See below for Summaries of our recent work!

(For a full list of publications, click HERE)

2019 

 1. Self-criticism impacts emotional responses to pain 

Reference: Fox, K. R., O'Sullivan, I. M., Wang, S. B., & Hooley, J. M. (2019). Self-criticism impacts emotional responses to pain. Behavior Therapy, 50(2), 410–420. https://doi.org/10.1016/j.beth.2018.07.008 

 

What question was asked?  

How does self-criticism impact emotional responses to pain in people with and without a history of NSSI?  

 

How did we test it? 

The sample was 109 young women — 44 had a history of NSSI and 65 did not. Two weeks before their study visit, the women completed an assessment to determine NSSI history. In the lab, they first completed questionnaires assessing self-criticism, self-punishment, depressive symptoms, and mood. They then wrote for five minutes about a time they failed, to increase negative mood (also called a negative mood induction). Finally, participants were instructed to complete a ‘pain task,’ where they used a pressure pain device (pictured below) alone in the experiment room. Participants were told to keep the device on their finger as long as they wanted to, and that they could remove the device whenever they wanted. Across the negative mood induction and the pain task, participants shared their ratings of mood and self-punishment desires on a computer.   

 

What were the results?  

Participants with a history of NSSI reported more depressive symptoms, emotional reactivity, self-criticism, and felt more deserving of pain and punishment at the start of the study. Approximately half of the participants with NSSI histories engaged in NSSI in the past year, with most using self-cutting as the method. Overall, higher levels of self-criticism were associated with mood improvements during pain in both groups.  

Why is this important?  

Results suggest that self-criticism may help to explain why some people engage in NSSI. Specifically, the study highlighted that people who are highly self-critical actually feel better during pain. 

2. Does Nonsuicidal Self-injury Prospectively Predict Change in Depression and Self-criticism?

Reference: Burke, T. A., Fox, K., Zelkowitz, R. L., Smith, D. M. Y., Alloy, L. B., Hooley, J. M., & Cole, D. A. (2019). Does nonsuicidal self-injury prospectively predict change in depression and self-criticism? Cognitive Therapy and Research, 43(2), 345–353. https://doi.org/10.1007/s10608-018-9984-8

What question was asked?  

Does recent NSSI predict increases in depressive symptoms and self-criticism in adults with varying histories of NSSI over time?  

 

How did we test it? 

Three independent studies were conducted — the first sample (Study 1) included 144 adults with a past-month history of NSSI, the second sample (Study 2) included 178 adults with a lifetime history of NSSI, and the third sample (Study 3) included 97 adults with a past-year history of NSSI. Across all three studies, participants completed a survey assessing their self-injurious thoughts and behaviors, depressive symptoms, and self-criticism. They then completed a follow-up assessment four weeks later (Study 1), one month later (Study 2), or six months later (Study 3) where they answered questions about their depressive symptoms and self-criticism. 

 

The researchers used linear regression to assess past month NSSI frequency as a predictor of both depressive symptoms and self-criticism. To better understand potentially conflicting results across the three studies, the researchers did an internal meta-analysis.  

 

What were the results?  

Overall, the studies collectively found that NSSI frequency did not significantly predict depressive symptoms or self-criticism four weeks or one month later. However, in Study 3, NSSI predicted increases in depressive symptoms six months later.  

 

 Why is this important?  

Most research on NSSI to-date focuses on predictors of NSSI. However, it’s important to remember that NSSI does not occur in a vacuum. This study set the stage to better understand how NSSI is related to other mental health outcomes, like depression and self-criticism. This is critical for building a complete understanding of NSSI. This is especially important considering NSSI is a strong predictor of future suicidal behavior.

3. Comparing self-harming intentions underlying eating disordered behaviors and NSSI: Evidence that distinctions are less clear than assumed  

Reference: Fox, KR, Wang, SB, Boccagno, C, Haynos, AF, Kleiman, E, Hooley, JM. Comparing self-harming intentions underlying eating disordered behaviors and NSSI: Evidence that distinctions are less clear than assumed. Int J Eat Disord. 2019; 52: 564– 575. https://doi.org/10.1002/eat.23041

 

What question was asked?  

How are suicidal and self-harm intentions similar and different across NSSI and restrictive eating, binge eating, and compensatory behaviors (i.e., eating disordered behaviors)? 

 

How did we test it? 

