The Geography of Pain: Country versus Country, Province versus Province, State versus State

A recent study comparing pain levels between the US and Canada highlights the importance of socioeconomic status as a contributor to pain, as do subnational analyses in each country.

Over the last several decades, many studies have documented a high prevalence of pain in continents and countries throughout the world. But the number of epidemiological investigations specifically designed to look at the geographical distribution of pain at a more fine-grained level – particularly at how pain levels differ not only between countries but within countries as well – are not as common. A new survey study now fills that gap, providing a detailed “geography of pain” between and within the US and Canada.

Using a measure of pain that includes the frequency of pain and pain interference over the last 30 days, researchers led by Anna Zajacova, University of Western Ontario, Canada, found significantly higher pain scores in the US compared to Canada. This was in part because of worse economic conditions in the US, pointing to the importance of socioeconomic status as a contributor to pain.

Further, while the geography of pain was fairly similar across Canadian provinces, pain levels differed more widely between US states, with the highest levels observed in the Deep South, Appalachia, and parts of the West.

“I really enjoyed this paper. The goal was very ambitious, and the survey that the authors used to address that goal is very interesting and unique,” said Dale Langford, a pain researcher at the Hospital for Special Surgery, New York City, US, who was not involved with the study.

“While the findings were as expected, based on what we know about differences in health and healthcare between the two countries and about the pervasive and important impact of socioeconomic factors, the way in which the authors conducted the study was creative and comprehensive, the statistical techniques they used were sophisticated, and they really did their best, within the confines of a survey study that aimed to capture nationally representative perspectives, to show how representative the respondents were,” Langford added.

As for the relevance of the findings to headache disorders, the survey did not collect information about the particular anatomical sites or causes of pain. However, the socioeconomic factors that may underlie the new findings, along with geographic differences in healthcare systems, health policy, and politics that could also influence pain levels, certainly affect people with headache disorders too.

Further, while there are epidemiological studies, such as the Global Burden of Disease study (see MSC related news article and a recent review of migraine epidemiology), providing information about the prevalence of headache conditions in different continents and countries, meticulous analyses specifically designed to assess between- and within-country differences in headache are harder to come by. Perhaps it’s time for the migraine field to consider such a study now.

The new research appeared in the December 2022 issue of the Journal of Pain.

A unique dataset
Zajacova, who is a demographer and sociologist, told Migraine Science Collaborative that her long-standing interest in macro-level determinants of health, such as a country’s approach to dealing with public health issues, was one reason why she did the study.

“The idea is that if there are major geographic differences in health, even when you adjust for the composition of the people who live there, then that suggests there is something bigger than the individual – something at the state level or the local level that is influencing people’s health.”

Another reason she undertook the study was the availability of an exceptional dataset. At the beginning of the coronavirus pandemic, Zajacova and colleagues at the University of Western Ontario conducted a cross-sectional online survey, called Recovery and Resilience COVID-19, asking Americans and Canadians questions about a range of sociopolitical, economic, and health-related factors, including questions about pain.

“We had truly unique data; there is no other dataset anywhere like it from the US and Canada. It allows us to look across provinces and states – more precisely, we actually know people’s ZIP codes,” she said.

In total, 2,124 Americans and 2,110 Canadians over the age of 18 completed the survey. The researchers aimed for the survey to be nationally representative of each country in terms of age, gender, and where people lived. To verify this, the group compared the composition of their sample to the main health survey in each country, either the National Health Interview Survey (NHIS) for the US or the Canadian Community Health Survey (CCHS).

In their survey, the authors calculated a pain score, on a 0-5 scale, based on respondents’ answers to how often they experienced pain in the past 30 days. Those who said they had pain at least rarely were then asked how much their pain interfered with general activities like work or household chores, on a 0-11 scale. By multiplying the pain frequency and pain interference scores, the team could calculate an overall pain score, which they used as their study outcome.

Country and subnational comparisons
At the national level, the average pain score for the entire study sample was 12.5 in the US and 10.7 in Canada, a result that was statistically significant. The authors say that equates to roughly 10 million extra US adults with pain.

When analyzing the results by subgroups such as gender, race/ethnicity, and immigrant status, among others, the researchers found that pain scores were always higher in the US than in Canada, and often significantly so. Further, in the US, advantaged subgroups – characterized by possessing a college education, being employed, having high incomes, and having no financial hardship – had higher pain levels than advantaged Canadians, though it’s unclear why.

One similarity between countries was that the highest average pain scores were seen in people who struggled financially because of COVID-19, and in those who said their main activity status was “disabled.”

The researchers then tried to pinpoint what explained the difference in mean pain score between the US and Canada. Their analysis showed that 35% of the difference stemmed from differences in the economic composition of each country. Here, survey respondents in the US had lower household income and more financial hardship than Canadian respondents.

“On average, the higher pain scores in the US appear to be explained if you control for people’s economic hardships. When we did that, we did not see more pain in the US. That, to me, speaks to these larger socioeconomic factors,” said Zajacova.

The investigators then turned to differences in pain scores between provinces in Canada, and between states in the US, finding a narrower range of pain scores in the former compared to the latter. In Canada, pain scores ranged from roughly 8 through 13. But in American states, pain scores ranged from 8 all the way to 23. Regions on the higher end of this range in the US – “pain hotspots” – included the Deep South, parts of Appalachia, and some western and northwestern states like Oregon and Nevada. Interestingly, the pain hotspots in the US overlapped with areas of high opioid use, though it is uncertain why that was the case.

The authors caution, however, that the subnational comparisons must be interpreted with substantial caution, considering that the number of survey participants for many states and provinces was very low. Still, they concluded that differences in pain across states and provinces were greater than between the two countries.

Weighted mean pain scores in U.S. states and Canadian provinces. Note. The map visualizes weighted mean pain scores (range 0–55) in U.S. states and Canadian provinces. Source: Recovery and Resilience (RR) data, N = 4,113. Credit: Figure 1 from Zajacova et al. The geography of pain in the United States and Canada. J Pain. 2022 Dec;23(12):2155-2166.

Acknowledging the limitations – and the need to act
Langford said that a strength of the paper is that the authors recognized the shortcomings in their methodology and data.

“They did a great job of acknowledging the limitations, which really apply to almost any survey study,” she said.

Along with the low number of respondents in certain states and provinces, and an overall low rate of response to the authors’ email invitation to participate in the study (17% in Canada and 25% in the US), another important limitation has to do with the measure of pain the team used as its outcome. Since survey respondents were asked to report only the pain they had over the past month, it’s unclear if the results would hold up for chronic pain. Further, the current study’s measure of pain differs from the measure that other pain epidemiology investigations use, so it is difficult to compare the results to previous research in the field.

How representative the population of survey respondents was in each country is also unclear. The authors did find some differences when they compared their US sample to the NHIS and the Canadian sample to the CCHS. For instance, in each country, the most disadvantaged groups seem to have been underrepresented in the current study.

The usual limitation of a cross-sectional study, where researchers can only identify associations between variables rather than causative factors, also applies here. For instance, it’s impossible to know from the current study whether pain leads to lower socioeconomic status, or the reverse.

Another potential limitation that Langford pointed to is the timing of the research.

“The survey was conducted during a pandemic, when stress was heightened, countries were managing the crisis in different ways, and there was a significant difference in death rates between the US and Canada. The finding of greater gaps in pain among advantaged groups might have been due to differences in stress – a factor known to be associated with pain – whereas socioeconomic factors may have served as an equalizer among disadvantaged groups. And so, other unmeasured contextual factors may also have played a role.”

Finally, with regard to headache conditions, while it is impossible to know if participants’ pain reported over the past 30 days stemmed from those conditions, Zajacova said that there is no reason to believe that the current study results would not apply in this instance.

“The sources of pain in our sample are likely to be no different than the sources of overall pain in the total population, both in the US and Canada,” she said.

With the new results in hand, what’s the next step?

“The study really highlights this pervasive and important impact of socioeconomic factors, and then the question is, How do we intervene?” Langford said. “And at which level should we intervene: the patient, healthcare, federal, state, or province level? We probably have to intervene at all levels. What do we do now, and is there a collective agenda for how we get there?”

Zajacova said the biggest question she has, based on her results, also raises the question of who is ready to act.

“What is it about the sociopolitical orientation of the South that’s causing people to live with more pain and disability? And, once we have a better sense of that, is anybody willing to do anything about it?”

Neil Andrews is a science journalist and executive editor of the Migraine Science Collaborative. Follow him on Twitter @NeilAndrews

Reference
The geography of pain in the United States and Canada.
Zajacova et al.
J Pain. 2022 Dec;23(12):2155-2166.