• Type:#article
  • Year read:#read2022
  • Subject: Insomnia Network theory
  • Bibtex: @lancee2022
  • Bibliography: Lancee, J., Harvey, A. G., Morin, C. M., Ivers, H., van der Zweerde, T., & Blanken, T. F. (2022). Network Intervention Analyses of cognitive therapy and behavior therapy for insomnia: Symptom specific effects and process measures. Behaviour Research and Therapy, 153, 104100. https://doi.org/10.1016/j.brat.2022.104100

Example citation

Key takeaways

  • Individual items from Insomnia Severity Index (ISI), plus process measures for both treatments.
  • Cognitive therapy (n = 65) vs Behavior therapy (n = 63)
  • A strength here is that process measures were included in the data, not just symptom measures.
  • CT-effects: worry, impaired QoL, dysfunctional beliefs, monitoring sleep-related threats
  • BT-effects: sleep efficiency, difficulty maintaining sleep, early morning awakening, time in bed, sleep incompatible behaviors, bed- and rise time variability

Very similar to a previous Blanken paper, Blanken2021 - Symptom-specific effects of cognitive therapy and behavior therapy for insomnia, A network intervention analysis. But here they had moderators as well.

The estimated networks will be visualized and can be interpreted as follows. All variables are included as nodes, and all continuous variables will be presented as circles, whereas the binary treatment allocation variable will be included as a square. The variables will be connected by edges that represent de conditional dependence relations among them, i. e., the unique association between two variables after conditioning on all other variables in the network. Blue edges indicate positive associations, and red edges indicate negative associations. Since we are contrasting two active treatment conditions (CT vs BT), any edge between the treatment allocation variable and another symptom or mediator, will delineate a treatment-specific effect. Any effect in favor of CT will be represented by a green edge, and any effect in favor of BT will be rep resented by a yellowedge. It is importantto keep in mind that any direct effects will specifically reflect differences between the two active treatments. Hence, if the two treatments affect a symptom or mediator similarly, this effect will not be represented in the network. Consequently, the absence of an edge does not indicate the absence of a treatment effect, but merely the absence of a differential treatment effect between CT and BT.

In the Network Intervention Analysis on sleep efficiency and the items of the Insomnia Severity Index, we observed that at baseline, as expected, there were no differences between the CT and BT condition.At mid-treatment, CT showed a symptom specific effect on ‘worry’ (green line) while BT showed a symptom specific effect on ‘sleep efficiency’ and ‘difficulty maintaining sleep’ (yellow lines). At post-test only the effect of BT on ‘sleep efficiency’ remained, and at 6-month follow-up no symptom specific effects were observed. At 12-month follow-up, BT showed a differential effect on ‘sleep efficiency’ and on ‘early morning awakening’. CT linked to ‘impaired quality of life’. Please see Fig. 1 for the networks.