Geographic Hotspots for Opioid Prescribing

Opioid prescribing rates vary dramatically across the country. Some states average rates 2-3x higher than others — patterns that have persisted for years despite national reform efforts. Understanding where opioids are prescribed most heavily, and why, is essential to evaluating prescriber risk and targeting interventions where they matter most.

Highest Opioid Prescribing States

The Appalachian Corridor

The highest-prescribing region in the United States follows the spine of the Appalachian Mountains from West Virginia through eastern Kentucky, southwestern Virginia, and into parts of Tennessee and Ohio. This is not a coincidence. For decades, the Appalachian economy depended on coal mining and heavy manual labor — industries that produce chronic musculoskeletal injuries at extraordinary rates. Workers with back injuries, knee damage, and shoulder problems were funneled into pain management systems that relied heavily on opioid prescriptions.

The disability rates in Appalachian counties tell the story. In some counties of eastern Kentucky and southern West Virginia, Social Security Disability Insurance (SSDI) enrollment exceeds 15% of the working-age population — rates three to four times the national average. These are communities where a generation of workers developed chronic pain from physically demanding jobs, and where opioid prescriptions became a form of de facto disability management. The coal industry's decline compounded the problem: as mines closed and jobs disappeared, the pain remained, but the economic resources to pursue alternatives to opioid therapy did not.

When Purdue Pharma launched OxyContin in 1996, its sales representatives specifically targeted regions with high rates of workplace injury and disability claims. Appalachia was ground zero. The region had a large population of patients with legitimate chronic pain, a limited supply of specialists, and primary care physicians who were under-resourced and overwhelmed. Opioids offered a straightforward solution to complex pain management problems. Internal Purdue documents later revealed that the company identified Appalachian providers as high-value targets precisely because they treated populations with elevated pain levels and had fewer non-pharmacological treatment options available.

By the mid-2000s, pill mills had proliferated across the region. Clinics in states like Florida, Kentucky, and West Virginia dispensed millions of doses with minimal oversight. Some providers wrote hundreds of opioid prescriptions per month. Pain clinics in parts of Appalachia operated on a cash-only basis, required no medical records, and performed no physical examinations — patients simply paid a fee and received a prescription. The business model was prescription volume, not patient care.

The DEA eventually shut many of these operations down through a series of high-profile enforcement actions in the late 2000s and early 2010s, but the prescribing culture they created proved far more durable than the clinics themselves. Even today, providers in Appalachian states prescribe opioids at rates well above the national average, and the region continues to lead the country in opioid-related claims per capita. The legacy of the pill mill era is baked into the patient population — thousands of Medicare beneficiaries in the region have been on chronic opioid therapy for over a decade, and tapering them safely is a clinical challenge that no guideline can simplify.

The Southern Belt

Beyond Appalachia, a broader pattern emerges across the American South. States like Alabama, Mississippi, Louisiana, Tennessee, and Arkansas consistently rank among the highest for opioid prescribing in Medicare Part D data. Several factors converge to explain this pattern.

First, the South has an older and sicker Medicare population on average. Rates of obesity, diabetes, cardiovascular disease, and arthritis are higher across the region, creating more demand for pain management. Second, the South has historically had fewer pain management specialists per capita, meaning that primary care providers shoulder more of the burden of treating chronic pain — often with the tools most readily available to them, which are opioid prescriptions.

Third, several southern states were late to adopt robust prescription drug monitoring programs. While states like New York and Kentucky implemented mandatory PDMP checks early, states like Alabama, Mississippi, and Missouri lagged behind, allowing high-volume prescribing to persist with less oversight. Finally, pharmaceutical industry marketing spending was disproportionately concentrated in the South, particularly for extended-release opioid formulations promoted for chronic non-cancer pain.

The result is a prescribing culture that developed over decades and has proven resistant to change. Even as national prescribing volumes decline, the relative ranking of southern states has remained largely stable. A provider in Alabama still operates in a fundamentally different prescribing environment than a provider in Minnesota.

Urban vs Rural Divide

One of the most consistent findings in opioid prescribing data is the gap between rural and urban areas. Rural counties across the country have opioid prescribing rates roughly 1.5 to 2 times higher than urban counties, even after adjusting for age and health status. This pattern holds within states — rural Alabama prescribes more than Birmingham, rural Oregon more than Portland.

