Draft_CMS Mini Plenary Slides 1.26.2018 - Eldridge slides only.pptx

Presentation at the 2018 CMS Quality Conference on the AHRQ QSRS Opioid Adverse Event measure (as it stood at that time) and preliminary data

1. Implementing Recommendations from the National Action Plan for Adverse Drug Event Prevention: Preventing Adverse Drug Events from Opioid Analgesics. February 13, 2018 1

2. Learning Objectives for Today 2 After this Session, you should be able to: Describe the key elements, targets, and metrics within the National Action Plan for Adverse Drug Event Prevention relevant to opioids. Identify and describe an online training tool for primary care physicians, pharmacists, and nurses. Apply lessons learned and tools from the VHA’s Opioid Safety Initiative. Describe national quality data systems that track inpatient opioid ADEs as well as trends in opioid-related ER visits and hospital admissions.

3. Overview Background: ADE Action Plan Pathways to Safer Opioid Use VHA’s Opioid Safety Initiative New Inpatient Opioid Adverse Event Measure in the AHRQ Quality and Safety Review System Clydette Powell Anna Gribble Friedhelm Sandbrink Noel Eldridge 3

4. 4 Noel Eldridge Center for Quality Improvement and Patient Safety Agency for Healthcare Research and Quality New Inpatient Opioid Adverse Event Measure in the AHRQ Quality and Safety Review System

5. ADEs in the Quality and Safety Review System 5 QSRS is a successor to the Medicare Patient Safety Monitoring System that has been in use by CMS and AHRQ beginning in 2002 MPSMS is the source of PfP and National HAC Rate data on anticoagulant and hypoglycemic adverse events, which demonstrated a combined 16% reduction from 2010 to 2014 QSRS v3.3 is now in Pilot Test and is producing data that we are reviewing. QSRS ADEs include: Suspected Allergic Reactions and Overdoses Anticoagulant Adverse Events (associated with both older medications and newer medications) Hypoglycemic Adverse Events Opioid Adverse Events (today’s topic) Other ADEs option for unprompted free text inputs not captured above, e.g., GI bleeds from NSAIDs, delirium associated with benzodiazepines, etc.

6. Opioid Adverse Event Description (Numerator) 6 Patient receiving opioids (e.g., morphine, fentanyl, meperidine) during hospital stay, and experiencing any of the following within 24 hours after opioid administration: Administration of intravenous (IV) naloxone, unless: IV naloxone was administered during a procedure or within 2 hours following a procedure, or IV naloxone was administered for pruritis , urinary retention, or constipation, or IV naloxone was administered only in combination with, or at the same time as, the opioid Respiratory arrest, unless: The respiratory arrest was described as due to the patient’s underlying condition or diagnosis, or The respiratory arrest was described as anticipated or normal, or responded to as if it had been anticipated by the hospital’s clinical personnel, based on the opioid dosage Unresponsiveness or response only to noxious stimulation, unless: The unresponsiveness was described as due to the patient’s underlying condition or diagnosis, or The unresponsiveness was described as anticipated or normal, or responded to as if it had been anticipated by the hospital’s clinical personnel, based on the opioid dosage

7. Opioid Adverse Event Description (Supporting Information on Denominators, without Data) 7 Risk Assessments and Preventive Actions, Preliminary data analysis: Patients diagnosed with Acute Myocardial Infarction (AMI), Heart Failure (HF), and Pneumonia are over-represented in this sample. For patients receiving opioids during the stay Not taking opioids prior to admission Taking opioids prior to admission Opioid was not a prescription drug Opioid was a prescription drug (even if was not prescribed for this patient). (Note: see list of prescription opioids) Whether the opioid was prescribed for this patient Whether the opioid was a drug that could be used for medication assisted therapy (MAT ) (Note: see list of MAT medications)

8. Opioid Selection of QSRS Medication Event Module Algorithm 8 Pre-decisional Information for Internal Government Use Only

9. Section of Algorithm on pre-admission opioid use

10. Section of algorithm for establishing an opioid adverse event based on naloxone use

11. Opioid Adverse Event Description (Supporting Information on Denominators, with Data) Risk Assessments and Preventive Actions, Preliminary data analysis: Patients diagnosed with Acute Myocardial Infarction (AMI), Heart Failure (HF), and Pneumonia are over-represented in this sample. Sample from QSRSv3.3 (3,454 patient charts) For patients receiving opioids during the stay ??? Not taking opioids prior to admission Taking opioids prior to admission ??? Opioid was not a prescription drug Opioid was a prescription drug (even if was not prescribed for this patient). (Note: see list of prescription opioids) ??? Whether the opioid was prescribed for this patient ??? Whether the opioid was a drug that could be used for medication assisted therapy (MAT ) (Note: see list of MAT medications) ???

12. Opioid Adverse Event Description (Supporting Information on Denominators, with Data) 12 Risk Assessments and Preventive Actions, Preliminary data analysis: Patients diagnosed with Acute Myocardial Infarction (AMI), Heart Failure (HF), and Pneumonia are over-represented in this sample. Sample from QSRSv3.3 (3,454 patient charts) For patients receiving opioids during the stay 2,344, 68% Not taking opioids prior to admission Taking opioids prior to admission ??? Opioid was not a prescription drug Opioid was a prescription drug (even if was not prescribed for this patient). (Note: see list of prescription opioids) ??? Whether the opioid was prescribed for this patient ??? Whether the opioid was a drug that could be used for medication assisted therapy (MAT ) (Note: see list of MAT medications) ???

13. It’s 68%, but it’s complicated, and it needs more analysis, which we will do… 13 Preliminary data analysis: Patients diagnosed with Acute Myocardial Infarction (AMI), Heart Failure (HF), and Pneumonia are over-represented in this sample.

14. Opioid Adverse Event Description (Supporting Information on Denominators, with Data) Risk Assessments and Preventive Actions, Preliminary data analysis: Patients diagnosed with Acute Myocardial Infarction (AMI), Heart Failure (HF), and Pneumonia are over-represented in this sample. Sample from QSRSv3.3 (3,454 patient charts) For patients receiving opioids during the stay 2,344, 68% Not taking opioids prior to admission Taking opioids prior to admission 805, 34% Opioid was not a prescription drug Opioid was a prescription drug (even if was not prescribed for this patient). (Note: see list of prescription opioids) ??? Whether the opioid was prescribed for this patient ??? Whether the opioid was a drug that could be used for medication assisted therapy (MAT ) (Note: see list of MAT medications) ???

15. It’s 34%, but it’s complicated and the sample size is getting small (n=805) 15 (This 5% is 1 out of 20) Preliminary data analysis: Patients diagnosed with Acute Myocardial Infarction (AMI), Heart Failure (HF), and Pneumonia are over-represented in this sample.

16. Opioid Adverse Event Description (Supporting Information on Denominators, with Data) Risk Assessments and Preventive Actions, Preliminary data analysis: Patients diagnosed with Acute Myocardial Infarction (AMI), Heart Failure (HF), and Pneumonia are over-represented in this sample. Sample from QSRSv3.3 (3,454 patient charts) For patients receiving opioids during the stay 2,344, 68% Not taking opioids prior to admission Taking opioids prior to admission 805, 34% Opioid was not a prescription drug Opioid was a prescription drug (even if was not prescribed for this patient). (Note: see list of prescription opioids) 786, 97.6% Whether the opioid was prescribed for this patient ??? Whether the opioid was a drug that could be used for medication assisted therapy (MAT ) (Note: see list of MAT medications) ???

17. 17 No graph for 97.6% 

18. Opioid Adverse Event Description (Supporting Information on Denominators, with Data) 18 Risk Assessments and Preventive Actions, Preliminary data analysis: Patients diagnosed with Acute Myocardial Infarction (AMI), Heart Failure (HF), and Pneumonia are over-represented in this sample. Sample from QSRSv3.3 (3,454 patient charts) For patients receiving opioids during the stay 2,344, 68% Not taking opioids prior to admission Taking opioids prior to admission 805, 34% Opioid was not a prescription drug Opioid was a prescription drug (even if was not prescribed for this patient). (Note: see list of prescription opioids) 786, 97.6% Whether the opioid was prescribed for this patient 771, 98.1% Whether the opioid was a drug that could be used for medication assisted therapy (MAT ) (Note: see list of MAT medications) ???

