Zondag 19 feb, John Sharp

661 views
604 views

Published on

Published in: Health & Medicine, Business
0 Comments
1 Like
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
661
On SlideShare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
5
Comments
0
Likes
1
Embeds 0
No embeds

No notes for slide

Zondag 19 feb, John Sharp

  1. 1. Cleveland  Clinic    1300  bed  main  hospital    9  Regional  Hospitals    54,000  admissions,  2  million  visits    Group  practice  of  2700  salaried  physicians   and  scientists    3000+  research  projects    Innovative  Medical  School    30  spin  off  companies    Office  of  Patient  Experience  
  2. 2. Lethal  Lag  Time    It  takes  an  average  of  17  years  to  implement   clinical  research  results  into  daily  practice    Unacceptable  to  patients    Can  Electronic  Medical  Records  and  Clinical   Decision  Support  Systems  change  this?  
  3. 3. Electronic  Medical  Records    Comprehensive  medical   information    Images    Communication  with  other   physicians,  medical   professionals    Communication  with   patients    3  million  active  patients,  10   years  
  4. 4. EMR  Inputs  and  Outputs  Inputs   EMR  Tools   Outputs  •  Clinical   •  Alerts   Secondary  Use  •  Labs   •  Best  practices   •  Data  sets  •  Devices   •  Smart  sets   •  Registries  •  Remote  monitoring   •  Workflow   •  Quality  reports  •  Pt  outcomes   •  Communication  to  •  Omics   other  providers,  •  Social  media?   patients  
  5. 5. Clinical  Workflow   Workflow  
  6. 6. Clinical  Decision  Support    Process  for  enhancing  health-­‐related   decisions  and  actions  with  pertinent,   organized  clinical  knowledge  and  patient   information      to  improve  health  and  healthcare  delivery.      Information  recipients  can  include  patients,   clinicians  and  others  involved  in  patient  care   delivery   http://www.himss.org/ASP/ topics_clinicalDecision.asp  
  7. 7. Like a GPS, CDS suppliesinformation tailored to the current situation, and organized for maximum value.
  8. 8. Diagnostic  Cockpit  
  9. 9. CDS  Example:  Order  Sets  
  10. 10. CDS  as  a  Strategic  Tool  •  CDS  should  be  used  as  a  strategic  tool  for   achieving  an  organization’s  priority  care  delivery   objectives.    •   These  objectives  are  driven  by  external  forces   such  as     •  payment  models     •  regulations  related  to  improving  care  quality  and  safety   •  internal  needs  for  improving  quality  and  patient  safety   •   reducing  medical  errors   •  increasing  efficiency  
  11. 11. EMR  Alert  Types   Clinical  Decision  Support   Target Area of Care   Example  Preventive care   Immunization, screening, disease management guidelines for secondary prevention  Diagnosis   Suggestions for possible diagnoses that match a patient’s signs and symptoms  Planning or implementing Treatment guidelines for specific diagnoses, drugtreatment   dosage recommendations, alerts for drug-drug interactions  Followup management   Corollary orders, reminders for drug adverse event monitoring  Hospital, provider efficiency   Care plans to minimize length of stay, order sets  Cost reductions and improved Duplicate testing alerts, drug formulary guidelines  patient convenience  
  12. 12. Clinical  Decision  Support  Examples    New  diagnosis  of  Rheumatoid  Arthritis,   prompted  to  refer  to  preventive  cardiology  
  13. 13. Clinical  Decision  Support  Examples    Age  >  50  and  a  fragile  fracture  diagnosis  –   order  set  for  bone  density  scan  and   appropriate  medication  regimen    Go  to  Smart  Set  
  14. 14. Clinical  Decision  Support   Examples    Solid  organ  transplant  –  chemoprevention  for   skin  cancer  
  15. 15. The  CDS  Toolbox    (more  examples)    Drug-­‐Drug  Interactions       Rules  to  meet  strategic    Drug-­‐Allergy  interactions     objectives  (core  measures,   antibiotic  usage,  blood    Dose  Range  Checking     management)    Standardized  evidence  based     Documentation  templates     ordersets       Relevant  data  displays      Links  to  knowledge  references       Point  of  care  reference    Links  to  local  policies     information  (i.e.  InfoButtons)       Web  based  reference   information       Diagnostic  decision  support   tools    
  16. 16. Virtuous  Cycle  of  Clinical   Decision  Support   Registry   Measure   Practice   Guideline   CDS  http://www2.eerp.usp.br/Nepien/DisponibilizarArquivos/tomada_de_decis%C3%A3o.pdf  
  17. 17. EMRs  and  Quality  of  Care  
  18. 18. EMR  and  Quality  of  Care    Diabetes  care  was  35.1  percentage  points  higher   at  EHR  sites  than  at  paper-­‐based  sites      Standards  for  outcomes  was  15.2  percentage   points  higher     Across  all  insurance  types,  EHR  sites  were   associated  with  significantly  higher  achievement   of  care  and  outcome  standards  and  greater   improvement  in  diabetes  care    Better  Health  Greater  Cleveland  
  19. 19. Meaningful    Use  
  20. 20. The  Role  of  Registries    EMR  data  available  to  create  a  registry  for   any  condition    Study  the  condition  –  progression,   treatments    Comparative  effectiveness  of  treatments    Recruit  for  clinical  trials    Develop  clinical  decision  support  
  21. 21. Chronic  Kidney     Disease  Registry    Chronic  Kidney  Disease  Registry    Established  2009    60,000  patients  from  the  health  system    Cohort  –  Adults  with  two  eGFRs  less  than  60   within  3  months,  outpatient  results  only,  or   diagnosis  of  CKD    http://www.chrp.org/pdf/ HSR_12022011_Slides.pdf  
  22. 22. Validation  Results    Our  dataset’s  agreement  with  EHR-­‐extracted   data  for  documentation  of  the  presence  and   absence  of  comorbid  conditions,  ranged  from   substantial  to  near  perfect  agreement.    Hypertension  and  coronary  artery  disease   were  exceptions      EMR  data  accurate  for  research  use  
  23. 23. Registry  Results    2011    5  out  of  5  abstracts  accepted  to     American  Society  of  Nephrology  annual   meeting    Three  papers  accepted  to  nephrology   journals    NIH  grant    Partnerships  with  other  research  centers  
  24. 24. Pediatric  Surgical  Site  Infection    Data  from  the  EMR  and  the  operative  record    When  did  antibiotics  start?    Was  pre-­‐op  skin  prep  done?    Was  the  time-­‐out  and  checklist  observed  in   the  OR    Post-­‐op  care  quality  
  25. 25. Patient  Reported  Outcomes    Understanding  the  outcomes  of  treatment   incomplete  without    Patient  Reported  Outcomes  Measurement   Information  System   http://www.nihpromis.org/    Patient-­‐Centered  Outcomes  Research   Institute  http://www.pcori.org/  
  26. 26. Patient  Reported  Outcomes    Quality  of  life    Activities  of  daily  living    Recording  weight,  diet,  exercise  using  apps    Quantified  Self  
  27. 27. Population  Health    New  tools  to  enable  the  study  of  disease   trends  and  epidemics    PopHealth  -­‐  submission  of  quality  measures   to  public  health  organizations   http://projectpophealth.org    Query  Health  –  standards  to  enable   Distributed  Health  Queries   http://wiki.siframework.org/Query+Health  
  28. 28. Predictive  Models    Predicting  6-­‐Year  Mortality  Risk  in  Patients   With  Type  2  Diabetes    Cohort  of  33,067  patients  with  type  2  diabetes   identified  in  the  Cleveland  EMR    Prediction  tool  created  in  this  study  was   accurate  in  predicting  6-­‐year  mortality  risk   among  patients  with  type  2  diabetes    Diabetes  Care  December  2008,  vol.  31  no.  12:  2301-­‐2306  
  29. 29. Postoperative nomogram based on 996 patients treated at The Methodist Hospital, Houston, TX, for predicting PSA recurrence after radical prostatectomy. Nomograms  bring  into  visual  perspective  the  effect    exerted  by  continuous  variables     against  measured  end  points   Kattan M W et al. JCO 1999;17:1499-1499©1999 by American Society of Clinical Oncology
  30. 30. Risk    Calculators  Type  2  Diabetes  Predicting  6-­‐Year  Mortality  Risk  
  31. 31. Algorithms  clevelandclinicmeded.com/  medicalpubs/micu/  