The reasons are structural. Rural areas have fewer providers overall, meaning patients often see the same physician for everything from diabetes management to chronic back pain. These generalist providers may have less training in multimodal pain management and fewer referral options for physical therapy, cognitive behavioral therapy, or interventional pain procedures. When the nearest pain clinic is two hours away, a prescription becomes the default.

Telehealth has narrowed this gap in some specialties, but pain management remains difficult to deliver remotely. Physical examinations, urine drug screens, and the clinical judgment required for opioid prescribing decisions are hard to replicate through a screen. Rural providers who want to refer patients to non-opioid pain management often find that the services simply do not exist within a reasonable driving distance. A 2023 analysis found that over 60% of rural counties in the United States lacked a single board-certified pain medicine specialist.

Rural areas also have dramatically fewer addiction treatment resources. Patients who develop opioid use disorder in rural counties face long drives to reach a provider authorized to prescribe buprenorphine, and inpatient treatment facilities are sparse. This creates a paradox: the areas with the highest opioid prescribing rates often have the fewest options for treating the addiction that prescribing can create.

Medicare data reflects this divide clearly. When we examine provider-level risk scores, rural providers are overrepresented among high-opioid prescribers — not necessarily because they are irresponsible, but because they operate in systems with fewer alternatives. Understanding this context is essential to interpreting any provider-level prescribing metric — flagging a solo rural family physician for an elevated opioid rate without accounting for the absence of referral options misses the point entirely.

The CDC Prescribing Guidelines

In 2016, the CDC published its first Guideline for Prescribing Opioids for Chronic Pain, a landmark document that recommended limiting opioid doses to under 90 morphine milligram equivalents (MME) per day, avoiding opioids as first-line therapy for chronic non-cancer pain, and prescribing the lowest effective dose for the shortest duration necessary. The guidelines were intended as recommendations, not regulations, but their impact was enormous.

Insurers, state medical boards, pharmacy chains, and health systems translated the guidelines into hard limits. Many imposed prior authorization requirements for opioid prescriptions above certain thresholds. Some states passed laws capping initial opioid prescriptions to three or seven days. The effect was a measurable decline in opioid prescribing volume nationally — total opioid prescriptions dispensed fell by over 40% between 2012 and 2022.

However, the 2016 guidelines also drew criticism for unintended consequences. Patients with chronic pain who had been stable on opioid therapy for years were abruptly tapered or cut off. Some providers, fearful of regulatory scrutiny, stopped prescribing opioids entirely. The CDC acknowledged these problems in its updated 2022 Clinical Practice Guideline for Prescribing Opioids, which removed the specific MME thresholds and emphasized individualized patient care, the importance of not abruptly discontinuing opioids in patients with established therapy, and a more nuanced approach to the benefits and risks of opioid treatment.

The geographic impact of these guidelines was uneven. States that already had lower prescribing rates saw modest further declines. States with the highest rates — the Appalachian and southern states — saw larger absolute drops, but their relative position in the rankings barely changed. The guidelines narrowed the gap somewhat, but did not close it.

State-by-State Variation

The tables above show the endpoints of a wide spectrum, but the middle tells an important story too. The difference between the highest-prescribing state and the lowest is not gradual — it is a steep curve with a long tail. A handful of states cluster at the top with dramatically elevated rates, a large group sits near the national average, and a smaller group of states (typically in the Northeast and upper Midwest) prescribe at rates well below the mean.

This variation matters for provider-level analysis. A family medicine physician in West Virginia who prescribes opioids at 15% of claims may be entirely typical for their state, while the same rate in Minnesota would place them in the top percentile. Raw prescribing rates without geographic context are misleading — they punish providers for practicing in high-prevalence areas and give a pass to providers in low-prevalence areas who may still be prescribing inappropriately relative to their peers.

State-level differences also reflect policy choices. States with mandatory PDMP queries, limits on initial opioid prescriptions, and requirements for co-prescribing naloxone tend to cluster at the lower end. States without these policies, or that implemented them later, tend to cluster higher. The data does not prove causation — states that pass aggressive opioid legislation may already have cultures that discourage opioid use — but the correlation is consistent across multiple years of Medicare Part D data.

Border State Patterns

States along the US-Mexico border — Texas, New Mexico, Arizona, and California — present a different pattern. Their opioid prescribing rates tend to fall near or below the national average for Medicare providers. However, these states face a distinct challenge: the availability of illicit fentanyl and heroin crossing the border means that prescription opioid data alone underestimates the full scope of opioid exposure in these communities.

In border states, the opioid crisis increasingly involves illicit supply rather than prescription diversion. Fentanyl seizures at the border have increased dramatically since 2019, and overdose deaths involving synthetic opioids have risen even as prescription opioid volumes declined. For Medicare beneficiaries specifically, the prescribing data we track on OpenPrescriber captures only the legal prescription side of this equation. A provider in El Paso may prescribe opioids conservatively, but their patients may still face significant exposure risk from non-prescription sources.

This pattern highlights a limitation of prescription-based analysis: it can only measure what goes through pharmacies. In areas where the illicit supply is dominant, prescription monitoring becomes a less complete picture of opioid risk. Our ML fraud detection models account for this by weighting prescription-based signals alongside other indicators of anomalous prescribing behavior.

What Changed Since 2016

The national story since 2016 is one of paradox: opioid prescribing volumes have fallen substantially, but opioid overdose deaths have risen to record levels. In 2016, approximately 214 million opioid prescriptions were dispensed in the US. By 2022, that number had fallen below 140 million — a reduction of over 35%. Yet annual opioid overdose deaths nearly tripled over the same period, driven almost entirely by illicit fentanyl and its analogues.

This divergence is critical for interpreting the geographic data on OpenPrescriber. A state that shows declining prescription opioid rates may not be getting safer — it may be transitioning from a prescription-driven crisis to an illicit-supply-driven crisis. West Virginia, for example, has seen meaningful reductions in opioid prescribing since its peak, but continues to lead the nation in overdose deaths per capita. The prescriptions went down; the deaths went up.

The composition of prescribing has changed as well. High-dose prescriptions (above 90 MME per day) have declined more steeply than low-dose prescriptions, which is a positive signal for patient safety. But the patients who were abruptly tapered from high-dose regimens did not all transition smoothly — some turned to illicit sources, some experienced undertreated pain, and some entered a cycle of emergency department visits that created its own set of costs and risks. The 2022 CDC guideline update explicitly addressed forced tapering as a harmful practice, but the damage from the post-2016 period is still visible in the data.

For Medicare beneficiaries specifically, the shift to illicit supply is less pronounced than in the general population — Medicare patients are more likely to obtain opioids through legitimate prescriptions than through street purchases. This makes prescription-level monitoring, like what OpenPrescriber provides, particularly valuable for the Medicare population. The prescribing data is a more complete picture for this demographic than for younger populations where illicit supply dominates.

The Treatment Desert

Perhaps the most troubling geographic pattern is the overlap between high-prescribing areas and low-treatment areas. Counties with the highest opioid prescribing rates per capita are often the same counties with the fewest addiction medicine providers, the fewest buprenorphine waiver holders, and the fewest inpatient treatment beds. SAMHSA data shows that over 40% of US counties have no buprenorphine prescriber at all, and these counties are disproportionately located in the same Appalachian and southern regions that lead the country in opioid prescribing.

This creates a devastating feedback loop. Patients in these areas are more likely to receive opioid prescriptions, more likely to develop opioid use disorder, and less likely to have access to evidence-based treatment when they do. Methadone, the other FDA-approved medication for opioid use disorder, is even less accessible — it can only be dispensed through certified opioid treatment programs (OTPs), which are concentrated in urban areas. A Medicare beneficiary in rural Mississippi who develops opioid use disorder may have no methadone clinic within a two-hour drive.

The 2023 elimination of the federal X-waiver requirement for prescribing buprenorphine was intended to address this gap by allowing any DEA-licensed provider to prescribe buprenorphine for opioid use disorder. However, early data suggests that uptake in rural and high-prescribing areas has been slow — the providers who could now prescribe buprenorphine are often the same overworked generalists who lack the time, training, and support to add addiction treatment to their practice. Removing a regulatory barrier is necessary but not sufficient when the underlying workforce shortage remains.

The state-by-state profiles on OpenPrescriber show this mismatch clearly. States at the top of the opioid prescribing rankings rarely appear at the top of addiction treatment provider rankings. Addressing this geographic mismatch — ensuring that the areas generating the most opioid prescriptions also have adequate resources to treat the consequences — is one of the most important policy challenges in the opioid crisis.

How Our Risk Model Uses Geography

OpenPrescriber does not flag providers simply for practicing in high-prescribing states. A raw geographic comparison would unfairly penalize every provider in West Virginia and give a free pass to outliers in New York. Instead, our risk model uses geography as a contextual adjustment — comparing each provider against peers in the same state, same specialty, and similar practice setting.

A family medicine provider in Alabama is compared against other family medicine providers in Alabama, not against dermatologists in Massachusetts. This specialty-and-geography-adjusted approach means that a provider flagged as high-risk in our model is prescribing more opioids than their local peers would expect — a signal that is meaningful regardless of whether the provider practices in a high-prescribing or low-prescribing state.

Geography enters the model in other ways as well. Providers in states without mandatory PDMP checks receive a slight contextual adjustment, since the regulatory environment is more permissive. Providers in designated Health Professional Shortage Areas (HPSAs) receive adjustments reflecting the limited referral options available to them. The goal is to distinguish providers who are genuinely prescribing anomalously from providers who are simply responding to the conditions of their practice environment. You can explore these adjustments in the Risk Explorer.

Policy Implications

The geographic persistence of opioid prescribing patterns suggests that national one-size-fits-all interventions have limited effectiveness. The CDC guidelines reduced overall volume, but the states that prescribed the most before the guidelines still prescribe the most after them. This has led to growing interest in geographically targeted interventions that account for regional differences in healthcare infrastructure, patient demographics, and prescribing culture.

State prescription drug monitoring programs (PDMPs) remain the primary tool for real-time prescribing oversight. However, PDMP effectiveness varies enormously. States that mandate providers check the PDMP before every controlled substance prescription see measurably lower prescribing rates than states where checks are voluntary. Interstate data sharing — allowing a provider in Tennessee to see a patient's prescriptions filled in Kentucky — has improved through platforms like PMP InterConnect, but remains incomplete. As of 2024, not all states share data in real time, and some border-region patients continue to fill opioid prescriptions across state lines without detection.

DEA enforcement actions have shifted from targeting individual pill mills (the approach of the 2010s) toward broader distributor-level interventions, investigating pharmaceutical wholesalers that shipped suspicious volumes to specific pharmacies and regions. The landmark settlements with distributors McKesson, Cardinal Health, and AmerisourceBergen, along with manufacturer settlements with Purdue Pharma and others, have directed billions of dollars toward affected communities — though the allocation and effectiveness of those funds remains contested. Many public health advocates argue that settlement funds have been diverted to general government budgets rather than invested in addiction treatment infrastructure where it is needed most.

State-level legislative approaches have also diverged. Some states have implemented prescribing caps — limiting initial opioid prescriptions to three, five, or seven days. Others have required co-prescribing of naloxone when opioid doses exceed certain thresholds. Still others have focused on expanding access to non-opioid pain management through insurance mandates covering physical therapy, acupuncture, and cognitive behavioral therapy. The evidence base for which approach is most effective is still developing, and the optimal policy mix likely varies by state.

At the provider level, the most promising interventions are peer comparison programs. When prescribers receive data showing how their opioid prescribing compares to peers in their specialty and region, prescribing behavior changes measurably. A randomized trial of peer comparison letters sent to the top 5% of opioid prescribers in Medicare found a sustained reduction in prescribing volume that persisted for over two years. This is precisely the type of comparison that OpenPrescriber enables — transparent, public data that allows providers, patients, and policymakers to see where prescribing patterns fall outside expected norms.

The Persistence Problem

Despite over a decade of interventions — CDC guidelines, state monitoring programs, prescriber limits, and enforcement actions — the geographic distribution of opioid prescribing remains remarkably stable. The same states that led in 2013 largely still lead today. Prescribing culture, once established, changes slowly. Provider training, patient expectations, local formulary norms, and regulatory environments all reinforce existing patterns.

This persistence is why geographic context matters so much in prescriber analysis. A provider's prescribing rate means very little without knowing where they practice. OpenPrescriber provides that context — not to excuse high prescribing, but to make comparisons meaningful and interventions targeted. The goal is not to eliminate geographic variation entirely, but to identify providers whose prescribing is anomalous even within their own context, and to give communities the data they need to understand and address the prescribing patterns in their region.