Applied Analytics: Medicaid’s Influence on the Opioid Crisis

Trying to explain the nation’s opioid crisis, Senator Ron Johnson from Wisconsin pointed to Medicaid expansion following 2013’s passage of the Affordable Care Act. “I’m not saying this is the primary cause,” he said at a Congressional hearing in January 2018. “I think what we are certainly saying is this is an unintended consequence.”

In this analysis, I seek evidence for this statement in data from the Healthcare Cost and Utilization Project (HCUP), a Federal-State-Industry partnership that is USA’s largest collection of longitudinal hospital care data. Spoiler alert: the numbers provide no support for the senator’s statement. Indeed, from 2005 to 2015, in states with the highest growth rate in opioid-related hospital treatment, the corresponding growth rate in the proportion of treatment paid for by Medicaid generally falls below the average.

HCUP defines “opioid-related hospital treatment” as treatment for opioid dependencies, abuse, and poisoning, among other opioid-related illnesses (for details, see https://www.hcup-us.ahrq.gov/faststats/OpioidUseServlet). HCUP breaks the data into inpatient stays and emergency visits; it facilitates comparisons among states by standardizing treatment rate on a per 100,000 population per quarter basis. For my analysis, I subset the data to the 14 states with the complete set of quarterly reports for inpatient stays and emergency visits from 2005 to 2015.

The increase in rates of opioid treatment for both inpatient stays and emergency visits is evident in Figure 1, which represents the average rate across all 14 states in the dataset from 2005–2015; the rate for inpatient treatment exceeds that for emergency visits.

Figure 1. Rate of opioid treatment per 100,000 population, averaged over 14 states (see listing in Figure 2) from 2005–2015.

Figure 2(a) shows the average annual growth rate of treatment for emergency visits at the state level. It indicates that states with the highest growth rate in treatment are Indiana, South Carolina, Missouri, Wisconsin, and Arizona. Of these, except for Wisconsin, the corresponding growth rate in the proportion paid by Medicaid is either at or below the mean, as shown by Figure 2(b).

Figure 2. (a) Ranking of states based on average annual growth rate of opioid treatment for emergency visits. (b) Ranking of states based on average annual growth rate of proportion of emergency treatment paid by Medicaid; red, dashed line is the population mean. Note that 4 out of the 5 states with the highest growth rate in treatment have growth rates in proportion paid by Medicaid falling below the population mean.

Figure 3 provides similarly scant support for the suggestion that Medicaid access increases opioid use. Figure 3(a) shows that states with the highest growth rates in inpatient, opioid treatment are Arizona, Georgia, Florida, Tennessee, and Missouri. For all five, the growth rate in the proportion paid by Medicaid is below the mean. Indeed, Figure 3(b) shows that Arizona, Georgia and Tennessee are among the states with the lowest growth in proportion paid by Medicaid.

Figure 3. (a) Ranking of states based on average annual growth rate of opioid treatment for inpatient stays. (b) Ranking of states based on average annual growth rate of proportion of inpatient treatment paid by Medicaid; red, dashed line is the population mean. Note that the 5 states with the highest growth rate in treatment have growth rates in proportion paid by Medicaid falling below the population mean.

With Wisconsin having the fourth highest growth rate for inpatient, opioid treatment, the senator from that state understandably worries about the opioid crisis. But to provide meaningful solutions, perhaps Senator Johnson should shift his focus away from Medicaid availability to factors that actually have an empirical basis.

Leave a Reply