid use. Endocrinology referral was recommended for hemoglobin A1c greater than 8, and procedures for these patients were canceled. Complete blood count, basic metabolic panel, and type and screen were ordered for all TJA patients including healthy patients younger than 65 years as well as with the following conditions: BMI greater than 35, obstructive sleep apnea, poor functional capacity (<4 Metabolic Equivalent of Task), age greater than 65, diabetes mellitus, hypertension, and coronary artery disease. Screening coagulation studies were only performed when clinically indicated, such as in patients with known or suspected coagulopathy, liver disease, or other relevant medical conditions. Urinalysis and chest X-rays were ordered only when symptoms indicated potential infection. Electrocardiograms were obtained for individuals with cardiometabolic risk or age greater than 65. Cardiac risk was assessed using the Revised Cardiac Risk Index, with further workup utilizing the National Surgical Quality Improvement Program Myocardial Infarction and Cardiac Arrest calculator or American College of Surgeons surgical risk calculators, if indicated. Cardiology was consulted for known coronary artery disease or abnormal electrocardiogram. On the day of surgery, urine pregnancy tests (women 12-52 years), international normalized ratio (if on warfarin), glucose, and potassium levels (if chronic kidney disease or on diuretics) were performed. Data analysis Prior to data extraction, detectable effect sizes were calculated based on the estimated number of patient charts eligible for inclusion. This estimate was derived from historic clinical volume for the study population and prescreening of the electronic medical record system based on key inclusion criteria. It was determined that approximately 6000 patient charts would be eligible, with a 70:30% PATC to PCP allocation. The effect size calculation was conducted using 80% power, and alpha = 0.05. It was determined that a sample size of 6000 charts would be sufficient to detect effect sizes of 0.08 or larger. This retrospective study used a priori power analysis based on an estimated 6000 eligible charts with a 70:30 PATC-to-PCP allocation. This estimated sample size would provide 81% power to detect effect sizes (h) ≥ 0.08 at α = 0.05. The actual number of eligible patient charts exceeded this estimation by ∼2700. All data were checked for normality of distribution and that assumptions were met for the tests performed. Assumptions were visually verified using histograms and Q-Q plots and statistically examined using the Shapiro-Wilk test. To ensure assumptions were met for regression modeling, data were examined for extreme outliers, and independent variables were inspected for multicollinearity using the variance inflation factor. Normally distributed continuous data are presented as the mean and standard deviation, and non-normally distributed continuous data are presented as median and interquartile range. Categorical data are presented as frequency and percentage. Demographic, clinical, and procedural variables were compared between PATC vs PCP with an independent samples t-test or Mann-Whitney U test for continuous variables and a chi-square or Fisher’s exact test for categorical variables. Cancellation outcomes and key clinical outcomes were adjusted for age, BMI, procedure type (hip or knee), and CCI using binary logistic regression for binary variables and multinomial logistic regression for variables with more than 2 categories. Multivariable linear regression was used for the dependent variable LOS. SPSS (IBM, Chicago, IL, USA) version 29.0 was used for statistical analysis, and statistical significance was defined as P < .05. For the total sample, a Bonferroni adjustment was applied to the alpha threshold for statistical interpretation for primary clinical outcomes including LOS, discharge disposition, 30-day complication rate, and 90-day complication rate. Therefore, P < .0125 was used as the level of statistical significance for these endpoints. Results There were 8784 cases that met the inclusion criteria. Among these, 5641 (64.2%) were preoperatively evaluated by the PATC, and 3143 (35.8%) were evaluated by a PCP alone. Patient demographics and clinical characteristics Patient demographics and clinical characteristics between groups are shown in Table 1. Patients evaluated by the PATC were slightly younger (P < .001), had a higher BMI (P < .001), and a greater CCI (P < .001). A greater proportion of PATC patients were Asian, Black or African American, or another race. Conversely, a greater proportion of patients evaluated by a PCP identified were White or Caucasian. PATC patients were more likely to have state Medicaid, whereas patients evaluated by a PCP were more likely to be covered by Medicare. In addition, a greater proportion of PATC patients underwent TKA, whereas THA was more commonly performed among patients evaluated by a PCP (P < .001). PATC patients had lower preoperative PROM scores across all measures—Patient-Reported Outcomes Measurement Information System-10, Knee Injury and Osteoarthritis Outcome Score for Joint Replacement, and Hip Osteoarthritis Outcome Score for Joint Replacement (all P < .001).Table 1Patient demographic and clinical characteristics.VariablePATC (n = 5641)PCP (n = 3143)Total (N = 8784)P valueAge, mean ± SD68.98 ± 9.3570.38 ± 9.0169.48 ± 9.26<.001Gender, n (%).474 Women3386 (60.0)1926 (61.2)5311 (60.5) Men2255 (40.0)1217 (38.7)3472 (39.5)Race, n (%)<.001 American Indian or Alaska Native7 (0.1)1 (0.0)8 (0.1) Asian53 (0.9)a15 (0.5)68 (0.8) Black or African American357 (6.3)a114 (3.6)471 (5.4) Native Hawaiian or other Pacific Islander5 (0.1)1 (0.0)6 (0.1) White or Caucasian4582 (81.2)2610 (83.0)a7192 (81.9) Other359 (6.4)a85 (2.7)444 (5.1) Patient refused195 (3.5)286 (9.1)a481 (5.5) Unknown83 (1.5)31 (1.0)114 (1.3)Body mass index, kg/m230.96 ± 5.5730.36 ± 5.4230.76 ± 5.42<.001Provider, n (%)<.001 Commercial1650 (29.3)891 (28.3)2541 (28.9) CT Medicaid198 (3.5)a38 (1.2)236 (2.7) Medicare3707 (65.7)2176 (69.2)a5883 (67.0) Other government16 (0.3)11 (0.3)27 (0.3) Worker’s compensation64 (1.1)24 (0.8)88 (1.0) Unknown6 (0.1)3 (0.1)9 (0.1)ASA rating, n (%).756 163 (1.1)35 (1.1)98 (1.1) 22486 (44.1)1383 (44.0)3869 (44.0) 33027 (53.7)1691 (53.8)4718 (53.7) 465 (1.2)34 (1.1)99 (1.1)Procedure, n (%)<.001 Knee2507 (44.4)1153 (36.7)3660 (41.7) Hip3134 (55.6)1990 (63.3)5124 (58.3)Charlson Comorbidity Index, median, IQR1.0, 2.01.0, 1.01.0, 2.0<.001Preoperative PROMIS-10, mean ± SDn = 526333.61 ± 6.32n = 286434.91 ± 6.16n = 812734.07 ± 6.29<.001Preoperative KOOS Jr, mean ± SDn = 296750.12 ± 12.22n = 182352.42 ± 11.99n = 479051.00 ± 12.19<.001Preoperative HOOS Jr, mean ± SDn = 222650.90 ± 13.15n = 102152.66 ± 13.55n = 324751.46 ± 13.30<.001ASA, American Society of Anesthesiologists; PROMIS-10, Patient-Reported Outcomes Measurement Information System-10; KOOS Jr, Knee Injury and Osteoarthritis Outcome Score for Joint Replacement; HOOS Jr, Hip Disability and Osteoarthritis Outcome Score for Joint Replacement; IQR, interquartile range; SD, standard deviation.aIndicates where statistical significance occurred. Table 1Patient demographic and clinical characteristics.VariablePATC (n = 5641)PCP (n = 3143)Total (N = 8784)P valueAge, mean ± SD68.98 ± 9.3570.38 ± 9.0169.48 ± 9.26<.001Gender, n (%).474 Women3386 (60.0)1926 (61.2)5311 (60.5) Men2255 (40.0)1217 (38.7)3472 (39.5)Race, n (%)<.001 American Indian or Alaska Native7 (0.1)1 (0.0)8 (0.1) Asian53 (0.9)a15 (0.5)68 (0.8) Black or African American357 (6.3)a114 (3.6)471 (5.4) Native Hawaiian or other Pacific Islander5 (0.1)1 (0.0)6 (0.1) White or Caucasian4582 (81.2)2610 (83.0)a7192 (81.9) Other359 (6.4)a85 (2.7)444 (5.1) Patient refused195 (3.5)286 (9.1)a481 (5.5) Unknown83 (1.5)31 (1.0)114 (1.3)Body mass index, kg/m230.96 ± 5.5730.36 ± 5.4230.76 ± 5.42<.001Provider, n (%)<.001 Commercial1650 (29.3)891 (28.3)2541 (28.9) CT Medicaid198 (3.5)a38 (1.2)236 (2.7) Medicare3707 (65.7)2176 (69.2)a5883 (67.0) Other government16 (0.3)11 (0.3)27 (0.3) Worker’s compensation64 (1.1)24 (0.8)88 (1.0) Unknown6 (0.1)3 (0.1)9 (0.1)ASA rating, n (%).756 163 (1.1)35 (1.1)98 (1.1) 22486 (44.1)1383 (44.0)3869 (44.0) 33027 (53.7)1691 (53.8)4718 (53.7) 465 (1.2)34 (1.1)99 (1.1)Procedure, n (%)<.001 Knee2507 (44.4)1153 (36.7)3660 (41.7) Hip3134 (55.6)1990 (63.3)5124 (58.3)Charlson Comorbidity Index, median, IQR1.0, 2.01.0, 1.01.0, 2.0<.001Preoperative PROMIS-10, mean ± SDn = 526333.61 ± 6.32n = 286434.91 ± 6.16n = 812734.07 ± 6.29<.001Preoperative KOOS Jr, mean ± SDn = 296750.12 ± 12.22n = 182352.42 ± 11.99n = 479051.00 ± 12.19<.001Preoperative HOOS Jr, mean ± SDn = 222650.90 ± 13.15n = 102152.66 ± 13.55n = 324751.46 ± 13.30<.001ASA, American Society of Anesthesiologists; PROMIS-10, Patient-Reported Outcomes Measurement Information System-10; KOOS Jr, Knee Injury and Osteoarthritis Outcome Score for Joint Replacement; HOOS Jr, Hip Disability and Osteoarthritis Outcome Score for Joint Replacement; IQR, interquartile range; SD, standard deviation.aIndicates where statistical significance occurred. Patient demographic and clinical characteristics. ASA, American Society of Anesthesiologists; PROMIS-10, Patient-Reported Outcomes Measurement Information System-10; KOOS Jr, Knee Injury and Osteoarthritis Outcome Score for Joint Replacement; HOOS Jr, Hip Disability and Osteoarthritis Outcome Score for Joint Replacement; IQR, interquartile range; SD, standard deviation. Indicates where statistical significance occurred. Cancellations The univariable and multivariable analysis results for cancellations between groups are shown in Table 2. Subgroup analyses for THA and TKA cases are displayed in Tables S1 and S3, respectively. Among the 8784 patients included, there were 217 (2.5%) cancellations, 110 (1.3%) of which were same-day cancellations (≤24 hours before surgery) and 107 (1.2%) were within a week (>24 hours and ≤7 days before surgery). The overall cancellation rate did not differ significantly between groups (overall: multivariable adjusted P = .637; THA: adjusted P = .751; TKA: adjusted P = .863). The most cited reasons for cancellation included acute change in medical condition, incomplete preoperative optimization, scheduling conflict, and other/unspecified. Additionally, reasons for cancellation did not significantly differ between groups (overall: P = .131; THA: P = .279; TKA: P = .242).Table 2Rate of cancellation and reasons for cancellation.VariablePATC (n = 5641)PCP (n = 3143)Total (N = 8784)PAdjusted PTotal cancellations (%)144 (2.6)73 (2.3)217 (2.5).505.637 Same d (%)73 (1.3)37 (1.2)110 (1.3).688.699 Within 1 wk (%)71 (1.3)36 (1.1)107 (1.2).415.739Status.705 Canceled42 (0.7)18 (0.6)60 (0.7)Ref Procedure not performed56 (1.0)27 (0.9)83 (0.9).743 Rescheduled46 (0.8)28 (0.9)74 (0.8).738Reason for cancellation.131 Acute change in patient’s medical condition60 (1.1)31 (1.0)91 (1.0)Ref Incomplete preoperative optimization36 (0.6)16 (0.5)52 (0.6).755 Scheduling conflict43 (0.8)9 (0.3)60 (0.7).350 Other5 (0.1)17 (0.5)14 (0.2).060Ref, reference category.Adjusted P when comparing groups while controlling covariates. Multivariable logistic regressions were used with study group, age, body mass index, procedure type (knee or hip), and Charlson Comorbidity Index as independent variables and cancellation outcome as the dependent variable.Within 1 week is defined as > 24 hours and ≤7 days before surgery; same day is defined as ≤ 24 hours before surgery.Reference category selected based on what could be considered the “normative category” or the largest category. Table 2Rate of cancellation and reasons for cancellation.VariablePATC (n = 5641)PCP (n = 3143)Total (N = 8784)PAdjusted PTotal cancellations (%)144 (2.6)73 (2.3)217 (2.5).505.637 Same d (%)73 (1.3)37 (1.2)110 (1.3).688.699 Within 1 wk (%)71 (1.3)36 (1.1)107 (1.2).415.739Status.705 Canceled42 (0.7)18 (0.6)60 (0.7)Ref Procedure not performed56 (1.0)27 (0.9)83 (0.9).743 Rescheduled46 (0.8)28 (0.9)74 (0.8).738Reason for cancellation.131 Acute change in patient’s medical condition60 (1.1)31 (1.0)91 (1.0)Ref Incomplete preoperative optimization36 (0.6)16 (0.5)52 (0.6).755 Scheduling conflict43 (0.8)9 (0.3)60 (0.7).350 Other5 (0.1)17 (0.5)14 (0.2).060Ref, reference category.Adjusted P when comparing groups while controlling covariates. Multivariable logistic regressions were used with study group, age, body mass index, procedure type (knee or hip), and Charlson Comorbidity Index as independent variables and cancellation outcome as the dependent variable.Within 1 week is defined as > 24 hours and ≤7 days before surgery; same day is defined as ≤ 24 hours before surgery.Reference category selected based on what could be considered the “normative category” or the largest category. Rate of cancellation and reasons for cancellation. Ref, reference category. Adjusted P when comparing groups while controlling covariates. Multivariable logistic regressions were used with study group, age, body mass index, procedure type (knee or hip), and Charlson Comorbidity Index as independent variables and cancellation outcome as the dependent variable. Within 1 week is defined as > 24 hours and ≤7 days before surgery; same day is defined as ≤ 24 hours before surgery. Reference category selected based on what could be considered the “normative category” or the largest category. Clinical outcomes Primary and secondary clinical outcomes between PATC and PCP groups are shown in Table 3 and Table 4, respectively. Subgroup analyses for THA and TKA cases are displayed in Tables S2 and S4, respectively. Patients evaluated by PATC had a slightly longer median LOS (overall: P = .001). A greater proportion of patients in the overall cohort and TKA subgroups were discharged to a skilled nursing facility (SNF) and had postoperative ED visit(s) (overall: adjusted P < .001; TKA: adjusted P = .003). There were no significant differences in postoperative complications in the overall sample (30-day: adjusted P = .321; 90-day: adjusted P = .212) or in subgroup analysis for THA (30-day complications adjusted P = .991; 90-day complications adjusted P = .867) or TKA (30-day complications adjusted P = .665; 90-day complications adjusted P = .065).Table 3Primary clinical outcomes.VariablePATC (n = 5641)PCP (n = 3143)Total (N = 8784)PAdjusted PLength of stay (h), median, IQR27.9, 5.327.5, 5.027.7, 5.2<.0010.001Discharge disposition, n (%)<.001 Home or self-care1528 (27.1)893 (28.4)2421 (27.6)Ref Home with health care services3890 (69.0)2177 (69.3)6067 (69.1)0.032 Skilled nursing facility223 (4.0)a73 (2.3)296 (3.4)<.00130-d complication, n (%)9 (0.2)1 (0.0)10 (0.1).1080.32190-d complication, n (%)40 (0.7)23 (0.7)63 (0.7).9040.212Ref, reference category; IQR, interquartile range.Reference category selected based on what could be considered the “normative category” or the largest category.aIndicates where statistical significance occurred. Adjusted P when controlling covariates. Multivariable logistic regressions were used for discharge disposition, complications as dependent variables, and study group, age, body mass index, procedure type (knee or hip), and Charlson Comorbidity Index as independent variables. Multivariable linear regression was used for length of stay as the dependent variable and study group, age, body mass index, procedure type (knee or hip), and Charlson Comorbidity Index as independent variables.Table 4Secondary clinical outcomes.VariablePATC (n = 5641)No PATC (n = 3143)Total (N = 8784)PComplications, n (%) AMI <7 d1 (0.0)1 (0.0)2 (0.0)1.00 Pneumonia <7 d2 (0.0)0 (0.0)2 (0.0).540 Sepsis <7 d5 (0.1)0 (0.0)5 (0.1).168 Surgical site bleeding <30 d0 (0.0)0 (0.0)0 (0.0)-- Pulmonary embolism <30 d1 (0.0)0 (0.0)1 (0.0)1.00 Death <30 d0 (0.0)0 (0.0)0 (0.0)-- Mechanical complication <90 d18 (0.3)14 (0.4)32 (0.4).346 Prosthetic joint infection <90 d15 (0.3)9 (0.3)24 (0.3).860Mortality, n (%)0 (0.0)0 (0.0)0 (0.0)--Emergency department visit, n (%)373 (6.6)126 (4.0)499 (5.7)<.001Readmission, n (%)69 (1.2)32 (1.0)101 (1.1).388AMI, acute myocardial infarction. Table 3Primary clinical outcomes.VariablePATC (n = 5641)PCP (n = 3143)Total (N = 8784)PAdjusted PLength of stay (h), median, IQR27.9, 5.327.5, 5.027.7, 5.2<.0010.001Discharge disposition, n (%)<.001 Home or self-care1528 (27.1)893 (28.4)2421 (27.6)Ref Home with health care services3890 (69.0)2177 (69.3)6067 (69.1)0.032 Skilled nursing facility223 (4.0)a73 (2.3)296 (3.4)<.00130-d complication, n (%)9 (0.2)1 (0.0)10 (0.1).1080.32190-d complication, n (%)40 (0.7)23 (0.7)63 (0.7).9040.212Ref, reference category; IQR, interquartile range.Reference category selected based on what could be considered the “normative category” or the largest category.aIndicates where statistical significance occurred. Adjusted P when controlling covariates. Multivariable logistic regressions were used for discharge disposition, complications as dependent variables, and study group, age, body mass index, procedure type (knee or hip), and Charlson Comorbidity Index as independent variables. Multivariable linear regression was used for length of stay as the dependent variable and study group, age, body mass index, procedure type (knee or hip), and Charlson Comorbidity Index as independent variables. Primary clinical outcomes. Ref, reference category; IQR, interquartile range. Reference category selected based on what could be considered the “normative category” or the largest category. Indicates where statistical significance occurred. Adjusted P when controlling covariates. Multivariable logistic regressions were used for discharge disposition, complications as dependent variables, and study group, age, body mass index, procedure type (knee or hip), and Charlson Comorbidity Index as independent variables. Multivariable linear regression was used for length of stay as the dependent variable and study group, age, body mass index, procedure type (knee or hip), and Charlson Comorbidity Index as independent variables. Table 4Secondary clinical outcomes.VariablePATC (n = 5641)No PATC (n = 3143)Total (N = 8784)PComplications, n (%) AMI <7 d1 (0.0)1 (0.0)2 (0.0)1.00 Pneumonia <7 d2 (0.0)0 (0.0)2 (0.0).540 Sepsis <7 d5 (0.1)0 (0.0)5 (0.1).168 Surgical site bleeding <30 d0 (0.0)0 (0.0)0 (0.0)-- Pulmonary embolism <30 d1 (0.0)0 (0.0)1 (0.0)1.00 Death <30 d0 (0.0)0 (0.0)0 (0.0)-- Mechanical complication <90 d18 (0.3)14 (0.4)32 (0.4).346 Prosthetic joint infection <90 d15 (0.3)9 (0.3)24 (0.3).860Mortality, n (%)0 (0.0)0 (0.0)0 (0.0)--Emergency department visit, n (%)373 (6.6)126 (4.0)499 (5.7)<.001Readmission, n (%)69 (1.2)32 (1.0)101 (1.1).388AMI, acute myocardial infarction. Secondary clinical outcomes. AMI, acute myocardial infarction. Discussion The authors hypothesized that PATC-based preoperative clearance prior to TJA would reduce cancellations and surgical complications through standardized optimization. In this retrospective analysis, PATC patients demonstrated comparable cancellation and complication rates to those undergoing traditional PCP-based clearance, despite representing a more medically complex cohort with a higher BMI, greater comorbidity burden, and lower baseline PROMs. Preoperative optimization and structured evaluation have been shown to play a critical role in reducing perioperative risk and improving outcomes in TJA [1,10]. Given the nonrandomized assignment to clearance pathways, these findings should be interpreted as reflective of real-world practice. Notably, comparable cancellation rates were observed among PATC patients despite clinical and sociodemographic characteristics historically associated with limited access to primary care [4,5,14,15]. Prior studies have demonstrated that surgical cancellations are frequently driven by inadequate preoperative optimization and underlying medical concerns [16,17], with additional risk factors including higher comorbidity burden and socioeconomic disparities [18]. This suggests that centralized, protocol-driven preoperative evaluation may help maintain surgical readiness in populations vulnerable to access-related delays, although the underlying mechanisms cannot be isolated in the current study. PATC patients experienced longer LOS and greater rates of postoperative ED visits than those evaluated by PCP; however, THA and TKA subgroup analyses showed similar cancellation rates and primary outcomes across clearance pathways. Increased LOS and resource utilization have been associated with higher BMI, greater comorbidity burden, and reduced baseline functional status [[19], [20], [21], [22], [23], [24], [25]]. Similarly, increased rates of ED utilization following TJA have been linked to medical complexity, socioeconomic factors, and barriers to outpatient care continuity [[26], [27], [28], [29], [30], [31]]. These findings support the role of PATC as a viable alternative pathway for