Participants were recruited online; those who were interested first completed a screening questionnaire asking about past month and lifetime NSSI, past month and past year restrictive eating engagement, and other risky behaviors to obscure inclusion criteria. A total of 151 adults aged 18 to 44 years were included in the study, with the majority being female (79.5%) and reporting a lifetime history of both NSSI and eating disordered behaviors. Participants completed questionnaires assessing frequency and severity of eating disordered behaviors, as well as self‐harming and suicide-related characteristics across average episodes of NSSI and eating disordered behaviors.  

What were the results?  

Overall, the study found that people engaged in NSSI and each eating disordered behavior with some intent to cause physical harm in the short- and long-term, and alongside suicidal thoughts and intentions. That said, there were some important differences across behaviors. Participants reported greater intent to cause physical harm in the moment via NSSI in comparison with all eating disordered behaviors. Participants reported greater intention to cause physical harm in the long-run via restrictive eating compared to binge eating and compensatory behaviors, and compared to NSSI. 

 

Why is this important?  

People may engage in both NSSI and eating disordered behaviors to cause themselves physical pain and harm in the short and long-term, and with the knowledge, hope, and even intent to die due to these behaviors. In contrast to most traditional classification systems, results suggest it may be more accurate to conceptualize self-harming intentions and behaviors as overlapping, rather than as set categories of “direct” or “indirect,” or “suicidal” or “non-suicidal.” 

Check Out Our Blog Post Here to learn more about This Article!

4. Model complexity improves the prediction of nonsuicidal self-injury

Reference: Fox, K. R., Huang, X., Linthicum, K. P., Wang, S. B., Franklin, J. C., & Ribeiro, J. D. (2019). Model complexity improves the prediction of nonsuicidal self-injury. Journal of Consulting and Clinical Psychology, 87(8), 684–692. https://doi.org/10.1037/ccp0000421

 

What question was asked?  

Can we improve the prediction of NSSI by using more advanced and complex statistical models?   

 

How did we test it? 

The sample included 1,021 young adults, recruited internationally online, reporting a suicide attempt in the past year and at least 2 self-cutting episodes without wanting to die in the past two weeks or frequent active suicide ideation in the past two weeks. Most of the sample was White, young adult females from the United States with a lifetime history of NSSI. Qualifying participants then completed a baseline survey lasting 50 minutes, including computer-based tasks and questions related to NSSI, suicidal thoughts and behaviors, and a range of mental health-related measures and assessments. Participants then took the same survey again 3, 14, and 28 days later.  

 

Next, we compared univariate (i.e., single variable) logistic regression models, multiple logistic regression models, and random forest models in their ability to predict NSSI engagement at each follow-up time point. Models included up to 39 predictors.  

 

What were the results?  

Univariate (single variable) models produced the poorest prediction. Multiple logistic regression models resulted in improved but still poor prediction. Random forest models produced much-improved prediction compared to both univariate and multiple logistic regression.  

 

In the random forest models tested, the 10 most important predictors were: self-injury, SITBs, psychopathology, self-disgust, agitation, and other clinically relevant items. However, random forest model performance (aka predictive accuracy) did not change when the most important factor in the original algorithm was removed, supporting NSSI as a complex rather than a factor-dependent phenomenon. 

Why is this important?  

Results suggest that model complexity was associated with greater predictive accuracy and that no single factor was sufficient for strong NSSI prediction. These results are similar to those found in other studies and indicate that NSSI engagement and its risk is complex, and can likely not be understood or predicted by specific risk factors. 

5. Predicting imminent suicidal thoughts and nonfatal attempts: The role of complexity

Reference: Ribeiro, J. D., Huang, X., Fox, K. R., Walsh, C. G., & Linthicum, K. P. (2019). Predicting imminent suicidal thoughts and nonfatal attempts: The role of complexity. Clinical Psychological Science, 7(5), 941–957. https://doi.org/10.1177/2167702619838464

What question was asked?  

How are the methods of long follow-ups and the application of simple conceptualizations of risk hurting our chances at predicting imminent suicidal ideation and nonfatal suicide attempts among high-risk individuals?  

 

How did we test it? 

The sample included 1,021 high-risk suicidal and/or self-injuring adults from around the world. Eligibility included being 18 or older, English fluency, recent, frequent nonsuicidal self-cutting, a nonfatal suicide attempt in the past year, or frequent, active suicidal ideation. Qualifying participants then completed an online assessment measuring items like self-injurious thoughts and behaviors, affect toward suicidal stimuli, hopelessness, agitation, insomnia symptoms, social functioning, and other psychological symptoms.  Most participants (90%) took the assessment again 3, 14, and 28 days after. For each assessment, participants had 48 hours to complete it. The researchers then used machine learning to examine the complexities of suicide risk.  

 

What were the results?  

At baseline, almost 90% of the sample reported NSSI engagement at least once in their lifetime, with self-cutting being the most common method. Suicidal behaviors like planning, attempt, and ideation overall were also common. Overall, suicidal ideation was the strongest univariate predictor.   

 

Why is this important?  

Results suggest that our ability to predict suicidal thoughts and behaviors can improve if we focus on short-term prediction and use methods that model the complexities of risk. These results have implications for how researchers understand suicidal thoughts and behaviors and how clinicians can predict and treat them. 

2020

6. Nonsuicidal Self-Injury: Diagnostic Challenges and Current Perspectives

Reference: Hooley, J. M., Fox, K. R., & Boccagno, C. (2020). Nonsuicidal self-injury: Diagnostic challenges and current perspectives. Neuropsychiatric Disease and Treatment, 16, Article 101-112. https://doi.org/10.2147/NDT.S198806

What question was asked?  

How has the conceptualization of NSSI evolved over time? 

 

How did we test it? 

This article analyzes the definitions of NSSI and self-harming behavior in general and the diagnostic criteria in the DSM-5. The researchers then discuss the various measures used to assess NSSI.  

 

What were the results?  

The addition of NSSI Disorder in the DSM-5 is a major improvement. However, there are still several inconsistencies in how we define, diagnose, and assess NSSI and other self-harming behaviors.  

 

Why is this important?  

Efforts to make a clear, consistent definition can help diagnose and assess NSSI better. This is especially important considering the link between NSSI and suicide risk.

7.  Longitudinal predictors of self-injurious thoughts and behaviors in sexual and gender minority adolescents 

Reference: Smith, D. M., Wang, S. B., Carter, M. L., Fox, K. R., & Hooley, J. M. (2020). Longitudinal predictors of self-injurious thoughts and behaviors in sexual and gender minority adolescents. Journal of Abnormal Psychology, 129(1), 114–121. https://doi.org/10.1037/abn0000483

 

What question was asked?  

What general and minority-specific factors are prospectively related to SITB engagement in sexual and gender minority (SGM) youth?  

 

How did we test it? 

The sample included 252 SGM adolescents aged 14 or 15 years old recruited from social media sites related to SGM populations or flyers in or around high schools. Eligibility included reporting female sex assigned at birth as well as identifying as homosexual, bisexual, questioning, and/or a gender minority. Qualifying participants then completed an online assessment measuring items like self-injurious thoughts and behaviors, depression, body image, self-criticism, family support and strain, and experiences of everyday discrimination. Later, participants completed a follow-up assessment asking about any SITB engagement during the past six months.  

 

What were the results?  

The models were highly accurate in predicting NSSI and suicidal ideation, accurate at predicting suicide plans, and fairly accurate at predicting suicidal behaviors. Some important SGM-specific factors included family support, self-criticism, body image, and depressive symptoms.  

 

Why is this important?  

Results suggest that minority-specific factors may predict SITB engagement in SGM youth. This study is important because SGM-identifying individuals are at increased risk for negative psychological and physical outcomes like SITBs.

8. Mental health among sexual and gender minority adolescents: Examining interactions with race and ethnicity

Reference: Fox, K. R., Choukas-Bradley, S., Salk, R. H., Marshal, M. P., & Thoma, B. C. (2020). Mental health among sexual and gender minority adolescents: Examining interactions with race and ethnicity. Journal of Consulting and Clinical Psychology, 88(5), 402–415. https://doi.org/10.1037/ccp0000486 

What question was asked?  

How do depressive symptoms and SITB engagement vary among adolescents at intersections of sexual orientation, gender, and race/ethnicity? 

 

How did we test it? 

This study had a racially/ethnically diverse sample of 2,948 adolescents aged 14 to 18 years old classified into three groups: gender minority (GM), sexual orientation minority (SM), and cisgender heterosexual. The researchers excluded those who identified as American Indian, intersex, and those assigned male at birth but were not GM to have a stronger analysis. In a self-report survey, participants shared their age, sex assigned at birth, race/ethnicity, sexual and gender identity, socioeconomic status, depressive symptoms, and any SITBs. GM participants answered additional questions about GM-specific psychosocial factors.  

 

What were the results?  

SM and GM participants reported higher depressive symptoms, suicide ideation and attempts, and more NSSI histories than the cisgender heterosexual participants. Also, GM adolescents reported the highest risk for depressive symptoms and SITBs overall compared with the other groups.  

 

Why is this important?  

These results suggest that risk varies across interactions of sexual orientation, gender, and race/ethnicity. This study is important because SGM-identifying individuals are at increased risk for negative psychological and physical outcomes like SITBs

Check Out Our Blog Post Here to learn more about This Article!

9. Self-Injurious Thoughts and Behaviors Interview—Revised: Development, reliability, and validity 

Reference: Fox, K. R., Harris, J. A., Wang, S. B., Millner, A. J., Deming, C. A., & Nock, M. K. (2020). Self-Injurious Thoughts and Behaviors Interview—Revised: Development, reliability, and validity. Psychological Assessment, 32(7), 677–689. https://doi.org/10.1037/pas0000819 

 

What question was asked?  

How can a revised version of the Self-Injurious Thoughts and Behaviors Interview (SITBI) address the limitations of the current version?  

 

How did we test it? 

The researchers first created a new version of the SITBI (the SITBI-R) by adding updated language and more behaviors. Eligibility included past-year SITB engagement, 18 or older, English fluency, and Internet access. Both studies had eligible participants complete self-report questionnaires before the other assessments.  

 

In Study 1, they compared the validity of the SITBI-R to the widely used Columbia Suicide Severity Rating Scale (C-SSRS) among 114 adults. Participants were randomly chosen to complete the in-lab or online part of the study first, set two weeks apart. In the lab, the SITBI-R and the C-SSRS were given as an interview, and the order of which test went first was randomized. In Study 2, they examined the reliability of the SITBI-R when given entirely online to an anonymous sample of 188 adults. Participants completed the assessment twice two weeks apart.  

 

What were the results? 

The majority of participants in both studies reported histories of lifetime NSSI, suicide ideation, and suicide methods. Study 1 found that the SITBI-R showed convergent validity in comparison with the C-SSRS. Interestingly, almost half of the participants in Study 1 reported they preferred disclosing their SITB history in person; however, most participants in Study 2 preferred online disclosure.  

 

Why is this important?  

Across both studies, reliability was strongest for more recent outcomes and more commonly assessed SITBs (ex: suicidal thoughts or plans) than for less commonly assessed behaviors (ex: aborted or interrupted suicide attempts). Overall, the SITBI-R online or in-person is a reliable and valid measure for a variety of SITBs.

10. Emotional changes following discrimination induction in gender-and sexuality-diverse adolescents

Reference: Smith, D. M., Fox, K. R., Carter, M. L., Thoma, B. C., & Hooley, J. M. (2020). Emotional changes following discrimination induction in gender- and sexuality-diverse adolescents. Emotion. Advance online publication. https://doi.org/10.1037/emo0000862
 

What question was asked?  

What is the relationship between experiences of discrimination, experimentally induced emotional reactivity, and depressive symptoms among SGM youth?  

 

How did we test it? 

The sample was 328 adolescents aged 14 or 15 years old recruited from social media sites related to the LGBTQ+ community. Eligibility included reporting female sex assigned at birth as well as identifying as heterosexual, a sexual minority (homosexual, bisexual, or questioning/other), or a gender minority (any identity different than assigned at birth).  

 

First, qualifying participants rated their mood and self-critical feelings on the Visual Analogue Scales the researchers created, then completed a computer task (AMP) measuring the implicit affect of self-criticism. Next, they completed questionnaires collecting information about their demographics, any self-injurious thoughts and behaviors, depression, self-criticism, and experiences of everyday discrimination. Then, participants wrote for five minutes to induce feelings of discrimination. They completed the VAS ratings and AMP again to document the effects of the induced discrimination. Finally, all participants experienced a positive mood induction to counteract any negative effects of the study.  

 

Why is this important? 

Results show that SGM adolescents reported higher levels of discrimination, self-criticism, experimentally induced emotion reactivity, depression, and NSSI compared with their cisgender heterosexual peers. However, many participants used cognitive restructuring in their written responses, which was associated with lower levels of depressive symptoms. Such restructuring may be a protective factor for adolescents at risk for negative mental health outcomes. 

11. Self-Critical and Self-Punishment Cognitions Differentiate Those With and Without a History of Nonsuicidal Self-Injury: An Ecological Momentary Assessment Study 

Reference: Burke, T. A., Fox, K., Kautz, M. M., Rodriguez-Seijas, C., Bettis, A. H., & Alloy, L. B. (2021). Self-critical and self-punishment cognitions differentiate those with and without a history of nonsuicidal self-injury: An ecological momentary assessment study. Behavior Therapy, 52(3), 686–697. https://doi.org/10.1016/j.beth.2020.08.006 

What question was asked?  

How do self-criticism and self-punishment cognitions play a role in NSSI?  

 

How did we test it? 

The sample was 123 diverse university students — 64 had a history of NSSI and 59 did not. Eligible participants completed a self-report questionnaire and an in-person interview to determine NSSI history. Starting the day after the interview, the students received four online surveys each day for 10 days. The surveys measured sleep, feelings of self-criticism and self-punishment. The students could choose when to receive the surveys throughout the day and were asked to complete them as soon as possible.  

 

What were the results?  

Almost half of the participants with NSSI histories engaged in NSSI in the past year, with most using self-cutting as the method. Additionally, the study found that participants with a history of NSSI demonstrated higher levels of self-critical and self-punishment cognitions in comparison to the group without a history of NSSI.  

  

Why is this important?  

Results suggest that self-criticism and self-punishment may be important to assess and target during treatment in hopes of reducing NSSI engagement.  

12. Interventions for suicide and self-injury: A meta-analysis of randomized controlled trials across nearly 50 years of research

Reference: Fox, K. R., Huang, X., Guzmán, E. M., Funsch, K. M., Cha, C. B., Ribeiro, J. D., & Franklin, J. C. (2020). Interventions for suicide and self-injury: A meta-analysis of randomized controlled trials across nearly 50 years of research. Psychological Bulletin, 146(12), 1117–1145. https://psycnet.apa.org/doi/10.1037/bul0000305

 

What question was asked?  

Have the interventions for suicide and NSSI improved over the past 50 years?  

 

How did we test it? 

This study analyzed 1,125 studies from 591 articles published from 1970 to 2018. Inclusion criteria included random assignment and an assessment of a SITB after treatment. The researchers excluded studies not published in English or that assessed other negative outcomes. The studies were coded by the researchers or trained research assistants for SITB outcomes, assessment time points, sample severity and age, intervention characteristics, and the study characteristics.  

 

What were the results?  

The majority of the articles had similar features, such as being published after 2000, data gathered posttreatment on a clinical sample of less than 500 adults, and interventions highlighting psychopathology and treatment with medication. Overall, this study found that intervention efficacy in reducing SITBs has not significantly improved over the last 50 years of research. 

 

Why is this important?  

The results suggest that more research is needed to assess the causes of SITBs, which will then help improve treatment.  

Check Out Our Blog Post Here to learn more about This Article!

13. Real-time monitoring of the associations between self-critical and self-punishment cognitions and nonsuicidal self-injury 

Reference: Burke, T. A., Fox, K.R., Kautz, M., Siegel, D. M., Kleiman, E., & Alloy, L. B. (2021). Real-time monitoring of the associations between self-critical and self-punishment cognitions and nonsuicidal self-injury. Behavior Research and Therapy, 131, 103775. https://doi.org/10.1016/j.brat.2020.103775

What question was asked?  

How can capturing in-the-moment data about NSSI better explain its relationship with  

self-criticism and self-punishment?  

 

How did we test it? 

The sample was 64 undergraduate students from a public, urban university with a history of repeated NSSI. Eligible participants completed a self-report questionnaire and an in-person interview to determine NSSI history. Starting the day after the interview, the students received four online surveys each day for 10 days. The surveys measured sleep, feelings of self-criticism and self-punishment, NSSI urge and behavior. The students could choose when to receive the surveys throughout the day and were asked to complete them as soon as possible.  

 

What were the results? 

The study found that 13 of the students engaged in NSSI and 20 had an NSSI urge within the past month. Throughout the 10 days of surveys, 9 students reported NSSI engagement and 28 reported an NSSI urge. After analysis, this study found that self-punishment cognitions (both temporary and stable), but not self-criticism, predicted NSSI urges during the 10-day period.  

 

Why is this important?  

The results suggest that self-critical and self-punishment cognitions accompany or follow NSSI urges, rather than as central risk factors.