Pain ratings aren't always effective

Pain is an extremely difficult subject to study, because individuals experience pain differently -- some people have a much greater tolerance for pain than others, and some people just seem to complain more. Doctors typically handle the problem with assessing pain by asking patients to rate the pain they are feeling on a scale of 1 to 10. This gets around some of the issues of measuring pain because the individual's subjective experience is accounted for by the rating scale.

But the 1 to 10 scale might also be problematic. People might exaggerate their pain just to get access to stronger painkillers. Worse, they might not be reporting pain because of problems with the measurement scale itself. A new study claims to be "the first to evaluate the accuracy of the widely-used numeric rating scale [NRS] as a screening test to identify primary care patients with clinically important pain." Here's what they found:

The authors found that, while the NRS is easy to administer, it fails to identify about a third of patients with pain serious enough to impair day-to-day functioning. Most patients in this study had long-standing pain, and many had more than one pain problem. The authors did not evaluate the accuracy of pain ratings in settings where short-term pain is more common, such as after surgery.

The researchers noted that because it focuses on current pain, the NRS may miss intermittent symptoms. They also reported that "pain" was not the preferred word for some patients. For example, one study participant indicated that he felt discomfort, but not pain.

So some patients experiencing significant pain are missed because they aren't in pain when the doctor happens to ask them about it, and others are missed because they don't share the doctor's understanding of the concept of "pain." To me this concept of impairing "day-to-day functioning" is also very important. Perhaps doctors need to be asking at least two questions about pain: First, the numeric rating scale, and second, a question about how they are functioning day-to-day.

Other CogDaily posts about pain:


More like this

by Kim Krisberg This is the first in a series exploring the intersections between effectively caring for people living with chronic pain and the rise in unintentional poisoning deaths due to prescription painkillers. The series will explore the science and policy of balancing the need for treatment…
Back when advocates of “alternative” medicine were busily trying to legitimize their quackery by first renaming it “complementary and alternative medicine” (CAM), long before CAM “evolved” into “integrative medicine,” they really believed that if their favorite woo were to be studied scientifically…
Dana Ullman is an idiot. Or maybe insane. I'm not sure which, but his latest article at the Huffington Post reveals such a severe defect in rational thought that it must be one or the other (charitably speaking). He calls it "Lies, Damn Lies (sic), and Medical Research," and the point of it is…
I sense a disturbance in the skeptical blogosphere. It is something that I half-expected, but, even so, it nonetheless somewhat surprised me when it arrived in the form of comments on my blog and e-mails from readers, fellow supporters of science-based medicine, and others asking me what I thought…

Yes, I really dislike the numeric pain scale thing. I suspect I've tended to under-report pain because when 10 is supposed to represent the worst pain imaginable... well, I have a rather vivid imagination, so I can usually think of several things that would have to hurt worse than my little kidney stone or whatever. My thought process goes something like, "What if my arm were slowly being cut off with a dull chain saw? That would probably hurt a LOT more than this kidney stone. So, I guess the kidney stone only ranks maybe a 6."

I've recently recalibrated my definition of the word "pain" after some treatment. I had always thought that "pain" started at the point where you could not suppress a grunt or scream, or could not fulfil a normal function like walking at all. Anything less was not "pain", but "hurt" or "discomfort" or "ache". I was quite surprised to find that my understanding was non-standard, and my doctor was equally surprised at my definition. If anybody had asked me to rate my pain, any pain, until recently, I would never have answered more than a 1 or 2 on a scale of 10. Now I have no idea at all how to assign a number to a pain.

I am amazed by the number of people who walk up to the nurses station at work with no visible signs of discomfort and then rate their pain as a 10. By the time you get to a rating of 8 I would expect someone to be shaking, sweating and almost crying.

I have always been infuriated by this 1-10 rating question. Even if I know what 1 and 10 are, it doesn't tell me what the ones in between are -- just degrees of increasing pain. Have there been any studies showing something measured by this number?

