Adding a New Tool to the Pain Management Toolbox

Anecdotal accounts and clinical studies alike indicate that primary care physicians and other clinicians who use opioid medications to treat patients for chronic pain are struggling to implement uniform and effective abuse and misuse assessment and risk management protocols. Urine drug testing is an important part of this process, but many non-pain specialists need additional guidance and resources to support their decision-making processes in terms of administering the tests and interpreting the results.

In this Q&A, Harry L. Leider, MD, MBA, FACPE, Chief Medical Officer and Senior Vice President of Ameritox, discusses the role of urine drug testing in pain management, and explains how Rx Guardian CD, a reference database and normalization algorithm, can help clinicians manage risk in their patients being treated with opioids.

What is the role of urine drug testing in monitoring patients for misuse, abuse, and diversion of opioid medications, and why is it such a useful tool for this purpose?

It’s pretty well established that even the best clinicians who treat patients with chronic opioid therapies cannot reliably determine whether their patients are consistently adhering to treatment, or whether they might be misusing, abusing, or diverting their medications, or even just being nonadherent to the medication plan because they’re having adverse side effects or are concerned about addiction. There are several good studies around that show that physicians are wrong 30-40% of the time, because patients often forget to take their medications as prescribed, and are very good at deceiving their physician. Because of this, organizations such as the American Pain Society (APS), the American Academy of Pain Medicine (AAPM), and the American College of Occupational and Environmental Medicine (ACOEM) have suggested that routine monitoring using urine drug testing (UDT) is recommended as standard practice.

The American Academy of Pain Medicine (AAPM), and the American College of Occupational and Environmental Medicine (ACOEM) have suggested that routine monitoring using urine drug testing (UDT) is recommended as standard practice.

ACOEM Opioid Guidelines http://hcp.lv/n39qnA

When clinicians talk about UDT, they are usually referring to what I call “firstgeneration” or “standard” UDT, where you’re looking for a “yes or no” result on tests for illegal drugs like cocaine, heroin, or amphetamines; tests for opioids that you have prescribed for the patient; tests for other opioids that the patient may have obtained illegally or from another physician; and tests for other substances that might be a problem when combined with opioids, such as benzodiazepines or barbiturates. The laboratory technologies that exist todayimmunoassay screening as the first step, followed by mass spectrometry if there’s a positive on the screen—are generally very good and very sensitive at picking up small amounts of any kind of drug, legal or illegal.

Both negative and positive test results can provide useful information to the treating physician? What does a positive result not tell the physician?

Generally, a negative test is helpful, because it can show that the patient has not taken that legal or illegal drug in the last several days. If you’ve been prescribing them an opioid, and the drug doesn’t show up on the test, then that’s a concern. Maybe it’s sitting unused in their medicine cabinet for now, but one day when they’re in a lot of pain they may take too much and overdose. Or maybe they’re diverting it.

The problem with the first-generation testing is that a positive result for a prescribed drug is not necessarily comforting. Let’s say you have a patient on chronic morphine or oxycodone and you do a standard UDT, and it comes back positive for that drug. It can’t tell you whether the patient is taking the full, correct dose you prescribed, or whether they figured out that if they take one pill the night before they come in for testing they’ll still test positive, and then they can sell the other 59 pills or give them to someone else. A positive result for the opioid you have prescribed, while reassuring you that the patient is taking something, doesn’t tell you that they are taking the correct amount, or they aren’t abusing, misusing, or diverting it.

Even the best clinicians who treat patients with chronic opioid therapies cannot reliably determine whether their patients are consistently adhering to treatment, or whether they might be misusing, abusing, or diverting their medications.

How can Rx Guardian help in this regard? What is it, and how does it work?

Both the first version of Rx Guardian and the latest enhancement, Rx Guardian CD, were developed to address an unmet need for physicians treating patients with chronic opioid therapy. It can help physicians get a better sense of whether a patient is taking the drug that they have been prescribed, and also provide additional information about whether the patient is likely to be taking it in the correct manner.

Rx Guardian is based on the work of a researcher who figured out that urine drug levels can vary due to the size of the person and how much fluid they drink. This researcher did some studies and came up with a formula that enabled clinicians to account for physiologic variation and other factors that can affect urine drug level. Ameritox licensed this proprietary algorithm and branded it as Rx Guardian. We’ve been offering this tool for the last five or six years at no extra charge to clinicians who use our lab for UDT, in order to help them make decisions about patient care. We’ve had good results with it, and people seem to value it. However, it relies on patient data that was accumulated in the mid 1990s, and without access to the data we had no way to further expand and enhance the tool.

