Chronic pain, which is defined as pain lasting for weeks or months, affects as many as 100 million adults in the US. Individuals who experience chronic pain are at substantially greater risk of suicidal ideation, attempt, and mortality than are those without pain. Understanding the processes that account for the association between chronic pain and suicidal behavior is therefore a critical public health issue. In particular, there are several key questions raised by existing research on chronic pain and suicidal behavior. First, pain occurs in many medical conditions, and it is not clear whether associations with suicidal behavior differ across the various chronic pain conditions. This question is particularly challenging because chronic pain conditions are frequently comorbid with one another: Individuals with one type of pain are likely to have others as well (e.g., back pain, headache). Second, it is well-established that there are familial influences on chronic pain as well as suicidal behavior. Consequently, their association could reflect the fact that chronic pain and suicidal behavior share underlying genetic risk factors rather than the possibility that the experience of chronic pain causes people to become suicidal. For this reason, genetically informative research is needed to evaluate the extent to which chronic pain increases risk of suicidal behavior.
The proposed research will address these questions using a prospective, population-based study of Swedish twins. We will link self-reported pain and other psychosocial factors from approximately 25,000 twins in the 2005-2006 Study of Twin Adults: Genes and Environment (STAGE) with nationwide administrative register data on health and mortality. The linkage will allow us to examine 11 years of suicidal behavior assessed from healthcare and death records through 2017. We will use a multivariate twin modeling framework to provide a rigorous examination of the role of chronic pain. In our first aim, we will apply factor analysis to separate the risk of suicidal behavior unique to specific chronic pain conditions from that shared generally across all painful conditions. Our second aim will take advantage of the genetic and environmental similarity between twins. By estimating differences in risk of suicidal behavior between twins differentially exposed to pain, we will rule out confounding from all sources of similarity—genetic and environmental—within a twin pair. Moreover, we will also include statistical controls for a wide range of other psychosocial risk factors to account for individual-level confounding.
We expect that this research will generate positive outcomes across the translational spectrum. To the extent that associations between chronic pain and suicidal behavior persist within twin pairs, our results would support the causal influence of chronic pain. Thus, our results would suggest that appropriate treatment for chronic pain, whether generally or for specific conditions, may be an important priority for suicide prevention. On the other hand, to the extent that shared genetic or otherwise familial factors drive the increased risk of suicide among people with chronic pain, understanding those underlying processes could help clarify alternative intervention targets.
