Outcome Domains and Measures in Acute and Chronic Pain Clinical Trials

~

An important feature in the planning of any clinical trial regards the decisions made as to the outcome criteria that will be used to evaluate treatment efficacy. Traditionally, the outcomes used in both acute and chronic pain research have been a reduction in pain intensity and presence of adverse effects (AEs). However, more recently, other important factors have been identified that should be included as outcome measures.

Several groups have reviewed and suggested outcomes to be considered in clinical trials, both general and well as disease specific (e.g. the Initiative on Methods, Measures, and Pain Assessment in Clinical Trials; IMMPACT), Core Outcome Measures in Effectiveness Trials (COMET), outcomes measures in rheumatology, NIH task force, and Veterans Health Administration Workgroup. The objective of COMET is the development and application of agreed standardized sets of outcomes, known as “Core Outcome Sets” (COS). They noted that the sets represent the minimum that should be measured and reported in all clinical trials of a specific condition and that COS are also suitable for use in routine care, clinical audit, and research, other than randomized trials. They stated that the existence or use of a COS does not imply that outcomes in a particular study should be restricted to those in the relevant core outcome set. COMET “aims to collate and stimulate relevant resources, both applied and methodological, to facilitate the exchange of ideas and information, and to foster methodological research in this area.” A representative example of the group’s publications to accomplish this primary aim is a COS on effectiveness trials.

This chapter will discuss the outcome domains and measures advocated by several groups, namely, IMMPACT, ^, pedIMMPACT, and APS-POQ-R as they provide comprehensive recommendations. The domains enumerated by IMMPACT, pediatric IMMPACT (pedIMMPACT), and American Pain Society-Post-operative Questionnaire-Revised (APS-POQ-R), (i.e. pain; physical functioning; emotional functioning; participant ratings of improvement and satisfaction with treatment, symptoms and adverse events; and participant disposition [adherence to the treatment regimen and reasons for premature withdrawal from the trial]) were developed by a set of individuals with experience in conducting clinical research, quality improvement, providing clinical care, and several lay advocates. It is important to acknowledge that patients may have different ideas regarding the most important outcomes for them. For example, when adults with chronic pain were surveyed, they affirmed that the set of domains recommended by these groups were important to them. However, they also described the importance of improving sleep quality, reducing fatigue, and improving the enjoyment of life following treatments. We will review domains and measures for acute and chronic pain for adult and pediatric populations specifically proposed by IMMPACT, pedIMMPACT, and APS-POQ-R. Moreover, patient satisfaction, a reflection of the interaction of the different domains, will be discussed in detail as a complement to the others proposed.

Besides considering important outcome domains, we will discuss general issues in selecting appropriate outcome measures to address the research questions of interest and to describe some of the most frequently used instruments in clinical trials conducted in anesthesiology and pain medicine. We do not intend to present an exhaustive list or provide a comprehensive critique of the psychometric properties and the strengths and weaknesses of all measurement tools enumerated. Instead, we will discuss several measures that have been selected by knowledgeable investigators of clinical trials in anesthesiology and pain medicine. We have included an extensive set of references to accompany this chapter and that are electronically available that readers can consult depending on their particular interests. Note, that although we focused on measures that can be considered for use in clinical trials, many of the measures described should be considered appropriate for use in clinical practice.

Selecting Outcome Measures for Use in Clinical Trials

Although the outcome domains for acute and chronic pain trials in both adults and children overlap, the measures selected will vary to some extent based on the age of the patients and their ability to comprehend the concepts entailed in the measures, the amount of patient burden required to respond, and the experimental context (e.g. developmental stage; cognitive limitations, postoperative vs. chronic pain state). There are instances where the same measures can be used across different ages, type of pain, and treatment contexts (e.g. perioperative vs. chronic). For example, a single question regarding pain severity or several questions regarding mood, ability to engage in functional activities, and the presence of AEs may be acceptable in both acute and chronic pain trials. Similarly, simple ratings of pain intensity may be used across certain ages (eight to 18) but not for all (e.g. neonates) or those with the inability to communicate (unconscious or otherwise cognitively compromised patients). The availability of appropriate normative data for the sample to be included in a trial becomes an essential consideration in selecting from available measures.

Throughout this chapter, we will describe a wide range of measures, particularly patient reported outcome measures (PROMs). These measures are essential in assessing a subjective state such as pain and the impact that pain has on all domains of an individual’s life—physical functioning, emotional, and broadly conceptualized as health-related quality of life (HR-QoL) as emphasized by IMMPACT, pedIMMPACT, and APS-POQ-R. In selecting from the vast array of general, disease specific, and treatment specific PROMs available, investigators, as well as clinicians, need to pay attention to a host of factors outlined in Table 84.1 . The set of concepts outlined are typically referred to as the psychometric or clinometric characteristics of the measure or procedure.

For measures dependent on physical assessment (1) the inter-rater reliability (Would the same result be obtained from two or more raters of the same patient?) and (2) intra-rater reliability (Would the same rater obtain comparable results if the procedure is performed on the same patient at two points in time?) need to be considered. For physical assessment, it is important to acknowledge the importance of training for examiners to maximize agreement.

Outcome Domains and Measures in Adult Acute Pain Trials

For acute pain trials, both IMMPACT and APS-POQ-R recommend that intensity and pain relief should be the primary outcome domains for and clinical trials and quality improvement (QI). IMMPACT suggested that an NRS be used. Verbal rating scales (VRS) for pain intensity (“no pain,” “mild pain,” “moderate pain,” “severe pain”) and amount of pain relief (“no relief,” “little relief,” “some relief,” “moderate relief,” “complete relief”) have also been frequently used. APS-POQ-R recommended using an 11-point NRS (zero, no pain, to ten, worst pain imaginable) for two questions—pain at its least and at its worst over the past 24 h (note that the interval could be shorter depending on the patient’s ability to respond and the research question, for example, 6 h). In addition, APS-POQ-R recommended a third question, the percentage of time the patient experienced severe pain in the past 24 h. An additional outcome of potential interest noted was time to first analgesic intake. Consideration should also be given to the timing of pain assessment during recovery. Pain ratings should also pain at rest (e.g. lying down) and when appropriate on specific movements (e.g. coughing, repositioning, ambulating).

The occurrence of AEs is an important outcome domain. The usual side effects should be observed and when patients are able directly reported by them. APS-POQ-R specifically suggests assessing the severity of four common postoperative side effects, nausea, drowsiness, itching, and dizziness, with each rated on an 11-point NRS when patients are capable of making such ratings. A caution regarding solicitation reports of side effects from patients is noteworthy. Asking about the presence of specific side effects may prompt patients to report the occurrence of these and increase the number, whereas an open-end question asking patients if they have experienced or are experiencing “any” side effects may lead to missing some that the patients do not consider reporting. Thus there are some trade-offs with either approach.

The impact of pain and treatments on other AEs have also been assessed. For example, studies have focused on the degree of the patient’s sedation (e.g. observer’s assessment of alertness/sedation scale, Ramsey sedation scale), sleep quality or interference (Pittsburgh Sleep Quality Index, Epworth Sleepiness Scale), or the presence of delirium. The Mini-Mental Status Examination (MINI) has been used to measure sedation as well as the cognitive ability of the patient. The Opioid-Related Symptoms Distress Scale has been employed to look specifically at symptoms related to opioid intake; these include difficulty passing urine, headache, feeling of lightheadedness, fatigue, or weakness. It should be noted that there is no single assessment tool designed to assess all the side effects, mostly the focus is on a specific one.

Both IMMPACT and APS-POQ-R suggest that physical functioning is an important outcome domain for acute pain clinical trials. APS-POQ-R suggests asking four questions related to the inference of pain with physical functioning, specifically, sedentary activities (i.e. turning, sitting up, repositioning), activities out of bed (i.e. walking, sitting in a chair, standing at the sink), and falling asleep and staying asleep. They suggest that each of these questions should be rated on an 11-point NRS. For perioperative studies, it was suggested that time to first bowel movement, length of intensive care unit (ICU) and/or hospital stay, and patient satisfaction be considered as appropriate. For longer duration acute pain studies as in the recovery room, intensive care, or emergency department, IMMPACT and APS-POQ-R recommended additional outcomes domains for physical function such as the timing of ambulation. Assessment of other functional activities to consider will depend on the context and duration of pain. Some of the measures used for assessing physical functioning in chronic pain discussed below may be appropriate to consider depending on the time interval.

Emotional functioning is the least studied outcome domain in postoperative pain studies, although it is important as it may influence the rate of recovery, length of hospital stay, and patient satisfaction with the treatment. APS-POQ-R recommends four questions rating emotional functioning, each on an 11-point NRS scale, namely how anxious, depressed, frightened, and hopeless patients felt. In the immediate postoperative period, these simple measures of emotional distress may be most reasonable to decrease patient burden. However, in longer duration acute pain trials (e.g. days, weeks), more comprehensive tools for assessing anxiety may be considered, such as the Yale Preoperative Anxiety Scale, the Hospital Anxiety and Depression Scale (HADS), and the Generalized Anxiety Disorder-7 (GAD-7). There are multidimensional tools used in chronic pain investigations (discussed below) that have been utilized in acute pain studies, when reasonable given the context of acute pain (e.g. emergency department, ambulatory care clinic), that assess several outcome domains ( Table 84.2 ). These include the Brief Pain Inventory (BPI), Medical Outcomes Study Short Form Health Survey SF-36 and its modifications (SF-12 and SF-8), and EuroQoL 5-D. These instruments have been shown to have very good psychometric properties when used with chronic pain samples. However, they should be cautiously used as they were not specifically designed or validated for use in the immediate postoperative period.

Another multi-domain measurement tool is the International Pain Outcomes Questionnaire (IPO), developed by the European Commission-funded PAIN OUT project. IPO includes domains of pain intensity, physical and emotional, functional interference, side effects, and perceptions of care. The questionnaire was finalized after two modifications: (1) phase one to five questions were added to the 14 item APS-QoR-R, (2) phase two to eight questions were removed from the questionnaire. As a result, the final version had high internal consistency reliability. However, although promising, to date, not many publications have reported results using the IPO.

Perhaps the comprehensive measure and one that has received the greatest amount of research is the set of quality of recovery (QoR) instruments. These were initially developed and validated for use in the perioperative period. The QoR is the most commonly used instrument in postoperative pain studies, and it appears that a shortened version, the QoR (QoR 9, QoR 15), may be a reasonable alternative, and thus we describe it in some detail.

Myles group formulated a comprehensive QoR40, simpler nine item QoR questionnaire, and more recently, a QoR15. The set of QoR tools measure five dimensions of health: pain, physical comfort, emotional state, psychological support, and physical independence. All of the QoR instruments have very good validity and reliability and excellent completion and return rates. The MCID, i.e. ­minimal change in score that indicates a meaningful change in a patient ­status —improvement or deterioration—was noted to be 0.92, 6.3, and 8.0 for the QoR, QoR40, and QoR15, respectively.

The QoR 15, which has fewer items than the QoR 40 and with similar psychometric properties, has fewer items and takes 3 min to complete ( Figure 84.1 ). Utilization of the QoR15 (or the QoR40), the IPO, or the APS-POQ-R includes most of the outcome domains recommended by the IMMPACT group.

While patient satisfaction is not directly addressed in the QoR instrument ( Table 84.2 ), the authors noted that quality of recovery is directly related to patient satisfaction. In one study, the investigators noted a strong association between the original nine item QoR and patient satisfaction. Since the QoR does not specifically assess patient satisfaction, investigators should consider including a measure of patient satisfaction when using the QoR.

Outcome Domains for Adult Chronic Pain Clinical Trials

As noted, based on a review of the literature, IMMPACT recommended that six core domains should be considered in chronic pain clinical trials: (1) pain; (2) physical functioning; (3) emotional functioning; (4) participant ratings of improvement and satisfaction with treatment; (5) symptoms and adverse events; and (6) participant disposition (adherence to the treatment regimen and reasons for premature withdrawal from the trial) ( Table 84.3 ). Moreover, IMMPACT indicated that other domains might be considered depending on specific research questions addressed in the trial (e.g. clinician or surrogate ratings of improvement, interpersonal functioning). The group emphasized that there are exceptions to using all the domains but the rationale for not including a domain should be explained. Additionally, the group acknowledged that statistically significant effects for all the domains were important to establish the efficacy of the intervention, however, investigators should indicate the primary outcome(s) of interest.

The IMMPACT domains depend primarily on patient self-report (PROMs). Self-reports assess experience and behavior at a single point in time or evaluate an experience averaged over a recalled period (e.g. last seven days). However, pain can fluctuate over time depending on the activity and mood of the patient and on environmental factors that can impact emotional or physical function. Thus more frequent ratings using daily diaries and real-time data acquired electronically may be useful ways to evaluate the impact of treatments as related to circumstances that are influenced by a variety of factors (e.g. pain intensity at rest versus during movement, time of day as symptoms may vary).