Interventions in Chronic Pain Management. 4. Monitoring Progress and Compliance in Chronic Pain Management

4.1 Educational Activity: To justify to a workers’ compensation nurse the use of functional and psychologic screening for managing a patient with a chronic work-related back injury 

PAIN IS A COMPLEX experience that is subjective and influenced not only by the pain’s intensity but also by its meaning to the patient; it is linked with the patient’s social, cultural, and psychologic background.1 It is a dynamic process that changes over time. Any instrument used to measure pain should be valid and consistent and should reflect pain changes associated with interventions. At least 4 dimensions of pain need to be assessed: intensity, affect, quality, and location. Intensity is how much a patient hurts, whereas affect is the emotional response to the experience of pain. Quality and location refer, respectively, to the physical sensations attendant to the pain and the anatomic area in which the pain is located.

Unidimensional pain scales measure pain intensity. Simple measures include the verbal rating scale (VRS), the numeric rating scale (NRS), and the visual analog scale (VAS).2, 3 These 3 scales are valid measures of pain intensity. They are all easy to use and score and reliably reflect changes that occur with treatment. The VRS is a list of adjectives that describes levels of pain intensity. It includes adjectives that describe the extremes of pain with less extreme descriptors in between. Weaknesses of the VRS include the patient’s facility with the language in which it is given and the ability or lack of ability to choose between the adjectives presented on the test. Also, ranking of adjectives may not accurately reflect the magnitude of the differences between various intensities of pain.4 A VAS layout consists of a line with adjectives and/or numbers along it. This scale may be better than the VRS at showing the magnitude of change with treatment. There are mechanical VAS scales with a sliding marker, and there are paper-and-pencil VAS scales. Some dexterity may be required. The NRS is a numeric scale that is simple to use and gives reproducible measurements, allowing the patient to be easily monitored during treatment. It may be easier than the verbal scales to use with older adults and children.

The main measures of pain quality are the Neuropathic Pain Scale (NPS) and the McGill Pain Questionnaire (MPQ). The NPS scores 8 pain descriptors on a 1-to-10 scale and also rates 2 global descriptors of pain (pain intensity, unpleasantness). Its strengths include an ability to distinguish between patients with different neuropathic pain diagnoses and its brevity.4

Pain location is generally evaluated with a pain drawing. Various symbols can be used to indicate sensations over the body.

Multidimensional pain scales include the MPQ and its variants. In contrast to the unidimensional measures described previously, the MPQ looks at quality and location of pain as well as pain intensity. It evaluates the patient’s affective response to pain. In addition to containing a picture of the body (for pain localization), the MPQ also has the following sections: pain rating index (PRI), number of words chosen, and present pain intensity (PPI). The PRI includes 20 classes of words that describe the affective response to pain. The words are ranked and summed into scores for sensory, evaluative, affective, and miscellaneous subscales. The PPI is a combination of numbers and words constructed to evaluate current pain intensity. It is useful, reliable, and has been used in patients with dental pain, malignancies, postherpetic neuralgia, phantom pain, rheumatologic pain, postoperative pain, low back pain, and obstetric pain. It can be used to compare diagnosis and treatment in various pain syndromes. Significant psychologic disturbance or anxiety may decrease the ability to discriminate between different syndromes. The short-form MPQ created by Melzack in 1987 is easier to give, and it also helps assess efficacy of treatment interventions and changes in self-esteem.1, 5

The 3 psychologic aspects of pain1 are the sensory-discriminative, the motivational-affective, and the cognitive-evaluative dimensions of the pain experience. Psychologic testing should be used in patients with significant psychologic distress, excessive pain behaviors, functional deficits, substance abuse, or excessive medication use. In the patient who has pain with underlying psychologic pathology or significant psychosocial issues as a result of his/her pain syndrome, testing goals include identification of psychosocial factors that may be impairing recovery, assessment of the efficacy of treatment strategies, and evaluation for implantable technologies and other approaches.

Major measures of depression (table 1) include the Beck Depression Inventory, Center for Epidemiologic Studies Depression Scale, Minnesota Multiphasic Personality Inventory, Millon Behavioral Health Inventory, and the Illness Behavior Questionnaire. The West Haven−Yale Multidimensional Pain Inventory and the Sickness Impact Profile are also used to evaluate disability and suffering in patients with chronic pain.6, 7

Table 1. Instruments Used to Measure and Screen for Depression

	Instrument	Purpose of Screen	No. of Items	Comment
	BDI	Depression	21	Evaluates somatic disturbances.
	CES-D	Depression	20	Less emphasis on somatic complaints and better for rheumatologic disorders. May be better at assessing change to cognitive-behavioral and medical treatment.
	MMPI	Depression	566	Elevations in hysteria, hypochondriasis, depression associated with perception of severe pain, affective disturbance and disruption of social, vocational, and marital function.6, 7
	MBHI	Psychologic function	150	Designed to evaluate medical patients and less emphasis on symptoms of physical illness.
	IBQ	Abnormal illness behavior	62	Looks at hypochondriasis, disease conviction, psychologic versus somatic focus of disease, affective disturbance and inhibition, denial, and irritability.
	WHYMPI	Functional disability	56	Evaluates perceived interference with work, home, and daily function as well as social interactions with spouse, family, and others.
	SIP	Functional disability	136	Assesses 12 dimensions of functional capacity including both activities and social and emotional response.

Abbreviations: BDI, Beck Depression Inventory; CES-D, Center for Epidemiologic Studies Depression Scale; IBQ, Illness Behavior Questionnaire; MBHI, Millon Behavioral Health Inventory; MMPI, Minnesota Multiphasic Personality Inventory; SIP, Sickness Impact Profile; WHYMPI, West Haven−Yale Multidimensional Pain Inventory.

4.2 Clinical Activity: You are contemplating the use of opioids to treat chronic pain in a 35-year-old man. You describe to him the use of screening procedures to identify potential for substance abuse and the role of a drug-use agreement 

Substantial debate exists about the use of opioids to treat chronic, nonmalignant pain. Abuse, diversion, and habituation without benefit are strong concerns of clinicians who prescribe opioids. Many physicians rely solely on the patient’s account regarding the efficacy and proper use of chronic opioid treatment. Unfortunately, relying exclusively on patient reporting is fraught with potential problems because patients often incorrectly report their use of opioids. Patients may inaccurately report the use of opioids without nefarious intention or they may intentionally deceive for a variety of reasons such as the desire to receive more medications for the purpose of diversion or to nourish an addiction. With this in mind, clinicians should have tools available to monitor adherence to prescribed treatment and to screen for the use of nonprescribed drugs.8, 9, 10, 11, 12, 13, 14, 15, 16

The U.S. Drug Enforcement Agency has taken the position that clinicians are justified in using opioids to treat malignant and nonmalignant pain when opioids are the best clinical choice of treatment.9 Accurate and thorough documentation should be kept on your rationale for prescribing the opioids, the patient’s analgesic response to the prescribed opioids, adverse effects of opioid treatment, and effects of opioids on the patient’s activity level. The medical record should document a suitable history and physical examination, discussion of responses to current and prior opioid and nonopioid pain therapies, discussion of potential risks and benefits of opioid treatment perhaps including addiction and dependence, and results of diagnostic testing.9 The physician and patient should maintain an open and frank discussion pertaining to opioid requirements, adverse effects, and adherence to prescribed treatment.

Before prescribing opioids, a written agreement between the physician and patient that clearly defines the rules regarding opioid treatment should be given strong consideration. It is important that the agreement stipulate that the patient can only receive opioid treatment from the designated practice as long as the agreement is in effect. The agreement should also discuss patient requests for early refills, stipulations for lost medications, toxicology screenings, diversion of medications, and repercussions for violation of the agreement.9, 10 Patients who use multiple pharmacies to fill opioid medications may arouse suspicion, and the agreement should specify the use of only 1 pharmacy for opioid prescriptions. When a single pharmacy is used, it is easier for the physician to monitor the filling of opioid prescriptions.

The patient who receives opioids for chronic pain should be reliable and sincere. However, all patients receiving opioids must be screened because of the potential for drug abuse and diversion; several screening instruments are available. One such tool, which is invaluable in monitoring opioid prescription use, is toxicology testing by means of urine and/or serum.8, 9, 10, 11, 12, 13, 14, 15, 16 Technologies are emerging for the use of saliva, hair, and sweat for toxicology testing in medical practices. Currently, testing is typically performed by using urine because it is readily available and collection is noninvasive and relatively inexpensive.11 Frequent toxicology testing can be performed in patients who arouse suspicion.

Many products are commercially available to assist patients who intend to cheat with urine testing. When the urine sample is received, characteristics of the urine sample such as temperature and pH should be checked to ensure that the sample is genuine. If there is strong concern for chicanery, then it may be necessary to actually watch the patient produce the urine sample. Having the patient place the urine specimen with the requisition form in a sealed container will discourage charges of tampering. There are practical and medicolegal implications to opioid toxicology testing. The physician must both exercise prudence and maintain documentation for collecting specimens, interpreting test results, communicating with the patient, and initiating actions related to toxicology test results.10, 11, 12

Random pill counts can also be used to monitor compliance. A larger than expected number of pills suggests that the patient is not taking the medication as often as expected, whereas fewer pills suggest the possibility of excessive medication use, diversion, or pseudoaddiction. Not finding the prescribed drug and/or its metabolites on a toxicology screen or finding a nonprescribed or illicit drug is of concern. This occurrence should prompt further discussion with the patient, a reevaluation of the treatment strategy, and perhaps discharge of the patient from the practice.9

Clinicians must continually assess patients for abuse, not only when opioid treatment is initiated but also on subsequent visits. Screening questionnaires such as the CAGE questionnaire have primarily focused on alcohol use and are of uncertain value with respect to opioids.10, 13 The following criteria have been proposed to indicate potential abuse or misuse of opioids: (1) excessive opioid needs, (2) deception or lying to obtain controlled substances, (3) doctor shopping, (4) low functional level, (5) exaggeration of pain, and (6) prescription forgery. Frequent telephone calls or patient visits regarding opioid treatment and multiple medication allergies or intolerances are also warning signals of problematic opioid use.9 A history of opioid abuse does not necessarily preclude the patient from receiving opioid treatment if justified, but close monitoring is advisable. The use of opioids has been shown to be more problematic in male patients and younger patients, such as the 35-year-old man mentioned earlier.14

The CAGE questionnaire is an acronym of 4 questions that were designed to evaluate for alcohol dependency. These questions may have applicability as a general screen for substance abuse. The clinician asks the patient if he/she feels the need to Cut down on substance use, gets Annoyed by others commenting on substance use, feels Guilty about substance use, or needs Eye openers.

The Opioid Risk Tool (ORT) is a brief screening tool to help physicians accurately assess which patients may develop opioid-related aberrant behaviors when prescribed opioids for chronic pain. The ORT characterizes patients as high, medium, or low risk for development of aberrant behaviors by measuring risk factors for substance abuse. Risks factors used in the ORT to calculate risk include a personal or family history of substance abuse, age, history of preadolescent sexual abuse, and certain psychologic diseases. In the low-risk classification, 94% of patients did not display aberrant behaviors, whereas 91% in the high-risk classification did display aberrant behaviors. The ORT was found to have a high degree of sensitivity and specificity.15

The Screener and Opioid Assessment for Patients with Pain (SOAPP) is a brief questionnaire that can help determine how much monitoring a patient on long-term opioid therapy may require. The long version of this questionnaire has 14 questions and typically takes 8 minutes or less to complete. The short version has 5 questions and typically takes 5 minutes or less to complete. The SOAPP is not intended to rule out patients for opioid treatment. Rather, it can help clinicians formulate a risk-benefit analysis with regard to opioid treatment.

A relationship exists between opioid misuse and psychopathology. Evaluation, awareness, and treatment of concurrent psychopathology may affect opioid-prescribing decisions. Toxicology testing alone is not sufficient to monitor compliance; it is also important that the prescribing clinician and office staff evaluate for aberrant behaviors suggestive of abuse or misuse of opioids. A biopsychosocial assessment is an important element in screening for potential misuse of opioids.13

4.3 Clinical Activity: To interpret the results of drug screening in that same patient 

Toxicology screens can detect certain drugs and their metabolites. These screens provide information about how much and how recently a drug was used. Initial testing is generally performed with class-specific immunoassay drug panels that are not able to identify specific drugs within a class. Often, more precise testing is subsequently performed with gas chromatography and mass spectrometry (GC-MS) to determine the presence or absence of specific drugs.9, 11, 12

Because considerable variability exists among the various tests, the clinician should be aware of nuances in specific tests used. For instance, laboratories may test for different drugs, and the minimal thresholds for detection and reporting may vary between laboratories. Some laboratories convey quantitative data, whereas others do not. The detection time for a drug in the urine varies depending on the laboratory, the drug, and each patient’s metabolism.9, 11, 12

Morphine, methadone, hydrocodone, and oxycodone can be detected after the most recent use for a few days or perhaps up to 2 weeks after it was last taken. Cocaine is not typically detected for more than a few days after the most recent use. Marijuana is detected for up to a month after use. Benzodiazepines and barbiturates vary considerably in detection, largely dependent on the individual medication half-life. Testing is not routinely performed for heroin because it has a very short half-life and is thus difficult to detect.9, 11, 12 Table 2 refers to commonly used drugs and expected detection time on a urine drug screen in patients taking the specific medication over an extended time.

Table 2. Detection Times and Cutoff Levels in Urine Tests of Commonly Abused Drugs

	Drug	Detectable in Urine⁎	Cutoff Level (ng/mL)†
	Morphine	1−3d(2wk)	300
	Methadone	2−4d(2wk)	300
	Hydrocodone	2−4d(2wk)	50
	Oxycodone	2−4d(2wk)	100
	Benzodiazepines	Up to 30d	300
	Barbiturates (short acting)	2−4d	300
	Barbiturates (long acting)	Up to 30d	300
	Marijuana (chronic use)	Up to 30d	50
	Cocaine	1−3d	300
	Amphetamines or methamphetamine	2−4d(2wk)	1000

NOTE. Reprinted with permission.9

⁎ Values in parentheses indicate the maximum time after the drug’s discontinuation that it can be detected in the urine. † The cutoff level is the smallest concentration of the drug detectable in a urine sample; the cutoff level may vary among laboratories.

It is essential that the interpreting clinician have knowledge about opioid metabolites and the potential for cross-reactivity. Without this knowledge, urine screens will be misinterpreted and erroneous actions may be taken. For example, a metabolite of hydrocodone is hydromorphone, and it is not unexpected to detect both in a patient prescribed hydrocodone. Oxymorphone is a metabolite of oxycodone, and both oxycodone and oxymorphone are likely to be present on drug testing in a patient prescribed oxycodone. On the other hand, the presence of hydrocodone in a patient prescribed hydromorphone is problematic, as is the presence of oxycodone in a patient prescribed oxymorphone. Accusing a patient of taking hydromorphone in addition to the prescribed hydrocodone or oxymorphone in addition to the prescribed oxycodone may erode the doctor-patient relationship or perhaps have even greater consequences.11, 12

Because codeine metabolizes to morphine, it is common to find morphine in a patient who is taking codeine but not codeine in someone taking morphine. A patient taking large amounts of codeine may show traces of hydrocodone that are unrelated to hydrocodone use. Likewise, patients taking large quantities of morphine may show traces of hydromorphone that is unrelated to hydromorphone use.11, 12

The presence of methadone, cocaine, oxycodone, and marijuana is diagnostic because these agents exhibit no cross-reactivity; their presence cannot be the result of metabolites from other drugs. Passive marijuana smoke inhalation rarely explains a positive marijuana test result, and the clinician should be skeptical of this explanation. Legal food products containing hemp seeds and coca tea leaves can, but should not, produce positive marijuana and cocaine test results, respectively. Some over-the-counter medications such as Vicks inhalers can produce drug test results that are positive for methamphetamines. Immunoassay screening for methamphetamines must be confirmed with GC-MS with a level that exceeds 200ng/mL of amphetamine to be positive for illicit use. Before testing, it is advised that physicians query patients about their use of over-the-counter medication and document it. Amphetamines may be detected in patients taking over-the-counter medications containing ephedrine or pseudoephedrine, such as decongestants. The only substance that can cause a false-positive opioid result is the poppy seed. A large quantity of poppy seed would have to be ingested to produce a false-positive result.11, 12

Patients on chronic opioid therapy may have blood levels as high as 5 times greater than the laboratory reference range and show no clinical evidence of intoxication or impairment. On the other hand, patients may have subtherapeutic serum opioid levels and show clinical signs of impairment. With this in mind, it is essential that clinicians treat the patient and not the serum opioid level, even if it is at a “toxic” level. Serum testing does have the advantage that tampering is far more difficult than with urine samples. However, serum samples are more invasive to collect, more inconvenient to the patient (urine screening can be performed more easily in the office), and testing is far more expensive. To ensure validity of the test, serum testing should be performed soon after the drug is ingested.16 Toxicology testing should be used primarily to improve the quality of care and to identify problems or potential problems. Interpretation and the ensuing reaction should be done with sensitivity to each patient’s biopsychosocial issues. Toxicology testing is subject to false-positive and false-negative results, like any other diagnostic test. Erroneous interpretation can have drastic consequences for a patient.9, 11, 12 Random urine drug testing has been shown to reduce illicit drug use in patients with chronic pain who are receiving opioids.8 A medical review officer can assist with drug test interpretation. A website with good resources on this topic is http://www.MROCC.com.

4.4 Clinical Activity: To assign a functional status to a 23-year-old construction worker with 6 months of lower back pain who has reached maximal medical improvement 

Patients with lower back pain from work-related injuries may reach a point where they are not making progress in terms of pain or function. The longer a patient is out of work, the lower the chance that he/she will return to work. Up to 90% of acute low back pain resolves within 1 month. In a person with acute back pain, rapid improvements in pain and disability usually occur over the first month. Improvement often continues for 3 months. After that, improvement remains steady over the next 12 months.17 Over 80% of workers who are injured return to work within 1 month, and 90% of workers return by 3 months. At 6 months, the chances of return to work decrease to 50%, and by 1 year only 25% of patients return to work. By 2 years, the return-to-work rate approaches zero. Five percent of patients never return to work.18, 19, 20, 21

Economic ramifications of lower back pain are based on direct and indirect costs. Direct costs include monies spent on items such as treatment (including hospitalization), outpatient physician visits, diagnostic testing, medications, therapy, and equipment. Indirect costs include monies spent to address the disability that is associated with the injury; these costs include lost wages, decreased productivity of the injured patient, transportation, and resources spent to manage the disabled worker.22 Frymoyer and Cats-Baril23 estimated that the total cost of low back pain was $50 to $100 billion (US dollars) in 1991. In 2000, the U.S. government spent $60 billion in disability insurance awards plus another $5 billion on the administration of disability programs within Social Security. Costs are expected to rise to $85 billion by 2010.24 Fewer than 5% of patients are responsible for 75% of the costs of occupational back pain.18, 19, 21

From a cost perspective, it becomes critical to identify when a patient is not progressing so that the costs pertaining to the work-related injury can be contained. These expenditures may include costs related to medications, devices, and sequelae related to such treatments that are provided in connection with the original injury. From a patient perspective, it is important to give a patient every chance to minimize pain and maximize functional improvement. Maximal Medical Improvement (MMI) is defined as follows: “A condition or state that is well stabilized and unlikely to change substantially in the next year, with or without medical treatment. Over time, there may be some change; however, further recovery or deterioration is not expected.”25 This determination is subjective. In the legal system, a physician is often expected to respond with a 51% or greater degree of certainty. Terminology may also include the phrase “to a reasonable degree of medical certainty.” Treatment is not complete until the standard of care has been offered to the patient. If the patient does not comply with appropriate treatment that may diminish pain and restore function, the determination to support disability must be reconsidered. On the other hand, if a patient has not been offered all reasonable options for care, then these other treatments must be considered before placing the patient at MMI.24, 26, 27 Litigation and workers’ compensation payments may influence recovery from illness. In these cases, it may be reasonable to state that MMI has been attained based on progress to date. In general, once a patient reaches MMI, further treatment and investigation is not necessary; however, if a new treatment option becomes available or if the patient’s clinical scenario changes dramatically, then these options must be reconsidered.

The evaluation of a patient’s functional abilities is not a straightforward task. Impairment, deconditioning, nociception and fear of nociceptive stimuli, illness behavior, secondary gain, and malingering may all play a part in influencing functional performance. Illness behavior is the “observable and potentially measurable actions and conduct which express and communicate the individual’s own perception of disturbed health.”28, 29, 30 Malingering is a very different issue. It is defined as “a condition or state that is willfully feigning or exaggeration of a disease or effect of an injury in order to obtain specific external gain.”25

Physical capacities can be determined in several ways. A functional history detailing the patient’s current activity level at home, on the job, and with recreational activities should be taken. This history should then be compared with what is seen on the physical examination and with the information available from the patient’s therapeutic program. Functional capacity evaluations (FCEs) are standardized formal assessments that provide objective information about the patient’s physical abilities. Many types of FCEs exist, administration of which may take up to 2 hours to 2 days. They may be limited by the evaluator’s skills as well as by the format of the test and by the patient’s ability to participate in the testing. The FCE provides a snapshot of the patient’s abilities and does not show how the patient will perform over time. Furthermore, inconsistent FCE results do not prove that a patient is malingering or has secondary gain or illness behavior. The coefficient of variation (CV) is a statistical measurement of stability that is the quotient of the standard deviation of a set of scores divided by the mean. It looks at intratask consistency. It does not portray the sincerity of effort. No agreement exists regarding the appropriate threshold CV for determining the submaximal effort. A high CV may occur for reasons other than patient compliance; these factors include the type of muscle contraction used, the presence of injury or pain, testing protocol or equipment, and the magnitude of the variable in question. A work-hardening trial may be a better way to show if the patient has the ability to perform a physical job that requires a lot of repetition.27, 31, 32, 33 A repeat FCE is not usually indicated but may be a consideration if the functional status has significantly changed. Once a patient has been determined to have achieved MMI, a permanent impairment rating can then be determined.25