Is the Opioid Overdose Crisis a Story of Supply or Demand? Depends Where You Look

By Shannon Monnat, an Associate Professor of Sociology and the Lerner Chair of Public Health Promotion at Syracuse University. Originally published at the Institute for New Economic Thinking website

Over the past two decades deaths from opioids and other drugs have grown to be a major U.S. health problem, but fatal overdose rates are much higher in some places than in others. Explanations for this variation are debated. One ongoing debate is whether geographic differences in drug mortality rates are driven mostly by opioid supply factors (Currie et al. 2018; Ruhm 2018) or socioeconomic distress, that is demand factors. (Case and Deaton 2015, 2017; Hollingsworth et al. 2017; Monnat 2018). Existing research in this area does not allow for the possibility that demand factors may matter more in some parts of the U.S. while supply factors might matter more in others. In particular, the relative contributions of economic distress and opioid supply factors in explaining variation in drug mortality rates in rural versus urban areas and across different labor markets have been ignored.

In my INET paper, I use National Vital Statistics System mortality data to examine relationships between county-level non-Hispanic white drug mortality rates for 2000-02 and 2014-16 and several socioeconomic and opioid supply measures across the urban-rural continuum and within different rural labor markets. My aim is to tease out the relative contributions of socioeconomic distress and opioid supply factors on geographic variation in drug mortality rates and identify characteristics of counties bearing the heaviest drug mortality burdens.

Where are Drug Mortality Rates the Highest?

Average non-Hispanic white drug mortality rates in 2014-16 were highest in large metro counties and decline precipitously as the continuum moves from more urban to more rural. Rates have increased most in large metro counties and least in the most rural counties since the early 2000. Counties characterized by more economic distress, family distress, persistent population loss, dependence on the mining or service industries, and opioid supply (proxied by exposure to prescription opioids and fentanyl) have significantly higher drug mortality rates and experienced the largest increases in rates between 2000-02 and 2014-16. Drug mortality rates are also higher among counties with neighbors with high drug mortality rates and high opioid prescribing rates, suggesting spatial spillover effects. Importantly, the introduction of opioid supply measures to regression models did not eliminate the significance or magnitude of the economic distress factors, suggesting that economically distressed counties do not have higher drug mortality rates simply because they also have a bigger opioid supply problem.

However, the relative contributions of socioeconomic distress and opioid supply factors vary across the urban-rural continuum and across different rural labor markets. In rural counties, economic distress appears to be a stronger predictor than opioid supply of drug mortality rates; that is, economic distress better explains variation in rates between rural counties than does opioid supply factors. Conversely, in urban counties, opioid supply factors better explain variation in drug mortality rates between urban counties than does economic distress. Ultimately, the highest drug mortality rates are disproportionately concentrated in economically-distressed mining and service sector dependent counties with high exposure to prescription opioids and fentanyl.

Conclusions drawn about the role of socioeconomic distress and opioid supply factors at the national scale cannot be assumed to be universally applicable across the U.S. In terms of the clear disadvantages experienced within mining- and service-dependent counties, numerous in-depth accounts show that dependence on a declining industry and the economic instability, population loss, and disinvestments following job losses in that industry tend to overlap and be mutually reinforcing, showing up in collective psychosocial distress, family and community breakdown, and social disorders like substance misuse (Alexander, 2017; Chen, 2015; Goldstein, 2017; Macy, 2018; Quinones, 2015; Sherman, 2009). Graham and Pinto (2018) show that poor rural whites are less hopeful and optimistic about their futures than any other group in the U.S. They suggest that people who are hopeful and optimistic about their future tend to invest in that future, whereas those with dampened hope may underinvest or simply give up. Mining-dependent counties in Appalachia have experienced pronounced and sustained declines in employment, wages, and population over the past several decades. As a result, these places were vulnerable to opioids—drugs that numb both physical and psychological pain—and were primed for high rates of addiction and overdose as opioids infiltrated these communities.

The cliché that “addiction does not discriminate” ignores the fact that drug mortality rates vary significantly by geography and are disproportionately clustered among places characterized by higher prevalence of socioeconomic distress, disadvantaged labor markets, and greater access to opioids. Though the physiological processes that underlie addiction may not discriminate, the factors that place individuals and communities at risk are not spatially random. Without a clear understanding of this geographic variation and without identifying the types of people and places at risk of high mortality rates, resources and policy solutions may be misdirected.

See original post for references

Print Friendly, PDF & Email

12 comments

  1. tegnost

    Generally informative but I can’t square this circle…
    Graham and Pinto (2018) show that poor rural whites are less hopeful and optimistic about their futures than any other group in the U.S. They suggest that people who are hopeful and optimistic about their future tend to invest in that future, whereas those with dampened hope may underinvest or simply give up.
    notice the neoliberal term invest here, and compare it to…

    drug mortality rates vary significantly by geography and are disproportionately clustered among places characterized by higher prevalence of socioeconomic distress, disadvantaged labor markets, and greater access to opioids.

    I think in a disadvantaged labor market they probably have nothing to invest? Not a direct comparison as the first quote presents other authors conclusions, while the second is this authors summary so maybe I’m being picky

    Reply
    1. Peter Pan

      I suspect that the reference to “invest” is taken from this portion of the referenced paper by Graham and Pinto (2018):

      “Having less hope about the future could increase one’s likelihood of premature death (e.g., via under-investment in one’s health and/or via simply giving up);…”

      My interpretation is that investing in this sense means exercising, smoking cessation, etc. But the authors note that these individuals face reduced access to safety nets, some of which is cultural peer pressure, difficult bureaucratic navigation, lack of availability and/or all of these in combination. I suspect that many live in states that do not have access to ACA’s expansion of Medicaid (pain city) that could help mitigate some of these health issues.

      Reply
      1. JBird4049

        Yes, in order to invest, one needs something to invest. Perhaps poor people being poor don’t have anything to invest?

        I realize that the United States differs greatly from area to area but sometimes when reading about our problems what is written reads like badly written article on an exploration of Greater Nairobi or the Great Basin.

        Reply
          1. JBird4049

            My apologies for being rude.

            As I am one of those poor people here in Bluest California, I was trying to point out that rather than drive 80 miles to Red California where being poor is the norm, or even just stepping right out the front door along much of the Financial District, or Sixth Street, or the Haigt, or under the freeways, or…And if I want I can drive to MLK drive in the East Bay for a thrill. Really, any town or city of any decent size in the fabulous dystopian wasteland of dreams that makes the Bay Area.

            I say this because too often the comfortable, or especially the successful fearful, pretend that the poverty that is slowly drowning us all is there fault, whoever “they” are and that somehow the poor inhabit a strange, mysterious land.

            Reply
    1. Conrad

      UK doctors and patients aren’t exposed to drug marketing? All those millions of dollars spent on marketing have to have some effect surely

      Reply
    2. el_tel

      Given the populations of the USA and Canada, their figures are broadly similar per capita. However, the UK is indeed very different. I have an anecdote that partly supports the comment by Conrad.

      In 1997 I worked for a (UK) consultancy which was contracted by the Canadian arm of the pharma company that produced the fentanyl patch. The project was to show cost-effectiveness compared to morphine and its derivatives when you took into account the much lower costs (need for) laxatives etc. The docs I interviewed worked in end-of-life care and were spread across Ontario and, curiously, most were British expats. All thought the UK was barbarous in terms of underuse of pain-killers in palliative/cancer care and were clearly very heavily “pro fentanyl patch”. Now, whilst this use is OF COURSE radically different from the subsequent opioid scandal (partly involving fentanyl), to me in retrospect it showed the radically different attitudes toward pain management in North America compared to the UK. NB I think the UK has traditionally under-used pain-killers so this isn’t cheer-leading – there is a happy medium here somewhere. Plus the UK has had its own scandal involving pregabalin.

      Indeed I wonder if the referenced study were done in the UK and looked at pregabalin we might see some parallels.

      Reply
      1. PlutoniumKun

        There are huge cultural differences in prescribing pain relief between the US and Europe, and indeed within Europe. A relative of mine is actually conducting research on this right now, it should be published this year I think.

        There is an arguable case I think that there is too much of a reluctance to give out opiates in a controlled environment – when I had an accident 10 years ago I was given one course and then flatly refused any more, even when I was in hospital (damn, they were good, they really worked). But there is an ongoing attempt by the pham industry to promote ‘pain clinics’ on this side of the Atlantic too.

        Reply
        1. el_tel

          Great that your relative has been investigating this and I agree totally that opiates have been used too little in controlled environments.

          I’ve heard about the promoted pain clinics. Whilst in practice they are, as you say, being used by the pharm industry to make us “more like the USA”, it irks me that in principle they could be doing a lot to alleviate pain using non-opiates whose properties are well-known. One example in primary care is how few GPs know that amitriptylline (at low, non-anti-depressant doses) has been a mainstay of pain relief in secondary care for decades and is a real wonder drug. Of course this is down to drug reps plugging new fancy meds, many of which have turned out to be of marginal benefit or downright awful in key subgroups of people…

          Reply
    1. Angie Neer

      This is anecdote, not data, but some years ago I was sent to an oral surgeon to have wisdom teeth extracted. The assistant who escorted me to the chair said, “OK, while the doctor is working on you we’ll set up a prescription for your medicine, so you can go straight to the pharmacy and pick it up afterward. What’s your favorite painkiller?” My favorite? Is that how we prescribe things now? I had never needed painkillers, so I said I don’t know. They went ahead and prescribed me one of those big-name opioids, I don’t remember which. But I never took one, because I didn’t have much pain. The incident illustrated to me that 1) they had adopted the big pharma line that these things are low-risk so you can hand them out like candy and 2) there is so much pressure to get the patient out the door, and so little consideration for cost, that it’s better just to ‘scrip us up on speculation.

      Reply

Leave a Reply

Your email address will not be published. Required fields are marked *