Trial By Error: “Effort Preference”? WTF?

By David Tuller, DrPH

When I was a young gay man in the 1980s (I’m 67), a common term for sexual orientation was “sexual preference.” This phrase always struck me as weird. Preference? The urges I felt were not a “preference.” Preference clearly implies a choice—as in, I prefer to live in the city rather than the countryside. I was not given a choice about my sexual orientation. From my earliest awareness as a sexual being, I was attracted to men. Period. “Preference” had nothing to do with it.

(For me, that is. Others might have experienced their sexual orientation differently. Having said that, I want to add that were I given a choice now, I’d choose to be gay. Despite all the heartache and loss of the AIDS epidemic and the many other challenges of being a member of an often-despised minority, my life has been rich in experience, love, pleasure, and joy—all intermingled with the pain that is integral to being a sentient being.)

I mention this as a way of highlighting my initial reaction to the long-awaited and long-delayed publication of the US National Institutes of Health’s study of 17 ME/CFS patients, along with 21 healthy controls. The paper, titled “Deep phenotyping of post-infectious myalgic encephalomyelitis/chronic fatigue syndrome,” was published in February in Nature Communications, a high-impact journal. It immediately triggered howls of protest.

One reason for concern was that the authors made sweeping claims for a paper based on such a small sample of patients, some of whom were too disabled to participate in every part of the study. (The study was originally supposed to include a much larger sample, but the coronavirus pandemic interrupted those plans.) Another was the controversial claim at its core–that the profound fatigue reported by patients is the result of a construct called “effort preference.” As noted in the third sentence of the abstract, “one defining feature of PI-ME/CFS was an alteration of effort preference, rather than physical or central fatigue.”

Members of the patient community have conducted extensive and robust analyses of the flaws of the “effort preference” construct. Jeannette Burmeister, a lawyer as well as a patient, posted a blockbuster four-part series about it on her blog, Thoughts About M.E. In an extensive thread, members of the Science for ME forum also whacked through the thicket of data shenanigans that led to the NIH’s conclusion. As of this posting, that thread includes almost 800 comments—many of them quite lengthy. (On the Health Rising forum, Cort Johnson offered a perspective that was more sympathetic to the research team.)

Last week on X (formerly Twitter), Burmeister reported that the NIH’s Agency Intramural Research Integrity Officer has begun assessing the complaints made about the study. Let’s hope this turns out to be an in-depth inquiry that figures out what exactly has gone on here. And let’s hope they clean up their EEfRT analysis and get rid of the “effort preference” mess. Unless that happens, it seems likely future research into ME/CFS and Long Covid will also adopt this approach.

Anybody who knows anything about the terrible history of ME (or CFS, or ME/CFS) should have realized the disastrous nature of suggesting the involvement of any sort of “preference.” Subsequent assurances from the NIH that “preference” refers to subconscious or unconscious or pre-conscious calculations by the brain and does not mean what it obviously means to the lay person—that is, a conscious choice–are pathetic, even if taken at face value.

It is hard to fathom how those leading this research team could be so out of touch and tone-deaf. We now have more than three decades of experience regarding the fallout of the decision by the US Centers for Disease Control and Prevention to name this devastating condition “chronic fatigue syndrome” in the 1988 study of the famous Lake Tahoe outbreak. Health officials seem to have learned nothing since then about the power of language to shape illness perceptions and negatively impact the lives of patients. That’s very troubling.

In response to the concerns, the NIH released a statement that sought to blunt the criticism. In describing the meaning of “effort preference,” the statement argued that “the term does not describe a psychological difference in people with ME/CFS and people without ME/CFS; it describes an actual abnormal finding in how the body’s neural system is supposed to function.”

Great! Explain that to everyone who will read the study and interpret the word “preference” in the normative sense, not in the convoluted and non-intuitive way that the NIH apparently thinks it should be taken. This includes clinicians who might be inclined to scold patients for “preferring” not to make an effort. For all we know, the study might already be generating this sort of result.

It is also unjustified for health officials to chastise patients for criticizing the study, as Avindra Nath, the senior and corresponding author, has done. During a public presentation of the findings, he said: “When you doubt our intentions and pick apart every single word, you also tear us apart. It causes pain and suffering on both sides and demoralizes us and shatters our goals.” Putting aside the issue of intentions, it is essential to “pick apart” words if they are, like “preference,” so poorly and ineptly chosen.

