In this chapter, we address substance use, the final topic for which future clinicians will need considerable mastery in handling mental health problems in medical settings. We’ll focus mainly on prescription-based disorders where the prescribing clinician has great control. Physicians unfortunately received little clinical training in chronic pain or in using opioids when in medical school or in most residencies, just one among many factors contributing to the current opioid crisis.
The Mental Health Care Model (MHCM) from Chapter 3 provides the overarching structure for managing prescription substance problems, including chronic pain. In this chapter, we’ll address the pharmacologic and related nonpharmacologic management of patients with problematic substance use. Let’s start with the opioid crisis, the worst of the prescription drug problems.
PRESCRIPTION OPIOID MISUSE
Considerable research shows there is little, if any, clinically significant benefit for opioid use in chronic noncancer pain.1-4 The Centers for Disease Control and Prevention (CDC) has recently underscored this fact, including that opioids should be prescribed for acute pain only and for no more than 3 to 7 days.1 For the unfortunate millions of people already addicted to or otherwise dependent on chronic opioid use, the CDC advises that the safe dose is no more than 50 morphine milligram equivalents (MME) per day, and that the maximum acceptable dose is 90 MME each day; more on how to calculate MME shortly. Unfortunately, we have seen people taking as much as 2200 MME and commonly see them taking 200 to 400 MME. These are lethal amounts for someone not accustomed to this high dose. For example, if a teenager steals them from an addicted parent and takes the same dose written on the bottle for their parent, it could be lethal.
You will, we believe, be encouraged to learn that there are effective ways to manage these patients. Informed by a rich body of research from psychiatry, multidisciplinary pain management, and primary care,5-15 the authors’ research group identified an evidence-based model for medical physicians and other clinicians, the MHCM, which you’ve already learned in Chapter 3. It is designed to guide clinicians in conducting mental health care in medical settings. This model includes treating chronic pain and co-occurring mental disorders, reducing and discontinuing opioids while replacing them with something effective, and handling patients demanding more narcotics.13,14,16-20 In addition to demonstrating that it is a very effective model to care for patients, we also have shown it is easily learned.16,21
Many opioid prescriptions are “misused,” which we define as use for nonprescribed reasons such as for recreation, to satisfy an addiction, to self-treat one’s pain, to give to someone else, or to sell.22 Importantly, there is no agreed-upon definition of misuse, abuse, or addiction,23,24 and we have chosen to use this as the most straightforward way to understand misuse. Here’s what makes the misuse problem harder for clinicians: it’s difficult to determine who is misusing opioids. Certainly, the majority who use opioids do not misuse them, but each year in America more than 11 million will.25,26 All told, at some point in their lives, 52 million adults have used prescription drugs for nonmedical reasons.27 Especially worrisome, about one-half of teenagers believe that prescription drugs are safer than street drugs, preferring them recreationally. Compounding this problem, over half of prescription drug users get them free from a friend or relative.27 Both facts have serious implications for deaths from suicidal overdoses and accidental overdoses when using the drug recreationally.
Let’s first look at some clinical clues suggestive of opioid misuse. These clues can raise your suspicion and help direct your questioning and subsequent investigation to arrive at a more definitive misuse diagnosis.
For patients using opioids for chronic pain and often associated symptoms of depression, anxiety, and insomnia, the greater the number of the patient behaviors in Table 6-1, the more likely the patient is misusing the prescribed opioid.22,28 These aberrant, often drug-seeking behaviors represent the impact of opioids on the patient’s personal and professional life. Certainly, they are not diagnostic, and patients may occasionally lose a prescription or have it stolen. Rather, it is the repetitive pattern that is the clue to potential issues.
In addition to detecting misuse, in taking a careful history, you’ll learn of important contraindications to any opioid use: prior history overdoses, especially on opioids, any substance use disorder (prescription, alcohol, illicit), or an untreated mental disorder.
While one is always patient-centered, obtaining substance-related information often requires the more specific, closed-ended questioning characteristic of clinician-centered inquiry.29 As well, important information about opioid use often comes from relatives or others familiar with the patient because they are concerned about the patient’s medication use.29 Knowing that these interactions can be a stressful, often guarded discussion for the patient, assure them that you ask these questions of everyone where opioids are being prescribed or requested—and that the information obtained is part of providing the best care possible. In particular, you need to determine the following information about possible risk factors30:
Use of opioids: first use; doses; duration; present use; time of most recent use; side effects and complications; problems with work, school, relationships, or the law; any accidental or deliberate overdosing
Other substances used now and in the past with similar details obtained, particularly alcohol and illicit drugs
Mental disorders and the details of these
Family history of drug or alcohol abuse or mental health problems
Specific medical/surgical problems sometimes associated with opioid misuse: hepatitis B and C, HIV, frequent accidents and trauma, and frequent skin infections
Pain that is unexplained, occurs after a motor vehicle accident, or occurs in more than 3 body areas
Careful as you may be in your history, we still need systematic drug screening for misuse in all patients prior to beginning opioids (or prior to the next refill if already on them). To identify aberrant, misuse behaviors, we recommend the Opioid Risk Tool (ORT) in Table 6-2.31 The revised Screener and Opioid Assessment for Patients with Pain (SOAPP-R)30,32 is much longer, but has good psychometrics and a higher rate of identifying patients who eventually evince misuse.
This tool should be administered to patients upon an initial visit prior to beginning or renewing opioid therapy for pain management.
Mark each box that applies
|Family history of substance abuse|
|Personal history of substance abuse|
|Age between 16 and 45 years||1||1|
|History of preadolescent sexual abuse||3||0|
|ADD, OCD, bipolar disorder, schizophrenia||2||2|
Observing aberrant behaviors, however, also is insufficient and may underestimate misuse.30 There are 2 necessary items of information you will need in all patients on or beginning on opioids: (1) the patient’s pattern of filling prescriptions from your state’s prescription drug monitoring program (PDMP); and (2) a urine drug screen. They are key elements in making a diagnosis of misuse and are reflected in the first 2 items in Table 6-3.
PDMPs are statewide electronic databases that collect data on controlled substances dispensed in that state. Data are available to individuals, such as clinicians or pharmacists, authorized to receive the information from the state agency administering the program. The intent is to provide a means of monitoring controlled substance use by both physicians and patients, to prevent and deter drug abuse and diversion, to identify patients with addiction problems, to inform public health initiatives, and to educate people about drug abuse. Although each system is state based, there are some that communicate with other states. All states and the District of Columbia have operating PDMPs. Although practices vary considerably, it has been demonstrated that mandatory checking of the PDMP before prescribing an opioid has reduced overdose deaths.33 Current PDMP contact information can be obtained from the Alliance of States with Prescription Monitoring Programs (www.nascsa.org/rxMonitoring.htm).34
Often required by state regulations, it is imperative to check the PDMP before prescribing opioids to your patient, particularly for any new patient requesting them. When pharmacists dispense controlled substances, they enter these data into the PDMP, increasingly doing so immediately so that data are current; unfortunately, some states still allow delays ranging up to a month, which reduces their value, but they still should be used.33,35 Most reports from the PDMP provide the time and date of all controlled substance prescriptions filled (and when it was written), the prescriber, the location, and the number of pills and dose actually filled.
From the PDMP you determine what prescriptions a patient has received, when, and from whom. For example, if you had a contract to be the only prescriber and for the patient to use only one pharmacy and the only controlled substance you prescribed was 3 tablets (10 mg) per day for 30 days of oxycodone, you should find this prescription to be the only entry into the PDMP for that period. In general, there are several possibilities from a PDMP report:
As in the example above, the opioid(s) prescribed is the only one listed in the PDMP and it is in the amount prescribed over the correct duration—this suggests adherence and no misuse.
If the opioid(s) prescribed is not present—this suggests diversion or other misuse, but the patient’s inability to obtain the medication must be reviewed with the patient; for example, pharmacy did not have it, patient could not pay for it.
If opioids (or other substances) not prescribed in your contract appear in the PDMP—this means there are additional prescribers (they would be listed if from your state) and strongly suggests misuse—again, this requires discussion with the patient for some alternative explanation; for example, went to the emergency department because of a fractured arm or the dentist prescribed opioids following a tooth extraction.
