Mediators of cognitive-behavioral therapy for insomnia: A review of randomized controlled trials and secondary analysis studies

Abstract

The examination of treatment mechanisms in randomized controlled trials (RCTs) has considerable implications for research and clinical practice. Insomnia is a highly prevalent and distressing disorder, associated with many adverse outcomes. Although extensive work has focused on the cognitive-behavioral treatment of insomnia (CBT-I), few studies have directly examined the mechanisms of this intervention. CBT-I is a short-term, multi-component treatment that has demonstrated strong efficacy in treating insomnia. The purpose of the present study is: (a) to investigate if CBT-I works in accordance with its proposed mechanisms, and (b) to evaluate how the field is progressing in its understanding of these processes. This study comprehensively reviewed CBT-I RCTs for their inclusion of mediator variables. Secondary analysis studies were also surveyed for relevant mediator variables. Results demonstrated that 21 RCTs (39% of the total RCTs) and 11 secondary analysis studies examined at least one of the proposed mediators. Results of this review highlight that, although CBT-I appears to be targeting the hypothesized sleep processes, more research is needed to better understand whether CBT-I works in accordance with its theorized mechanisms. Inclusion of mediational analyses in future RCTs and secondary analysis studies would allow for further refinement of CBT-I and improved treatment outcomes.

Introduction

The importance of examining mechanisms of treatment in randomized controlled trials (RCTs) has been emphasized throughout the literature (Kazdin, 2007, Kraemer et al., 2002). Treatment mechanism research may lead to a number of considerable outcomes, including a better understanding of the processes leading to therapeutic change, the refinement of current treatment protocols, and increased generalizability of treatment effects from research to practice. Understanding treatment mechanisms may also help to clarify the nature and etiology of disorders, thereby informing future research and clinical practice. According to Kraemer et al. (2002), mechanisms are causal links between treatment and outcomes and mediators explain why and how treatments have effects. Although mediators have less specificity than mechanisms, they identify possible mechanisms of treatment and are therefore an important first step in understanding the processes underlying therapeutic change (Kazdin, 2007).

In light of the implications of treatment mechanism research, there is a growing movement within the cognitive-behavioral therapy (CBT) literature to examine the processes of change throughout treatment. A number of research areas, including CBT for depression (e.g., Hollon et al., 1987, Quilty et al., 2008), anxiety disorders (e.g., Hofmann, 2004, Hofmann et al., 2007, Smits et al., 2004), eating disorders (e.g., Murphy et al., 2009, Spangler et al., 2004), and medical illnesses (e.g., Knoop et al., 2012, Turner et al., 2007, Wiborg et al., 2011) have made efforts to examine mediators of therapeutic change. To date, extensive research has focused on the cognitive-behavioral treatment of insomnia (CBT-I); however, few studies have examined the mechanisms of this intervention. An understanding of these processes is critical given the high prevalence of insomnia and the significant distress associated with it.

In a U.S. epidemiological study (n = 10,094), prevalence of insomnia varied depending on the criteria used. Prevalence estimates were 22.1% for DSM-IV-TR, 14.7% for Research Diagnostic Criteria/International Classification of Sleep Disorders, Second Edition (RDC/ICSD-2), and 3.9% for International Classification of Diseases, Tenth Revision (ICD-10) (Roth et al., 2011). Insomnia is present across a wide range of psychological illnesses, including depression and anxiety (Taylor, Lichstein, Durrence, Reidel, & Bush, 2005), as well as medical illnesses, including heart disease, cancer, high blood pressure, neurologic disease, breathing problems, urinary problems, chronic pain, and gastrointestinal problems (Taylor et al., 2007). It is associated with worsened quality of life, marked functional impairment (Katz and McHorney, 2002, Kyle et al., 2010, Ohayon, 2002), and considerable financial costs (Daley, Morin, LeBlanc, Grégoire, & Savard, 2009). A Canadian study examining the economic burden of insomnia estimated that the average annual per-person costs were $421 for good sleepers compared to $5010 for individuals with insomnia. Such elevated costs were a result of direct expenses (e.g., increased health-care utilization, prescription medications, and over-the-counter products) and indirect expenses (e.g., insomnia-related absenteeism, productivity losses/diminished job performance, and increased proneness to accidents) (Daley et al., 2009).