Improving Naloxone Distribution in the Opioid Epidemic A cost-effectiveness analysis of naloxone distribution to first responders and laypeople Tarlise Townsend , Freida Blostein, Tran Doan, Sammie Madson-Olson, Paige Galecki, and David Hutton University of Michigan School of Public Health Department of Health Management and Policy
59,000 to 69,000 overdose deaths Drug overdose deaths, 1980-2016 60,000 (2016) Peak car crash 50,000 deaths (1972) Peak HIV deaths (1995) 40,000 Peak gun deaths (1993) 30,000 20,000 10,000 1980 1985 1990 1995 2000 2005 2010 2015 The New York Times
How can we distribute naloxone for maximum benefit given scarce resources?
How can we distribute naloxone for maximum benefit given scarce resources? Distribution to first responders: EMS, firefighters, Distribution to laypersons: users & others & law enforcement likely to witness an overdose
How can we distribute naloxone for maximum benefit given scarce resources? Pros - Earlier administration than in ER Distribution to first responders: EMS, firefighters, & law enforcement
How can we distribute naloxone for maximum benefit given scarce resources? Pros - Earlier administration than in ER - Most overdoses witnessed Distribution to first responders: EMS, firefighters, & law enforcement
How can we distribute naloxone for maximum benefit given scarce resources? Pros - Earlier administration than in ER - Most overdoses witnessed - First responders trained for such situations Distribution to first responders: EMS, firefighters, & law enforcement
How can we distribute naloxone for maximum benefit given scarce resources? Pros - Earlier administration than in ER - Most overdoses witnessed - First responders trained for such situations Cons - Barriers to calling 911 Distribution to first responders: EMS, firefighters, & law enforcement
How can we distribute naloxone for maximum benefit given scarce resources? Distribution to laypersons: users & others likely to witness an overdose
How can we distribute naloxone for maximum benefit given scarce resources? Pros - Most overdoses witnessed Distribution to laypersons: users & others likely to witness an overdose
How can we distribute naloxone for maximum benefit given scarce resources? Pros - Most overdoses witnessed - Laypeople administer effectively Distribution to laypersons: users & others likely to witness an overdose
How can we distribute naloxone for maximum benefit given scarce resources? Pros - Most overdoses witnessed - Laypeople administer effectively - Earlier administration than FRs Distribution to laypersons: users & others likely to witness an overdose
How can we distribute naloxone for maximum benefit given scarce resources? Pros - Most overdoses witnessed - Laypeople administer effectively - Earlier administration than FRs - Benefits a population not available to FRs Distribution to laypersons: users & others likely to witness an overdose
How can we distribute naloxone for maximum benefit given scarce resources? Pros - Most overdoses witnessed - Laypeople administer effectively - Earlier administration than FRs - Benefits a population not available to FRs - Very cost-effective (Coffin & Sullivan, 2013) Distribution to laypersons: users & others likely to witness an overdose
How can we distribute naloxone for maximum benefit given scarce resources? Pros - Most overdoses witnessed - Laypeople administer effectively - Earlier administration than FRs - Benefits a population not available to FRs - Very cost-effective (Coffin & Sullivan, 2013) Cons - Less politically palatable Distribution to laypersons: users & others likely to witness an overdose
How can we distribute naloxone for maximum benefit given scarce resources? Pros - Most overdoses witnessed - Laypeople administer effectively - Earlier administration than FRs - Benefits a population not available to FRs - Very cost-effective (Coffin & Sullivan, 2013) Cons - Less politically palatable - Higher number needed to treat Distribution to laypersons: users & others likely to witness an overdose
The Question - More resources are allocated to naloxone for first responders than for laypeople
The Question - More resources are allocated to naloxone for first responders than for laypeople - Layperson distribution is highly cost-effective
The Question - More resources are allocated to naloxone for first responders than for laypeople - Layperson distribution is highly cost-effective …So is our disproportionate emphasis on first responder distribution merited?
