Diagnostic criteria for malingering

Hutan

Moderator
Staff member
Diagnostic criteria for assessing malingering are used in insurance medicine. A recent review found that 61% of the studies found assessing the percentages of people judged to be malingerers used this one:

Diagnostic criteria for malingered neurocognitive dysfunction: proposed standards for clinical practice and research, 1999, Slick, Sherman and Iverson Paywall

Abstract
Over the past 10 years, widespread and concerted research efforts have led to increasingly sophisticated and efficient methods and instruments for detecting exaggeration or fabrication of cognitive dysfunction. Despite these psychometric advances, the process of diagnosing malingering remains difficult and largely idiosyncratic. This article presents a proposed set of diagnostic criteria that define psychometric, behavioral, and collateral data indicative of possible, probable, and definite malingering of cognitive dysfunction, for use in clinical practice and for defining populations for clinical research. Relevant literature is reviewed, and limitations and benefits of the proposed criteria are discussed.
 
Last edited:
Multidimensional Malingering Criteria for Neuropsychological Assessment: A 20-Year Update of the Malingered Neuropsychological Dysfunction Criteria, 2020, Sherman, Slick and Iversen

The same authors as the 1999 diagnostic criteria appear to have made an update.

Abstract
Objectives
Empirically informed neuropsychological opinion is critical for determining whether cognitive deficits and symptoms are legitimate, particularly in settings where there are significant external incentives for successful malingering. The Slick, Sherman, and Iversion (1999) criteria for malingered neurocognitive dysfunction (MND) are considered a major milestone in the field’s operationalization of neurocognitive malingering and have strongly influenced the development of malingering detection methods, including serving as the criterion of malingering in the validation of several performance validity tests (PVTs) and symptom validity tests (SVTs) (Slick, D.J., Sherman, E.M.S., & Iverson, G. L. (1999). Diagnostic criteria for malingered neurocognitive dysfunction: Proposed standards for clinical practice and research. The Clinical Neuropsychologist, 13(4), 545–561). However, the MND criteria are long overdue for revision to address advances in malingering research and to address limitations identified by experts in the field

Method
The MND criteria were critically reviewed, updated with reference to research on malingering, and expanded to address other forms of malingering pertinent to neuropsychological evaluation such as exaggeration of self-reported somatic and psychiatric symptoms.

Results
The new proposed criteria simplify diagnostic categories, expand and clarify external incentives, more clearly define the role of compelling inconsistencies, address issues concerning PVTs and SVTs (i.e., number administered, false positives, and redundancy), better define the role of SVTs and of marked discrepancies indicative of malingering, and most importantly, clearly define exclusionary criteria based on the last two decades of research on malingering in neuropsychology. Lastly, the new criteria provide specifiers to better describe clinical presentations for use in neuropsychological assessment.

Conclusions
The proposed multidimensional malingering criteria that define cognitive, somatic, and psychiatric malingering for use in neuropsychological assessment are presented.

External incentives for malingering may be financial, such as monetary settlements in personal injury litigation and wage replacement in disability and workers’ compensation claims, but may also include avoidance of duties or responsibilities, such as discharge from military service, and avoidance of criminal prosecution or harsher criminal sentencing. Outside of legal and forensic settings, certain clinical diagnoses bring with them financial support and access to services (e.g., intellectual disability), accommodations in academic and work settings (e.g., ADHD, learning disability), or access to controlled substances including narcotics or stimulants (e.g., chronic pain, ADHD).

The percentage of examinees who feign, exaggerate, or fabricate cognitive deficits during neuropsychological evaluation is substantial, and although estimates vary depending on the sample, ranges from less than 10% in medical populations without external incentives (e.g., Wodushek & Domen, 2018) up to 40% in personal injury and disability evaluations (Larrabee, Millis, & Meyers, 2009; Mittenberg, Patton, Canyock, & Condit, 2002; Ruff, Klopfer, & Blank, 2016), up to 60% in social security applicants (Chafetz, 2008), up to 50% or higher in criminal justice, penal, and military settings (Ardolf, Denney, & Houston, 2007; Jones, 2016), up to 50% in pain clinics (Greve, Binder, & Bianchini, 2009), and up to 50% in college settings where ADHD is assessed (Marshall et al., 2010; Musso & Gouvier, 2014; Suhr, Hammers, Dobbins-Buckland, Zimak, & Hughes, 2008; Sullivan, May, & Galbally, 2007).
Those are astonishing claims of malingering - 'up to 50% in pain clinics' for example. It must be a sad to work in a pain clinic when you think half of your clientele don't really have significant pain, surely that would undermine your trust in people generally?
 
