Comments on Past history serves as a better predictor of future outcomes than does client prediction for self-injurious thoughts and behaviors

Mike Anestis commented on 'Past history serves as a better predictor of future outcomes than does client prediction for self-injurious thoughts and behaviors'

2009-07-08T15:48:50Z

I think this is a really good point for several reasons.

First, just in terms of insightful comments on potential third variables, I think this was really strong. The authors were very careful to temper their interpretations and mention that these findings are preliminary and intended to inspire future research in this area, but it is very useful to point out variables that might contribute to such findings.

Second, I think this offers up a great idea for a future study that would help clarify the relationship between current ideation and attempts/behaviors in the immediate future (6 months) as well as the role of specific therapeutic approaches in moderating that relationship.

The study itself did not involve treatment, but many of the participants were either involved in some form of psychotherapy, prescribed psychopharmacological treatments, or both. With a larger sample size, I think that your hypothesis could have been tested indirectly. Hopefully a follow-up study is in the works!

Jen Hames commented on 'Past history serves as a better predictor of future outcomes than does client prediction for self-injurious thoughts and behaviors'

2009-07-07T17:22:53Z

Just some thoughts about the role of treatment:

I was curious about the data looking specifically at participants' self-injurious thoughts and behavior at the time they were asked to predict their future thoughts because I was thinking that their current levels of self-injurious thought might have an impact on both their predictions and their actual future thoughts. Consider for example, two subgroups of participants: (1) particiapnts who were experiencing self-injurious thoughts at the time of their prediction, and (2) participants who were not experiencing self-injurious thoughts at the time of their prediction

Participants who were experiencing self-injurious thoughts at or close to the time they were asked to predict their future thoughts may have been more likely to receive treatment for their thoughts, potentially contributing to a reduction in future self-injurious thoughts. Given their current thoughts, these participants may also have predicted that they would continue to have self-injurious thoughts, therefore leading to inaccuracy in their predictions.

Participants, however, who were not experiencing self-injurious thoughts at or close to the time they were asked to predict their future thoughts were likely not receiving treatment directed at having recent self-injurious thoughts, so their current lack of self-injurious thoughts may have contributed to their prediction that they would not have these thoughts in the future. Given (1) their past history of self-injurious thoughts and (2) the fact that they are likely not receiving direct treatment for their thoughts (because the thoughts are not currently present), these people may be at the most risk for having actual future thoughts. Assuming prior history of self-injurious thoughts and behavior, it would be interesting to see if people's current level of these thoughts and behavior is able to predict their actual future thoughts and behavior.

Mike Anestis commented on 'Past history serves as a better predictor of future outcomes than does client prediction for self-injurious thoughts and behaviors'

2009-07-06T19:47:53Z

Hey Jen -

Great to hear from you. I hope your final weeks up north are going well.

The authors did provide some data along those lines. What they provided was a percentage number indicating what proportion of the sample had experienced some level each variable over the course of their lifetime. I would have liked to have seen those numbers broken down by severity, frequency, etc...although the sample size was too low to effectively run that type of analysis. More importantly, I would have preferred to see those levels reported for the previous 12 months, given that this was the actual independent variable (unless I am misinterpreting something, which is entirely possible).

Anyway, here's the data they presented:

- Lifetime history of suicide thoughts: 93.8%
- Lifetime history of suicide plans: 53.1%
- Lifetime history of suicide attempts: 42.2%
- Lifetime history of non-suicidal self-injury: 100%

Jen Hames commented on 'Past history serves as a better predictor of future outcomes than does client prediction for self-injurious thoughts and behaviors'

2009-07-06T18:50:06Z

This is a great article, and a humbling reminder that there are tools and methods that have predictive ability beyond that of our own judgment.

I am curious, though, did the authors present any data looking specifically at participants' levels of self-injurious thoughts and behavior at the time they were asked to predict their future thoughts (apart from their 12-month history)?