Compared to the status quo, what’s most cost-effective? Some first responders have naloxone -/- Few laypeople have naloxone (status quo)
Compared to the status quo, what’s most cost-effective? Some first responders have More first responders have naloxone naloxone -/- -/+ Few laypeople have naloxone (status quo)
Compared to the status quo, what’s most cost-effective? Some first responders have More first responders have naloxone naloxone -/- -/+ Few laypeople have naloxone (status quo) +/- More laypeople have naloxone
Compared to the status quo, what’s most cost-effective? Some first responders have More first responders have naloxone naloxone -/- -/+ Few laypeople have naloxone (status quo) +/- +/+ More laypeople have naloxone
More Basics - Population: Users of heroin or other illicit opioids and misusers of prescription pain relievers
More Basics - Population: Users of heroin or other illicit opioids and misusers of prescription pain relievers - Perspectives: Societal (productivity)
More Basics - Population: Users of heroin or other illicit opioids and misusers of prescription pain relievers - Perspectives: Societal (productivity) - Horizon: Lifetime
More Basics - Population: Users of heroin or other illicit opioids and misusers of prescription pain relievers - Perspectives: Societal (productivity) - Horizon: Lifetime - Data sources: National databases, one-off studies, expert interviews
Simplified Markov Model Currently Not currently misusing opioids misusing opioids
Simplified Markov Model Currently Not currently misusing opioids misusing opioids All other causes Dead Overdose + all other causes
Integrated Decision Tree (Simplified) Currently misusing opioids
Integrated Decision Tree (Simplified) Overdoses Witnessed Currently misusing opioids
Integrated Decision Tree (Simplified) Overdoses Witnessed Currently misusing opioids
Integrated Decision Tree (Simplified) Overdoses Witnessed Currently misusing opioids
Integrated Decision Tree (Simplified) Overdoses Witnessed Currently misusing opioids
144 Increased FR Status quo Cost (billions of dollars) 138 132 72.4 72.8 73.2 73.6 74.0 Effectiveness (millions of QALYs)
144 Increased FR Status quo Cost (billions of dollars) 138 Increased LP 132 72.4 72.8 73.2 73.6 74.0 Effectiveness (millions of QALYs)
144 Increased FR (Dominated!) Status quo Cost (billions of dollars) 138 Increased LP 132 72.4 72.8 73.2 73.6 74.0 Effectiveness (millions of QALYs)
144 Increased FR (Dominated!) Status quo Cost (billions of dollars) 138 Combined ($5800 per QALY) Increased LP 132 72.4 72.8 73.2 73.6 74.0 Effectiveness (millions of QALYs)
3,600 Threshold: Net Monetary Benefit (billions) Prevalence of LP naloxone required for net 18% of LPs benefit equal to High LP, High FR High LP, Low FR 75% FRs Low LP, High FR Low LP, Low FR equipped 3,480 0.04 0.5 1.0 Prevalence of LP naloxone in high-LP condition
Results highly robust to sensitivity analyses - No strategy surpassed $50,000 per QALY - No meaningful changes in rankings
3,720 Implications Implications of of the Net Monetary Benefit (billions) Increase of 14% the “Safety Net” “Safety Net” Hypothesis Hypothesis High LP, High FR High LP, Low FR Low LP, High FR Low LP, Low FR 3,480 0.8 0.9 1.0 1.1 1.2 Relative Risk of Overdose
Takeaways - Compared to increased first responder distribution, increased layperson distribution entails greater gain at less cost. - Results are highly robust to sensitivity analyses. - Therefore, our current imbalance is not merited.
Thank You… Coauthors: Sammie Tran Doan David Hutton Freida Blostein Paige Galecki Madson-Olson - Alice Bell, Prevention Point Pittsburgh Informants: - Leo Beletsky, Northeastern University - Robert Childs, North Carolina Harm Reduction Coalition - Rebecca Haffajee, University of Michigan - Brandon Hool, The Grand Rapids Red Project - Jimena Loveluck, Michigan Unified - Dominick Zurlo, New Mexico Department of Health Harm Reduction Program
SUPPLEMENTARY SLIDES
Recommend
More recommend
