Toward a causal model of chronic back pain: Challenges and opportunities

Abstract

Chronic low back pain (cLBP) afflicts 8. 2% of adults in the United States, and is the leading global cause of disability. Neuropsychiatric co-morbidities including anxiety, depression, and substance abuse- are common in cLBP patients. In particular, cLBP is a risk factor for opioid addiction, as more than 50% of opioid prescriptions in the United States are for cLBP. Misuse of these prescriptions is a common precursor to addiction. While associations between cLBP and neuropsychiatric disorders are well established, causal relationships for the most part are unknown. Developing effective treatments for cLBP, and associated co-morbidities, requires identifying and understanding causal relationships. Rigorous methods for causal inference, a process for quantifying causal effects from observational data, have been developed over the past 30 years. In this review we first discuss the conceptual model of cLBP that current treatments are based on, and how gaps in causal knowledge contribute to poor clinical outcomes. We then present cLBP as a "Big Data" problem and identify how advanced analytic techniques may close knowledge gaps and improve clinical outcomes. We will focus on causal discovery, which is a data-driven method that uses artificial intelligence (AI) and high dimensional datasets to identify causal structures, discussing both constraint-based (PC and Fast Causal Inference) and score-based (Fast Greedy Equivalent Search) algorithms.

Keywords: back pain; causal (structural) model; clinical trials; data science; pain.

Figure 1

Biological factors associated with chronic low back pain (cLBP).

Figure 2

Outcomes following back surgery. Y^{t = 1} = outcome with, Y^{t = 0} = outcome without.

Figure 3

Example of Exchangeability. y¹ = outcome, y⁰ = potential outcome.

Figure 4

Example of Confounding Variable. C is a common cause of T and Y.

Figure 5

Paths in a Causal Directed Acyclic Graph (DAG). Circles are variable types, arrows indicate direction of effect.

Figure 6

cLBP potential causal DAG. (A) Constraint-based causal algorithms begin with undirected graph, with all variables connected. (B) A potential causal DAG in which edges are pruned and a more parsimonious causal graph is created, that can then be furthered tested and refined.

Figure 7

Fast Causal Inference Algorithm. (A) FCI begins with undirected graph. (B) FCI then removes edges and identifies possible “V” structures, where two variables are condtionally dependent on a third (colliders). (C) FCI also mitigates confounder bias by identifying “Y” structures, where two variables (a and b) are found to be independent of a third (y), conditional on a fourth variable x.