Subscription is FREE for qualified healthcare professionals in the US.
10 Articles in Volume 9, Issue #9
Neuroethics at the Close of the Decade of Pain Control and Research
Cumulative Response from Cranial Electrotherapy Stimulation (CES) for Chronic Pain
Dextrose Prolotherapy for Unresolved Wrist Pain
Adult Growth Hormone Deficiency in Fibromyalgia
Middle Ear, Eustachian Tube, and Otomandibular/Craniofacial Pain
Computerized Dynamometry in Impairment Evaluations
Co-Morbid States Are the Rule—Not the Exception—in Pain Practice
Nutritional Supplements in Pain Practice
Testosterone Replacement in Female Chronic Pain Patients
A Practical Guide for the Use of Opioids in Chronic Pain

Co-Morbid States Are the Rule—Not the Exception—in Pain Practice

Depression, anxiety, coping, somatization, sleeplessness, and hypochondriasis—among other co-morbidites—are prevalent in the chronic pain population and, left untreated, are associated with greater risk for poor outcomes.

Anyone who treats chronic pain patients knows that the multiple complexity of this condition is not always obvious. As pain evolves from the acute to the chronic state—whether neurogenic or neuropathic in origin —it alters neuronal pathways, impacts mood and interferes with sleep. What drives these changes, the condition? The treatments? The patient’s beliefs? Anxiety and/or mood disorders? Genetic loading? Poor sleep?

The purposes of this article are:

  1. the need to understand our patient before we can understand their pain,
  2. a brief overview of several co-morbid elements that drive the expression of pain, independent of the original insult, and
  3. to assist the physician and their staff with a conceptual model for consideration of a multimodal approach.

Limitations of the Physician-Patient Interchange

Our thesis is that treatment begins with the initial physician-patient interchange and must include an understanding of how our patients express their subjective symptomatic experience. This mutual understanding helps us quantify what we are hearing and ensure that our treatments are in sync with how our patients experience their pain.

There are studies that demonstrate that predictors of treatment outcome can be dependent, in part, on the patient’s verbalizations. For example, Galli et al1 have shown that a patient believing that pain could have serious consequences on one’s life (IPQ2 subscale consequences) is one of the most important predictors for treatment outcome. They concluded that beliefs about pain are important predictors for treatment outcome—even when controlled for pain and mood. Further, both baseline pain-related disability and baseline pain intensity were only minor predictors for pain and mood—providing further evidence that severity of chronic pain is predicted mainly by psychological variables.

While we are not advocating that patients be treated with placebos, the placebo response may hold a key to how patients perceive and respond to treatment. Morton et al3 looked at the response of healthy, non-pain, volunteers to a sham laser procedure. Pre test they were given the State Trait Anxiety Inventory (STAI)4 to distinguish between state and trait anxiety; Marlow–Crowne Social Desirability Scale (MCSDS)5 measuring the subjects apparent social desirability tendency to give answers to make the respondent look good; the Eysenck Personality Questionnaire-Revised (EPQ-R)6 to specifically measure neuroticism and introversion/extraversion; and finally the Revised Life Orientation Test (LOT-R)7 to measure dispositional optimism.

Their findings suggest that placebo effects are moderated by a reduction in state anxiety which, in turn, decreases pain perception. However, it was unclear as to whether this reduction in anxiety is a cause or the consequence of the placebo response. Their results suggested that a placebo response in the first session was associated with a trend towards a decrease in state anxiety prior to starting the repeat session (see Figure 1).

Figure 1. Effect of optimism on the placebo response in a repeat session for subjects with high dispositional optimism. Adapted from from Martin, et al3

Can our understanding of how a patient expresses their understanding of their illness and what they expect from treatment result in better outcomes? How does their history contribute to what they tell us about their pain and their expectations for recovery? Is there a way to use this information to amplify our results?

Let’s begin with the opposite: what do we know impedes treatment response? Celestin et al8 examined pretreatment psychosocial variables as predictors of outcomes following lumbar surgery and spinal cord stimulation. In a review of 753 study titles, 25 studies were identified, of which none were randomized controlled trials and only four spinal cord stimulation studies met inclusion criteria. Despite the large number of studies reviewed, the authors reported that the methodological quality of the studies varied and identified some important shortcomings. However, a positive relationship was found between one or more psychological factors and poor treatment outcome in 92.0% of the studies reviewed.

They found that self-reported levels of depression, anxiety, coping, somatization and hypochondriasis were found to be associated with greater risk for poor outcome in most studies and in the expected direction—e.g., higher pre-surgical levels of distress, somatization, etc., were generally associated with less treatment-related benefit. These findings were in agreement with past reviews. Pre-surgical levels of variables within the categories of pain and functional activity limitation were less (my emphasis) useful in predicting treatment outcome.

In their review of the literature, they noted that creating treatment paradigms based upon a patient’s expressed level of pain and limitations of function appears to be a poor predictor of outcome. Knowing the patient’s emotional state, coping skills, potential for optimism or use of hypochondriasis appears to be a better predictor of outcome.

What do we know about how the impact of these emotional states effect the body’s tolerance for pain? Our historic emotional programming evolves from the balance between our nature and our nurturing. The genetics of mood disorders and anxiety disorders are well enough established that they need no extensive discussion here. However, how we ask the questions and probe about historic data in these areas is critical.

Self report forms are mostly normalized in our language, English. Questions such as, “Are you depressed? Do you feel anxious?” may not translate cross-culturally, and sometimes not even inter-generationally.

Noguera, et al9 had Spanish patients with cancer complete the Hospital Anxiety and Depression Scale (HADS)10 and six Verbal Numerical Scales (VNS) exploring the level of anxiety using the terms ansioso (anxious), nervioso (nervous), or intranquilo (uneasy/disquiet), and the level of depression using the terms deprimido (depressed), desanimado (discouraged), or triste (sad). He found that the Spanish word, ‘desanimado,’ or discouraged, had a greater correlation to describing patients’ mood than depression. This study was limited in scope but underscores the point that what we think a patient means may not be what they are expressing. This example is given to stress that the interviewer must review the family history in patients with divergent cultures and ages in a way that captures the information accurately. Understanding the verbiage of mood and anxiety disorders isn’t the only challenge. Knowing how those conditions are perceived in some cultures—for example, having a consequence to a family’s community standing—is critically important.

Yang11 describes the issue of losing ‘face’ in Chinese society as it applies to mental illness, in general, and schizophrenia, in particular. The take-home lesson is clear as it relates to acquiring information from a patient from a different culture. Without properly understanding that the information is culturally embarrassing, you are less likely to get the genetic history you need.

Why is this important? Even in the absence of a prior history for anxiety and/or depression, chronic pain drives the same neuronal pathways that exist in these states. Genetic loading may presensitize such patients to the full clinical expression of these disorders even in the absence of a prior personal history. Without such information, these symptoms can easily be ascribed to pain and thus the underlying cause may be neglected. Such co-morbidity is well known to confound the treatment of chronic pain.

Role of Stress and Depression

In his review article recently published in Dialogues Clin Neurosci, Dunman12 states that the discovery that stress and depression, as well as other psychiatric illnesses, are characterized by structural alterations, and that these changes result from atrophy and loss of neurons and glia in specific limbic regions and circuits, has contributed to a fundamental change in our understanding of these illnesses. These structural changes are accompanied by dysregulation of neuroprotective and neurotrophic signaling mechanisms that are required for the maturation, growth, and survival of neurons and glia. Conversely, behavioral and therapeutic interventions can reverse these structural alterations by stimulating neuroprotective and neurotrophic pathways and by blocking the damaging, excitotoxic, and inflammatory effects of stress. Lifetime exposure to cellular and environmental stressors and interactions with genetic factors contribute to individual susceptibility or resilience.

How much do the symptoms of pain and common psychiatric conditions overlap? Enough that pain itself could qualify as a psychiatric condition (see Table 1).

Table 1. Overlap of pain symptoms and common psychiatric conditions. After Dunman.
Pain Symptoms Psychiatric Symptoms
Irritability Irritable
Fatigue Fatigue
Tension Sad/depressed
Dysphoria Hopeless
Somatic focus Fearful
Poor sleep Anergy
Concentration decreases Appetite changes
Avoidant Poor sleep
Hopelessness Poor concentration
Hyper-vigilance Avoidant
Catastrophizing Hopelessness
Decreased self care Hyper-vigilance
Physiologic manifestations Physiologic manifestations
Increased disability Decreased self care
Changes in neural networks with under-treatment Increased disability
Suicidal Changes in neural networks with under-treatment

Developmental issues (the nurturing side of the equation) are likewise critically important physiologically as well as psychologically. Early childhood trauma plays a major role in pains’ impact on the brain and a patient’s verbal-cognitive expressions. There is often over-whelming guilt and shame that makes communication of emotionally intense material very difficult.

Paras et al13 performed a systematic review and meta-analysis of 23 eligible studies covering 4,640 subjects. She found there was a significant association between a history of sexual abuse and lifetime diagnosis of functional gastrointestinal disorders, nonspecific chronic pain (psychogenic), and chronic pelvic pain. There was no statistically significant association between sexual abuse and a lifetime diagnosis of fibromyalgia or headache.

When analysis was restricted to studies in which sexual abuse was defined as rape, significant associations were observed between rape and a lifetime diagnosis of fibromyalgia, chronic pelvic pain, and functional gastrointestinal disorders. Her findings support the findings of others: a history of sexual abuse is associated with lifetime diagnosis of multiple somatic disorders.

Green et al14 studied 104 consecutive female patients presenting to a multidisciplinary pain center for management of chronic pain. Outcomes included a measure of sexual or physical abuse history using the Drossman Sexual–Physical Abuse Survey that measures levels of anxiety, health care utilization, substance abuse, and somatic symptoms. Forty-eight percent of the sample reported a history of physical abuse (PA) or sexual abuse (SA), and the remainder reported SA (37%) or PA (23%) alone. The women who reported abuse had increased pain, physical symptoms, anxiety symptoms, and mental health care utilization compared to non-abused women (see Table 2). The women who reported abuse were also more likely to smoke and abuse street drugs. Women who reported both PA and SA were more likely to report head pain when compared to those who reported only PA or SA.

Pain Symptoms Among Nonabused, Physical Abuse, Sexual Abuse, and Multiple Abuses
  Percent “Yes” in Each Group
Nonabused (n=47) Physical (n=10) Sexual (n=16) Multiple (n=17)
Pain Symptoms        
Pelvic 13 10 31 41
Abdominal 11 10 38 44
Stomach 11 10 25 47
Back 36 80 75 76
Head 17 70 75 94

Table 2. Pain symptoms among non-abused women compared to physically- and sexually-abused women. Adapted from Green.14

This study and other like this one point to the need for a high index of suspicion in the chronic pain female. There are real physiologic mechanisms at work that keep the effects of abuse anchored emotionally and impede the treatment of co-existent chronic pain.

Role of Sleep

DSM IV lists PTSD as an anxiety disorder, but PTSD also has elements of depression and needs no prior genetic history of either depression or anxiety to become chronic if un- or under-treated. PTSD’s impact on sleep has some controversy depending upon the population and the parameters being studied. Given the link between pain and PTSD, and how PTSD alters sleep, sleep must be a primary focus of discussion with the chronic pain patient.

The role of, and alterations in, sleep in all chronic illnesses cannot be understated. Disturbed phase 2/3 and REM sleep decrease pain threshold, impact immunity, decrease insulin sensitivity, and undermine the condition under treatment.

Mellman15 performed a meta-analysis on 20 PSG studies of PTSD and found increases in stage 1 (onset of sleep cycle-light sleep), decreased stage 3 sleep ( restorative, delta wave sleep), and increased REM density. REM is rapid eye movement state where active dreaming appears to occur.

Roth et al (2007) showed that with the fragmentation of REM sleep there is evidence for increased arousals and awakenings from this stage of sleep with PTSD. These arousals coincide with increases in heart rate variability, increases in circulating catecholamines, and increases in cortical activity on EEG. The effects of this condition on the HPA axis are self-evident.

Roehrs et al16 expanded on the observations that disturbed sleep is observed in association with acute and chronic pain and that some data suggest that disturbed and shortened sleep enhances pain. They studied healthy, pain-free, individuals using modest reductions of sleep time and specific loss of rapid eye movement (REM) sleep and showed that these effects produced hyperalgesia the following morning. Specifically, the loss of four hours of sleep and specific REM sleep loss resulted in hyperalgesic the following day. These findings imply that pharmacologic treatments and clinical conditions that reduce sleep and REM time during the normal sleep cycle may increase pain. Sleep and pain generate disturbances in each other and that cycle keeps pain thresholds low. Pain can also amplify anxiety and depressive symptoms as well as impair neuro-cognition.

The culprit is not only pain, however. Many of the medications we prescribe impair sleep. There are many articles that detail a wide variety of effects of CNS drugs on the structure of sleep. We must first inquire about sleep in such a way to eliminate the major primary sleep disorders such as sleep apnea, restless leg syndrome, etc. Short acting opioids are more disruptive to sleep than long acting ones. Benzodiazepines do not restore normal sleep and, especially the short acting drugs in this class, disrupt the normal stages of sleep. The report that a patient has slept for six to eight hours does not mean that they have had restorative sleep.

We argue that, in the equation of pain treatment, sleep is the “holy grail.” Clearly, sleep has direct effects on pain tolerability and the pulsing of neuro-hormonal events critical to maintaining health. And there is a strong cognitive behavioral component that is affected by, and affects, sleep.

Affleck et al17 did a sleep diary study of 50 fibromyalgia patients and recorded sleep quality, pain intensity, and attention to pain on a computer programmed as an electronic interviewer. As you would expect, a poor night’s sleep was followed by a more painful day. What is most interesting, however, was that, even in the absence of significant changes in pain during the day, the mere anticipation through increased attention to pain impacted sleep that night. The implications for cognitive–behavioral interventions are clear.


Addiction is a major problem for our patients, our society, and the sanctity of our practices. Determining addictive or diverting behavior is simply not possible without monitoring. Our laboratory has several papers (in press) showing that confirmatory testing of all negative values from Point of Care Testing (POC) in the office offers substantial proof that there are patients who take more than we prescribe. Without such information, regardless of the provider, we just can’t predict who they are.

In one sample of over 4,000 patients, we detected that 28% were taking unprescribed benzodiazepines and double the number detected by POC were using cocaine. The Substance Abuse and Mental Health Services Administration (SAMHSA) has recently lowered the cutoffs values for detecting abusable drugs and those changes will become apparent in the next year.

We will examine, in future articles, the issues surrounding drug abuse and the possibility it may be a form of self-medication for co-morbid conditions such as depression and anxiety. Unfortunately, after these drugs have been used for extended periods, it is often difficult to make this determination until patients are detoxified from the offending agent.


This article has attempted to bring together information from divergent fields that impacts our care of these very complicated patients. Recent changes to reimbursement and regulations governing our practice have stressed the pain physician in ways similar to what we have described in our patients. You are not alone in your frustration on how best to care for your patients. Hopefully this has stimulated your thinking and will raise questions we can all benefit by discussing.

Last updated on: January 28, 2012
close X