Patients with Prior Overdose Still Prescribed Opioids & Benzos

 

 

 

 

I read an interesting article in the latest issue of Journal of Addiction Medicine, titled “Prescribing of Opioids and Benzodiazepines Among Patients With History of Overdose,” by Griggs et al.

This article described a retrospective chart review of patients who had a previous history of opioid or benzodiazepine overdose. They identified patients who were prescribed either an opioid or benzodiazepine in a one-month period, in 2015, then reviewed their charts to see how many of these patients had a previous overdose. Then they studied the patients and prescribing situations to see what they had in common.

This study was done at a large healthcare system based in Charlotte, NC. The system is based at the same hospital where I did my residency in Internal Medicine about a billion years ago. OK, maybe it was only thirty-two years ago, but it feels like another lifetime. This hospital system has a robust Addiction Medicine department now, led by Dr. Stephen Wyatt, an addictionologist of national and perhaps international renown, who co-wrote this study.

The article began by reminding us of the recent increase in morbidity and mortality with opioid use disorder. Then it cited another article that I have written about (see my blog of January 23, 2016) authored by Larochelle et al., 2015, where it was found that in patients who survived an opioid overdose, 91% resumed opioid prescription within the next nine months.

Based on those previous findings, this study proposed to examine the prevalence of prior overdose among patients being prescribed benzodiazepines and/or opioids, and to examine patient and healthcare characteristics in these circumstances.

The study found 543 patients with prior opioid or benzodiazepine overdose history who were prescribed benzodiazepines or opioids during the designated month of the study. All the providers involved in this study use the same electronic medical record (EMR) which contained information about prior overdoses from 2007 forward, though no specific alerts appeared in the EMR.

Interestingly, opioids were involved in just under half of the overdose episodes among these identified patients, and benzodiazepines without opioids were involved in just over half of the overdoses.

Most of the identified patients received opioid or benzodiazepine prescriptions within two years of their documented overdose. Opioids accounted for around 72% of these prescriptions, with benzodiazepines accounting for around 23%, and 5% of the patients got both an opioid and a benzodiazepine, which is a particularly worrisome combination.

Of the patients prescribed opioids and/or benzodiazepines who had a prior overdose, 70% were between the ages of 35 to 64 years old. The leading cause, at 51%, of the prior overdose was unintentional, though 40% were suicide attempts. Many patients had mental health diagnoses: 54% had an anxiety disorder, 55% had depression, and 24% had bipolar disorder. Nearly a third, at 29%, had a diagnosis of substance use disorders.

Around a third of the opioid prescriptions were given for chronic pain issues, despite the prior overdose history. Over 25% of the opioid prescriptions were for more than 50mg daily morphine milligram equivalents. Around half of the patients had a prior drug screen in their record that was positive for marijuana, cocaine, or alcohol.

Most of the post-overdose prescriptions for opioids or benzodiazepines were given in outpatient clinics or emergency departments, but over a fourth of the prescriptions were issued after a medical phone call consultation. Only 5% of opioid or benzodiazepine prescriptions were issued from behavioral health providers, and less than 1% were from cancer care providers.

In the discussion section, the authors of this study voiced surprise that in a fourth of the patients, benzos and/or opioids were prescribed after a telephone consultation. The authors appropriately caution prescribers against this practice.

Having practiced in primary care for ten grueling years, I understand the telephone consult. Heaven help me, but sometimes I was tempted to allow medication to be called in because it would save me time and effort. It would also spare me the unpleasantness of having to see the patient in my office, and the extra time required.

I’m not intentionally being insulting to patients, but I felt patients who repeatedly asked for controlled substances were often miserable people who weren’t fun to take care of. They hurt, both physically and emotionally. I felt hopeless when I saw them, like nothing I could do or say would help them anyway, so where’s the harm in giving them a much-desired controlled substance?

Of course, now that I’m older, wiser, and better educated, I suspect many of these patients had treatable substance use disorder and/or mental health disorders.

The authors of the study concluded that providers for this patient group could have done a better job of identifying higher risk patients. The prescribers could benefit from an electronic tool, which according to the article is presently being developed, to support decision making processes and quantify the risk for a given patient.

I’ve talked in this blog before about the perils of labeling patients as “frequent flyers” or “drug seekers,” pejorative terms that create obstacles between needy patients and their providers. That old kind of labeling fosters the outdated idea that people with substance use disorders are bad, rather than sick. With that old system, patients can receive bad care, because providers stop thinking and start judging.

Instead, this article describes a better idea – one that provides information about the degree of risk for a given patient, before potentially harmful medications are prescribed. It sounds like this sort of tool can help providers mitigate risks for some patients, while not denying them appropriate medical care.

In other words, a high-risk patient with an acute pain situation, like a broken bone, may still need opioids, but fewer pills might be prescribed, with more frequent follow up, than patients at lower risk for overdose.

I don’t know if the tool this healthcare system developed is proprietary; I think I will ask for an example of how it works. I don’t work in primary care any more (addiction medicine is so much more fun), but I like to stay informed about these things.

3 responses to this post.

  1. Posted by Steve Wyatt on November 12, 2019 at 8:55 pm

    Jana, Thanks for pointing out this article. The screening tool was established by way of a CDC grant and is now being used in various forms on the Cerner EHR platform in other parts of the country beyond the Atrium Health System. It initially preformed a rapid screening of the chart for certain variables when one initiates a controlled substance prescription. The five variables include; a previous overdose with an opioid or benzodiazepine, a positive toxicology screen, 2 or more visits to the ED in which the patient received and opioid or benzodiazepine, greater than 50% of an opioid or benzodiazepine prescription should be remaining, three or more Rxs for an opioid and a benzodiazepine. It has provided realtime information to the provider that can open up a conversation they otherwise likely would not have had particularly in an emergent care situation. The functions of this tool have continued to expand. The EHR can be a frustration but this is an example of it working to help keep patients safe.

    Reply

  2. Posted by Alan Wartenberg MD on November 13, 2019 at 4:36 pm

    Previous articles on this topic have included many cases where the outpatient prescriber receives no information on the overdose because it is in a different medical system, and nobody’s computer systems communicate with many of the other systems (as well as 42 CFR part II and HIPPA issues). This tool appears to be of great value, but it needs to be accompanied by systems changes that allow the information to get to other providers in our health care system so they may be aware of the risks. Great work, Steve!

    Reply

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: