It is unethical to recruit patients to studies that are not sensitive to the parameter being tested. The patients' time and effort is wasted and they may have to endure suboptimal treatment unnecessarily. Studies that draw negative conclusions because they are insensitive rather than because the treatment is ineffective mislead clinicians and can distort literature summaries, such as meta-analysis.49

Multiple design details can influence the sensitivity of a study. The "floor effect'' occurs when the initial pain level is low so that it is difficult for a study to detect a decrease. Studies should therefore be on "properly painful conditions'' (equivalent to 30 mm on a 100-mm visual analog scale (VAS)).50 Assessment during standardized provocative movement rather than at rest may result in greater

Table 14.3 Measures of efficacy and effectiveness in analgesic studies.


Example of suitable measures


Physical functioning

Emotional functioning

Participant rating of global improvement and satisfaction with treatment Symptoms and adverse events

Participant disposal

11-point (0-10) numerical rating scale of pain intensity Use of rescue analgesics

Categorical rating of pain intensity (none, mild, moderate, severe) in circumstances in which numerical ratings may be problematic Multidimensional Pain Inventory (MPI)39 - (interference scale) Brief Pain Inventory (BPI)40,41 (interference items) Beck Depression Inventory (BDI)42 and/or Profile of Mood States (POMS)43 Patient Global Impression of Change (PGIC)44

Passive capture of spontaneously reported adverse events and symptoms and use of open-ended prompts

Detailed information regarding participant recruitment and progress through the trial, including all information specified in the CONSORT guidelines45

study sensitivity. Additionally, pain scores can be calibrated to take into account patient's previous experience of pain using individual responder analyses51 - this may increase sensitivity of trials in diverse populations. In clinical trials aiming to demonstrate a difference in level of pain, concurrent treatments will be discouraged in order to maximize sensitivity (whereas in clinical practice, any measure that reduces symptoms will be encouraged).

Other strategies for maximizing study sensitivity include:

• eliminating placebo responders from the study by using a placebo run-in period in all groups;

• excluding patients who may have incentive not to improve (e.g. seeking or receiving disability compensation); and

• studying only severe and clearly defined cases.

This is particularly important where diagnosis is contentious, such as with complex regional pain syndrome or fibromyalgia.

It is important that investigators delivering treatment are trained to be effective, particularly when psychosocial interventions are delivered52 and that patients are fully trained in completing the tests before and during the study.53 There are subtle interpersonal effects at the point of assessment - if the experimenter has elevated professional status or is of the opposite sex to the subject, pain scores will be lower and pain threshold higher.54 All these factors should, ideally be standardized in any study of efficacy.

Sensitive studies will be more powerful in detecting a genuine difference between two treatments. Wide patient variability accounts for lack of power in many analgesic studies and when the difference between the control and experimental event rate is small - studies may detect only statistically rather than clinically significant differences.55 Patient management often changes in subtle ways around a clinical study. Improvements in measures that can be attributed only to the fact that a study is being conducted are referred to as the Hawthorne effect.56


Simple brief measures are attractive and have in some cases been validated. A four-grade (none/mild/moderate/ severe) scale compares favorably with a VAS in adult migraneurs and is easier to administer,57 and the cut points (equivalent on a numerical scale) of this type of scale are generally reproducible.58 Many factors will influence the level of rating:

The action of arriving at a rating is better conceptualised as an attempt to construct meaning, influenced by and with reference to a range of internal and external factors and private meanings, rather than as a task of matching a distance or number to a discrete internal stimulus.59

Simple measures may therefore approach the validity and utility of more detailed and intensive measures of chronic pain60 or acute pain,61 particularly when large populations are studied.


Prompt recording of data (contemporaneous) is important for accuracy because delayed recall increasingly introduces interference from the condition at the time of recall.62 Electronic recordings using hand-held or internet-based devices or via automated telephone systems have replaced many paper systems and are reliable in adults63 and children.64


In general, parallel group studies should conform to CONSORT guidelines;45 however, as we have seen, there are many considerations relating to chronic pain. Surgical treatments avoid most problems of compliance and specific checklists for studies have been advocated for tri-geminal neuralgia.29 Progress with measures for complex, multicomponent treatments, such as pain management programmes, has also been reviewed and there are many difficulties remaining.65 It is no surprise, therefore, that it is exceptionally difficult to compare surgical and non-surgical treatments.66

Naturally Cure Your Headaches

Naturally Cure Your Headaches

Are Headaches Taking Your Life Hostage and Preventing You From Living to Your Fullest Potential? Are you tired of being given the run around by doctors who tell you that your headaches or migraines are psychological or that they have no cause that can be treated? Are you sick of calling in sick because you woke up with a headache so bad that you can barely think or see straight?

Get My Free Ebook

Post a comment