Faulty Opioid Overdose Data

“The fault lies not in our stars, but in ourselves,” is a commonly referenced line from Shakespeare’s Julius Caesar. It means we’re responsible for our actions. Fate is just an excuse. When we apply this logic to data, we can just as well infer that the inputs we use and how we format the data tell more than the data itself.

This is particularly true for data on the opioid crisis. Currently, we’re enamored with mortality and overdose rates. We focus on these two because they’re easy to measure and readily available. But they are a poor gauge of the crisis.

Addiction is a complex and multifaceted condition. It’s challenging to accurately measure. It can involve both physical and psychological dependence. And it can manifest in different ways in different people. We still don’t have a consensus understanding of the term. What we call an addiction is really a dependency. Where one ends and the other begins is hotly contested. But that distinction matters. How an addiction forms matters more than how it ends. Yet we only focus on the most glaring outcomes. As a result, our focus diverts away from identifying more meaningful metrics.

Addiction is best explained as a series of decisions with social and clinical implications, all overlapping in complex, dynamic ways. To measure this, we have to challenge our notions of clinical data. We need new ways of thinking so we can create new metrics to measure.

This includes factors like the prevalence of social stigma in different areas and varying access to treatment centers. Some struggling with addiction are more likely to seek treatment but cannot due to a lack of availability, while others may have easy access to treatment but might be reluctant to seek help because of social stigma. In both cases, the patients are struggling with addiction, but the underlying behaviors are different, so the mode of treatment should be different as well.

These dynamic differences in behavior are only a part of the issue. The other is how we understand addiction as a medical condition. Surprisingly, it varies across the country.

These discrepancies lead to differing diagnostic criteria and classification systems of addiction. So patients receive varying quality of care in different regions. While the diagnostic criteria for addiction will change as our understanding of addiction evolves, it should change uniformly and according to clinical studies, not through inconsistent diagnostic patterns.

Never has it been more important to find new methods of measuring addiction. Opioid settlements have already begun disseminating, but few truly understand how it should be allocated. States designated committees to decide where funds should go, but so much remains unknown – because we still don’t know how to measure addiction.

The severity of this problem is exemplified in the following decision – which would net a greater benefit to a community, investing in a new treatment facility or offering more training for law enforcement. Unfortunately, there’s no consistent answer. Each community has different forms of addiction emerging out of unique patterns of behavior, even for communities afflicted with the same substance of abuse. In some communities, a new treatment facility would be useful. In others, greater access to harm reduction medications would be beneficial. There’s no single answer because that’s not how addiction works.

This may appear novel to some, but to those living through the opioid crisis, this is common knowledge. But, as the saying goes, common sense is anything but common. Those with the greatest insights, the lived experiences, are often the ones left out of key decisions.

If we don’t have input from those most affected by this crisis, then we risk using the same metrics with the same data models. It’s inevitable. Policy doesn’t change because the same decision-makers develop new ways of thinking. Policy changes with new input from new sources. But with the same decision-makers in place, we’ll have the same results. And they’ll justify their actions through their data.

It’s not the data’s fault. The fault lies with those who wrongly interpret and format the data. If we don’t include the input of those with lived experiences to improve how we measure data, then the fault will lie only in ourselves.