19. 19 No graph for 98.1% 

20. Opioid Adverse Event Description (Supporting Information on Denominators, with Data) 20 Risk Assessments and Preventive Actions, Preliminary data analysis: Patients diagnosed with Acute Myocardial Infarction (AMI), Heart Failure (HF), and Pneumonia are over-represented in this sample. Sample from QSRSv3.3 (3,454 patient charts) For patients receiving opioids during the stay 2,344, 68% Not taking opioids prior to admission Taking opioids prior to admission 805, 34% Opioid was not a prescription drug Opioid was a prescription drug (even if was not prescribed for this patient). (Note: see list of prescription opioids) 786, 97.6% Whether the opioid was prescribed for this patient 771, 98.1% Whether the opioid was a drug that could be used for medication assisted therapy (MAT ) (Note: see list of MAT medications) 30, 4%

21. A Table for 4% 21 Opioid was of a type used in medication-assisted treatment (MAT)   OR Non-OR   Age OR Male OR Female Non-OR Male Non-OR Female Total 18-44 0% 11% 0% 10% 7% 45-64 6% 1% 10% 4% 5% 65-74 10% 2% 10% 2% 5% 75-84 0% 0% 0% 0% 0% 85+ 0% 0% 4% 0% 1% Total 6% 2% 7% 2% 4% Total 3% 4% 4% Sample Size 10 20 30 All Male (all-age and both settings) 6% All Female (all-age and both settings) 2% Preliminary data analysis: Patients diagnosed with Acute Myocardial Infarction (AMI), Heart Failure (HF), and Pneumonia are over-represented in this sample.

22. Current QSRS data (v3.3) is in line with some earlier QSRS data (from v3.1 and v3.2) Preliminary data analysis: Patients diagnosed with Acute Myocardial Infarction (AMI), Heart Failure (HF), and Pneumonia are over-represented in this sample.

23. The numerator: Opioid Event Description Patient receiving opioids (e.g., morphine, fentanyl, meperidine) during hospital stay, and experiencing any of the following within 24 hours after opioid administration: Administration of intravenous (IV) naloxone, unless: IV naloxone was administered during a procedure or within 2 hours following a procedure, or IV naloxone was administered for pruritis , urinary retention, or constipation, or IV naloxone was administered only in combination with, or at the same time as, the opioid Respiratory arrest, unless: The respiratory arrest was described as due to the patient’s underlying condition or diagnosis, or The respiratory arrest was described as anticipated or normal, or responded to as if it had been anticipated by the hospital’s clinical personnel, based on the opioid dosage Unresponsiveness or response only to noxious stimulation, unless: The unresponsiveness was described as due to the patient’s underlying condition or diagnosis, or The unresponsiveness was described as anticipated or normal, or responded to as if it had been anticipated by the hospital’s clinical personnel, based on the opioid dosage

24. The numerator: Opioid Event Description 24 Patient receiving opioids (e.g., morphine, fentanyl, meperidine) during hospital stay, and experiencing any of the following within 24 hours after opioid administration: Patients diagnosed with Acute Myocardial Infarction (AMI), Heart Failure (HF), and Pneumonia are over-represented in this sample. Sample from QSRSv3.3 (3,454 patient charts) Administration of intravenous (IV) naloxone, unless: IV naloxone was administered during a procedure or within 2 hours following a procedure, or IV naloxone was administered for pruritis , urinary retention, or constipation, or IV naloxone was administered only in combination with, or at the same time as, the opioid Respiratory arrest, unless: The respiratory arrest was described as due to the patient’s underlying condition or diagnosis, or The respiratory arrest was described as anticipated or normal, or responded to as if it had been anticipated by the hospital’s clinical personnel, based on the opioid dosage Unresponsiveness or response only to noxious stimulation, unless: The unresponsiveness was described as due to the patient’s underlying condition or diagnosis, or The unresponsiveness was described as anticipated or normal, or responded to as if it had been anticipated by the hospital’s clinical personnel, based on the opioid dosage

25. The numerator: Opioid Event Description 25 Patient receiving opioids (e.g., morphine, fentanyl, meperidine) during hospital stay, and experiencing any of the following within 24 hours after opioid administration: Patients diagnosed with Acute Myocardial Infarction (AMI), Heart Failure (HF), and Pneumonia are over-represented in this sample. Sample from QSRSv3.3 (3,454 patient charts, with 2,344 patients receiving inpatient opioids) Administration of intravenous (IV) naloxone, unless: IV naloxone was administered during a procedure or within 2 hours following a procedure, or IV naloxone was administered for pruritis , urinary retention, or constipation, or IV naloxone was administered only in combination with, or at the same time as, the opioid Respiratory arrest, unless: The respiratory arrest was described as due to the patient’s underlying condition or diagnosis, or The respiratory arrest was described as anticipated or normal, or responded to as if it had been anticipated by the hospital’s clinical personnel, based on the opioid dosage Unresponsiveness or response only to noxious stimulation, unless: The unresponsiveness was described as due to the patient’s underlying condition or diagnosis, or The unresponsiveness was described as anticipated or normal, or responded to as if it had been anticipated by the hospital’s clinical personnel, based on the opioid dosage 20 adverse events 11 adverse events 5 adverse events 6 adverse events

26. Next Steps? 26 Perform additional analyses on data Pull selected small groups of charts to confirm findings, check for accuracy -- e.g., and “don’t knows” vs. real “noes” Focus on determining if there are issues with too many false positives and false negatives in the numerator data Follow up with other members of Opioid FAWG and their agencies Follow up internally at AHRQ (HCUP team has data on Opioid ED visits and Opioid-related admissions from ICD data) to look for synergy Considerations for QSRS v3.4 (e.g., add question(s) re opioid prescription at discharge for inpatients receiving opioids?) Are numerator adverse event definitions in v3.3 appropriate and workable? Others?

27. Additional Backup Information 27

28. HCUP Data: National rate of opioid-related inpatient stays and emergency department visits, 2005-2014 28 Stat Brief #219, Figure 1, January 2017 Line graph that shows the rate of inpatient stays and emergency department visits per 100,000 population from 2005 to 2014. Inpatient stays: 2005, 136.8; 2006, 164.2; 2007, 159.0; 2008, 165.7; 2009, 181.4; 2010, 197.1; 2011, 207.8; 2012, 210.4; 2013, 213.7; 2014, 224.6, for a 64.1% cumulative increase and a 5.7% average annual growth rate. Emergency department visits: 2005, 89.1; 2006, 91.8; 2007, 82.6; 2008, 94.1; 2009, 107.4; 2010, 117.5; 2011, 131.2; 2012, 146.8; 2013, 166.2; 2014, 177.7, for a 99.4% cumulative increase and an 8.0% average annual growth rate. .9 goes to 1.8 per 1,000 1.4 goes to 2.2 per 1,000

29. HCUP data: National rate of opioid-related inpatient stays and ED visits by patient location, 2005-2014 29 Stat Brief #226, Figure 1, July 2017 .9 to 2 per 1,000 2 to 2.5 per 1,000 .7 to 1 per 1,000 1.5 to 2 per 1,000

30. HCUP Data: National rate of opioid-related inpatient stays and ED visits by patient sex, 2005-2014 Stat Brief #224, Figure 1, June 2017 Two line graphs that show the national rate of opioid-related inpatient stays and ED visits per 100,000 population from 2005 to 2014 by patient sex . Inpatient stays: Male: Increased from 145.6 in 2005 to 178.2 in 2006, decreased to 167.5 in 2007, and then increased steadily to 225.0 in 2014; cumulative increase, 54.5% . Female: Increased from 127.8 in 2005 to 150.5 in 2006, decreased to 150.2 in 2007, and then increased steadily to 224.1 in 2014; cumulative increase, 75.3% . ED visits: Male : Increased from 99.9 in 2005 to 102.9 in 2006, decreased to 91.9 in 2007, and then increased steadily to 202.8 in 2014; cumulative increase, 103.1% . Female : Increased from 78.6 in 2005 to 81.0 in 2006, decreased to 73.6 in 2007, and then increased steadily to 153.2 in 2014; cumulative increase, 94.8%.