Wednesday, March 6, 2013

Ratings of Pain Intensity: A Numbers Game

Briefly NotedThe classification of pain intensity as mild, moderate, or severe is important for everyday clinical practice and in research studies of pain therapies. However, since pain and interpretation of its severity are subjectively perceived by patients, defining thresholds between the 3 levels of pain intensity on a numeric rating scale can be difficult. A recent study examined the variability in how pain intensity is interpreted across patients and in those with different pain conditions.

In their study report, researchers from Germany writing in the journal PAIN® observe that defining cut points for mild, moderate, and severe pain intensity on the basis of differences in functional interference has an intuitive appeal and has been widely used in other research [Hirschfeld and Zernikow 2013]. They devised a trial in a large sample of patients with chronic pain to explore the variability associated with optimally defined “cut points” — ie, threshold values distinguishing between mild, moderate, and severe pain levels on a 0-10 numeric rating scale, or NRS. Ratings of maximum pain intensity and associated pain-related disability were collected from 2,249 youngsters with chronic pain being managed in a tertiary pain clinic; mean age was 12.45 years; 61% were girls.

The data were analyzed from 3 perspectives:

  • First, the "optimal" cut points — that is, thresholds distinguishing between mild-to-moderate and moderate-to-severe pain and disability — were determined for the sample as a whole.

  • Second, the variability of these cut points was quantified by a “bootstrap” technique. This is a statistical modeling process whereby the same analysis is repeated hundreds of times on randomly chosen subsets of data and the variation of results is examined.

  • Third, the variability of scores was also assessed in homogeneous subsamples of the study group: 650 children with constant pain, 430 children with chronic daily headache, and 295 children with musculoskeletal pain.

This approach revealed 3 main findings:

  1. The optimal cut points on the NRS between mild-moderate, and moderate-severe, pain in the whole sample were 4 and 8, respectively.

  2. The variability of these cut points within the whole sample was very high, identifying the optimal cut points only 40% of the time.

  3. Similarly large variability was also found in subsamples of patients with a homogeneous pain etiology. The cut points between mild-moderate and moderate-severe pain, respectively, were 5 and 8 in patients with constant pain; 4 and 8 for chronic daily headache; and 2 and 8 for musculoskeletal pain.

The researchers conclude that clear optimal cut points could not be identified, and the thresholds are strongly influenced by random fluctuations and chance variation within any sample of patients. Research studies and clinical assessments that aim to interpret differences between groups or the statuses of individual patients need to take into account considerations of randomness and variability in those measurements.

COMMENTARY: The NRS has been a very popular pain-intensity assessment tool used in research and clinical practice. It is preferred by patients with chronic pain and has high test-retest reliability [Hawker et al. 2011]. This present study, however, demonstrated that the cut points, or thresholds, distinguishing between mild, moderate and severe pain intensity are not straightforward.

The current study for the most part involved preteen children and adolescents, an age group in whom the NRS has been well-validated. The scale has been applied equally effectively at the other end of the age spectrum in adults older than 60 years [Wood et al. 2010].

In this study by Hirschfeld and Zernikow [2013], the lower threshold between mild-moderate variously ranged from 2 to 5 points; whereas, the estimated upper cut point between moderate-high remained stable at 8 points in the total population and across subgroups. However, the researchers provide a data table showing that in various other study samples reported in the literature the cut point between moderate-severe pain has ranged from 6 to 8 point.

Which cut points are used can make important differences for interpreting effects of treatments in research studies; eg, two studies of the same therapy may report significant differences in treatment efficacy and clinical improvement depending mainly on the cut points used. In this regard, without firmly established cut points, comparisons across research studies can become difficult, especially when numeric ratings are merely categorized for nonparametric statistical analyses as pain that is either mild, moderate, or severe.

On an individual patient basis, it is important to consider that many analgesics have distinct indications for mild-to-moderate pain, while others are indicated for moderate-to-severe pain. If objective tools like NRS scales are used to assess pain intensity level, but there are unclear distinctions between pain categories, will prescribing decisions be made appropriately?

In the United States, the need for well-delineated distinctions has become more urgent due to a petition submitted last summer to the Food and Drug Administration requesting that the labeling for opioid medications be changed to limit their use in noncancer pain to only “severe” pain [see the most recent discussion of the “PROP Petition” here]. Yet, this current research suggests that distinctions between moderate and severe pain may vary by up to 27%, or 3 points on the 11-point (0-to-10) NRS, depending on how thresholds are defined.

This sort of ambiguity could no doubt lead to misprescribing of these opioid analgesics, depending on whether practitioners adopt conservative or more liberal interpretations of pain intensity thresholds. Certainly, there is a need for a more definitive threshold value denoting moderate versus severe pain intensity, but it is also possible that different values might be needed for distinct pain conditions and groups of patients.

> Hawker GA, Mian S, Kendzerska T, French M. Measures of adult pain. Arth Care Res. 2011(63(S11):S240-252 [
> Hirschfeld G, Zernikow B. Variability of "optimal" cut points for mild, moderate, and severe pain: neglected problems when comparing groups. PAIN. 2013;154(1):154-159 [
> Wood BM, Nicholas MK, Blyth F, et al. Assessing pain in older people with persistent pain: the NRS is valid but only provides part of the picture. J Pain. 2010;11(12):1259-1266 [

eNotifications Don’t Miss Out. Stay Up-to-Date on Pain-Topics UPDATES!
Register [here] to receive a once-weekly e-Notification of new postings.