COI/Disclosures Predictive Analytics: Making Adult Chris Ames, MD - PDF document

COI/Disclosures Predictive Analytics: Making Adult  Chris Ames, MD has financial interests to disclose. Spinal Deformity Surgery Sustainable  Royalty: Biomet Zimmer, Stryker, Depuy Synthes, Christopher P Ames MD K2M, Next Spine, Medicrea, Astura Professor of Neurosurgery and Orthopaedic Surgery  Consulting: Medtronic, Biomet Zimmer, Depuy Synthes, K2M, Medicrea Director of Spinal Deformity and Spinal Tumor Surgery  Research: Titan Spine, Depuy Synthes ISSG  Editorial Board: Operative Neurosurgery University of California San Francisco  Grant Funding: SRS Benzel AANS 2019  Executive Committee: ISSG How much more can spine How much has implant surgeon technical performance innovation changed complication improve? rates and improved outcomes since the first multiaxial screw was designed?  Will the next generation be more technically facile than Ed Benzel or Volker Sonntag?  Bounds of human technical performance can be predicted using data analytics  Filippo Radicci (Indiana) predicted exactly the new 100m record of Usain Bolt at 9.63 s and using analytics predicts the ultimate bound of human performance is 8.28 s

Why is disruptive technology Economic Burden of Aging needed now? Musculoskeletal System  53 million people over age 65 now and  Total Health care cost increasing 3.5 trillion 2017  80 million over 65 by 2050  Musculoskeletal disease cost >800  60% prevalence of spinal deformity (cobb billion/year greater than 10 degrees)  Spinal  32 million people with ASD in US Deformity $80 billion (2011) Number of USA ASD Procedures increased by 157% in 10 years Number of discharges with at least one diagnosis of spinal curvature' (ICD ‐ 9 code 737.0 to 737.9) 250,000 200,000 150,000 Children Adult 100,000 50,000 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Healthcare Costs and Utilization Project (HCUP http://hcupnet.ahrq.gov),

Complexity Increasing Do not go gentle … Utilization of wedge osteotomies # Wedge Osteotomies Wedge Osteotomies by age group (77.29 ICD ‐ 9 ‐ CM) 100% 90% 800 80% 700 70% 60% 600 >65 50% 45 ‐ 64 500 40% 18 ‐ 44 400 30% 20% 300 10% 200 0% 2003 2004 2005 2006 2007 2008 2009 2010 2003 2004 2005 2006 2007 2008 2009 2010 Increases on 275% in less than 10 years Increase proportion of patients >65yo ~250 procedures in 2003 ~20% in 2003 ~700 procedures in 2012 ~40% in 2012 Modern expectations of high function in old age Surgery improves disability Disease State PCS; MCS; mean mean NBS NBS points points US Total 50 49.9 Population Spine J 2014 US Healthy 55.4 52.9 Population ASD 40.9 49.4 Back Pain 45.7 47.6 Cancer 40.9 47.6 Depression 45.4 36.3 Diabetes 41.1 47.8 Heart Disease 38.9 48.3 Hypertension 44.0 49.7 Limited Use 39.0 43.0 Arms Legs Lung Disease 38.3 45.6

Failure Prevention  Double pelvis  Double rods  VCR rod  BMP-2  Ligament repair  Vertebroplasty  2 surgeons  Plastic Surgery Failures destroy cost effectiveness  Eliminates provider variability  Appropriateness criteria for all surgeons  Transparency  Multidisciplinary  Best practices  3 fold improvement in the worst complications  12 fold decrease in return to surgery in the first three months postop

Reduce Complications by Collective Intelligence Limiting Care  The 56 person group  Of course we decrease complications by average better than any operating on more robust patients individual and came  But, patients who experience major within 3% of total complications still do well  Only 1 individual  Most disabled patients with high frailty scores “guessed” better improve the most  Eliminates Approved: Low risk outliers High Risk but Good Outcome Older had Greater improvement Big Data-Datify the Patient after PSO in general health “Painting true picture of patient with many data points”

