Thursday, June 24, 2010

Suffering from the pain of social rejection? Feel better with TYLENOL®

It's not just for headaches anymore! The active ingredient in TYLENOL® (acetaminophen, also known as paracetamol) has been shown to ease the pain of social rejection. Wouldn't it be great if you could pop an over-the-counter medication to lessen the hurt of being excluded from that grad student party? Of being ostracized by all your old friends? Even disowned by your family and becoming homeless? The journal article, which was promoted by press release six months ago, has finally appeared online (Dewall et al., 2010). An excerpt from the December 2009 press release is below.
A Pill for Psychological Pain?

. . .

“The idea—that a drug designed to alleviate physical pain should reduce the pain of social rejection—seemed simple and straightforward based on what we know about neural overlap between social and physical pain systems. To my surprise, I couldn’t find anyone who had ever tested this idea,” [psychologist C. Nathan] DeWall said.
Perhaps because there's no clear mechanistic basis for such an idea? The authors themselves never proposed one either. One might expect that a psychopharmacological experiment with a drug that can cause serious liver damage would be conducted with a specific hypothesis in mind and some basic knowledge about how the drug is thought to work, but we didn't see that here. Granted, that would not be typical fare for Psych Sci. So instead the rationale given by Dewall et al. (2010) is partially linguistic, partially based on a neuroimaging study (Eisenberger et al., 2003):
Studies suggest that the similar linguistic descriptions of social and physical pain extend beyond metaphor, and demonstrate overlap in the neurobiological systems underlying physical pain and social pain (DeWall & Baumeister, 2006; Eisenberger, Lieberman, & Williams, 2003; Way, Taylor, & Eisenberger, 2009). In the present experiments, we examined one functional consequence of the hypothesis that social and physical pain rely on shared neurobiological systems—whether acetaminophen, a common physical pain reliever, also reduces social pain.
The "shared neurobiological systems" are thought to be located in the dorsal anterior cingulate cortex (ACC), a brain structure that contains discrete regions responsive to physical pain (Kwan et al., 2000). Interestingly, externally applied vs. self-administered thermal pain activate anatomically distinct areas of the ACC (Mohr et al., 2005). Furthermore, it is not at all clear whether the same regions of ACC represent social pain and the affective components of physical pain. In a study designed to dissociate expectancy violations from social rejection, the dorsal ACC was activated when expectations were violated, while ventral ACC (quite distant from the physical pain regions) was activated by social rejection (Somerville et al., 2006).

Figure 2 (Somerville et al., 2006). Differential ACC response to expectancy violation and social feedback. (a) A three-dimensional rendering of the medial surface of the brain illustrates a functional dissociation between dorsal (dACC) and ventral (vACC) anterior cingulate. A whole-brain voxel-by-voxel ANOVA was used to identify voxels that showed a significant main effect (P less than 0.001, uncorrected) of expectancy violation (blue) and a main effect of feedback type (yellow).

At any rate, participants in the Eisenberger et al. (2003) study took part in a computerized ball-tossing game while being scanned. Initially, two fictitious players included the scanned subject in the game, but then started to exclude him/her. This was the “social exclusion” condition, which was compared to the inclusion condition. But it happens to be the case that this paper was singled out as one of the worst of the "voodoo correlation" violators by Vul and his colleagues [PDF], since it reported a statistically unlikely value based on a non-independent analysis:
Eisenberger, Lieberman, and Williams (2003), writing in Science, described a game they created to expose individuals to social rejection in the laboratory. The authors measured the brain activity in 13 individuals at the same time as the actual rejection took place, and later obtained a self-report measure of how much distress the subject had experienced. Distress was correlated at r=.88 with activity in the anterior cingulate cortex (ACC).
A correlation of r=.88 between dACC activity and self-reported distress is implausibly high... But I'll stop here, and point to Lieberman, Berkman, and Wager's (2009) reply to Vul et al.

