Opioid abuse can have harmful consequences, such as stigmatization, opiophobia, and the undertreatment of pain [28]. Hence, it is important that the practice of opioid prescribing strikes a balance between the extremes of widespread opioid use and opioid avoidance, wherein risk stratification is used for patient selection and the principles of addiction medicine are applied during ongoing treatment. Exposure to drugs does not create drug addicts. Rather, only vulnerable individuals who are exposed to drugs have a risk of addiction. Only individuals exposed to alcohol or opioids, who have the genetic, social, and/or psychological predisposition to addiction, actually develop a problem. There is no problem inherent to the chemical nature of opioids; rather, the growing problem of prescription drug abuse is due to the increasing use of prescription drugs among individuals not screened for risk of drug abuse. The recent problems associated with oxycodone have stemmed from the prescription of the opioid to individuals who were not assessed for their risk of drug abuse and then were treated in the context of a low-risk drug treatment paradigm. The sustained-release preparation of oxycodone approved by the Food and Drug Administration (FDA) in 1995 was thought to have much lower abuse potential, leading to the unsubstantiated belief that the risk of opioid addiction was obviated with this slower-release oxycodone [6]. Truly, abuse deterrence is only tested and proven on the streets once the product is made available.

Opioid risk management techniques must be implemented to understand the risk of drug abuse of an individual in order to better guide the decision of whether or not opioids should be used for pain control and, if so, how best to deliver the analgesic and to tailor therapy accordingly. The assessment is directed at determining whether an individual will likely take their medication as prescribed and derive better function from the ∼30–60 % pain relief that the opioids provide or whether the opioid will be used as a coping mechanism for other issues and will not lead to psychosocial gains. If the individual has a penchant for recreational drug use, prescription of opioids could lead to the abuse and or diversion of the analgesics and, at worst, addiction. Several patient factors have been found to be predictive of a patient’s risk for opioid misuse or abuse. A mental health disorder is a moderately strong predictor of opioid abuse, while a history of illicit drug and alcohol abuse or legal problems is also predictive of future aberrant drug behaviors according to a survey of 145 patients being treated for chronic pain and a systematic review of the literature [29, 30]. Tobacco use is highly prevalent among substance misusers, and the Screening Instrument for Substance Abuse Potential (SISAP) and the Screener and Opioid Assessment for Patients with Pain (SOAPP) include tobacco use as a factor in determining risk [16, 31]. While smoking has been found to increase the desire to abuse drugs in an addict population (N = 160), alternatively, smoking can be used as a form of substance replacement in those trying to abstain from drug use [32, 33]. Furthermore, individuals who have chronic pain smoke at higher rates than the general population [34]. Cigarette use has been correlated with nonspecific low back pain, fibromyalgia, and headache disorders [35–37].

All patients being considered for opioid therapy need an individualized risk assessment. Patients considered for opioid therapy to treat chronic pain need to be assessed for risk of addiction with a validated tool; there have been many devices developed to assess addiction risk in order to help clinicians make better informed decisions regarding treatment for their patients [38]. As described above, the ORT and SOAPP are good choices for many clinics [15–17]. However, whatever tool may be chosen, it is important to approach this with patient from a standpoint that there are no right or wrong answers and that this is an important step in determining a treatment plan.

Delivering opioid therapy at a lower risk begins with learning how to document cases well. Chart reviews of primary care patients in pain management indicate that oftentimes the notes are not complete enough to support continuing opioid treatment. Typical notations such as “pain stable; renew hydrocodone #240” need to be modified to include the 4 A’s of pain treatment outcomes discussed above in reference to the PADT [18–20]. This approach helps to broaden the focus of opioid effects beyond analgesia to other important aspects such as physical functioning. An example of a good chart note is, “Mr. Jones is taking hydrocodone for his chronic low back pain. His pain has reduced from severe to moderate and he is now able to attend church with his wife and help with household chores. Constipation had been noted, but he is responding to a bowel regimen; there is no evidence of aberrant drug-related behaviors.” This documentation, along with a pain-focused physical examination and corroboration from a source other than self-report such as a significant other, caregiver, urine toxicology screen, or prescription monitoring report, is enough to thoroughly support pain management with opioids.

Addiction is a disorder characterized by craving, continued use despite harm, and compulsive and out-of-control behaviors. Behaviors that are common and sometimes ambiguous can be clearly associated with addiction when they continually reoccur. Fleming demonstrated in a primary care patient population that patients who self-report four or more aberrant drug behaviors over a lifetime are more likely to have a current substance use disorder [39]. Therefore, it is important to document the occurrence of even less predictive aberrant drug behaviors, because additively, they may indicate an addiction problem. Toward this end, there is a tool available called the Addiction Behaviors Checklist that has been designed and validated to longitudinally track behaviors potentially suggestive of addiction in patients taking long-term opioid therapy for chronic pain [40]. In the meantime, the available data suggests that patients should be given a second chance when one of these behaviors less predictive of addiction is noted; only when the problem reoccurs over a 6–12 month period should opioid therapy discontinuation or a referral be considered. Some individuals with chronic pain are treatable by primary care physicians, while others may require comanagement with a specialist or complete management by a clinician with addiction medicine training.

Aberrant behaviors can be due to several different etiologies, such as pseudoaddiction, in which poorly treated pain causes patients desperate for relief to appear as if they are addicted to their medication. Although pseudoaddiction was a concept first reported in the literature as a case study two decades ago, it has not been empirically validated, and it has been overextended. Moreover, sometimes in the face of circumstances that could be due to pseudoaddiction, dosages are escalated to unsafe levels. Instead, in cases where patients exhibit aberrant behaviors and complain of unrelieved pain, alternative approaches to pain control can be pursued instead of continued dose escalation [24, 41]. For example, for a patient who is seemingly unable to take their oxycodone as prescribed due to unrelieved pain, they can instead be prescribed a drug with a lower street value, such as sustained-release morphine. In addition, the patient could be given a urine toxicology screen and scheduled for an appointment with a psychologist in order to address the behavioral problems in opioid drug taking. Pseudoaddiction is a behavioral syndrome that needs to be addressed along with improving pain control. On the other hand, other individuals exhibit aberrant drug-taking behaviors when self-medicating to address a psychiatric issue, while still others may be selling their opioid medications. Individuals involved in drug diversion will be negative for opioids in a urine toxicology screen and will have no medicine to show when called in early for a pill count.

Tolerance and physiological dependence are not signs of addiction in an individual exposed to opioids for medical purposes. Although there are many behaviors that can be indicative of a developing drug addiction, such as stealing another patient’s drugs or injecting an oral opioid formulation, most of these obvious signs are not reported by patients [42]. Meanwhile, other types of behaviors, such as early dosing, drug hoarding, and increasing the dose without physician’s consent, are less predictive of addiction and very common. Opioid drug studies indicate that approximately 15–20 % of patients exhibit multiple behaviors possibly indicative of addiction. Therefore, with higher risk patients who are prone to engaging in aberrant behaviors, other systems of monitoring should be incorporated within their pain management program.

Once opioid treatment has begun, prescription monitoring program data can be an invaluable way to identify patients who are “doctor shopping”; however, this type of data varies in accessibility and quality from state to state, with many states now having operational prescription drug monitoring programs [43]. Still, a high-quality national database is yet needed to monitor opioid prescriptions in order to identify patients who doctor-shop across state lines. Prescribers who do not have the availability of a statewide program to track controlled substance prescriptions should at minimum develop a system to minimize duplicate prescriptions within their group practice.

Compliance monitoring with urine toxicology screens is also needed to corroborate patient claims due to the well-recognized unreliability of self-reported information [44]. The results of this drug testing can indicate whether the patient is taking their opioid medications as prescribed, whether they are obtaining controlled substances from another source, and whether they are concurrently taking illicit drugs. Urine toxicology screens are used for long-term monitoring in order to reduce the risk of a potentially serious adverse event, such as addiction or medication misuse, but are not intended to police patients. With chronic pain management, as with other chronic conditions such as heart disease and cancer, the ongoing consequences of treatment must be monitored to ensure safety. Observations of only patient behaviors are often not enough information to clearly indicate the presence or absence of drug misuse or abuse. Even pain and addiction specialists fail to identify a problem in one in five patients as indicated by surprise urine testing that showed drug use outside of the prescribed opioid and dose [44]. New technology developed for urine screening provides results in <5 min for 12 illicit or controlled substances at the point of care. The Federation of State Medical Board furnishes a strong foundation for support of the reimbursement of urine toxicology screening costs. Their model policy strongly recommends urine toxicology screens in patients who are considered high risk for nonadherence to taking their medications as prescribed and as an occasional screening tool to corroborate patient reports.