PAD Cost Effectiveness

Location-Specific Cost Effectiveness of Public Access Defibrillation: 

Location-Specific Cost Effectiveness of Public Access Defibrillation SAEM, San Francisco 2006

Ontario Prehospital Advanced Life Support Study : 

Ontario Prehospital Advanced Life Support Study

Slide3: 

VJ De Maio D Coyle IG Stiell K O'Grady C Vaillancourt L Nesbitt GA Wells Canadian Association of Emergency Physicians Canadian Health Services Research Foundation Medtronic Physio-Control University of North Carolina Ottawa Health Research Institute University of Ottawa

Public Access Defibrillation (PAD): 

Public Access Defibrillation (PAD) The effectiveness of rapid defibrillation for cardiac arrest is undisputed Rationale for PAD:  availability of AEDs  faster defib times   survival Little research guiding optimal placement of AEDs

PAD Considerations: 

PAD Considerations 3-minute response time interval Multiple AEDs may be necessary per site Patient, system and clinical factors affect likelihood of an individual cardiac arrest being amenable to defibrillatory shock Widespread PAD potentially expensive, divert attention and resources from other programs

What is the cost-effectiveness of PAD when added to existing EMS?: 

What is the cost-effectiveness of PAD when added to existing EMS?

In which specific locations would it be cost-effective to install AEDs?: 

In which specific locations would it be cost-effective to install AEDs?

Objective: 

Objective To compare the incremental cost-effectiveness of the addition of a PAD program to the standard EMS response for a number of public location categories

OPALS Study: 

OPALS Study Prospective before-after clinical trial 10 years, 20 Ontario study communities Studied impact of early defib/ALS on cardiac arrest survival, neuro outcomes, quality of life Phase II: EMS and firefighter rapid AED Phase III: Full ALS Utstein style

Patients: 

Patients All adults suffering a prehospital cardiac arrest of presumed cardiac etiology prior to EMS arrival Exclusions: andlt; 16 years Obvious deaths as per Ambulance Act Trauma victims Other non-cardiac etiology

Design: 

Design Economic Evaluation Compare the total costs and life expectancy of treating cardiac arrest patients with and without an on-site automated electronic defibrillator (AED)

Data Collection: 

Data Collection

Data Collection: 

Data Collection A priori, property type codes grouped into 18 location categories Roll provided total number of sites, per location type, within the study boundary

Primary Outcome: 

Primary Outcome Incremental Cost Effectiveness Ratio The relative mean dollar cost of PAD in each location category per additional quality-adjusted life year (QALY). ICER=(Costi – Costc) / (QALYi – QALYc) i.e., the additional cost per life year gained

Decision Analysis Model: 

Decision Analysis Model Estimated life expectancy and treatment costs of patients by survival status, gender, age at cardiac arrest. Adapted data within Weinstein model for survival estimates and disease progression Future life expectancy and costs discounted at a rate of 5%. Monte Carlo simulation estimated the uncertainty surrounding the ICERs.

Design of Decision Model: 

FIRST 30 DAYS AFTER CARDIAC ARREST FOLLOWING 11 MONTHS AND SUBSEQUENT YEARS Design of Decision Model Cardiac Arrest Survived to Hospital DIED before Hospital DIED within 30 Days Survived to 30 Days Arrest only Arrest plus MI New Event MI Arrest DIED Arrest DIED Arrest plus MI Arrest only No New Event MI only Update disease history CONTINUES AS PER ARREST WITH MI WITH UNIQUE SET OF PROBABILITIES

Input Data: 

Input Data Average n of cardiac arrests/yr (OPALS) Proportion of patients who die at scene, in hospital and survive to discharge without an on-site AED by gender, age (OPALS) Survival benefit from access to on-site AED in ≤ 3 min (OR = 3.0, 95%CI = 2.3-4.0) Annual cost of AED materials, training (5-yr amortization = $1319.01/yr) Resource utilization costs were based on a representative Canadian sample with incident coronary heart disease

Patient Characteristics(N=7707): 

Patient Characteristics (N=7707) Age 69 Male 67% Bystander Witnessed 49% Initial Rhythm: VF or VT 37% PEA 21% Asystole 42% Survival 4.0%

EMS Characteristics (N=7707): 

EMS Characteristics (N=7707) Bystander CPR 16% Fire/police CPR 39% PAD response 0.2% Fire first 49% Defibrillation 43% Defibrillation response interval: Mean (min) 5.4 Response andlt; 8 min 91%

OPALS Cardiac Arrest Locations(N=7707): 

OPALS Cardiac Arrest Locations (N=7707) Large public 7% Outdoors 3% Small residential 56% Large residential 29% Small public 5%

Slide21: 

Average Years Between PAD Use (per Site)

Slide22: 

Casino 28 2 2.800 $542 Non-acute hospital 42 42 0.200 $30750 Nursing home 457 460 0.199 $45926 Indoor shopping mall 77 394 0.039 $67690 Penal institution 6 21 0.057 $128783 Hotel 65 604 0.022 $143530 Golf course 9 156 0.012 $205990 Recreation/Assembly/Community 165 3206 0.010 $205407 Restaurant/Bar 48 1410 0.007 $347954 Airport/Heliport/Rail/Bus station 4 83 0.010 $368608 Water/Boat/Marina 5 240 0.004 $478647 School/College/University 36 1770 0.004 $598210 Single store/Strip mall 231 14956 0.003 $925784 Medical office/Clinic 41 2399 0.003 $955614 Office building 96 7276 0.003 $990511 Stadium/Fairground 1 238 0.001 $1910193 Sports field/Park 14 3139 0.001 $4104539 Factory/Industrial/Railway/Docks 56 17261 0.001 $4323180 Location Arrests Sites Arrest/Site/Yr ICER

Slide23: 

Casino 100% 54.8 Non-acute hospital 99% 1.7 Nursing home 67% 1.1 Indoor shopping mall 2% 1 Penal institution 0% 1 Hotel 0% 1 Golf course 0% 1 Recreation/Assembly/Community 0% 1 Restaurant/Bar 0% 1 Airport/Heliport/Rail/Bus station 0% 1 Water/Boat/Marina 0% 1 School/College/University 0% 1 Single store/Strip mall 0% 1 Medical office/Clinic 0% 1 Office building 0% 1 Stadium/Fairground 0% 1 Sports field/Park 0% 1 Factory/Industrial/Railway/Docks 0% 1 Probability Cost Effective # AEDs Cost Effective Location

Cost Effectiveness Acceptability Curves: 

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 0 10000 20000 30000 40000 50000 60000 70000 80000 90000 100000 Value of an Additional Year of Life Probability PAD Program is Cost Effective Casinos Non-acute Hospitals Nursing Homes Cost Effectiveness Acceptability Curves

Limitations: 

No large metropolitan centers (andgt;1 million) in our population Few large sporting venues in this Canadian study region Canadian costs Assumptions of economic analysis Some would argue for willingness to pay threshold greater than $50,000/life year gained Limitations

Conclusions: 

Conclusions Only 3 locations where PAD considered cost-effective based on willingness to pay $50K/life year gained Does not appear cost-effective to install AEDs in any of the other 16 locations No convincing evidence to support widespread implementation of PAD

Conclusions cont’d…: 

EMS and public health directors should consider methods for improving survival for ALL cardiac arrest patients (e.g., improving citizen CPR, optimizing the traditional EMS response, targeted responder programs) Conclusions cont’d…

Slide28: