Industry Voices—5 ways longitudinal data can help fight the opioid crisis while still treating pain

Hydrocodone opioid pills
Health data can help physicians identify patients at risk of opioid abuse. (Getty/smartstock)

In spite of all the opioid prescriptions that clinicians have written and the focused efforts to tackle the country’s opioid crisis, there has been no overall change in the amount of perceived pain people are experiencing.

William Kirsh
William Kirsh, D.O.
(Courtesy of William Kirsh)

We’ve ripped off the bandage, but there is still a gaping wound.

According to a Harvard Health webcast, we are left with a persistent problem—a pain epidemic. While we’ve acknowledged the need to treat acute pain with limited to no opioid use, treatment of chronic pain—a clinically distinct entity—is limited by an individual’s insurance plan.

Current health delivery organizations are challenged to provide realistic integrated treatments that balance clinical interventions with increasing comorbid diagnoses in patients living longer with chronic conditions. A recent Washington Post article confirms that our patients are susceptible to inappropriate use of opioids to impact acute pain, coupled with minimal covered options to control chronic pain by other medications or modalities. With the advent of big data analytics, we have vast systems of knowledge and longitudinal data (anonymized information from many patients over many years) that we can use to identify and help ameliorate many of these issues. Here are five suggestions:

FREE WEBINAR | DECEMBER 13, 2018

Employer-Based Insurance: Top Priorities for 2019

Join Blue Health Intelligence (BHI) and Midwest Business Group on Health (MBGH) to hear how key descriptive, predictive and prescriptive analytics capabilities can drive new cost and quality insights for health plans, employers and benefit consultants.

1. Stay in control by developing best practices for your patients.

Using a longitudinal database can help you identify all the patients on opioids in your patient population over time. In your own practice, you can identify patients with specific diagnoses that include pain as a comorbid state. Then, take a deeper look at the patient data. How is their pain being treated? How well controlled is it? Are there other options to better manage their pain? Could the off-label use of other classes of medications (gabapentin, pregabalin or biologics like Etanercept) be just as effective?

Consider using new risk-scaling scores to document and better manage your patients with multiple comorbidities, with tools such as the “Neuropathy Pain Scale” by Bradley S. Galer, M.D., or a risk aggregation tool such as the Opioid Risk Tool (ORT) by Lynn Webster, M.D.  Some patients have chronic conditions that may necessitate long-term opioid therapy. By aggregating data longitudinally from the electronic medical record (EMR), you will be able to practice medicine on your terms. Your own EMR longitudinal database helps validate your clinical judgment. By harvesting your data, you can meet the needs of those patients who need chronic opioid therapy monitoring and treatment.

2. Identify patients at risk of abuse.

You can also use data to identify patients at risk for opioid abuse. As the Framingham study was key to identifying risk factors for developing cardiovascular disease, the ability to evaluate patient data over time could help us create a risk stratification tool for those patients with certain social determinants of health or medical conditions that may predispose them to an opioid problem. Living with chronic pain changes the brain over time—chemically and physically—leading to inherent problems with pain control due to an increased pain threshold.

In 2001, the Joint Commission first established standards for pain assessment and treatment, “in response to the national outcry about the widespread problem of undertreatment of pain.” Although this was well intended, there was no clear risk stratification of patients. We were encouraged to use interventions for pain without the full knowledge of the ramifications. While we have standardized pain scales, they say nothing about risk factors a patient may have for abuse or dependence, specific elements of past medical history or social determinants of health that may affect how they experience and tolerate pain.

Not everyone’s pain is the same. It’s time to rethink the use of a pain scale without having a comparable risk stratification, to reexamine how we capture pain scores and to use the longitudinal data we have to help categorize and risk-stratify our patients.

Distribution of opioid overdose deaths in the U.S. in 2016, by age
Click here for a full-size image. (Source:
CDC; Kaiser Family Foundation / Statista)​​​​​

3. Identify patients at financial risk.

Most patients with comorbid states have chronic pain. Unfortunately, they may also have minimal to no insurance coverage. This translates into few alternatives for pain management. If they cannot afford gym memberships, healthy food and safe housing, they will most likely not have an insurance plan that covers alternative modalities for treating chronic pain.

In short, they can’t control their pain well, they don’t feel well and many times they end up in the hospital for extensive and expensive workups. Insurance coverage (or lack thereof) predicts the behavior of your patients because it limits the scope of what the patient is able to do. If you prescribe a medication or therapy they cannot afford, your scope of practice is limited.

Until risk stratification of patients with chronic pain becomes part of a public policy to help change behavior, care and outcomes, the best way to help them is by using longitudinal data to identify those at risk to better manage the chronic conditions that may lead to costly hospitalizations. In a 2016 Pain News Network article, American Medical Association board chair Patrice Harris, M.D., said, “Insurers must cover non-opioid and non-pharmacologic therapies that have proved effective. Insurers must take a broader view to give patients and physicians more choices. These policies will save lives. That's the bottom line.”

4. Identify patients with comorbid states.

In an article from the Journal of Managed Care and Specialty Pharmacy, researchers found: “The costliest patients diagnosed with opioid abuse had high rates of preexisting and concurrent chronic comorbidities and mental health conditions, suggesting potential indicators for targeted intervention and a need for greater awareness and screening of comorbid conditions.” Some of the most common comorbidities included mental health problems, diabetic neuropathy, chronic musculoskeletal conditions and cancer.

Longitudinal datasets allow us to hone in on these patients and begin to consider better risk stratification tools for them. Even those patients coming in for an acute issue many times have an underlying chronic disease. It’s vital to be diligent with each patient interaction as we seek to understand their comorbid states and individual risk factors. The charts included in this article, from the Centers for Disease Control and Prevention and Kaiser Family Foundation, illustrate the distribution of overdose deaths and main reasons for consumption of opioids among U.S. adults.

Share of main reasons for consumption of opioids among U.S. adults as of 2017
Click here for a full-size version. (Source:
Statista Survey)

5. Find the right resources for your patients.

By using longitudinal datasets, we’re able to identify patients’ chronic conditions, comorbid conditions and social determinants of health. Having this information empowers all of us as providers to direct our patients to the best solutions for them—a  podiatrist for diabetic foot care, a psychologist for mental health concerns, an orthopedist for treating chronic back pain or a social worker to help navigate the system to find better housing or sources for healthy meals. We must, as providers, challenge ourselves to seek out resources that empower our patients while considering their unique social determinants of health.

An American Pain Society monograph noted: “Pain is subjective, so no satisfactory objective measures of pain exist. Pain is also multidimensional, so the clinician must consider multiple aspects (sensory, affective, cognitive) of the pain experience. . . no single approach is appropriate for all patients or settings.”

We must focus on our patients and their longitudinal data and use tools to determine their risk appropriately. Only then can we begin to tackle the interwoven pain and opioid crises.

William Kirsh, D.O., M.P.H., is the chief medical officer at Sentry Data Systems. He is board certified in family practice, geriatrics and hospice and palliative medicine and has been practicing medicine for more than 25 years. He is a founding partner of Sentry.