The Associations among Distress Tolerance, Unhelpful Coping Behaviors, and Symptoms of Depression: A Network Analysis

Though we use the term emotional distress tolerance to note its distinction from other types of distress tolerance (e.g., physical pain tolerance), emotional distress tolerance likely primarily measures tolerance for negative emotions, despite the fact that many may experience distress due to positive emotions (Lass & Winer, in press; Winer & Salem 2016).

The references provided here evidence the relationships between rumination, avoidance, and coping with stress in relation to symptoms of depression, and not distress tolerance.

It is important to note that, though Bernstein and colleagues (2017) provide potential directed feedback loops, directed acyclic graphs generated from cross-sectional data cannot be interpreted in the same way as directed networks that are generated from data with multiple time points (Jordan et al., in press).

This process is explained in much more detail in the data analytic plan.

Depressive symptoms often co-occur with other disorders, and behaviors such as nonsuicidal self-injury, substance use, and obsessions can also be found in depressed individuals. We are simply noting that distress tolerance has not been widely examined in the depression literature.

We used the Cognitive Style Questionnaire (CSQ; Haeffel et al., 2008) as a way to assess response validity. In this questionnaire, participants are asked to imagine themselves in several hypothetical situations (e.g., “You take an exam and receive a low grade on it,” then are asked to provide qualitative responses that explain the cause of that event. This questionnaire was included to determine eligibility for a laboratory study separate to this study, and was not used in the present set of analyses, other than to assist in identifying careless responders. Participants were excluded from analyses if they failed to provide a qualitative response or their response did not make sense or answer the question, (e.g., responding with “I don’t know”).

The inclusion of an item that asks about suicidal ideation was approved by the Institutional Review Board where data collection took place, along with a plan to mitigate suicide risk of our participants and provide resources. To monitor the risk of participants in the study, the first author of this study downloaded and extracted data on the suicide item daily. Participants provided consent to be contacted by researchers for this study as part of the informed consent. As such, individuals who selected either “I would like to kill myself” or “I would kill myself if I had the chance” received an email from the first author that provided community mental health resources, emergency resources, and the contact information for the first and fourth author, in case participants had questions about how to access these resources. Additionally, the fourth author maintained an active license as a Licensed Psychologist in the state where data collection took place and made himself available for consultation, as needed.

Reliability analyses for the DTS subscales can be found in the supplemental materials.

Reliability analyses for the BADS subscales can be found in the supplemental materials.

BDI item 6, skew = 3.41, kurtosis = 11.99; BDI item 19, skew = 4.14, kurtosis = 18.27.

These values were used to determine redundancy because (1) they have been used in previous work (Bernstein et al. 2019) and (2) they are the default values for this analysis.

Betweenness and closeness have also been used to assess a node’s centrality. We did not examine betweenness or closeness in our study due to concerns regarding their stability (Epskamp et al., 2018) and suitability in psychopathology networks (Bringmann et al., 2019).

Please note that, we are not interpreting a longitudinal, dynamical relationship from our data; however, edges may be considered pairwise interactions between two nodes (Haslbeck et al., 2019), as the relationship of both A → B and B → A is included in the covariance matrix that calculates the strength of the edges.