Catherine H. MacLean

Catherine H. MacLean

Auto-generated profile, based on publicly available information

Catherine H. MacLean

Hospital for Special Surgery, New York, New York

Research Focus & Metrics

Academic Career Length

22Years (2005-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
10 May 2018 in New England Journal of Medicine
Verified
Journal Article
Closed access
PDF file
CMCatherine H. MacLean
Catherine H. MacLean
EKEve A. Kerr
Eve A. Kerr
AQAmir Qaseem
Amir Qaseem
Performance measurement in the U.S. health care system has expanded dramatically over the past 30 years. The National Quality Measures Clearinghouse now lists more than 2500 performance measures. These measures are used in various quality-reporting, accountability, and payment programs sponsored by commercial payers, government agencies, and independent quality-assessment organizations. The Centers for Medicare and Medicaid Services (CMS) aims to base 90% of Medicare fee-for-service payments to clinicians on “value” by the end of 2018 by using performance scores.

1 November 2021 in Journal of Bone and Joint Surgery
Verified
Journal Article
Closed access
KKKyle N. Kunze
Kyle N. Kunze
MFMark A. Fontana
Mark A. Fontana
CMCatherine H. MacLean
Catherine H. MacLean
SLStephen Lyman
Stephen Lyman
AMAlexander S. McLawhorn
Alexander S. McLawhorn
Background: It is essential to quantify an acceptable outcome after total joint arthroplasty (TJA) in order to understand quality of care. The purpose of this study was to define patient acceptable symptom state (PASS) thresholds for the Knee injury and Osteoarthritis Outcome Score, Joint Replacement (KOOS JR) and the Hip disability and Osteoarthritis Outcome Score, Joint Replacement (HOOS JR) after TJA. Methods: A receiver operating characteristic (ROC) curve analysis, leveraging 2-year satisfaction of “moderate improvement” or better as the anchor, was used to establish PASS thresholds among 5,216 patients who underwent primary total hip arthroplasty and 4,036 who underwent primary total knee arthroplasty from 2007 to 2012 with use of an institutional registry. Changes in PASS thresholds were explored by stratifying and recalculating these thresholds by age at the time of surgery (<70 or ≥70 years of age), sex (men or women), body mass index (BMI; <30 or ≥30 kg/m2), and baseline Short Form-36 (SF-36) physical and mental component scores (<50 or ≥50). Results: The HOOS JR PASS threshold was 76.7 (area under the ROC curve [AUC] = 0.91), which was achieved by 4,334 patients (83.1%). The KOOS JR PASS threshold was 63.7 (AUC = 0.89), which was achieved by 3,461 patients (85.8%). Covariate stratification demonstrated that PASS thresholds were higher in men compared with women, and in those with higher preoperative SF-36 physical and mental scores (≥50) compared with lower SF-36 scores (<50). Results differed between instruments for BMI and age: higher BMI was associated with a lower PASS threshold for the HOOS JR but a higher PASS threshold for the KOOS JR. The HOOS JR PASS threshold was higher in patients who were <70 years of age compared with those who were ≥70 years of age, but was equivalent for the KOOS JR. Conclusions: The PASS thresholds for the HOOS JR and KOOS JR at 2 years after TJA were 76.7 and 63.7, respectively. The PASS thresholds were associated with certain preoperative covariates, suggesting that an acceptable symptom state after TJA is influenced by patient-specific factors. Level of Evidence: Prognostic Level IV. See Instructions for Authors for a complete description of levels of evidence.

12 February 2025 in Clinical Orthopaedics and Related Research
Verified
Journal Article
Open access
ZTZachary A. Bs Trotzky
Zachary A. Bs Trotzky
RSRyan G. Ba Smolarsky
Ryan G. Ba Smolarsky
SMSophia J. Ba Madjarova
Sophia J. Ba Madjarova
OJOlivia M. Jochl
Olivia M. Jochl
BRBenjamin F. Ricciardi
Benjamin F. Ricciardi
ESErnest L. Sink
Ernest L. Sink
Background: The utility of patient-reported outcome measures (PROMs) has been well established, but their interpretation relies on population-specific definitions of meaningful improvement. As such, the minimum clinically important difference (MCID), substantial clinical benefit (SCB), and patient-acceptable symptom state (PASS) thresholds have become prominent metrics in the orthopaedic evidence to ascribe clinical relevance to numeric PROM scores. Studies assessing outcomes of periacetabular osteotomy (PAO) relative to the MCID and PASS have previously evaluated patients against thresholds defined for hip arthroscopy for the treatment of femoroacetabular impingement or distribution-based MCID calculations. These scores may not accurately reflect the status or expectations of patients with a different symptom profile undergoing open hip preservation surgery. Questions/purposes: For patients treated with PAO, we sought to (1) define the MCID, SCB, and PASS threshold values for the mHHS (modified Harris hip score) and International Hip Outcome Tool 12 (iHOT-12) using anchor-based methods; (2) assess the validity of MCID and SCB estimates against minimal detectable change (MDC) values; and (3) determine the proportion of patients who achieved a clinically meaningful threshold. Methods: Between February 2011 and May 2023, a total of 690 patients underwent PAO for symptomatic acetabular dysplasia at one institution and were included in a longitudinally maintained hip preservation registry. The cohort used to define and validate MCID, SCB, and PASS threshold values consisted of those with a completed postoperative anchor questionnaire, which yielded 456 patients as potentially eligible. An additional 139 patients were excluded because of missing mHHS or iHOT-12 scores during the eligibility window (1 to 2 years postoperatively), leaving 70% (317 of 456) of patients to define and validate MCID, SCB, and PASS at a mean ± SD of 1.0 ± 0.3 years of follow-up. A minimum 1-year follow-up was chosen to reduce recall bias. The cohort for defining MCID, SCB, and PASS (94% [298 of 317] women, mean ± SD age at time of surgery 27 ± 8 years) included 21% (68 of 317) of patients with prior ipsilateral surgery. From those registry patients without complete anchor questionnaires, 37% (137 of 373) were identified with pre- and postoperative PROM scores at a mean ± SD of 1.0 ± 0.9 years of follow-up to form the sample for assessing the proportion of patients achieving a clinically meaningful threshold. The MCID, SCB, and PASS thresholds for the mHHS and iHOT-12 were calculated through an anchor-based approach, using area under the receiver operating curve to determine cut points that best identified positive responses, according to quality of life–based anchor questions. The MDC was calculated with confidence intervals (CIs) reflecting 80%, 90%, and 95% certainty to determine the smallest change in the PROM scores that can be considered above the level of measurement error. The validity of MCID estimates was assessed by confirming that they exceeded corresponding MDC values. The validity of SCB estimates were assessed by confirming that they exceeded corresponding MCID values. The proportion of patients achieving a clinically meaningful threshold was determined by calculating the percentage of patients who met the defined anchor-based scores. Results: The MCID, SCB, and PASS thresholds for the mHHS were 18, 23, and 71, respectively. The MCID, SCB, and PASS thresholds for the iHOT-12 were 26, 42, and 65, respectively. The MDC ranged from 8 to 12 for the mHHS and 10 to 16 for the iHOT-12. The MCID values for the mHHS and iHOT-12 exceeded corresponding values of the MDC at all CIs. The SCB thresholds exceeded all corresponding MDC and MCID values. Across the mHHS and iHOT-12, the proportion of patients who achieved an MCID at the first time point ranged from 60% to 73%, the proportion of patients who achieved the SCB ranged from 49% to 56%, and the proportion of patients who achieved the PASS threshold ranged from 55% to 79%. Among the cohort for defining MCID, SCB, and PASS, the proportion of patients achieving any MCID, SCB, or PASS was 79%, 66%, and 81%, respectively. Among the sample for assessing the proportion of patients achieving a clinically meaningful threshold, the proportion achieving any MCID, SCB, or PASS threshold was 74%, 58%, and 72%, respectively. Conclusion: We found that using a sample of patients undergoing PAO, the anchor-based values for the MCID and SCB were generally larger than previous distribution- and anchor-based scores that have been defined for hip preservation, whereas PASS threshold scores were similar. All MCID and SCB thresholds exceeded corresponding MDC values, confirming these scores to be valid estimates. These metrics provide more rigorous, procedure-specific definitions for the evaluation of treatment success and failure after PAO. As anchor-based metrics are defined based on patients’ perceptions, they should be used preferentially for postoperative assessment over distribution-based scores. Level of Evidence: Level III, therapeutic study.

31 May 2023 in Clinical Orthopaedics and Related Research
Verified
Journal Article
Closed access
KKKyle N. Kunze
Kyle N. Kunze
MSMiranda M. Mbs So
Miranda M. Mbs So
DPDouglas E. Padgett
Douglas E. Padgett
SLStephen Lyman
Stephen Lyman
CMCatherine H. MacLean
Catherine H. MacLean
MFMark Alan Fontana
Mark Alan Fontana
Background: Unplanned hospital readmissions after total joint arthroplasty (TJA) represent potentially serious adverse events and remain a critical measure of hospital quality. Predicting the risk of readmission after TJA may provide patients and clinicians with valuable information for preoperative decision-making. Questions/purposes: (1) Can nonlinear machine-learning models integrating preoperatively available patient, surgeon, hospital, and county-level information predict 30-day unplanned hospital readmissions in a large cohort of nationwide Medicare beneficiaries undergoing TJA? (2) Which predictors are the most important in predicting 30-day unplanned hospital readmissions? (3) What specific information regarding population-level associations can we obtain from interpreting partial dependency plots (plots describing, given our modeling choice, the potentially nonlinear shape of associations between predictors and readmissions) of the most important predictors of 30-day readmission? Methods: National Medicare claims data (chosen because this database represents a large proportion of patients undergoing TJA annually) were analyzed for patients undergoing inpatient TJA between October 2016 and September 2018. A total of 679,041 TJAs (239,391 THAs [61.3% women, 91.9% White, 52.6% between 70 and 79 years old] and 439,650 TKAs [63.3% women, 90% White, 55.2% between 70 and 79 years old]) were included. Model features included demographics, county-level social determinants of health, prior-year (365-day) hospital and surgeon TJA procedure volumes, and clinical classification software-refined diagnosis and procedure categories summarizing each patient’s Medicare claims 365 days before TJA. Machine-learning models, namely generalized additive models with pairwise interactions (prediction models consisting of both univariate predictions and pairwise interaction terms that allow for nonlinear effects), were trained and evaluated for predictive performance using area under the receiver operating characteristic (AUROC; 1.0 = perfect discrimination, 0.5 = no better than random chance) and precision-recall curves (AUPRC; equivalent to the average positive predictive value, which does not give credit for guessing “no readmission” when this is true most of the time, interpretable relative to the base rate of readmissions) on two holdout samples. All admissions (except the last 2 months’ worth) were collected and split randomly 80%/20%. The training cohort was formed with the random 80% sample, which was downsampled (so it included all readmissions and a random, equal number of nonreadmissions). The random 20% sample served as the first test cohort (“random holdout”). The last 2 months of admissions (originally held aside) served as the second test cohort (“2-month holdout”). Finally, feature importances (the degree to which each variable contributed to the predictions) and partial dependency plots were investigated to answer the second and third research questions. Results: For the random holdout sample, model performance values in terms of AUROC and AUPRC were 0.65 and 0.087, respectively, for THA and 0.66 and 0.077, respectively, for TKA. For the 2-month holdout sample, these numbers were 0.66 and 0.087 and 0.65 and 0.075. Thus, our nonlinear models incorporating a wide variety of preoperative features from Medicare claims data could not well-predict the individual likelihood of readmissions (that is, the models performed poorly and are not appropriate for clinical use). The most predictive features (in terms of mean absolute scores) and their partial dependency graphs still confer information about population-level associations with increased risk of readmission, namely with older patient age, low prior 365-day surgeon and hospital TJA procedure volumes, being a man, patient history of cardiac diagnoses and lack of oncologic diagnoses, and higher county-level rates of hospitalizations for ambulatory-care sensitive conditions. Further inspection of partial dependency plots revealed nonlinear population-level associations specifically for surgeon and hospital procedure volumes. The readmission risk for THA and TKA decreased as surgeons performed more procedures in the prior 365 days, up to approximately 75 TJAs (odds ratio [OR] = 1.2 for TKA and 1.3 for THA), but no further risk reduction was observed for higher annual surgeon procedure volumes. For THA, the readmission risk decreased as hospitals performed more procedures, up to approximately 600 TJAs (OR = 1.2), but no further risk reduction was observed for higher annual hospital procedure volumes. Conclusion: A large dataset of Medicare claims and machine learning were inadequate to provide a clinically useful individual prediction model for 30-day unplanned readmissions after TKA or THA, suggesting that other factors that are not routinely collected in claims databases are needed for predicting readmissions. Nonlinear population-level associations between low surgeon and hospital procedure volumes and increased readmission risk were identified, including specific volume thresholds above which the readmission risk no longer decreases, which may still be indirectly clinically useful in guiding policy as well as patient decision-making when selecting a hospital or surgeon for treatment. Level of Evidence: Level III, therapeutic study.

2 June 2023 in Journal of the American Academy of Orthopaedic Surgeons
Verified
Journal Article
Open access
CMCatherine H. MacLean
Catherine H. MacLean
VAVinicius C. Antao
Vinicius C. Antao
ACAmy S. Chin
Amy S. Chin
AMAlexander S. McLawhorn
Alexander S. McLawhorn
The intersection of big data and artificial intelligence (AI) has resulted in advances in numerous areas, including machine learning, computer vision, and natural language processing. Although there are many potentially transformative applications of AI in health care, including precision medicine, this industry has been slow to adopt these technologies. At the same time, the operations of health care have historically been system-directed and physician-directed rather than patient-centered. The application of AI to patient-reported outcome measures (PROMs), which provide insight into patient-centered health outcomes, could steer research and healthcare delivery toward decisions that optimize outcomes important to patients. Historically, PROMs have only been collected within research registries. However, the increasing availability of PROMs within electronic health records has led to their inclusion in big data ecosystems, where they can inform or be informed by other data elements. The use of big data to analyze PROMs can help establish norms, evaluate data distribution, and determine proportions of patients achieving change or threshold standards. This information can be used for benchmarking, risk adjustment, predictive modeling, and ultimately improving the health of individuals and populations.

19 March 2019 in JBJS Open Access
Verified
Journal Article
Open access
MFMark Alan Fontana
Mark Alan Fontana
SLStephen Lyman
Stephen Lyman
WIWasif Islam
Wasif Islam
CMCatherine H. MacLean
Catherine H. MacLean
Background: Volume-outcome relationships are well established for coronary artery bypass grafting and total joint arthroplasty surgery. Although the U.S. Centers for Medicare & Medicaid Services (CMS) Overall Hospital Quality Star Ratings program includes outcome quality measures for these procedures, these outcome quality measures are not counted toward the star ratings for low-volume hospitals. We sought to assess whether excluding low-volume hospitals from surgical quality measures with known volume-outcome relationships affects the star ratings. Methods: We identified quality measures used in CMS’s star ratings that are related to surgical procedures with a known volume-outcome relationship and tested for the presence of the volume-outcome association for each of these measures. We then imputed missing values for low-volume hospitals for each measure and otherwise identically repeated the CMS calculations in order to assess the percentages of hospitals with the same, better, or worse ratings. Results: Among the measures used to calculate star ratings, we identified 4 quality measures (2 related to coronary artery bypass grafting and 2 related to total joint arthroplasty) with known volume-outcome relationships that were excluded from the calculations of the star ratings for low-volume hospitals. We confirmed a volume-outcome association in the CMS data for all 4 measures. When total joint arthroplasty complications were imputed for low-volume hospitals and then included in the calculation of the star ratings, over one-third of hospitals received a different rating; both low-volume and other hospitals were more often hurt than helped. Imputing the other 3 quality measures among low-volume hospitals left the ratings unchanged. Conclusions: The CMS star ratings do not fully represent the risks of undergoing procedures at low-volume hospitals, potentially misrepresent quality across facilities, and hence are of uncertain utility to consumers.