1
|
Arora A, Lituiev D, Jain D, Hadley D, Butte AJ, Berven S, Peterson TA. Predictive Models for Length of Stay and Discharge Disposition in Elective Spine Surgery: Development, Validation, and Comparison to the ACS NSQIP Risk Calculator. Spine (Phila Pa 1976) 2023; 48:E1-E13. [PMID: 36398784 PMCID: PMC9772082 DOI: 10.1097/brs.0000000000004490] [Citation(s) in RCA: 5] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/06/2022] [Accepted: 09/12/2022] [Indexed: 11/19/2022]
Abstract
STUDY DESIGN A retrospective study at a single academic institution. OBJECTIVE The purpose of this study is to utilize machine learning to predict hospital length of stay (LOS) and discharge disposition following adult elective spine surgery, and to compare performance metrics of machine learning models to the American College of Surgeon's National Surgical Quality Improvement Program's (ACS NSQIP) prediction calculator. SUMMARY OF BACKGROUND DATA A total of 3678 adult patients undergoing elective spine surgery between 2014 and 2019, acquired from the electronic health record. METHODS Patients were divided into three stratified cohorts: cervical degenerative, lumbar degenerative, and adult spinal deformity groups. Predictive variables included demographics, body mass index, surgical region, surgical invasiveness, surgical approach, and comorbidities. Regression, classification trees, and least absolute shrinkage and selection operator (LASSO) were used to build predictive models. Validation of the models was conducted on 16% of patients (N=587), using area under the receiver operator curve (AUROC), sensitivity, specificity, and correlation. Patient data were manually entered into the ACS NSQIP online risk calculator to compare performance. Outcome variables were discharge disposition (home vs. rehabilitation) and LOS (days). RESULTS Of 3678 patients analyzed, 51.4% were male (n=1890) and 48.6% were female (n=1788). The average LOS was 3.66 days. In all, 78% were discharged home and 22% discharged to rehabilitation. Compared with NSQIP (Pearson R2 =0.16), the predictions of poisson regression ( R2 =0.29) and LASSO ( R2 =0.29) models were significantly more correlated with observed LOS ( P =0.025 and 0.004, respectively). Of the models generated to predict discharge location, logistic regression yielded an AUROC of 0.79, which was statistically equivalent to the AUROC of 0.75 for NSQIP ( P =0.135). CONCLUSION The predictive models developed in this study can enable accurate preoperative estimation of LOS and risk of rehabilitation discharge for adult patients undergoing elective spine surgery. The demonstrated models exhibited better performance than NSQIP for prediction of LOS and equivalent performance to NSQIP for prediction of discharge location.
Collapse
Affiliation(s)
- Ayush Arora
- Department of Orthopaedic Surgery, University of California, San Francisco, San Francisco, CA, USA
| | - Dmytro Lituiev
- Bakar Computational Health Sciences Institute, University of California, San Francisco, San Francisco, CA, USA
| | - Deeptee Jain
- Department of Orthopaedic Surgery, Washington University in St. Louis, St. Louis, MO
| | - Dexter Hadley
- Department of Pathology, University of Central Florida, FL, USA
| | - Atul J. Butte
- Bakar Computational Health Sciences Institute, University of California, San Francisco, San Francisco, CA, USA
- Department of Pediatrics, University of California, San Francisco, San Francisco, CA, USA
- Center for Data-driven Insights and Innovation, University of California Health, Oakland, USA
| | - Sigurd Berven
- Department of Orthopaedic Surgery, University of California, San Francisco, San Francisco, CA, USA
| | - Thomas A. Peterson
- Department of Orthopaedic Surgery, University of California, San Francisco, San Francisco, CA, USA
- Bakar Computational Health Sciences Institute, University of California, San Francisco, San Francisco, CA, USA
| |
Collapse
|
2
|
Marinone Lares SG, Clark S, Mathy JA, Chaplin J, McIvor N. Evaluation of a novel database for quality assurance at a head and neck service in New Zealand: an audit of free flap head and neck reconstruction. ANZ J Surg 2020; 90:1386-1390. [PMID: 32436238 DOI: 10.1111/ans.15974] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/01/2020] [Revised: 04/17/2020] [Accepted: 04/25/2020] [Indexed: 11/28/2022]
Abstract
BACKGROUND Clinical audit is a critical quality improvement exercise, yet efficient audit tools are lacking. The main objective of this study was to evaluate a recently deployed database in facilitating the process of clinical audit, and the secondary objective was to evaluate the outcomes of free flap reconstruction of the head and neck at our centre. METHODS A head and neck cancer-specific database was customized to suit the needs of our head and neck multidisciplinary team. Data has been entered prospectively into this database since March of 2018. An audit of free flap reconstruction of the head and neck over a 12-month period was performed using the database and analysed as a case study to examine its efficacy as a clinical audit tool. Additionally, the outcomes of free flap reconstruction at our centre were compared to those reported in the international literature. RESULTS The database allows flexible and specific queries, analysis and export of data, and can provide immediate results. However, issues with data quality and completeness were identified. In this audit, the overall 30-day complication rate and 30-day mortality in patients undergoing free flap reconstruction of the head and neck were 58% and 3%, respectively. CONCLUSION The database is fit for its intended purpose as an audit tool. Outcomes of free flap reconstruction of the head and neck at our centre are comparable to those of institutions overseas.
Collapse
Affiliation(s)
| | - Sita Clark
- Department of Surgery, The University of Auckland, Auckland, New Zealand
| | - Jon A Mathy
- Department of Surgery, The University of Auckland, Auckland, New Zealand.,Department of Otolaryngology-Head and Neck Surgery, Auckland City Hospital, Auckland, New Zealand.,Auckland Regional Plastic Surgery Unit, Auckland, New Zealand
| | - John Chaplin
- Department of Otolaryngology-Head and Neck Surgery, Auckland City Hospital, Auckland, New Zealand
| | - Nick McIvor
- Department of Otolaryngology-Head and Neck Surgery, Auckland City Hospital, Auckland, New Zealand
| |
Collapse
|
3
|
Hanick A, Meleca JB, Fritz MA. Early discharge after free-tissue transfer does not increase adverse events. Am J Otolaryngol 2020; 41:102374. [PMID: 31883753 DOI: 10.1016/j.amjoto.2019.102374] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/01/2019] [Accepted: 12/08/2019] [Indexed: 11/17/2022]
Abstract
INTRODUCTION Demonstrate that carefully selected free flap patients may be discharged early after surgery without increasing the rates of postoperative complications or readmissions. METHODS Based on a published article in Laryngoscope 2016 of 51 free-tissue transfers, a retrospective chart review was performed on an expanded cohort who underwent free-tissue transfer for head and neck reconstruction between February 2010 and May 2018 and discharged by postoperative day 3. RESULTS 101 patients who underwent 104 free flaps with average age of 56 (3-84) years old were reviewed. Free flap indications included orbital and maxillary defects (n = 22), palatal defects (n = 16), nasal and septal defects (n = 16), cranioplasty and scalp defects (n = 16), mandibular defects due to osteoradionecrosis (n = 14), facial contouring and parotid defects (n = 12), and complex postsurgical and radiotherapy wounds or fistula closure (n = 8). Free flaps performed were anterolateral thigh (n = 97), radial forearm (n = 2), serratus (n = 2), latissimus (n = 1), fibula (n = 1) and supraclavicular (n = 1). The recipient vessels used via minimal access approaches were facial (n = 43), superficial temporal (n = 29), angular (n = 20) and others. There were 3 flap failures (2.9%) recognized in follow-up. No flap failures or perioperative complications were associated with early discharge. There were only 2 patients readmitted and 1 watched in observation within 30 days postoperatively. CONCLUSION An updated review of our institutional experience with more than double the cohort size substantiates previous conclusions that early discharge after free-tissue transfer is a safe option in select patients. Moreover, earlier discharge is a critical management choice that reduces cost and decreases hospital-related adverse events.
Collapse
Affiliation(s)
- Andrea Hanick
- Cleveland Clinic, Head and Neck Institute, Cleveland, OH, USA
| | - Joseph B Meleca
- Cleveland Clinic, Head and Neck Institute, Cleveland, OH, USA.
| | - Michael A Fritz
- Cleveland Clinic, Head and Neck Institute, Cleveland, OH, USA
| |
Collapse
|
4
|
Tierney W, Shah J, Clancy K, Lee MY, Ciolek PJ, Fritz MA, Lamarre ED. Predictive value of the ACS NSQIP calculator for head and neck reconstruction free tissue transfer. Laryngoscope 2019; 130:679-684. [PMID: 31361334 DOI: 10.1002/lary.28195] [Citation(s) in RCA: 6] [Impact Index Per Article: 1.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/30/2019] [Revised: 06/15/2019] [Accepted: 07/05/2019] [Indexed: 11/07/2022]
Abstract
BACKGROUND Predictive models to forecast the likelihood of specific outcomes after surgical intervention allow informed shared decision-making by surgeons and patients. Previous studies have suggested that existing general surgical risk calculators poorly forecast head and neck surgical outcomes. However, no large study has addressed this question while subdividing subjects by surgery performed. OBJECTIVES To determine the accuracy of the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) Surgical Risk Calculator in estimating length of hospital stay and risk of postoperative complications after free tissue transfer surgery. STUDY DESIGN A retrospective chart review of patients at one institution was performed using Current Procedural Terminology codes for anterolateral thigh (ALT) flap, fibula free flap (FFF), and radial forearm free flap (RFFF) reconstruction. Output data from the ACS NSQIP surgical risk calculator were compared with the observed rates in our patients. METHODS Incidences of cardiac complications, pneumonia, venous thromboembolism, return to the operating room, and discharge to skilled nursing facility (SNF) were compared to predicted incidences. Length of stay was also compared to the predicted length of stay. RESULTS Three hundred thirty-six free flap reconstructions with 197 ALT flaps, 85 RFFFs, and 54 FFFFs were included. Brier scores were calculated using ACS NSQIP forecast and actual incidences. No Brier score was <0.01 for the entire sample or any subgroup, which indicates that the NSQIP risk calculator does not accurately forecast outcomes after free tissue reconstruction. CONCLUSION The ACS NSQIP failed to accurately forecast postoperative outcomes after head and neck free flap reconstruction for the entire sample or subgroup analyses. LEVEL OF EVIDENCE 4 Laryngoscope, 130:679-684, 2020.
Collapse
Affiliation(s)
- William Tierney
- Cleveland Clinic, Head and Neck Institute, Cleveland, Ohio, U.S.A.,Cleveland Clinic Lerner College of Medicine of Case Western Reserve University, Cleveland, Ohio, U.S.A
| | - Janki Shah
- Cleveland Clinic, Head and Neck Institute, Cleveland, Ohio, U.S.A
| | - Kate Clancy
- Department of Otolaryngology-Head and Neck Surgery, Case Western Reserve University, Cleveland, Ohio, U.S.A
| | - Maxwell Y Lee
- Cleveland Clinic Lerner College of Medicine of Case Western Reserve University, Cleveland, Ohio, U.S.A
| | - Peter J Ciolek
- Cleveland Clinic, Head and Neck Institute, Cleveland, Ohio, U.S.A
| | - Michael A Fritz
- Cleveland Clinic, Head and Neck Institute, Cleveland, Ohio, U.S.A
| | - Eric D Lamarre
- Cleveland Clinic, Head and Neck Institute, Cleveland, Ohio, U.S.A.,Cleveland Clinic Lerner College of Medicine of Case Western Reserve University, Cleveland, Ohio, U.S.A
| |
Collapse
|
5
|
Gupta A, Sonis ST, Schneider EB, Villa A. Impact of the insurance type of head and neck cancer patients on their hospitalization utilization patterns. Cancer 2017; 124:760-768. [PMID: 29112234 DOI: 10.1002/cncr.31095] [Citation(s) in RCA: 22] [Impact Index Per Article: 3.1] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/06/2017] [Revised: 09/14/2017] [Accepted: 10/02/2017] [Indexed: 12/29/2022]
Abstract
BACKGROUND Head and neck cancer (HNC) patients with Medicaid, Medicare, or no insurance show poor outcomes in comparison with privately insured patients. It was hypothesized that nonprivate insurance coverage biases the selection of the treatment site to favor hospitals that are not associated with optimum treatment outcomes. This study assessed the relation between the insurance type of HNC patients and the hospital type for inpatient care. METHODS Adult HNC patients were identified from the Nationwide Inpatient Sample (2012 and 2013). The primary exposure was the insurance provider type. The outcome was the hospital type, which was classified by the hospital's ownership and its location and teaching status. Multivariate multinomial logistic regression models were constructed to control for the patient's age, sex, race, income, mortality risk, and geographic location. The analysis was weighted and was adjusted for multiple comparisons. RESULTS In all, 37,466 HNC patients representing 187,330 patients nationally were identified. After adjustments for age, sex, race, income, and mortality risk, in comparison with privately insured patients, Medicaid, Medicare, and uninsured patients demonstrated 1.14 to 2.29 increased odds of undergoing treatment at rural, urban nonteaching, private investor-owned, or government (nonfederal) hospitals (P < .05). This trend remained apparent even after adjustments for the geographic location. CONCLUSIONS Uninsured patients or patients insured by government programs predominantly underwent care for HNC at hospital types most often associated with inferior survival outcomes. This finding could explain some proportion of insurance-related disparities in HNC outcomes. Further studies are warranted to determine whether interventions to promote equitable access to optimal hospital settings for patients, regardless of their insurance type, might improve outcomes among nonprivate insurance holders. Cancer 2018;124:760-8. © 2017 American Cancer Society.
Collapse
Affiliation(s)
- Avni Gupta
- Center for Surgery and Public Health, Brigham and Women's Hospital, Boston, Massachusetts
| | - Stephen T Sonis
- Division of Oral Medicine and Dentistry, Brigham and Women's Hospital, Boston, Massachusetts.,Dana-Farber Cancer Institute, Boston, Massachusetts.,Department of Oral Medicine, Harvard School of Dental Medicine, Boston, Massachusetts
| | | | - Alessandro Villa
- Division of Oral Medicine and Dentistry, Brigham and Women's Hospital, Boston, Massachusetts.,Dana-Farber Cancer Institute, Boston, Massachusetts.,Department of Oral Medicine, Harvard School of Dental Medicine, Boston, Massachusetts
| |
Collapse
|
6
|
Krouse JH. Presenting OTO Open. OTO Open 2017; 1:2473974X16685560. [PMID: 30480170 PMCID: PMC6239047 DOI: 10.1177/2473974x16685560] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/24/2022] Open
Affiliation(s)
- John H. Krouse
- Department of Otolaryngology/Head and
Neck Surgery, Lewis Katz School of Medicine at Temple University, Philadelphia,
Pennsylvania, USA
| |
Collapse
|