All research is subject to legitimate criticism, no matter how much time and effort has gone into it. Complaining about the “pain and suffering” of researchers whose work is scrutinized closely is a form of emotional blackmail. To wield government authority in order to shame a long-abused patient community into shutting up is unwarranted—especially when it was the NIH, not patients, that selected a problematic term implying patients “prefer” to do less.

**********

The weird use of the EEfRT

The issues with “effort preference,” however, do not seem limited to the nomenclature. The construct emerged through the NIH’s use of something called the Effort-Expenditure for Rewards Task (EEfRT). This approach, which was designed to test motivation in depressive patients with anhedonia, was first described in a 2009 paper in PLoS One. The EEfRT is quite complicated, so I won’t go into it at length here. Suffice it to say that it involved requiring participants to choose between so-called “hard” and “easy” tasks in a series of button-pushing trials.

As the 2009 paper explained, “Successful completion of hard-task trials required the subject to make 100 button presses, using the non-dominant little finger within 21 seconds, while successful completion of easy-task trials required the subject to make 30 button presses, using the dominant index finger within 7 seconds.” The tasks involve varying amounts of small monetary rewards that participants might receive upon successful completion, as well as different probabilities for whether they would actually receive the designated reward for completing particular trials.

While the EEfRT has been adapted for use in subsequent studies, it is unclear why the NIH decided that a system designed with anhedonic depressives in mind would be applicable to and appropriate for ME/CFS. Given that Brian Walitt* [I originally misspelt this as Wallitt, here and in two other mentions], the lead author of the paper, has a long history of psychologizing complex chronic illnesses, perhaps the choice should not have been a surprise. Back when Walitt was tapped for this slot, many patients expressed dismay and noted his history of offensive statements regarding these conditions. Their concerns have been borne out.

This statement from the paper pretty much sums it up: “Effort preference is how much effort a person subjectively wants to exert.” And this one: “Together these findings suggest that effort preference, not fatigue, is the defining motor behavior of this illness.” And then there’s this statement from Walitt himself: “Rather than physical exhaustion or a lack of motivation, fatigue may arise from a mismatch between what someone thinks they can achieve and what their bodies perform.”

So the conclusion is that a patient’s subjective—and apparently inaccurate—feelings or thoughts about their capabilities, and nothing else, are the key drivers of ME/CFS. On its face, this assertion seems little different from the claims of the PACE authors that “unhelpful beliefs” about illness are a major cause of illness. Given Wallitt’s past claims, it is challenging to accept that these statements are just clumsy or inartful turns of phrase.

I’m not a statistician and can’t vet every argument made by Burmeister or commenters on the Science for ME forum. But their analyses have highlighted so many apparent departures from standard research practices in the conduct of the EEfRT and the presentation of the results that the validity and credibility of the study is now in question. Besides the fact that the EEfRT was designed for purposes completely unrelated to ME/CFS, Walitt and his colleagues seem to have ignored a central tenet: the tasks, even the “hard” ones, are supposed to be easy enough for everyone to complete without fatigue playing a role.  

Here’s a key passage from the 2009 paper on “the effects of fatigue during the EEfRT”:

“An important requirement for the EEfRT is that it measure individual differences in motivation for rewards, rather than individual differences in ability or fatigue. The task was specifically designed to require a meaningful difference in effort between hard and easy-task choices while still being simple enough to ensure that all subjects were capable of completing either task, and that subjects would not reach a point of exhaustion.”

This was not apparently the case in the way the NIH applied the test. In fact, the data indicate that patients did have more problems completing the tasks than the healthy controls. That’s obviously a problem—and only one of many outlined in the critiques. Unfortunately, it seems that the EEfRT is likely to be used in studies of Long Covid as well. (I urge readers to review both Burmeister’s opus and the long thread on the Science for ME forum to assess the issues for themselves.)

**********

A related side note

In related news, Nath made the following statement during a podcast interview earlier this year: “The thing is that there’s so much overlap between that and long COVID, and the population affected by long COVID is so much larger, and the symptoms are so much more severe compared to ME/CFS, that we decided that what we should do is focus on long COVID.”

Really? On what basis did Nath decide that Long Covid symptoms “are so much more severe compared to ME/CFS”? Is he aware that the most severe patients could not even have dreamt of traveling to the NIH for two weeks to participate in his mammoth study? This assertion is mind-boggling—and a sign of serious problems in whatever is going on at the NIH complex in Bethesda, Maryland.