That brings us to the present study by Dewall et al. (2010). In Experiment 1, 30 participants (24 women, 6 men) took one 500 mg acetaminophen pill immediately after waking up and another 500 mg an hour before going to sleep (1,000 mg per day for 3 weeks). The other 32 participants (24 women, 8 men) took the same dosing of placebo for 3 weeks. Each evening, subjects filled out the the Hurt Feelings Scale (the "today" version) to report how much social pain they had experienced that day. Despite the fact that the half life of acetaminophen is 4 hours, it took about 10 days for the drug group to report significantly lower hurt feelings than the placebo group. The difference on day 21 was greatest (p < .005). However, the difference in change-over-time slopes between the two groups was only marginally significant (p ≤ .10). The explanation of the time course for these effects was unclear:
Acetaminophen has a relatively short half-life, lasting approximately 4 hr, which means that it is unlikely that acetaminophen had a cumulative effect in our experiments. Our finding that acetaminophen reduced hurt feelings over time could be due to a combination of not feeling hurt and having a greater ability to reappraise the rejection experience.
In Experiment 2, the dose was upped to 2,000 mg acetaminophen per day for 3 weeks. Instructions were given to refrain from drinking entirely, since alcohol can potentiate liver damage when taken with acetaminophen. In 2009, an FDA panel made a recommendation to lower the maximum daily dose from 4,000 mg (to an unspecified value). The panel also endorsed limiting the maximum single dose of the drug to 650 mg, down from the current 1,000 mg dose (which was given in Exp. 2). At the end of the three week period, the cyberball exclusion fMRI study was run. The acetaminophen group showed less activity in dACC in response to social exclusion, but they did not report lower hurt feelings.

Hmm. As an aside, here's another puzzling observation. If it's been claimed that "social exclusion hurts" (Macdonald & Leary, 2005), then why does the experience of social exclusion result in higher tolerance for physical pain and higher pain thresholds (DeWall & Baumeister, 2006)?

Next, I have a series of questions for the authors:
  • Did you consider the negative consequences of acetaminophen?
  • Did your IRB have anything to say about this (there was NO info in the paper on institutional approval or the signed consent procedure with participants)?
  • How did you decide on your dosing regimen?
  • Why 3 weeks, when the half-life is only 4 hrs?
  • Instructions say "do not use for more than 10 days without doctor's permission." Was an MD involved in the study?
  • What do you know about the mechanism of action? For more info, see Acetaminophen from Frank J. Dowd.
  • And most critically, why did you choose acetaminophen, rather than aspirin, ibuprofen, or naproxen?
  • Related to this, why did the first version of the manuscript have "Tylenol" in the title? 1
I was also alarmed by the wild extrapolation from exclusion in a laboratory video game to purported increases in aggressive violence:
Furthermore, many studies have shown that being rejected can trigger aggressive and antisocial behavior, which could lead to further complications in social life (DeWall, Twenge, Gitter, & Baumeister, 2009; Warburton, Williams, & Cairns, 2006). If acetaminophen reduces the distress of rejection, the behavioral consequences of rejection, such as antisocial behavior, may be reduced as well. Indeed, our fMRI results showed that acetaminophen diminished reactivity in the dACC and amygdala, brain regions that have been linked to aggression (Denson, Pedersen, Ronquillo, & Nandy, 2009; Eisenberger, Way, Taylor, Welch, & Lieberman, 2007). It would therefore be worthwhile to explore whether acetaminophen reduces the aggressive consequences of social rejection.
I'm sure the 24 women in the placebo condition felt like committing mass murder after being excluded from a game of cyberball. Better put them on TYLENOL®.

The Medscape article on the study had prominent kudos from Bruce G. Charlton, MD, who:
...applauded the investigators' research efforts.

"It is particularly difficult to get research funding to study old, cheap, unpatented, over-the-counter drugs, so I congratulate the authors on doing this," he said.

Dr. Charlton, who [was] editor-in-chief of Medical Hypotheses and professor of theoretical medicine at the University of Buckingham, United Kingdom, agreed that different sorts of pain are often related, so there is good reason to assume that acetaminophen or paracetamol may benefit those who suffer any type of pain of unpleasant feelings, including some types of depression.