FICO Score….Preop Risk Score? Frailty is a Predictive ROS Results 1) Bladder incontinence ☐ Yes ☐ No 20) Would you say your current health is: 31) Travel more than 1 hour 2) Bowel incontinence ☐ Excellent or Good ☐ Moderate/Little/No difficulty ☐ Yes ☐ Fair or Poor ☐ Extreme difficulty/Require assistance or assistive device/Unable t ☐ No Major Complication Incidence How much difficulty do you have with each of the following activities: 32) Perform all personal care 3) Leg weakness ☐ Moderate/Little/No difficulty ☐ Yes ☐ Extreme difficulty/Require assistance or assistive device/Unable t 70% 21) Climbing 1 flight of stairs ☐ No ☐ Moderate/Little/No difficulty ☐ Extreme difficulty/Require assistance or assistive device/Unable to do How often in the last month have you experienced the following: 4) Loss of Balance ☐ Yes 33) Feeling downhearted and depressed 22) Driving a car 60% ☐ No ☐ Moderate/Little/No difficulty ☐ All or most of the time ☐ Some, little or none of the time ☐ Extreme difficulty/Require assistance or assistive device/Unable to do 5) Do you currently smoke? ☐ Yes 23) Getting dressed 34) Feeling so down in the dumps you cannot cheer up no matter what you ☐ No ☐ Moderate/Little/No difficulty ☐ All or most of the time 50% ☐ Extreme difficulty/Require assistance or assistive device/Unable to do ☐ Some, little or none of the time 6) Are you currently on disability? ☐ Yes 24) Getting in and out of bed 35) Feeling tired/exhausted ☐ No ☐ All or most of the time ☐ Moderate/Little/No difficulty ☐ Some, little or none of the time 40% ☐ Extreme difficulty/Require assistance or assistive device/Unable to do 7) Current height and weight (BMI) ☐ <18.5 36) Feeling worn out/used up 25) Walking 100 yards ☐ 18.5-30 ☐ Moderate/Little/No difficulty ☐ All or most of the time ☐ >30 ☐ Extreme difficulty/Require assistance or assistive device/Unable to do ☐ Some, little or none of the time 62% 30% 8-18) Medical History (check all that 26) Get around the house without an assistive device 37) Difficulty remembering things you used to have no trouble with apply): ☐ All or most of the time ☐ Moderate/Little/No difficulty ☐ Cancer ☐ Some, little or none of the time ☐ Extreme difficulty/Require assistance or assistive device/Unable to do 44% ☐ Heart Disease 20% ☐ Diabetes 38) Feeling like your thinking is slow or clouded 27) Performing light activity (vacuuming, playing golf) ☐ Hypertension ☐ Moderate/Little/No difficulty ☐ All or most of the time ☐ Liver disease ☐ Extreme difficulty/Require assistance or assistive device/Unable to do ☐ Some, little or none of the time 25% ☐ Lung disease ☐ Kidney disease 10% 28) Bathing yourself 39) What is your current level of activity? ☐ Osteoporosis ☐ Bedridden or primarily no activity ☐ Moderate/Little/No difficulty ☐ Peripheral vascular disease ☐ Extreme difficulty/Require assistance or assistive device/Unable to do ☐ Light to full sports/activities ☐ Prior DVT/PE/Stroke (blood clot) 40) How is your social life? 29) Normal work or schoolwork or housework 0% ☐ Greater than 3 medical problems ☐ Moderate/Little/No difficulty ☐ My social life is restricted to my home or non-existent ☐ My social life is normal or mildly restricted ☐ Extreme difficulty/Require assistance or assistive device/Unable to do Not Frail Pre-Frail Frail 19) Would you say your current health is: 30) Lift medium weight objects ☐ the same or better than last year ☐ Moderate/Little/No difficulty ☐ worse than this time last year Pearson Chi2 = 29.7 Pr = 0.000 ☐ Extreme difficulty/Require assistance or assistive device/Unable to do

Augmented Intelligence EHR Research to Implementation work flows Spine Frailty is now in UCSF EHR Ok for RISK but ….What drives OUTCOMES ?  Previous work has sought answers in correlations Datify the Procedure….  Outcome driven by alignment Mirza ASD-S ASD-R R 2 EBL 0.22 0.28 0.34  But what does the new p value 0.0012 <0.001 <0.001 EBL Information Age and R 2 Op AI tell us?? 0.18 0.26 0.34 time  There is much more p value 0.007 0.0002 < 0.0001 OP time Neurosurgery 2017

Baseline SVA vs ODI 2yr SVA fused to Pelvis vs ODI R2 = .19  All pts, op and nonop, n=1622  All 2yr follow up Op pts, n= 502 R2 = .04 Predictive Analytics First Generation Models-Q/O All data fields analyzed separately  3 successful binary output models constructed 25% MCID pain 10% MCID appearance  Proximal junctional kyphosis/failure *Spine 2016 50% revision Pt  Major intra/periop complications *JNS Spine 2017 5% medical complication 90% play tennis Apical Fusion  Oswestry Disability Index (ODI) minimal clinical + T10-pelvis important difference *Spine Deformity 2018 T3-pelvis PSO 60% MCID pain  Methods Frailty 80% MCID appearance 10% revision  5 different bootstrapped decision trees 25% major complication  Internal validation 70:30 data split for 5% play tennis Complication training/testing Avoidance  Accuracy, and the area under a receiver operator characteristic (ROC) calculated