By stuart allen (not verified) on 24 Sep 2007 #permalink

An adult friend, dying of cancer, declined to use the 1-10 pain scale offered by her hospice case manager. She rejected it utterly, in favor of the more qualitative Roo to Eeyore scale, in which the most miserable day is an Eeyore, a slightly less troublesome day a Wol...and on through Piglet, Pooh, and the rest all the way up to the joyous Tigger and utterly blissful Roo days. Since both parties took care to thoroughly understand and document the revised scale, there was no loss of information.

Admittedly, at first there were some heated discussions about whether a Tigger should actually score better than a Roo, and some fine distinctions had to be made between Pooh-with-honey days and Pooh-out-of-honey, all of which led to the scale itself actually reducing the Eeyorishness of the evaluations.

I hear ya brother. I have MS and my pain is neurogenic, so it can't be scaled from 1 to 10. I mean where does "feels like over a barbeque, covered with bees" or "disembodied, stabbing, itching" rate?

Great article - thanks


There are similar serious problems with fatigue rating scales. They reach a ceiling too soon and don't differentiate properly between the more extreme fatigue states and the next level or two down. It is causing major problems in the research (and of course for overall patient assessment).

By Obdulantist (not verified) on 25 Sep 2007 #permalink

In my physical therapy practice and in the university where I teach, we often use a more comprehensive inventory of pain, The McGill Pain Questionnaire. It is a lot longer than just the NRS, but offers insight into some psychological and functional components of what the patient is experiencing as well. It even includes "torturing" type pain as a choice of descriptor for our friend with the active imagination!

I hate the 1-10 pain scale, as a patient. Presumably, if I were in the 8-10 range, or so, nobody would be asking, because I'd be so visibly incapacitated that there'd be no doubt. Then I start worrying about philosophical concerns, like if I can accurately recall or imagine pain levels, and psychological ones, like how the doctor will interpret my response (don't want to exaggerate or low-ball it, you'll get the wrong sort of care). I even end up with mathematical problems, since nobody said the scale was linear, perhaps it's logarithmic, like the Richter scale (which actually might make sense, but would shift the answers up quite a lot).

In the end, I've just decided to refuse to answer on the scale, and give a functional answer, like "makes it impossible to do X" or "much better than yesterday" or "like the time Y happened" or whatever's appropriate. Seems both more honest and more reliable.

By Matthew L. (not verified) on 26 Sep 2007 #permalink

As a medical student, I find it very difficult to assess pain.

In my pediatrics rotation, I found the faces scale very helpful. Kids could never give a number, but when handed the card with faces on it, would study it and give a well-calculated response.

Now that I ask adults about pain, the faces scale is not so useful. When someone says 7 or 8, I ask the question again comparing unmedicated labor (if a woman) as 7-9 and a kidney stone as a 9 or 10. Is that correct?

I will start to ask, "Does the pain interfere with any of your daily activities?" I agree that this is really the indicator of impairment, and therefore of treatment!

Stuart Allen et seq., yeah, from what I hear test experts are kind of dubious about 1-10 scales in general -- people handle 1-7 or even 1-5 much better.

For myself, I've managed to avoid any injuries worse than a sprained wrist or a couple of cracked ribs, but even so it's clear that my experience of pain is non-standard. I get a lot of phantom pains, and (relatedly) often don't notice minor injuries.

By David Harmon (not verified) on 26 Nov 2007 #permalink

I've always struggled with the pain scale too. And it's also a poor reflector of an injury or disability's seriousness. Eg My chronic pain is usually an annoyance rather than an impediment, because I pace my activities through the week to prevent it getting to bad, and I just don't do things that I know would make things worse (eg a 2 hour car trip) without a really good reason for it. If I lived a 'normal' life, then my pain levels be consistently quite high.

In terms of the extreme end - as someone else said, if it's a 10, everyone around you will know it! It also depends on what you have experienced, and in fact practitioners will often explain it by saying something like '0 is no pain and 10 is the worst pain you've experienced'. That's a personal scale then, and no-one else will know what it means! Eg few people have really experienced a 10.

In my case, my 10 was when having a spinal procedure done under sedation. The resultant non-verbal noises from my mouth led to them sedating me into unconsciousness, and apparently I was still making gargling noises of some kind, and I had a very sore throat for a couple of days.