Sleep and Pain: A Systematic Review of Studies of Mediation

Abstract

Objectives: A relationship between sleep and pain is well established. A better understanding of the mechanisms that link sleep and pain intensity is urgently needed to optimize pain management interventions. The objective of this systematic review was to identify, synthesize, and critically appraise studies that have investigated putative mediators on the path between sleep and pain intensity.

Methods: A systematic search of 5 electronic bibliographic databases (EMBASE, MEDLINE, CINAHL, PsycINFO, and the Cochrane Central Register of Controlled Trials) was conducted. Eligible studies had to apply a formal test of mediation to variables on the path between a sleep variable and pain intensity or vice versa. All searches, data extraction and quality assessment were conducted by at least 2 independent reviewers.

Results: The search yielded 2839 unique articles, 9 of which were eligible. Of 13 mediation analyses, 11 investigated pathways from a sleep variable to pain intensity. Putative mediators included affect/mood, depression and/or anxiety, attention to pain, pain helplessness, stress, fatigue, and physical activity. Two analyses investigated pathways from pain intensity to a sleep variable, examining the potentially mediating role of depressive symptoms and mood. Although evidence supported a mediating role for psychological and physiological aspects of emotional experiences and attentional processes, methodological limitations were common, including use of cross-sectional data and minimal adjustment for potential confounders.

Discussion: A growing body of research is applying mediation analysis to elucidate mechanistic pathways between sleep and pain intensity. Currently sparse evidence would be illuminated by more intensively collected longitudinal data and improvements in analysis.

FIGURE 1

Prototypical case of a single mediating variable on the path from impaired sleep to higher pain intensity.

FIGURE 2

Flow of information through different phases of the systematic review.

FIGURE 3

Graphical summary of factors identified as mediators on the path between sleep variables and pain intensity. *Temporal associations have not yet been well established, therefore the figure should be interpreted as hypothetical and not necessarily reflecting causality. Cross-sectional studies included in the figure are denoted with an asterisk. Statistically significant mediation has not been identified through positive affect or physical activity. However, because of methodological limitations of research undertaken to date, their role in the path from sleep to pain is far from determined. Potential confounders adjusted for: age,,, sex, ,, baseline level of cortisol, negative affect, duration of exposure to a cold pressor task, maternal education, Sickle Cell Disease type, aggregated person means for sleep and pain variables across the course of study.

FIGURE 4

Graphical summary of factors identified as mediators on the path between pain intensity and sleep variables. *Temporal associations have not yet been well established, therefore the figure should be interpreted as hypothetical and not necessarily reflecting causality. Cross-sectional study included in the figure denoted with an asterisk. Potential confounders adjusted for: Annual income, age, sex, maternal education, Sickle Cell Disease type, aggregated person means for sleep and pain variables across the course of study.