Blog & News
New Data on Drug Overdose Deaths Highlight the Need for State-Level AnalysisMay 15, 2019:
In a rare positive sign regarding the national opioid crisis, new provisional data from the U.S. Centers for Disease Control and Prevention (CDC) suggest that the persistent increases in overdose deaths may be slowing, with a possible leveling out or even a small drop in U.S. opioid-related overdose deaths in late 2017 and early 2018.[i] Those data were recently highlighted by U.S. Secretary of Health and Human Services Alex Azar, who cautiously pointed to the encouraging development while acknowledging that more progress is needed and that “we are so far from the end of the epidemic.”[ii]
But a closer look at provisional overdose data from states underscores a vital point: because the specifics of the opioid crisis vary across states, we can’t truly understand and effectively respond to the opioid crisis without examining state-level data.[iii]
Too early to call a turnaround for the U.S.
The CDC does not make provisional data on opioid overdoses available at the state level, but because most drug overdoses involve an opioid, the data on provisional drug overdoses that are published by the CDC can provide a useful proxy.
At the national level, provisional CDC data indicate that drug overdoses began to level off around September 2017 when they reached a total of 70,636 reported deaths for the 12-month period ending that same month. In the months since then, rolling 12-month reported death totals have remained relatively stable or even declined slightly to a low of 68,690 in March 2018 (the most recent data available). Although that directional change offers hope, especially because the trend until then showed consistent increases since at least 2014, it is worth acknowledging two important caveats:
- First, a seven-month period of relatively stable or declining deaths is still quite short compared to the nearly two-decade rise of drug overdose deaths, and it is yet to be seen whether that shift in trend will persist.
- Second, even accounting for recent slight declines in deaths from drug overdoses, the overall annual trend hasn’t yet turned positive—reported drug overdose deaths for the 12-month period ending in March 2018 were still 2.7 percent higher than March 2017.
Some states show progress, while others worsen
At the state level, trends are more varied—illustrating the pitfalls of assuming that individual states track the national shifts. Some states offer bright spots of declining drug overdose deaths, possibly indicating successful policy strategies for tackling the crisis. However, other states have seen overdose deaths continue to rise, a possible warning that the crisis is continuing to spread from those states that were struck early and hard by opioids to those that previously had been more insulated.
Among the states, 21 and the District of Columbia have seen declines in reported drug overdose deaths for the 12 months ending March 2018 compared to March 2017. One such example is Ohio, which saw reported drug overdose deaths decline 3.4 percent, from 4,788 to 4,623. Ohio historically has had among the highest opioid overdose death rates in the country but in recent years has made concerted efforts to intervene in the crisis.
Figure 1. CDC Provisional Reported Drug Overdose Deaths
Source: U.S. Centers for Disease Control and Prevention (CDC). Provisional drug overdose death counts. National Center for Health Statistics. 2018.
The remaining 29 states have continued to see their drug overdose deaths increase, and some of the highest increases are found in states that historically have had low rates of opioid overdose death rates—a concerning sign. For example, Nebraska has historically had among the lowest rates of opioid-related overdose deaths, but its drug overdose deaths increased 48.2 percent in the 12 months period ending March 2018 compared to March 2017 (163 vs. 110 deaths). Other examples include the large-population states of California and Texas, which each have historically had low rates of opioid overdose deaths but have seen increases in drug overdose deaths of 6.9 percent and 5.3 percent, respectively, during the same 12 month periods.
Take-away message for states
Although early signs of a potential plateau or small reversal in the long-term national trend of increasing drug overdose deaths are encouraging, states shouldn’t assume that trend applies to them and become complacent in efforts to intervene in the opioid crisis. Many of the states seeing early signs of progress in reducing drug overdose deaths are those that have historically had high rates of opioid overdose deaths but have adopted policies aimed at stemming that growth, such as reducing inappropriate prescribing of opioids and increasing access to treatment of substance use disorders. Unfortunately, though, most states continue to see increases in drug overdose deaths.
An additional concerning sign in provisional CDC data is early evidence that overdose deaths involving cocaine and psychostimulants (e.g., methamphetamine) have increased in the past few years. That shift could be an indication that the opioid crisis is evolving again—similar to the recent shift from prescription-type opioids to illicit synthetic opioids—but this time potentially extending to non-opioid illicit drugs.
Notes: Reported provisional counts for 12-month ending periods are the number of deaths received and processed for the 12-month period ending in the month indicated. Drug overdose deaths are often initially reported with no cause of death (pending investigation), because they require lengthy investigation, including toxicology testing. Reported provisional counts may not include all deaths that occurred during a given time period. Therefore, they should not be conisdered comparable with final data and are subject to change. Predicted provisional counts represent estimates of the number of deaths adjusted for incomplete reporting (see Technical Notes). Deaths are classified by the reporting jurisdiction in which the death occurred. Percent change refers to the relative difference between the reported or predicted provisional number of deaths due to drug overdose occurring in the 12-month period ending the month indicated compared with the 12-month period ending in the same month of the previous year. Drug overdose deaths are identified using ICD-10 underlying cause-of-death codes: X40-X44, X60-X64, and Y10-Y14.