preoperative clearance, particularly in medically and socially complex populations. This study builds upon prior research evaluating internal preoperative clinics. A previous single-institution study also reported comparable complication and cancellation rates for patients evaluated at an internal clinic, despite presenting greater risk factors, compared to those cleared by a PCP [8]. Our findings extend these results by demonstrating external validity across 2 high-volume community hospitals. Unlike the antecedent study [8], which did not control for variables that could have influenced results, our study adjusted findings for age, BMI, CCI, and procedure type. Prior research shows cancellations are often driven by medical concerns or inadequate optimization [16,17] and risk factors like high BMI, non-White race, men, and elevated CCI [18]. The PATC is specifically designed to assess surgical readiness using standardized, protocol-driven screening, with integrated referral pathways for abnormal findings and coordinated multidisciplinary care. This model facilitates rapid identification of modifiable risks, such as poorly controlled diabetes, anemia, or cardiac concerns, and expedites targeted interventions through embedded specialty services. By consolidating patient counseling, testing, and multidisciplinary specialty consultation into a coordinated visit, PATC may help minimize delays and maintain surgical readiness in high-risk or medically complex populations. PCP-based clearance may introduce additional administrative burden related to external record acquisition, whereas PATC may reduce inefficiencies associated with fragmented documentation, although these administrative impacts were not directly measured in the present study. Our study offers novel insight by assessing endpoints not included in the previous investigation [8], such as postoperative ED visits, LOS, discharge disposition, and baseline PROMs. Although LOS and SNF discharge were higher in the PATC cohort, these findings likely reflect baseline differences in comorbidity burden rather than pathway effects. Prior literature has reported associations between higher BMI [[19], [20], [21]], comorbidities [[22], [23], [24], [25]], lower functional scores, public insurance, racial minority status, and extended LOS with increased postacute care use [[32], [33], [34]]. Similarly, patients with higher comorbidities, lower socioeconomic status, and limited access to a PCP are more likely to return to the ED postoperatively [[26], [27], [28], [29], [30], [31]]. Furthermore, lower preoperative PROM scores have been associated with particularly challenging recovery trajectories and reduced postoperative function [[35], [36], [37], [38], [39]]. The PATC cohort in the present study had lower preoperative PROMs, higher rates of Medicaid coverage, greater racial diversity, and a longer LOS. These factors may account for the higher resource utilization observed in the PATC group in terms of discharge to SNF and ED visits. Internal clinics offer an ideal setting for initiating tailored interventions for higher-risk patients. Notably, regular outpatient follow-up has been shown to be an effective way to reduce ED utilization after orthopaedic surgery [40] and could help to further optimize surgical care. These findings underscore the importance of enhanced discharge planning, targeted patient education, and structured postoperative follow-up to help reduce reliance on postacute care among vulnerable patient populations. Comprehensive cost analyses, including program costs, downstream resource utilization, and administrative burden, are needed to fully assess the economic impact of PATC. Finally, our findings support the role of institutional preoperative clinics like PATC in offering a standardized approach to surgical clearance offered to all prospective TJA patients. It is estimated that 1 in 5 adults faces barriers to primary care, especially individuals with low income, mental health conditions, or those born outside of the United States [14]. In addition, communities with majority Black or African American populations are 67% more likely to experience PCP shortages [15]. In our study, PATC patients were more likely to be Medicaid-insured and identify as Black or African American, suggesting that internal clinics may offer a valuable alternative for surgical clearance in settings where timely PCP access is limited, particularly for populations that have historically faced barriers to primary care. Given projections of increasing THA and TKA procedures by 176% and 139%, respectively, by 2040 [41], expanding PATC access could help promote universal opportunities for patients regardless of PCP availability and potentially provide timely, high-quality surgical optimization as demand increases. Future studies should evaluate whether integrating PCP referral within preoperative clinic pathways improves continuity of care and long-term outcomes beyond the surgical episode. Limitations The authors acknowledge limitations of the current study, beginning with its retrospective design. Availability, proximity, transportation, and patient preference likely contributed to the decision of where patients received their clearance, which could introduce referral bias. A future randomized, controlled trial is needed to confirm our results and remove this limitation. Clearance pathway selection was at the surgeon’s discretion, and higher-volume TKA surgeons may have preferentially used PATC. Because allocation was not randomized, this referral pattern could introduce selection bias and explain the greater proportion of TKA patients in the PATC cohort and THA patients in the PCP cohort. Additionally, cancellation data prior to PATC implementation were inconsistently tracked, precluding pre- and post-implementation comparison. Consequently, baseline cancellation rates and the extent to which limited PCP access contributed to cancellations prior to PATC implementation could not be assessed, limiting our ability to isolate access-related benefits of the program. Evaluation of surgery postponement occurring earlier than 1 week prior to the scheduled date due to newly identified or inadequately optimized medical conditions was beyond the scope of the present analysis and warrants future investigation. While this study assessed patients from 2 community hospitals under a unified institution, it remains part of a single regional health system. Thus, our results may have limited generalizability across different healthcare systems with varying resources. Finally, we did not have data on access to transportation or patient health literacy, which may have been potential confounders. Future studies on this topic should consider these factors when examining cancellation rates. Conclusions These findings support the use of institutional preoperative clinics as a viable alternative to traditional PCP-based clearance pathways prior to TJA. Despite managing a higher-risk population, PATC evaluation yielded comparable complication and cancellation rates to traditional PCP-based clearance. This suggests that structured, protocol-driven optimization may help maintain surgical readiness even among patients with greater comorbidity burden. Future studies should prospectively evaluate clinical outcomes, functional recovery, cost-effectiveness, provider workflow impact, and patient satisfaction across preoperative pathways, while also assessing how internal clinics can further reduce disparities and enhance perioperative care. Conflicts of interest J. Bernstein is a paid employee for Johnson & Johnson MedTech and is a board member/committee appointments for AAHKS, AAOS, and EOA; all other authors declare no potential conflicts of interest. For full disclosure statements refer to https://doi.org/10.1016/j.artd.2026.102145. CRediT authorship contribution statement Jake Laverdiere: Writing – review & editing, Writing – original draft, Data curation. Swaroopa Vaidya: Writing – review & editing, Validation, Data curation. Gregory Panza: Writing – review & editing, Validation, Software, Methodology, Formal analysis, Data curation. Thomas Giannasca: Writing – review & editing, Writing – original draft, Data curation. Jenna Bernstein: Writing – review & editing, Supervision, Methodology, Investigation, Conceptualization.