  32. 32. Against  Diagnosis    The  act  of  diagnosis  requires  that  patients  be  placed   in  a  binary  category  of  either  having  or  not  having  a   certain  disease.     These  cut-­‐points  do  not  adequately  reflect  disease   biology,  may  inappropriately  treat  patients     Risk  prediction  as  an  alternative  to  diagnosis    Patient  risk  factors  (blood  pressure,  age)  are   combined  into  a  single  statistical  model  (risk  for  a   cardiovascular  event  within  10  years)  and  the  results   are  used  in  shared  decision  making  about  possible   treatments.    Annals  of  Internal  Medicine,  August  5,  2008vol.  149  no.  3  200-­‐203  
  33. 33. Information  Overload    New  information  in  the     Information  about  an   medical  literature   individual  patient     PubMed    adding  over     Lab  results   670,000  new  entries  per     Vitals   year     Imaging     Genomics  
  34. 34. Personalized  Medicine     The  boundaries  are  fading  between  basic   research  and  the  clinical  applications  of   systems  biology  and  proteomics    New  therapeutic  models     Journal  of  Proteome  Research  Vol.  3,  No.  2,  2004,  179-­‐196.  
  35. 35. Example–Parkinson’s  Disease    New  Cleveland  Clinic  partnership  with   23andMe  to  collect  DNA  from  Parkinson’s   patients    Looking  for  Genome  Wide  Associations   (GWAS)    23andme.com/pd/  
  36. 36. Precision  Medicine     ”state-­‐of-­‐the-­‐art  molecular  profiling  to   create  diagnostic,  prognostic,  and   therapeutic  strategies  precisely  tailored  to   each  patients  requirements.”     ”The  success  of  precision  medicine  will   depend  on  establishing  frameworks  for  … interpreting  the  influx  of  information  that  can   keep  pace  with  rapid  scientific   developments.”    N  Engl  J  Med  2012;  366:489-­‐491,  2/  9/2012  
  37. 37. Artificial  Intelligence  in  Medicine    Developing  a  search  engine  that   will  scan  thousands  of  medical   records  to  turn  up  documents   related  to  patient  queries.    Learn  based  on  how  it  is  used    “We  are  not  contemplating  ―   unless  this  were  an  unbelievably   fantastic  success  ―  letting  a   machine  practice  medicine.”    http://www.health2news.com/ 2012/02/10/the-­‐national-­‐library-­‐of-­‐ medicine-­‐explores-­‐a-­‐i/  
  38. 38. IBM  Watson    Medical  records,  texts,  journals  and  research   documents  are  all  written  in  natural  language   –  a  language  that  computers  traditionally   struggle  to  understand.  A  system  that   instantly  delivers  a  single,  precise  answer   from  these  documents  could  transform  the   healthcare  industry.    “This  is  no  longer  a  game”    http://tinyurl.com/3b8y8os  
  39. 39. Digital  Humans   Convergence  of:     Genomics     Social  media     mHealth     Rebooting  Clinical  Trials  
  40. 40. Conclusion  -­‐  1    EMR  as  the  platform  for  the  future  of   medicine    Data  incoming     Clinical     Patient  Reported     Genomic     Proteomic     Home  monitoring  
  41. 41. Conclusion  -­‐  2    Exploit  all  uses  of  the  EMR  to       Improve  practice  efficiency     Ensure  patient  safety     Learn  about  your  patients     (registries)     Compare  treatments     Engage  with  patients  
  42. 42. Conclusion  -­‐  3    Understand  Personalized     and  Precision  medicine    How  will  we  integrate     genomic  data  in  clinical   practice  in  the  future?  
  43. 43. Conclusion  -­‐  4    Predictive  models  inform  care    How  do  we  integrate  these  into  practice  in   the  EMR?  
  44. 44. Conclusion  -­‐  5    How  can  we  reduce  the  lethal  lag  time?    Getting  medical  findings  into  practice  more   rapidly    How  can  we  engage  patients?    Real  time  data  on  populations    New  technology  for  Big  Data  in  health  care  
  45. 45. Contact  me    @JohnSharp    Ehealth.johnwsharp.com    Linkedin.com/in/johnsharp    Slideshare.net/johnsharp  ______________________    ClevelandClinic.org    @ClevelandClinic    Facebook.com/ClevelandClinic    youtube.com/clevelandclinic  

×