Pain is an internal, subjective experience that cannot be measured by the use of physiologic markers or bioassays. The assessment of pain, therefore, relies largely (often exclusively) on the use of self-report. What is more, individual differences in pain sensitivity have long remained a perplexing and challenging clinical problem. For instance, pain assessment can often vary among individuals based on patient demographic (e.g., culture, gender) and facial expression cues. Therefore self-report measures following a multidimensional approach aim at filling this gap with determination of the following: chronicity, severity, quality, associated factors, and location.In special populations, limited cognitive or language skills may influence pain measures, as may the positive or negative consequences of an individual’s behaviors associated with pain. More recently, real-time data capture using electronically based pain assessments (e.g., pain apps) represents another promising method for the field.
Keywordsbehavioral observation, individual differences, pain assessment, scales, self-report measures, special populations, technology
By its very definition, pain is an internal, subjective experience that cannot be directly observed by others or measured by the use of physiologic markers or bioassays. The assessment of pain therefore relies largely (often exclusively) on the use of self-report. Though self-report of pain or any other construct is subject to a number of biases, a good deal of effort has been invested in testing and refining self-report methodology within the field of human pain research. The purpose of this chapter is to provide an overview of this research, to critically evaluate pain assessment tools, and to assist clinicians and researchers in selecting the pain assessment methods best suited to serve their purposes.
Challenges of Pain Measurement
Assessing pain requires measurement tools that are valid and reliable, as well as an ability to communicate (using language, movements, etc.). However, even when these basic requirements are met, additional challenges abound. For example, over what time frame is pain to be measured? By nature, most pain conditions are fairly variable, and it is sometimes unclear how representative (of a patient’s global pain experience) ratings of current or recent pain might be. Many ratings scales do query current pain, or pain over the past week, but longer time frames are often used, and these may introduce additional memory biases. In addition, pain is a multidimensional experience incorporating sensory and affective components that are correlated, but which may be assessed separately. Generally, most self-report pain assessment tools described here focus on pain intensity ratings over a relatively brief and recent period of time (e.g., the past week).
One major challenge in pain assessment is the inability of an evaluator to corroborate, using objective signs or indicators, a patient’s subjective report of pain. Clinicians have been trained to ask patients about their pain level at each encounter, and use that number as a vital sign. These methods, used within a health care environment that encourages patient-centered care, have potentially led practitioners to overprescribe pain medications, as argued in recent reviews of the field. Measurement of pain should be multifaceted, and may benefit from the inclusion of functional capacity evaluations (FCE). FCE protocols generally include anthropometric measurements, range of motion testing, static strength testing, cardiovascular endurance, muscle recruitment, stance, and end range of motion movements. FCE can be used for global function measurement or more specifically for selected body regions. Results can be analyzed using both normative and criterion-referenced testing. Moreover, FCE allows assessment of the level of physical effort provided by the subject during an evaluation, which affects both the reliability and the validity of the result. Effort testing at its peak gives the reader of the functional evaluation report confidence that the results represent a true picture of the individual’s ability on the day of the evaluation. Reports compromised by less than full effort are read as representing the injured person’s minimum ability.
Types of Self-Report Pain Scales
A variety of pain assessment scales are available for evaluating the intensity of acute and chronic pain. Multiple types of scales are widely used and well validated in both research and clinical settings. The three most commonly used methods to quantify the pain experience (pain intensity, usually) are verbal rating scales (VRSs), numerical rating scales (NRSs), and visual analog scales (VASs).
Verbal Rating Scales
A VRS generally consists of a series of adjectives (or phrases), ordered from least intense (or unpleasant) to most intense (or unpleasant). An adequate VRS should span a maximum possible range of the pain experience (e.g., from “no pain” to “extremely intense pain”). Patients are asked to select the adjective or phrase that best characterizes their level of pain. Dozens of VRS have been described and validated; one of the more common examples appears in Table 5.1 .
In general, a VRS is scored by assigning each adjective or phrase a number according to its rank (e.g., 0–4 in the example in Table 5.1 ). The strengths of the VRS include simplicity, ease of administration and scoring, as well as face validity (i.e., they appear to directly measure exactly what they purport to measure, such as the intensity of pain). In addition, because they are so easy to comprehend, compliance rates for the VRS can be superior to the rates obtained with other scales, especially within certain populations such as the elderly. The VRS has demonstrated good reliability (e.g., consistency over short periods of time) in a number of studies. The validity of the VRS has also been repeatedly established; these scales correlate positively with other self-report measures of pain intensity and with pain behaviors.
Despite its substantial strengths, the VRS also exhibits a number of weaknesses, based on which other pain researchers have hesitated to recommend these scales. First, the scoring method for VRS assumes equal intervals between adjectives. That is, the change in pain from “none” to “mild” is quantified identically with the change in pain from “moderate” to “severe.” This assumption is rarely tested, and is likely often violated. This property of the VRS poses difficulties in both the interpretation and analysis of VRS-derived data. Second, to properly use a VRS, a patient must both be familiar with all of the words used on the scale, and must be able to find one that accurately describes his or her pain. Some past reviews of the pain assessment literature have indicated that the VRS is being used less often in pain outcome research than has been the case in the past.
Numerical Rating Scales
An NRS typically consists of a series of numbers with verbal anchors representing the entire possible range of pain intensity. Generally, patients rate their pain from 0 to 10, from 0 to 20, or from 0 to 100. Zero represents “no pain,” whereas 10, 20, or 100 represents the opposite end of the pain continuum (e.g., “the most intense pain imaginable,” “pain as intense as it could be,” “maximum pain”). See Fig. 5.1 for an example. Like VRSs, the NRS have well-documented validity; they correlate positively with other measures of pain and show sensitivity to treatments that are expected to affect pain. The NRS can be administered verbally or in a written format, is simple and easily understood, and is easily administered and scored. The principal weakness of the NRS is that, statistically, it does not have ratio qualities. That is, numerically equal intervals on the scale (e.g., the difference between 1 and 3 and the difference between 7 and 9) may not represent equivalent intervals in terms of scaling the intensity of pain. One other limitation of most NRS measures of pain is that individuals’ ratings of a given pain experience can be altered in idiosyncratic ways by the choice of anchors on the upper end of the scale. For example, women and men use systematically different events to contextualize the anchor of “most intense pain imaginable,” and this can significantly affect studies of gender differences in the experience of pain.
Visual Analog Scales
A VAS consists of a line, often 10 cm long, with verbal anchors at either end, similar to an NRS (e.g., “no pain” on the far left and “the most intense pain imaginable” on the far right). The patient places a mark at a point on the line corresponding to the patient’s rating of pain intensity. The line may be depicted with a horizontal or vertical orientation, though a horizontal line is generally preferred ( Fig. 5.2 ). Recent versions include the mechanical VAS, which uses a sliding marker superimposed on a horizontal VAS drawn on a ruler, and is easily scored from the back, which includes numbers for each marker placement. The VAS has often been recommended as the measure of choice for assessment of pain intensity. Substantial evidence supports its validity, and the VAS is sensitive to treatment effects. Though most studies suggest minimal differences in sensitivity among rating scales, significant differences that do emerge generally favor a VAS over a VRS or an NRS. In addition, VAS scores correlate with pain behaviors, and VAS scores do show ratio-level scoring properties. The VAS does possess some limitations, however. It can be difficult to administer to patients with perceptual-motor problems, which are rather common in the context of chronically painful conditions. In addition, a VAS is generally scored using a ruler (the score is the number of centimeters or millimeters from the end of the line), making scoring more time consuming and adding additional possible sources of bias or error. Finally, relative to other rating scales, use of a VAS produces higher noncompletion rates among certain populations, primarily among those with cognitive limitations and among elderly samples (discussed later).
Mcgill Pain Questionnaire
The McGill Pain Questionnaire (MPQ) and its brief analog, the short-form MPQ, are among the most widely used measures of pain. In general, the MPQ is considered to be a multidimensional measure of pain quality; however, it also yields numerical indices of several dimensions of the pain experience. Researchers have proposed three dimensions of the experience of pain: sensory-discriminative, affective-motivational, and cognitive-evaluative. The MPQ was created to assess these multiple aspects of pain. It consists of 20 sets of verbal descriptors, ordered in intensity from lowest to highest. These sets of descriptors are divided into those assessing the sensory (10 sets), affective (5 sets), evaluative (1 set), and miscellaneous (4 sets) dimensions of pain. Patients select the words that describe their pain, and their word selections are converted into a pain-rating index, based on the sum of all of the words after they are assigned a rank value, as well as the total number of words chosen. In addition, the MPQ contains a present pain intensity VRS (i.e., the Present Pain Intensity [PPI]), ordered from “mild” to “excruciating.”
The more frequently used short form of the MPQ consists of 15 representative words that form the sensory (11 items) and affective (4 items) categories of the original MPQ. Each descriptor is ranked on a 0 (“none”) to 3 (“severe”) intensity scale. The PPI, along with a VAS, are also included ( Fig. 5.3 ). The short form correlates highly with the original scale, can discriminate among different pain conditions, and may be easier than the original scale for geriatric patients to use.
Studies of interventions designed to reduce pain often include a posttreatment assessment of pain relief in addition to measures of pain intensity obtained at both baseline and posttreatment. Pain relief is often measured using a VAS, a VRS with gradations of relief (e.g., “none,” “slight,” “moderate,” “complete”), or an NRS assessing the percentage of relief. Studies showed VAS and VRS responses to be highly correlated and the scales demonstrated similar effect sizes after treatment, though administration and interpretation of VRS was reported to be easier than with other types of scales. Although conceptually attractive, measures of pain relief have demonstrated problems with validity. For example, a significant minority of patients reports at least moderate relief on these scales when an analysis of sequential pain ratings (i.e., pretreatment compared with posttreatment) reveals increases in reported pain intensity. In one trial, whereas average pain ratings increased by 28% early in the study, approximately 90% of patients reported some degree of relief on a VAS. This phenomenon (i.e., the apparent overreporting of relief) seems to be due in part to a memory for past pain as being substantially greater than previous ratings would indicate.
Multidimensional Assessment in Clinical Trials
Although a full exploration of recommendations for outcome assessment in analgesic trials is beyond the scope of this chapter, the Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT) has produced a series of publications on this issue. Interested readers should consult one of the IMMPACT position papers (e.g., Dworkin et al.). In brief, this group of experts has reviewed measures of pain intensity, physical functioning, emotional functioning, and other pain-relevant outcome domains, making recommendations for the selection of outcome measures for clinical trials of pain treatments.
Differentiating Types of Pain: There has been a good deal of interest in the development of self-report measures of neuropathic pain; indeed, the MPQ has been studied in this context, and over recent years several screening tools for distinguishing neuropathic from nociceptive pain have been validated. The Pain DETECT assessment system, which relies on a set of self-report questions about symptoms, was designed to detect neuropathic pain in patients with low back pain; it has been validated in large studies with thousands of patients, and has been reported to achieve reasonable sensitivity and specificity in identifying patients with neuropathic back pain. Other questionnaires such as the Leeds Assessment of Neuropathic Symptoms and Signs (LANSS) scale and the Neuropathic Pain Questionnaire (NPQ) have also been studied as indicators of the presence of neuropathic pain. However, whereas some validity studies suggest good results, several persistent issues have plagued the research in this area. The first is that while the definition of neuropathic pain indicates that a lesion or dysfunction must be present in the nervous system, this can often be difficult to establish in patients with chronic pain, creating a questionable “gold standard” against which the diagnostic accuracy of a questionnaire can be measured. Second, multiple studies strongly suggest that the endorsement of classically “neuropathic” symptoms (e.g., shooting pain, numbness and tingling, etc.) is strongly influenced by other patient characteristics, such as emotional distress, indicating that a wide variety of factors are likely to contribute to self-report of the presence of neuropathic pain.
Daily Diaries: In trials of pain treatments, daily diaries are gradually becoming the standard for assessing pain-related symptoms to minimize memory biases that threaten the validity of global retrospective ratings of pain. Participants are generally asked to complete measures of pain and related symptoms one or more times per day, often for 1 to 2 weeks. Because pain reports can have substantial day-to-day variability, aggregated (averaged) ratings have been demonstrated to be more reliable and sensitive to treatment effects than retrospective measures of pain. In general, recent research has favored electronic diaries; in comparison with paper-and-pencil diaries, electronic diaries (e.g., usually implemented using a PDA, cell phone, or similar tool) have been repeatedly shown to demonstrate superior compliance rates and patient satisfaction. Electronic diaries incorporate several features that enhance reliability, including: date/time stamping of all diary entries, and automatic rejection of erroneous data. Pain drawing has also been applied to the field of pain evaluation, offering patients the ability to provide information on the location and type of pain they are experiencing on a human body map. Such topographical representation of pain was suggested to be very useful in summarizing patients’ nature of their pain in an interpretable way for clinicians, and it can be further used to keep track of changes in patients’ pain experiences. More recently, a novel three-dimensional (3D) approach is used for the purpose of reporting pain characteristics as an alternative to the conventional 2D pain drawing.