Everybody Hurts: How 2,000 Scots Became 150 Million Americans in Pain

Published on
July 30, 2012

In 1999, The Lancet published “The epidemiology of chronic pain in the community” by Drs. Alison Eliott and Blair Smith at the University of Aberdeen and coauthors.

It was a survey of 5,036 patients from 29 general practices in one region in Scotland. The study received responses from 3,605 people. About half of them said they had chronic pain. And that has since been used as the basis for the claim that 75 million to 150 million Americans also have chronic pain but that very few of them – 3 million to be exact – are treated for it.

Here’s my attempt to figure out how that happened.

For the Scotland study, respondents filled out and mailed in questionnaires that included, as the study explains, “case-screening questions, a question on the cause of the pain, the chronic pain grade questionnaire, the level of expressed needs questionnaire, and sociodemographic questions.”

The paper on the Scotland study breaks down the findings this way:

1817 (50·4%) of patients self reported chronic pain, equivalent to 46·5% of the general population. 576 reported back pain and 570 reported arthritis; these were the most common complaints and accounted for a third of all complaints. Backward stepwise logistic-regression modelling identified age, sex, housing tenure, and employment status as significant predictors of the presence of chronic pain in the community. 703 (48·7%) individuals with chronic pain had the least severe grade of pain, and 228 (15·8%) the most severe grade. Of those who reported chronic pain, 312 (17·2%) reported no expressed need, and 509 (28·0%) reported the highest expressed need.

The researchers on the Scotland study never say there are 150 million people in pain.

They never say there are 75 million people with pain.

They never even use the word “million” at any point.

So where do those numbers come from? And why are people citing this paper as evidence that 75 million to 150 million people are in pain?

I asked Dr. Dennis C. Turk, an anesthesiology professor at the University of Washington who cited the Scotland study for the claim that 75 million to 150 million people are in chronic pain. He did not address how estimates of Americans in pain came from the Scotland study specifically, but said that pain statistics are difficult to pin down.

There are a lot of assumptions and extrapolations that are required in deriving estimates from epidemiological data. Much of the variation in estimates will depend on how questions are asked and the anchors used (e.g., time interval of question, impact on function) and the sample selection.

Turk recommended that I look at an Institute of Medicine report that was published in 2011 – six years after his article was published. (I will address that in a future post.)

I could not find a way to derive the 75 million figure from the pain estimates in the Scotland paper, although there are no doubt people better adept at math who could do so. Here’s my best guess on the 150 million figure.

The study estimates that based on this survey 46.5% of the general population of Scotland suffers from chronic pain.

If you extrapolated that to the entire population of the U.S. – about 309 million people – that would mean about 144 million people in the U.S. suffer from pain. Rounding up, you get 150 million people.

Does it seem like a reasonable leap to make assumptions about 150 million people in the U.S. based on responses from fewer than 2,000 people in Scotland who may have been more motivated to reply to a survey about pain because they were, in fact, suffering from chronic pain?

In all fairness, similar leaps from small samples to large populations are made all the time. Researchers can’t possibly persuade every person in a city – let alone an entire country – to fill out questionnaires about their health. There are about 5 million people in Scotland. The survey tapped about 0.07% of the population for this study. That’s not far off from the largest telephone survey in the world, the Behavioral Risk Factor Surveillance System (BRFSS) by the Centers for Disease Control and Prevention, which captures about 0.12% of the population annually. Much of what we know about health trends in the U.S. comes from the BRFSS.

But the leap across the pond is a little more problematic. We’ll get into that in Friday’s post.

Next: Why Pain Is Not the Same at Every Age and in Every Place

Image by Giorgio Quassi via Flickr