About a year and a half ago, we learned that clinicians at the Marshfield Clinic in Wisconsin (http://hcp.lv/nqKsT0) had been doing some interesting research in UDT and risk monitoring, and had established a database that included every pain patient treated at the clinic. They conducted a rigorous assessment of whether each patient was taking his or her opioid correctly; they asked patients how and when they took their medications, looked at patients’ prescription refill history, screened for aberrant behaviors and investigated all occurrences, looked at addiction history and criminal records, and even checked with patients’ local pharmacies to see if there were any problems. Because they followed most of these patients over time, they could track results from a series of urine drug tests and look for evidence of nonadherence (ie, positive screens for illegal drugs, positive and/or negative screens for prescription opioids or other controlled substances).

Patients in the program who got through all those questions and had the expected results from UDT were flagged as highly likely to be adherent to their prescribed opioid regimen. The Marshfield Clinic accumulated a database of more than 1,000 patients on opioids who met the criteria. From that database, they normalized patients’ urine drug levels, adjusted them with a patented algorithm that uses urine creatinine as the variable to correct for hydration levels. They patented this algorithm. From that, Ameritox did some additional analysis to create reference ranges that were commercialized as Rx Guardian CD.

With this new tool, samples sent to Ameritox undergo standard testing, which identifies a quantitative drug level using mass spectrometry. This gets run through the algorithm, which calculates an adjusted/normalized drug level, and compared to the reference range of drug levels in the Marshfield database for adherent patients on that same drug.

Rx Guardian CD creates a statistic called the standard score for that drug, which is plotted on a bell-shaped curve. Ninetyfive percent of adherent patients in the database are between -2 and +2 standard scores on the bell-shaped curve; 99% are between -3 and +3. That allows clinicians to benchmark or judge their patients’ level of adherence. One key caveat to keep in mind is that Ameritox makes it quite clear that this data should not substitute for sound clinical judgment; you still have to monitor and assess the patient for aberrant behaviors and look for other potential problems. This is just additional information that provides some sense of the likelihood that a patient’s urine drug levels resemble other adherent patients prescribed that drug.

It’s not forensic, but rather more of a highly refined red flag?

That is correct. The reality is that busy physicians, whether primary care or pain specialists, have limited time in which to evaluate patients. Especially pain specialiststhe demand for their services is very high, and if they’re doing procedures, it’s even more time-consuming. Physicians who treat patients for pain have a relatively short amount of time to make a decision: Am I going to ask a patient a few basic questions about how things are going? Or am I going to spend an additional 15 minutes to really investigate how they’re taking their meds, screen for aberrant behaviors, and investigate any potential addiction risks? Rx Guardian CD is a tool that can help busy clinicians flag situations where it makes sense to invest that time, knowing that they’re sometimes going to find a problem, and sometimes not.

What about false positives, false negatives, and issues relating to sensitivity and specificity?

We’ve been doing some validation work on sensitivity that is likely to be presented this fall, but the sensitivity of this test to identify adherence looks to be quite high. We’re still working on defining the false negative rates. The vast majority of patients who are adherent should fall within the reference range, so a value outside of it means that there is a low likelihood that that patient is adherent and may require further inquiry. But you also can’t say that every patient who is within +2 or -2 is definitely adherent. You’re going to get some false positives.

There’s always a tradeoff between sensitivity and specificity. Inherently, a test that has good sensitivity to predict adherence, as it looks like this test does, is always going to have a bit lower specificity. We’re exploring that, too. The practical thing to remember is that for patients who are outside of the reference ranges, statistically the likelihood that they are adherent is low, but not zero. That’s why it’s a useful tool for clinicians who work with higher-risk populations. If a patient is within the reference range, but your clinical suspicions are highthey’ve got aberrant behaviors, or there are other reasons for concernyou don’t want to assume everything’s fine just because the patient’s within the reference range.

The growing consensus is that you should test all patients before you start chronic opioid therapy, and then monitor low-risk patients a minimum of twice a year, and highrisk patients a minimum of four times a year.

Does Ameritox offer resources to help clinicians interpret the test results?

We have multiple layers of resources and educational tools to help with interpretation, including a color-coded Rx Guardian interpretation guide that helps clinicians understand the differential diagnosis, as well as potential follow-up steps for an abnormal test result. All of our clinicians have access to this tool. We have a team of master’s-level toxicology specialists who are available by telephone to answer clinicians’ questions; you can call the Toxicology Hotline while the patient is still in your office, and ask a question or get an interpretation of the report. We also have a group of medical science liaisons who are PharmD- or PhD-level experts who can meet with clinicians to educate them about the science of UDT and help them make better use of the technology.

What are your thoughts on patient selection for urine drug monitoring? Should UDT be used with all patients who will be treated with opioid medications for pain, or only with those patients who may be at higher risk for abuse, misuse, and diversion?

The 2009 APS/AAPM guidelines for chronic opioid therapy management (see sidebar) say you should consider routine monitoring in low-risk patients, and strongly consider monitoring in high-risk patients. That being said, the guidelines don’t really define high-risk vs. low-risk patients, nor do they define what “routine” means. The growing consensus across pain management experts and state regulatory bodies is that you should test all patients before you start chronic opioid therapy, and then monitor low-risk patients a minimum of twice a year, and high-risk patients a minimum of four times a year. The reason for this is that the literature suggests that physicians cannot accurately determine the behavior of patients. The state of Florida’s board of medicine requires testing twice a year. Washington State, Utah, and Louisiana have some broad language about routine testing. Some professional societies recommend testing at least a couple of times a year. We have a group of scientific advisors who are some of the leading pain management experts in the country who have supported the idea of initial testing and follow-up monitoring 2-4 times per year.

So, there’s a growing consensus of regular monitoring. The key questions are what frequency is optimal for testing, and are the tools we have today for assessing risk (ORT, SOAP, etc) sufficient?

If that’s the case, why do you think more clinicians, especially primary care physicians, are not using urine drug testing to assess their pain patients for risk of misuse and abuse?

I’m a primary care physician by background, and I’ve learned a lot about pain in the last few years working with Ameritox. I think the reason for low rates of UDT use in many primary care practices is simply lack of knowledge. There are also several practical considerations involved—a pain specialist might be managing 300-500 patients, 60-80% of whom are on opiates, whereas a primary care doctor might have a panel of 2,500-3,000 patients, only a small percentage of whom may be on chronic opioid therapy. Many primary care physicians are not reading the journals that propagate the guidelines, and they’re not going to the national meetings around chronic pain and opioid therapy. It’s certainly not that they’re negligent; it’s just that they don’t always have the time to acquire this knowledge because they’re also managing patients with diabetes, heart disease, asthma, and a host of other conditions. Chronic pain is just a small slice of what they do.

Study shows Rx Guardian CD is applicable across diverse patient populations

Study results presented at PAINWeek 2011 (http://hcp.lv/pCwWMa) show that Rx Guardian, “when combined with other clinical indicators of non-adherence, should enhance clinicians’ ability to help identify potential non-adherent patients using chronic opioid therapy” (http://hcp.lv/o4kRwZ).

There is a learning curve to figuring out what to do with pain patients, interpreting results of UDT and talking to patients about those results. In primary care, the low frequency of chronic pain patients combined with this learning curve means that many physicians don’t know how to do this or don’t invest the necessary time. For pain management specialists—who are going to the meetings and reading the journals; who are aware of the concerns over abuse, misuse, and diversion, and who understand that the data shows that you can’t rely on intuition alone to identify high-risk patients—it’s a whole different dynamic. With upwards of three-quarters of patients on chronic opiates, specialists have a reason to put into place a systematic risk assessment and monitoring program.

More primary care doctors are starting to do this. We have an active effort underway to talk to primary care practices and help educate them about all of this. Approximately 40% of our current users are primary care physicians.

How do you respond to critics of the use of urine drug testing who claim that it creates an atmosphere of mistrust and puts a strain on the patient-physician relationship?

It’s been our experience that it depends on the relationship between the patient and physician. The clinicians I’ve met for whom this is not problematic are the ones who build testing and monitoring into patients’ expectations before starting therapy with opioids, and include it as part of the pain management agreement.

They’ll tell their patients, “Because other medications and other therapies have failed to provide the pain relief you need, you might be a good candidate for opioid medications. We should give opiates a try if you’re up for it, but we can’t be sure how they’re going to work for you and how you’re going to tolerate them. In order to do this, we’re going to have to agree on a few things: you’re going to get your pain medications from only one doctor; you’re going to get your medications from only one pharmacy so we can keep track of everything and make sure you don’t get into trouble with side effects; you’re not going to change the dose of your medication without discussing it with me, because that can be dangerous; and periodically I might test you, not because I don’t trust you, but because I need to be comfortable that the combinations of drugs and doses of drugs you’re taking aren’t dangerous, especially if you’re getting medications from other physicians.”

When you don’t have a uniform protocol in place, when you don’t do this up front, then you run the risk of the patient perceiving urine drug testing as punitive, and not as an essential component of pain care when prescribing drugs like opioids that have a lot of potential for misuse and abuse.