You can see that the PDMP is imperfect, so (also imperfect) laboratory testing of the urine for drugs is needed in all patients. In fact, from 30% to 50% of all patients on chronic opioids have abnormal UDTs.30 The idea is similar to that of the PDMP: ensuring that what is prescribed is present and that there are no nonprescribed controlled substances present and no illicit drugs. But there are pitfalls in interpretation of UDTs.
Urine drug testing is often conducted in 2 phases: (1) screening immunoassay and (2) confirmatory chromatography/mass spectrometry. All guidelines recommend that a positive test on the former be confirmed by the latter to exclude the very common false positives.30,36-41 Some believe that only the confirmatory test should be done in most instances because of the low sensitivity (high false negative rates) and low specificity (high false positive rates) of the immunoassay.30,41
Screening immunoassay: The attributes of the immunoassay are that it is immediately available (point of care or laboratory testing), detects drugs used in the preceding 5 to 7 days, and is inexpensive using test cups or test strips. Its immediate availability is sometimes important, say, in new patient requesting opioids where the PDMP from another state is not available—and you need to know if the drug reported is indeed present. Unfortunately, however, the immunoassay has so many false positives and false negatives that confirmatory testing is required in the opinion of most, unanimously when screens are positive for a substance of interest.30,36-40 As well, however, the high false negative rate also means that negative screens would benefit from confirmatory testing.
Confirmatory mass spectrometry testing: These tests are done in the laboratory only (no point of care option), are more expensive, detect substances used in the preceding 1 to 2 days, and may take several days to obtain a report. But confirmatory testing has much greater sensitivity and specificity, is able to differentiate among drugs of the same class, is able to measure semisynthetic and synthetic opioids (such as hydromorphone, fentanyl, and oxycodone), and has the ability to check for drug metabolites that sometimes are the key evidence of use.41 Many believe confirmatory testing should be obtained once or twice yearly in all patients on chronic opioids.30 The objection to confirmatory testing by mass spectrometry has been cost, but recent evaluations propose that immunoassay is not more cost-effective and should be discarded. The main issue of cost is whether the laboratory invests in the mass spectrometry equipment or sends specimens out for analysis.41
There are many important issues in UDT to attend to
You must know your laboratory, what its routine panel contains, how to order specific drugs that match your needs, what algorithms are used, and what cutoff values are used.41 The latter are particularly important because cutoffs are often set too high, resulting in false negatives.36 A standardized set of cutoffs recently was reported that identifies controlled substances and illicit substances in 97.5% of patients; see Table 6-5 in the reference for cutoffs.37
Likely already obtained during the history, it is critical to know the patient’s recent drug history, including the specific times when they last took each medication during the preceding week; this history should include alcohol and marijuana use, which often must be measured via urine testing due to under-reporting or failure to disclose use.
One must be very systematic in how the urine is collected to preclude dilution or adulteration, whether deliberate or inadvertent.30,36,38 Urine should be collected in the clinic or laboratory, preferably early in the morning and evaluated within 3 to 4 minutes of collection if immunoassay is done. If possible, use a dedicated toilet facility with no running water and add a blue color to the water in the toilet. Table 6-4 reviews the characteristics of normal and adulterated urine samples. If the urine is not completely normal, a repeat urine is obtained under direct supervision. The effect of a diluted urine is to decrease the concentration of the drug being measured, while adulteration can destroy the drug.36,37
|Urine Test||Normal||Abnormal||Likely Adulteration or Dilution|
|Temperature||90-100°F||<90°F||Cold substance, like tap water, added|
|pH||4.5-8.0||<3.0 or >11.0||Adulteration|
|Specific gravity||1.002 and 1.030||<1.001||<1.001 is not human urine|
|Creatinine, random||>20 mg/dL||<20 mg/dL|
<20 suggests dilution
<5 is not human urine
|Appearance||Variations of yellow||Nonyellow||Small bubbles suggest soap|
If the urine appears normal, it can be evaluated by immunoassay and/or sent to the lab for confirmatory testing (often collected at the lab). The result should be known before prescribing any opioids. If an immunoassay result is positive for opioids, confirmatory testing is obligatory to exclude false positives. If the result is negative and there is high suspicion of some substance use, confirmatory testing also should be ordered.
Although quantification of results is reported, it cannot be used to judge the actual doses taken by patients.30 Table 6-5 reviews some possible unexpected results from UDT. It is essential in all cases to discuss them with the patient, always using the patient-centered skills central to the MHCM.
|Prescribed drug absent||Diversion|
Patient could not afford
Drug stolen or lost
Over use and ran out early
|Consult with patient|
|Nonprescribed drug present|
Used others’ drug for recreation or self-medication
|Legitimate other prescriber, such as emergency room or dentist for acute problem|
Consult with patient
Refer to addiction specialist
|Illicit drug present||Illicit drug use|
Consult with patient
Refer to addiction specialist
|Low specific gravity and creatinine||Deception|
Renal tubular abnormality
Consult with patient and review medical issues
Refer to addiction specialist
|Very high concentration of drug prescribed||Additional drug taken||Metabolic variability|
Consult with patient
Refer to addiction specialist
|Low concentration of unexpected drug||Remote use|
Consult with patient
Follow-up to ensure clears
With information from the history and your screening with the ORT and UDT, we construct a risk profile. This includes the dose and type of opioid medication, polypharmacy (especially opioids and benzodiazepines), history of mental or substance use disorder, and other data. The profile guides in determining the frequency of visits and refills and how often to obtain UDTs. Table 6-6 outlines the features of high-, moderate-, and low-risk patients.30,38
We recommend initially monitoring everyone after 1 month (sometimes sooner if high risk) until we are clear on stability of the risk stratification.30 Then continue to monitor high-risk patients at monthly intervals (including UDT) until stable, including getting the opioid dose to 90 MME or lower. When this is achieved, see them at 2 and then 3 month intervals as long as stable, hopefully moving high-risk patients into the moderate- and lower-risk groups. Moderate risk groups are followed every 4 months and low-risk groups every 6 months. We will return to the details of management soon.
Other items in Table 6-3 also provide evidence of potential misuse. Tolerance means that there is less effect from the same amount of drug and/or that increased doses are needed to achieve the desired effect.42 And opioids may affect body systems differently, for example, motor coordination, respiratory depression, and sedation. Laboratory testing often is needed to make the diagnosis, observing high drug levels with little evidence of effect. Withdrawal is the development of symptoms when the substance is stopped—and relief of symptoms when resuming it.42 The remaining items in Table 6-3 are infrequent and often come to your attention from legal sources or from concerned relatives.
It is important to reiterate that there are few, if any, data suggesting clinically significant, long-term benefit from using opioids in chronic pain, and that they are causing great harm from overdose deaths, suicidal deaths, misuse, diversion,1-4 and the severe side effects outlined in Table 6-7.3,43-47 Before addressing actual treatment with opioids, there are 2 key tasks: calculating MMEs and how to ascertain the patient’s response to opioids if you use them.
Most adverse effects relate directly to the opioid dose.38 The intent in calculating MME is to provide a common denominator for estimating opioid risk. From a widely varying literature and relying mostly on a report from the Centers for Disease Control and Prevention (CDC), we have guidelines for “safe” doses and “dangerous” doses.1
We calculate doses in MMEs from research showing the relative potency on a milligram basis of all different opioids. This means that the dose of each opioid preparation is converted to the equivalent amount of morphine as a common reference point. Some prescription monitoring programs will provide you with this information, but it is important for you to be comfortable with the calculation yourself for quick, bedside assessment. With an agent like hydrocodone, for example, its potency on a milligram basis is the same as morphine. But many opioids are more potent on a milligram basis, and they need to be converted by a multiplier to MME. For example, oxycodone is nearly twice as potent as morphine, meaning 20 mg are equal to 40 MME. Similarly, hydromorphone is 4 times as potent, so we multiply by 4 to get MME. Adapted from the CDC, we developed Table 6-8 to assist you in calculating MME.
|Opioid (mg/d except fentanyl)||Multiply to Get MME|
|Hydrocodone (Vicodin, Norco)||1×|
(1.5× from CDC)
|Fentanyl transdermal (in µg/h)|
(2.4× from CDC)
|Codeine (Tylenol 3 or 4)||0.15×|
With a variable literature and to keep the calculations conservative and simple, we slightly modified the CDC recommendations: for oxycontin, use a multiplier of 2× rather than the 1.5; for fentanyl, use a multiplier of 3 rather than 2.4. In doing this, the clinician does not need to go to a calculator or website to determine the MME. Instead, they simply memorize the whole-number multipliers to calculate on the spot the MME a patient is taking:
1 × hydrocodone dose in mg = MME
2 × oxycodone dose in mg = MME
3 × fentanyl dose in µg/h = MME
4 × hydromorphone dose in mg = MME
Here’s an example:
A patient was taking 10 mg of oxycodone 4 times daily, a total of 40 mg each day. Multiply this by 2 (from the table) and this is 80 MME. He also used fentanyl skin patches with a dose of 25 µg/h; multiply this by 3 and this is 75 MME. Combining 80 and 75 MME, he is taking a total of 155 MME daily, a dangerous level.
A safe dose is defined as 50 MME or less, and borderline safe doses are up to 90 MME.1 Beyond that, the risk of death and other complications increases in direct proportion to MME doses. Noted earlier, we have seen (tolerant) patients taking as many as 2200 MME and often see them taking in the range of 200 to 400 MME. Here are a couple sobering figures. In a patient taking a low dose of opioids, the later development of the DSM-5 Opioid Use Disorder42 is 15 times more likely than someone not on opioids; for those on high doses (> 120 MME), the risk is 122 times greater.38 Further, a person taking >100 MME per day is 9 times more likely to overdose than a patient taking < 20 MME. One overdose in 7 is fatal.38 The MME is an important number to know as you are prescribing opioids, and it’s also important for patients to know when you are working with them to reduce their dose level.
If you prescribe opioids, their impact should be monitored, just as you would, for example, monitor blood sugars and glycohemoglobins in patients taking insulin. While the old standard of having patients rate their pain on a 1 to 10 scale can be used, it provides no sense of the functional impact of a patient’s pain—just the subjective severity. For example, a level 2 severity might be associated with the inability to work, but we would not know that and assume the patient was doing well. Alternatively, we might hear of a pain rating score of 8 and wonder about increasing the dose without learning whether the patient’s enjoyment in life and general activity were satisfactory or improved.
The Pain Intensity, Enjoyment of Life, and Interference with General Activity (PEG) Tool in Table 6-9 takes little additional time and informs you of the impact of pain on enjoyment of life and on general activity.38,48 We learn about the impact of pain as well as its subjective severity.
1. What number best describes your pain on average in the past week?
|No pain||Pain as bad as you can imagine|
2. What number best describes how, during the past week, pain has interfered with your enjoyment of life?
|Does not interfere||Completely interferes|
3. What number best describes how, during the past week, pain has interfered with your general activity?
|Does not interfere||Completely interferes|
Because you will see patients using harmful doses of opioids, the PEG allows you to determine if, from the patient’s perspective, there is any benefit—defined as a 30% reduction in the overall score from the time the opioid was initiated or the dose increased. This method defines the minimal clinically significant improvement we can expect if, indeed, the opioids actually worked. Failure to achieve this minimal level of improvement does not mean that the dose needs to be increased but, rather, that the opioid is not working and needs to be tapered and discontinued.
We have indicated that opioids have little, if any, value in chronic noncancer pain1,38 but, nonetheless, they are often used. You will encounter many patients already taking opioids, and the material here provides guidelines for evaluating and treating them. We begin by outlining where opioids should not be prescribed without advice from a mental health and/or addiction consultant. We then describe in detail how to use them, which includes how to taper and discontinue them.
Comorbid mental and nonopioid substance use disorders unless they are under effective treatment for this problem and opioid treatment is integrated into the care plan. These DSM-5 disorders are presented elsewhere in this book.42
History of a suicidal attempt, especially with opioids. Unfortunately, we know that opioids continue to be prescribed to many following a nonfatal opioid overdose, in which case fatal drug-associated outcomes 1 year later are 132 times higher than in the general population. These patients need careful follow-up and care, not more opioids.49
History of nonsuicidal overdose, especially with opioids.