This is from the 2020 paper. The authors praise their 1999 work:
The Slick, Sherman, and Iverson criteria for malingered neurocognitive dysfunction (MND) (Slick, Sherman, & Iverson, 1999) are the most widely accepted model for identifying malingering of cognitive deficits. The Slick et al. criteria are considered a major milestone in the field’s operationalization of cognitive malingering and have continued to stand the test of time as the malingering criteria with the most empirical research.
...
No other malingering frameworks have been studied to this degree in verified malingerers and in people with a variety of clinical conditions (i.e., known groups).

It seems that the criteria were criticised for being too biased to military and forensic settings, and not so appropriate for assessing children. Also, that they were focussed on cognitive dysfunction, but also needed to assess self-reported somatic and psychiatric symptoms. So this update broadens the concept of a benefit from the malingering, and extends out the coverage of kinds of disability or deficit that the person might present with.

The kinds of evaluations associated with external incentives for malingering include those related to personal injury litigation, determination of disability benefits and workers’ compensation, social services eligibility, criminal proceedings, military evaluations, and evaluations for specific clinical diagnoses that are associated with external incentives, such as those for brain injury, intellectual disability, chronic pain and related conditions, unexplained medical or neurological symptoms, ADHD, and learning disability, among others.


There are three sorts of criteria:

Criteria A: a reason for the malingering
In the original malingering framework, the term “MND” was defined as “the volitional exaggeration or fabrication of cognitive dysfunction for the purpose of obtaining substantial material gain or avoiding or escaping formal duty or responsibility” (Slick, Sherman, & Iverson, 1999, p. 552). Substantial material gain was defined as anything of nontrivial value, such as financial compensation for personal injury. Formal duties were defined as actions people are legally obligated to perform, such as military service, and formal responsibilities were those that involved accountability in legal proceedings, such as competency to stand trial.
2020 update
A clearly identifiable and substantial external incentive for feigning or exaggeration of deficits or symptoms is present at the time of examination.

External incentives for malingering include access to a desirable outcome such as financial settlement, disability payment, wage replacement, social assistance, access to services or accommodations in community, academic, or work settings, or access to prescription medication.

External incentives may also include avoidance of an undesirable outcome such as those related to criminal proceedings (e.g., avoiding being deemed competent to stand trial or avoiding criminal sentencing), military service (e.g., avoiding deployment), or work or school settings (e.g., avoiding probation, suspension, expulsion, or termination). Avoidance of an undesirable outcome in the context of malingering may also be adaptive (e.g., feigning illness to avoid being returned to an abusive situation). External incentives for malingering may also include avoiding having to fulfill more basic duties and responsibilities such as avoiding work, school examinations, or home responsibilities.


Criteria B: invalid presentation in a clinical examination
On examination of the examinee, there is either (a) compelling inconsistencies indicative of deliberate exaggeration or feigning of deficits or symptoms or (b) psychometric evidence of exaggeration or feigning of deficits or symptoms on performance validity tests (PVTs) or symptom validity tests (SVTs).
(e.g., unequivocal demonstration of disputed capacity when the examinee thinks he or she is unobserved; clear discrepancies between skills observed during the interview or while in the evaluation setting that are highly implausible and that indicate feigning, dissimulation, or distortion of symptoms).

Criteria C: marked discrepancies apparent in other evidence
1. Natural history and pathogenesis of the condition in question.
Information obtained by self-report or through tests or scales is markedly discrepant from currently accepted models of normal and abnormal neurological, medical, or psychiatric functioning in a way that suggests feigning or exaggeration of deficits or symptoms.

2. Records and other media.
Information obtained by self-report or through tests or scales is markedly inconsistent with records or other documented history (e.g., audio, video, social media) in a way that suggests feigning or exaggeration of deficits or symptoms.

3. Reliable collateral informant report.
Information obtained by self-report or through tests or scales is markedly discrepant from day-to-day level of function described by at least one reliable collateral informant with minimal stakes in the outcome of the evaluation, in a way that suggests feigning or exaggeration of dysfunction.

Criteria D: no other reason for the feigning identified in the clinical exam
This covers people with psychiatric symptoms and developmental issues that "result in significantly diminished capacity to appreciate laws or mores against malingering or inability to conform behavior to such standards." That's things like dementia and intellectual disability.
Malingering can co-occur in conditions associated with cognitive deficits including mild intellectual disability, mild dementia, or mild cognitive impairment. Similarly, malingering can co-occur in psychiatric or neurological conditions defined by somatoform symptoms (e.g., somatic symptom disorder, conversion disorder/functional neurological symptom disorder, factitious disorder, unexplained medical symptoms) and in the presence of other psychiatric conditions (e.g., depression).
 
After all that, I'm still not understanding how malingering is determined. It seems pretty circular and subject to the bias of the assessor. All of the assessments and categories seem to be just adding a veneer of 'sciencey-ness' to what boils down to the feeling of the assessor.

When the original criteria were created, we expected that a significant proportion of malingering examinees would present with significantly below-chance scores on PVTs. However, subsequent research did not bear this out. A minority of malingering examinees actually meet the Definite MND criteria; research now indicates that significantly below-chance performance on PVTs misses many—if not most—bona fide malingerers (e.g., Chafetz, 2008; Greve, Binder, & Bianchini, 2009).

I get the impression that their clever tests designed to trap the malingerer just tripped up vulnerable sick and disabled people who were struggling to work out what was going on, while the rare few genuine malingerers tended to be able to work out what answers they needed to give.
 
On compelling inconsistencies
compelling inconsistencies are instances of feigning or exaggeration of neurocognitive, somatic, or psychiatric dysfunction that are directly documented by the examiner, as opposed to being detected by PVTs or SVTs or found in records and documentation.

Examples of compelling inconsistencies are the demonstration of a disputed capacity when the examinee thinks he or she is unobserved, such as an examinee stuttering throughout a clinical interview after a minor concussion but later being observed to speak normally to the office receptionist and on their cellphone, or implausible inconsistencies evident during the examination, such as an examinee describing their medical history, medical diagnoses, pain medications, and pain symptoms in great detail but who is then vague and evasive regarding prior mental health diagnoses and medications, the examinee who reports severe and incapacitating memory problems but who describes memory lapses in extreme and lengthy detail, or the examinee who selectively denies or omits critical information indicative of a known prior condition or distorts prior work or academic history even when repeatedly questioned.

Compelling inconsistencies are not the typical, milder discrepancies seen in neuropsychological assessment, such as the examinee who reports word-finding problems yet speaks relatively normally during the interview. Rather, these are stark contradictions found either on observation or clinical interview that are so extreme or improbable that deliberate dissimulation, exaggeration, or feigning is determined to be the most reasonable cause.

Other examples are the examinee reporting catastrophic cognitive deficits who drives independently to and arrives on time for an appointment in a busy, unfamiliar location and is noted to be well groomed and articulate, or the examinee complaining of severely slowed processing speed and crippling visual problems after a minor motor vehicle collision who is able to expertly use electronic devices to show pictures of a crash scene and of news articles involving the collision to the examiner.
A lot of the examples of inconsistencies given could easily be explained by a medical condition that fluctuates in severity or affects only specific tasks. Someone might come across as vague and evasive about a prior mental health diagnoses if they don't agree with that diagnosis, or because the person being assessed doesn't think it is relevant to their current condition and they correctly assume that it will colour the assessor's evaluation.

On 'unconscious feigning' versus malingering
Importantly, compelling inconsistencies are not definitive evidence of malingering but rather of feigning or exaggeration. Malingering requires meeting additional criteria, including the presence of an external incentive and consideration of exclusionary criteria and of other conditions that can co-occur with malingering but that may better explain the feigning or exaggeration, for example, “unconscious” feigning/exaggeration in somatic symptom disorder or deliberate feigning/exaggeration in factitious disorder.
This suggests that a diagnosis of malingering excludes the "unconscious" feigning/exaggeration in somatic symptom disorder, although the process of diagnosis is the same.
 
On Presentation Validity Tests
PVTs are tests designed to detect invalid cognitive performance. Most PVTs are designed to identify scores that are not credible by virtue of being too low to be believable; that is, they identify scores that are indicative of exaggeration of cognitive problems to an extent that cannot be attributable to a bona fide cognitive, neurological, medical, or psychological condition either due to their statistical improbability based on binomial probability theory or to their clinical improbability based on studies of clinical groups, experimental malingerers, and verified malingerers.
PVTs are not “malingering tests”; the determination of malingering depends on meeting accepted, multidimensional malingering criteria that encompass not only PVTs but also external incentives and the totality of evidence available (i.e., observational, documentary, informant-based).
Although the state of the science at that time was relatively limited, the original MND criteria did not clearly define what kind of PVTs or neuropsychological tests could be used other than “forced-choice tests”; the criteria also used vague wording to refer to PVT data such as “evidence from neuropsychological testing” and evidence indicative of “response bias.” Although the criteria mentioned that tests should be well validated, there was no mention of how this should be defined or operationalized.
'the original MND criteria' are, more or less, what has been used in much of the malingering literature. It was very vague and the authors admit this..

In the revised criteria, this wording has been replaced by explicitly defined criteria that reflect current standards for using PVTs in malingering detection. Most importantly, the revised criteria operationalize the definition of a validated PVT and attempt to address the complex issue of PVT false positives with consideration of the number of PVTs administered, the issue of collinearity among PVTs, and the larger issue of the omnibus false-positive rate, critical concepts now known to be important dimensions of PVT interpretation that were not included in prior malingering models and which are discussed later in this paper.

It's a bit laughable. 'Experimental malingerers' So they give healthy people a cognitive test and ask them to answer as if they were a malingerer? 'Verified malingerers' - again all sorts of problems, especially with circularity.

This next quote is long and complicated, but it seems to be saying - 'those tests we used were not good at identifying malingerers, but our new tests are. That raises the question - is any review based on old literature that used these poor tests going to be fundamentally flawed? It also raises questions about who might be liable for all the despair created by people being falsely diagnosed as feigning when the old tests were applied to them?

I don't get any sense that these authors have any real understanding of what it might be like to be diagnosed as malingering or feigning, when actually you are sick.

A related point concerns the definition of PVT failure in prior malingering models. As already noted, the 1999 psychometric criteria for Definite MND was defined by the presence of below-chance performance “on a forced-choice test.” Much of the research at that time relied on binomial probability theory for identifying exaggeration on cognitive tests; the older conceptualization of noncredible performance based on binomial probability theory can still be seen in some test manuals published in that time period. For example, the Victoria Symptom Validity Test (VSVT; Slick, Hopp, Strauss, & Thompson, 1997) defines significantly below-chance performance as “Invalid,” chance-level performance as “Questionable,” and above-chance performance as “Valid” (Slick, Hopp, Strauss, & Thompson, 1997). As research has progressed, it has become clear that there are other ways of identifying noncredible performance with a very high degree of certainty other than significantly below-chance performance on forced-choice tests. We also now know that cutoffs ranging into the “Valid” range based on binomial probability are actually much more effective at detecting the greatest proportion of malingering examinees on forced-choice tests because only a minority of malingerers produce significantly below-chance or even chance performance (e.g., Jones, 2013). Thus, most modern forced-choice PVTs now define failure using specific empirically derived cutoffs with known sensitivity and specificity, not on cutoffs based solely on binomial probability theory. In addition, numerous PVTs that use methods other than the forced-choice procedure now exist, either as stand-alone tests or embedded within existing neuropsychological tests. The new model incorporates these psychometric advancements in the wording and overall approach to PVT criteria.
 
There's a whole lot more on Presentation Validity Tests. To me, it looks like people arguing about how many angels dance on the head of a pin e.g. here they are trying to work out how many tests have to be failed in order to be a malingerer - or is it the ratio of the number of tests? And, someone really shouldn't be classed as a malingerer on the basis of just one test - unless of course it's a really compelling result....
Thus, it is more appropriate to consider the ratio of PVT failures to total PVT scores rather than the absolute number of PVTs administered (Bianchini, personal communication, July 2019) with the assumption that most neuropsychologists will administer multiple PVTs in the neuropsychological evaluation to appropriately rule in/rule out malingering and to adequately sample validity throughout the assessment. For example, failing two of seven PVT scores would appear to meet criteria for invalid responding, as would failing four of 14 PVT scores; failing two of 14 PVT scores likely would not (i.e., because this would be equivalent to failing one out of seven PVTs). Using only one PVT score in the decision-making process for determining the presence of malingering would be considered insufficient unless that score indicates significantly below-chance performance; neuropsychological evaluations that contain no PVTs at all would not meet acceptable testing standards.

They briefly note that testing the validity of the tests on healthy volunteers pretending to be malingerers probably isn't ideal. They again note that the tests aren't very good at identifying malingerers. Their solution to that problem is to do more tests.

Thus, as already noted, because most PVTs are only moderately sensitive (i.e., do not detect most malingering cases), sensitivity will increase as more PVTs are administered, providing more impetus for administering multiple PVTs during the neuropsychological examination.
 
I've got to stop. That 2020 paper is incredibly long and I haven't got to the end of it yet. The revised version still seems to be focussed on assessing the feigning of cognitive capacity, in order to, for example, avoid being judged mentally capable in a criminal trial.

Here's a box with some examples of Criteria C issues. And I can see that some of these would be compelling evidence, but I can also see how many could be misunderstandings of the deficit the person is actually suffering from.

Again, it seems to me that the assessment of someone as feigning is very ad hoc, despite all of this effort to make it seem otherwise.

Box 2. Examples of Marked Discrepancies Between Test Data/Symptom Report and Other Sources of Evidence (Criterion C).
Discrepancies between test data/symptom report and the following

  1. Natural history and pathogenesis of the condition in question

    An examinee performs in the severely impaired range on memory tests after a mild concussion.

    An examinee recalls the details of a motor vehicle collision but claims anterograde amnesia for several months after the collision.

    An examinee sustains a mild traumatic brain injury and claims a loss of autobiographical memory.

    An examinee with alleged PTSD describes highly detailed, stereotypic, repetitive nightmares of increasing frequency beginning two years after a traumatic event.

    An examinee in a criminal forensic setting reports hearing a single exaggerated/malevolent voice commanding them to perform a criminal act but has no other psychiatric symptoms such as other hallucinations, delusions, negative symptoms, or disorganized speech.

    An examinee reports pain symptoms of sufficient severity to preclude work but routinely engages in recreational sports and social activities.
  2. Records and other media

    An examinee obtains severely impaired memory scores after a motor vehicle collision, but emergency, hospital, and family doctor records indicate no loss of consciousness or cognitive problems at the scene or subsequently.

    An examinee reports pervasive and lengthy post-traumatic amnesia, but hospital records and nursing notes indicate full alertness, normal behavior, and no post-traumatic amnesia.

    An examinee denies previous brain injury or psychiatric history yet has medical records that document prior personal injury litigation for brain injury and a long-standing history of mental health problems.

    An A-student on academic scholarship undergoing baseline concussion testing obtains severely impaired scores in almost all domains tested.

    An examinee who reports incapacitating neck pain interfering with activities of daily living after a minor collision is discovered to be a social media influencer with frequent posts documenting athletic accomplishments.

    An examinee reports a permanent decline in GPA after concussion attributed to a motor vehicle collision, but review of school records reveals marginal grades in high school and academic probation in college courses both before and after the collision.
  3. Reliable collateral informant report

    An examinee is unable to perform simple math problems in testing but performs well as an accountant according to an employer.

    A patient reports severe memory deficit impairing work performance but has above-average work performance according to a work supervisor.

    An examinee reports excellent grades and no accommodations before an accident, but teacher questionnaires indicate severe learning issues.

    A high school student reports a long-standing history of severe and impairing ADHD, but standardized questionnaires from a teacher who knows the student well indicate no ADHD-type problems.
 
I have not read any of the detail, but I suspect that this framework of ‘malingering’ where it might have a significant benefit such as avoiding military service or avoiding a criminal prosecution is also generalised to other situations where the cost of being disabled far outweighs the putative gains of accessing any benefits.

For the majority I suspect the supposed gains of being disabled do little to counter the substantial financial and societal losses incurred.
 
We have gained nothing.

We have lost, hugely, by every meaningful measure of quality of life.

That is the reality.

I defy any of those who think otherwise to trade places with one of us for a while. The vast majority of people will quickly realise how utterly wrong they were.
 
The thing about so called malingering is the idea of 'secondary gain'.

I think it is incumbent on people who accuse pwME of malingering to demonstrate what we're supposed to have gained.

Is it the dismissal of our reality by clinicians, the suspicion, disbelief and rejection by employers, colleagues, friends, even family members? Is it the poverty, is it the lost opportunities to live a full life, is it the isolation, the loneliness? Is it the dependence on others, the vulnerable and sometimes demeaning need for personal care, is it the barriers and trauma of the benefit application process?

Tell us, where is the secondary gain in all that?
 
Back
Top Bottom