However, he noted that the same effect would likely apply to aspirin, nonsteroidal anti-inflammatory drugs, and opiates, "about which there is more evidence," he said.
One last question arose when I checked funding for the study:
This work was funded by grants from the National Institute of Mental Health (MH-65999) and the Gulf Atlantic Group, Inc.
The grant number was misreported, it's MH065559, not MH-65999.

And funding from the Gulf Atlantic Group, Inc.?? Trying to trace that entity has been like navigating a maze, an endless series of shell corporations:

But maybe it's really Gulf Atlantic Funding Group Inc ? -- no, that's mortgages.

OR how about: Gulf Atlantic Group Incorporated in Tallahassee, FL (where one of the authors is located)?

Or maybe it was just another typo...


1 Original name of the article was: DeWall C. N., MacDonald, G., Webster, G. D., Masten, C., Baumeister, R. F., Powell, C., Combs, D., Schurtz, D. R., Stillman, T. F., Tice, D. M., & Eisenberger, N. I. (in press). Tylenol reduces social pain: Behavioral and neural evidence. Psychological Science.

Here are two other papers that use the brand name Tylenol:

DeWall, C. N., Stillman, T. F., MacDonald, G., Webster, G. D., Finkel, E. J., Tice, D. M., & Baumeister, R. F. (2010). Can Tylenol boost self-esteem? Effects of acetaminophen on perceived social threat and social self-esteem. Manuscript in preparation. (Intended outlet: Journal of Personality and Social Psychology).

DeWall, C. N. (2008). Effects of Daily Acetaminophen on Social Emotions: Can Two Tylenol Overcome Heartbreak? Paper presented at the Annual Meeting of the Society of Personality and Social Psychology. Albuquerque, NM.


DeWall CN, Baumeister RF (2006). Alone but feeling no pain: Effects of social exclusion on physical pain tolerance and pain threshold, affective forecasting, and interpersonal empathy. Journal of Personality and Social Psychology, 91, 1–15.

Dewall CN, Macdonald G, Webster GD, Masten CL, Baumeister RF, Powell C, Combs D, Schurtz DR, Stillman TF, Tice DM, & Eisenberger NI (2010). Acetaminophen Reduces Social Pain: Behavioral and Neural Evidence. Psychological Science PMID: 20548058

Eisenberger NI, Lieberman MD, Williams KD. (2003). Does rejection hurt? An FMRI study of social exclusion. Science 302:290-2.

Kwan CL, Crawley AP, Mikulis DJ, Davis KD. (2000). An fMRI study of the anterior cingulate cortex and surrounding medial wall activations evoked by noxious cutaneous heat and cold stimuli. Pain 85:359-74.

Lieberman M, Berkman E, Wager T. (2009). Correlations in Social Neuroscience Aren't Voodoo: Commentary on Vul et al. (2009) Perspectives on Psychological Science, 4 (3), 299-307

Macdonald G, Leary MR. (2005). Why does social exclusion hurt? The relationship between social and physical pain. Psychol Bull. 131:202-23.

Mohr C, Binkofski F, Erdmann C, Büchel C, Helmchen C. (2005). The anterior cingulate cortex contains distinct areas dissociating external from self-administered painful stimulation: a parametric fMRI study. Pain 114:347-57.

Somerville LH, Heatherton TF, Kelley WM. (2006). Anterior cingulate cortex responds differentially to expectancy violation and social rejection. Nat Neurosci. 9, 1007-1008.

Vul E, Harris C, Winkielman P, Pashler H (2009). Puzzlingly High Correlations in fMRI Studies of Emotion, Personality, and Social Cognition [PDF]. Perspectives on Psychological Science 4(3), 274-290.

Way BM, Taylor SE, Eisenberger NI (2009). Variation in the mu-opioid receptor gene (OPRM1) is associated with dispositional and neural sensitivity to social rejection. Proceedings of the National Academy of Sciences 106, 15079–15084.

Subscribe to Post Comments [Atom]


At June 26, 2010 7:24 PM, Blogger Michael Cohn said...

Regarding the use of acetaminophen rather than other drugs: Naproxen, ibuprofen, and all other over-the-counter pain relievers are non-steroidal anti-inflammatories (NSAIDs), which work by reducing inflammation, which is often a cause of pain. Acetaminophen is an analine analgesic, the only one currently in use. In addition to having some NSAID-like anti-inflammatory properties, It actually reduces the subjective experience of pain, probably via activity at anadamide receptors. The pain in social rejection studies is not hypothesized to result from peripheral inflammation, so one would predict that even if acetaminophen works, other pain relievers would not.

I actually don't find the hypothesis that outrageous. It's pretty common for evolution to piggyback new capacities on older ones, as in the case of moral disgust that looks a lot like physical disgust, or adult attachment showing many similarities to one's infant attachment. Why not check to see if various painful emotions respond to pain-relieving drugs? (and acetaminophen is the only good choice here, since showing that morphine reduced social pain wouldn't exactly surprise anyone).

At June 26, 2010 7:24 PM, Blogger Michael Cohn said...

(sorry to multi-post, but it looks like blogger was rejecting excessively long comments.)

Regarding the details of informed consent: Psychological Science is a short-report format journal. If you page through an issue, I doubt you will find any articles that devote more than half a sentence to informed consent procedures. Part of the review process for journal publication is affirming that your study was approved by an accepted IRB. As for your question about whether there was oversight by a physician, there may well have been. It's pretty common for pharmaceutical studies to be required to have a medical professional physically present or available on call. Also, I think that warnings about consulting a doctor before prolonged medication use are in large part concerned with making sure the medication isn't masking a more serious problem (e.g., someone using acetaminophen to try and treat pain that's being caused by a serious infection).

If you think that the authors may have skipped some aspect of human subjects approval, or written their consent information in a way that misrepresented the risks of acetaminophen, you can probably FOIA their original consent forms (since it's federally funded research) and ruin their careers. NIH takes this kind of thing seriously.

The risk of liver damage from acetaminophen use was probably not very high. Acute toxicity requires a dose of ~10,000mg, or ~6,000 per day for several days. A large meta-analysis found no evidence of acute liver failure or of substantial liver damage in prospective studies with healthy subjects and controlled dosing. Some of these studies used a daily dose of 4,000mg, four times higher than that used in this study. That said, I do hope that the authors screened for known, suspected, or elevated risk of liver damage while recruiting -- it would have been phenomenally stupid not to.

At June 28, 2010 4:58 PM, Blogger Maia Szalavitz said...

Hmm. As an aside, here's another puzzling observation. If it's been claimed that "social exclusion hurts" (Macdonald & Leary, 2005), then why does the experience of social exclusion result in higher tolerance for physical pain and higher pain thresholds (DeWall & Baumeister, 2006)?

The same reason experiencing physical pain does: tolerance. If you experience pain over long periods of time, you get better able to manage it (unless you sensitize and develop a chronic pain disorder).

This study would have made sense if they used opioids, which are known to reduce the emotional aspect of physical pain. There's also a high concentration of opioid receptors in the cingulate. Of course, the result wouldn't have been novel or surprising: junkies wouldn't exist if opioids didn't kill emotional pain.

Indeed, if acetaminophen could numb emotional pain, this would have been discovered by addicts by now. The fact that the drug remains boringly OTC suggests that this effect is either so small it can only be detected in the lab or nonexistent as the blog suggests.

At June 28, 2010 5:21 PM, Blogger Michael Cohn said...

I'd consider it important if acetaminophen turned out to have any detectable effect on social pain at all. This would demonstrate a linkage between two systems that was not previously established, and open the door to many important research questions (and potentially to the development of more targeted drugs that would be practically useful).

I actually don't know whether to assume that people should have already noticed any psychoactive effects of acetaminophen. I can't think of any other drugs that have psychological effects with no subjective physical, emotional, or cognitive effects (all it does is _stop_ something from happening). That could mean that: 1) drugs like that don't exist, or they're very rare, or 2) there are others; we're just not very good at noticing them.

I don't think tolerance is a good explanation for the loneliness / pain findings -- the study induced acute loneliness, rather than studying people who'd had a chance to get used to it. The authors do go into some detail regarding their hypothesis that loneliness activates a regulatory system that reduces all kinds of pain, in the same way that an injury to one part of your body could lead to an opioid response that dulls pain in other parts of your body as well. They highlighted the fact that the differences caused by the loneliness induction looked more like reduced sensitivity or numbness, rather than like increased executive control (which I imagine is what the original poster was thinking about when expressing skepticism that one kind of pain would increase tolerance for another kind).

Any vagueness here is primarily my fault -- my neuropsych is weak, but the article goes into a fair bit of detail.

At June 29, 2010 5:09 AM, Blogger Maia Szalavitz said...

It's possible that if tylenol has a delayed effect that numbs emotional pain, it wouldn't be noticed and picked up by addicts. If it was associable within hours of taking the drug, however, reduced social pain is something people notice profoundly and addicts seek intentionally.

This is a primary effect of SSRI's: and to those who are oversensitive to rejection and otherwise socially sensitive, lifting that is an amazing relief. This is why SSRI's help in cases of addiction driven by depression that is primarily about social fears.

SSRI's aren't "addictive" in the sense of causing compulsive use despite negative consequences (the DSM addiction definition, basically) however because they take weeks to take effect-- so you don't connect the drug and the effect. If there was an instant SSRI, I'd bet money it would soon be classified as a controlled substance.

At June 30, 2010 9:04 AM, Anonymous Rick Heller said...

Regardless of the weakness of individual papers in their statistical analyses (which I am unqualified to judge) the overlap between the affective component of physical pain and the suffering of emotional pain in the dorsal anterior cingulate is well-replicated. Here's another reference:
Marco L. Loggia, Jeffrey S. Mogil, Catherine Bushnell
Empathy hurts: Compassion for another increases both sensory
and affective components of pain perception Pain 136 (2008) 168–176
Thus, any analgesic that works centrally is likely affect both physical and emotional pain. Obviously, Tylenol doesn't have as big an impact as morphine, which is why this interesting but minor fact has escaped people's attention.

At June 30, 2010 9:34 PM, Blogger Mike Mike said...

It is possible that when people report things like "loneliness" and "emotional pain", they're reporting a very general negative affective state, which could be caused in part by cumulative aches and pains. Also, it's known that negative affect can cause a greater sensitivity to normal, marginally painful sensation. Maybe provoking loneliness for 10 days increased people's sensitivity to their normal, slightly painful sensations, and then paracetamol helped this, which improved their mood over the people who didn't get the drug.

Sorry, I didn't have time to hunt for the paper, just my thoughts based on your post.

At July 02, 2010 2:41 AM, Blogger Neuroskeptic said...

The mechanism of action of paracetamol is interesting because it's metabolized into AM404 which modulates the endocannabinoid system.

I don't know how important that is in the real world analgesic effects, but it does suggest it at least might have central properties.

Also, that could cause its mechanism of action to be longer than the 4 hrs would suggest...

But this Gulf Atlantic Funding Group is rather mysterious.

At July 25, 2010 11:44 PM, Anonymous Anonymous said...

The hypothesis that "social pain" is experienced through a "mirror activation" of the so-called "pain matrix" is based on a very strong assumption: that this so-called "pain matrix" is actually pain-specific...

See "From the neuromatrix to the pain matrix, and back" (Mouraux & Iannetti 2010 EBR) for a critical review of the specificity of the brain regions "responding to pain".

At July 27, 2010 12:44 AM, Blogger The Neurocritic said...

Thanks for providing a link to this new article, which is challenging indeed to the notion of a "Pain Matrix".


Post a Comment

<< Home

eXTReMe Tracker