Not currently taking them. There are 2 groups of patients:
Never have had opioids for chronic pain
Have had opioids for chronic pain in the past but have been off for a week or more when you first see them
Two of the following occur within a 12-month period and lead to clinically significant impairment or distress.
An exception to these recommendations, implemented only in consultation with a mental health and/or an addiction specialist, is for a patient who has tried all other treatment described later in this chapter and by the MHCM (Chapter 3) and obtained incomplete relief. This would particularly apply to a patient who has responded in the past to safe doses of opioids but is not now taking this. The nonopioid treatment outlined shortly is always tried first and continued if you add an opioid. Thus, one would never begin opioids until after exhausting all other means of treatment.
The CDC has indicated that opioids in acute pain should be used for no more than 3 to 7 days.1 Further, they advise that you rely on nonsteroidal analgesic drugs and use opioids only when these are incompletely effective, that is, opioids are a secondary choice, not the first treatment provided to an acute pain patient. This practice targets clinicians (including dentists) who prescribe opioids routinely for any pain problem, especially when they provide a 1-month supply with refills. Encouragingly, a recent emergency department study demonstrated that treatment of severe, acute extremity pain with 400 mg of ibuprofen combined with 1000 mg of acetaminophen was as effective as three different opioid-acetaminophen combinations in relieving pain.50 If opioids are used in the acute setting, the state’s PDMP is first queried to determine if they already are taking them or have been in the past.
Acute treatment is straightforward if the patient has never used opioids regularly and has a clear-cut new acute pain problem, such as an ankle fracture or sprain, acute low back strain, acute appendicitis, or recent dental surgery. We strongly recommend that acute pain patients requesting refills after the initial 3-7-day prescription is gone be evaluated instead of filling the prescription by phone even though the latter may be easier. This is especially the case in patients with acute back pain where the natural course of resolution often lasts more than 1 week. In general, if a patient is not back to full activity within 2 to 4 weeks, they require careful evaluation to be certain of the diagnosis and of the appropriate, nonopioid treatment needed; for example, physical therapy and osteopathic manipulative therapy.
There is a good reason that we devote so much time to the management of acute pain. You can prevent chronic pain. One study found that more than 50% of patients on opioids for 90 or more days remained on them for years, and that the factors most associated with continuation were intermittent prior exposures, doses of 120 MME per day or greater, and indications of misuse.51 As well, you will not be surprised to learn, there is little, if any, clinically significant benefit from opioids for chronic low back pain where misuse is found in 24%.52 Be similarly wary in patients with what may have begun as an acute migraine headache or other acute problems that frequently recur. Properly treated, acute problems will improve over the short term. When patients are not improving on narcotics, it means the opioid is not effective, not that the dose is increased or the drug refilled.
But it’s not all this easy! There is a large group of patients who repeatedly present with acute problems, such as low back pain, abdominal pain, chest pain, or headache, yet there is no objective evidence of currently active disease even though the patient may have had, for example, a myocardial infarction in the past. Many of these patients also take opioids chronically and request increased doses for the acute flare, often repeatedly hospitalized where they typically receive very liberal opioid dosing, sometimes intravenously. If the patient is not on opioids regularly and the investigative work-up shows no objective disease explanation for the pain, whether in the office or in the emergency department, treatment is symptomatic and narcotics are used for no more than 3 to 7 days at less than 50 MME. For patients already taking narcotics, the opioid is not increased but is continued at the present dose, and the patient is entered into the treatment regimen that follows later in this chapter. For patients hospitalized who have received large doses of narcotics, often intravenously, they are quickly tapered over 1 to 2 days to the dose they took prior to admission and entered into the treatment regimen below. Similar principles apply in patients undergoing surgery, tapering the narcotic before discharge, if possible, or rapidly when they return home. If already on an opioid, the same or a slightly increased dose can be used to address the acute surgical needs, and the patient is then tapered back to their usual home dose. Recent data indicate an increased use of chronic opioids after perioperative exposure, and that switching from intravenous to oral or subcutaneous routes of administration works just as well in the perioperative period and reduces total opioid exposure.53
For the patients already on chronic opioids, we have therapeutic options leading to 3 possible outcomes: