CAN PHYSICAL THERAPISTS IDENTIFY MALINGERED PAIN · PDF fileCAN PHYSICAL THERAPISTS IDENTIFY MALINGERED PAIN IN THE ... Many physiotherapists use tools in an attempt to detect malingering;

August 2011

1

CAN PHYSICAL THERAPISTS IDENTIFY MALINGERED PAIN IN THE CLINICAL

SETTING?

Monica Caton MPT1, Jaskiran Chohan MPT1, Carlos Velez MPT1, Juny Wu MPT1, David Walton BScPT

PhD1

1: School of Physical Therapy, The University of Western Ontario, London Canada

Disclosure statement:

The authors disclose no conflict of interest, financial or otherwise, with the material contained in this

manuscript.

August 2011

2

ABSTRACT

Background: Many physiotherapists use tools in an attempt to detect malingering; however, there are

discrepancies within the literature as to their ability to identify patients who malinger.

Purpose: To survey the literature available on diagnostic tools that could be used in physiotherapy

practice and their ability to detect malingering of pain.

Methods: We conducted a review of 4 international databases, with a specific focus on review articles.

Search terms included malingering, pain, symptom amplification and exaggeration.

Results: Our search revealed a number of tools being used to detect malingering in the clinical setting:

Pain Patient Profile (P3), Modified Somatic Perception Questionnaire (MSPQ), Symptom Checklist 90-

Revised (SCL-90-R), Minnesota MultiPhasic Personality Inventory 2 (MMPI-2), Symptom Validity Tests

(SVT), Waddell Signs, and Manual Muscle Tests (MMT). There was no strong empirical support for any

of the tools. Consistent limitations interfered with our ability to label any one of them superior to the

others or in fact as valid means of identifying malingered pain.

Conclusion: Our review suggests that there are currently no tools that can be confidently supported as a

consistently strong test of malingered pain. The absence of both a clear gold standard and a consensus

definition of malingering pose a challenge to the development and validation of tools to identify

malingering.

Keywords: malingering, symptom exaggeration, pain, symptom amplification, screening

August 2011

3

INTRODUCTION

Malingering is “the intentional (conscious) production of false or grossly exaggerated physical or

psychological symptoms motivated by external incentives”.1 The prevalence of malingering has been

estimated to range from 1.25-10.4% amongst chronic pain samples.2 Researchers have argued that failure

to detect malingering is responsible for the diversion of a considerable amount of limited health care

dollars.3

Investigators have attempted to develop tools that clinicians, including physiotherapists, can use

to identify intentionally malingered pain. While many of these tools are reportedly sensitive to

malingering, their accuracy has been questioned.2,4 Fishbain et al.2 conducted a review of the literature

and concluded that “there are currently no reliable methods to identify malingering”. Conversely, in their

review, Bianchini et al.3 argued that multiple tools are able to accurately detect and diagnose malingering

if the limitation of false positive error rates is prioritized over sensitivity. Thus, controversy exists about

whether the tools currently used to detect malingering pain are reliable or valid.

The value of validating or discrediting a patient’s subjective report is expressed throughout the

scientific literature.2,3,4 Clinicians and researchers alike have been seeking methods of accurately

identifying “malingerers” as it is argued that the presence of financial incentives may adversely influence

outcome through incentivized disability.5 For the clinician, providing unnecessary treatments can

potentially harm the patient and take time away from those who have legitimate conditions. The issue of

malingered pain is also important among personal injury claimants who may experience depressed

emotional, physical and financial wellbeing if labelled a ‘malingerer’. Even if conservative estimates of

the prevalence of malingered pain are accurate, it still represents a considerable burden to health care.2

There is a clear need to use only those tools that can accurately and consistently detect

malingered painThe aims of this review are to survey the literature available on tools that can be used by

physiotherapists to detect malingered pain and todescribe the current state of knowledge regarding their

clinimetric properties.

METHODS

August 2011

4

We searched the literature for the most recent review articles between 2000 and 2011. Four

international databases (Pubmed/MEDLINE, CINAHL, PsychINFO, and SCOPUS) were searched using

the following terms: malinger*, exaggerat*, amplif*, and pain, with results limited to review articles or

meta-analyses. We first searched the terms malinger*, exaggerat* and amplif* using the Boolean logic

“OR”. These terms were then combined with “pain” using the Boolean logic “AND”. After results from

all of the databases were obtained and duplicate articles were eliminated, 54 review articles

remained. Two authors reviewed the abstracts and excluded articles that did not meet our inclusion

criteria. Studies were included if: (1) they were in the English language, (2) the tool described in the study

could be used to detect malingering and (3) the tool could be practically implemented in a physiotherapy

clinical setting. This resulted in 24 review articles. The articles were then divided between the five

authors and through group discussion, a consensus was reached on the tests that were relevant in the

physiotherapy setting. A secondary search was performed where each of the tests identified from the

primary search was searched on Pubmed/MEDLINE using the following search template: (malinger* OR

exaggerat* OR amplif*) AND (test name “OR” alternative names). Relevant articles that met our

inclusion criteria were then selected for our study. Finally, we conducted a secondary search of the

reference lists of the included articles where more information on properties of the tools described in the

review was required to form a valid opinion on its usefulness.

RESULTS

Seven tools were chosen from the 24 review articles selected: Pain Patient Profile (P3), Modified

Somatic Perception Questionnaire (MSPQ), Symptom Checklist-90-Revised (SCL-90-R), Minnesota

Multiphasic Personality Inventory 2 (MMPI-2), Symptom Validity Tests (SVT), Waddell’s Signs, and

Manual Muscle Testing (MMT). The Appendix describes each of the tests in more detail. The following

information was gathered regarding each tool: original author, construct measured, its application,

number of items, special requirements, time required to administer, discriminative accuracy (where

reported), and cost of the tool (Table 1). The results from these studies including clinimetric properties are

presented in Table 2.

August 2011

5

P3, MSPQ, SCL-90-R

The P3 is a 44-item self-report questionnaire that is intended to capture depression, anxiety, and

somatization associated with pain. It includes a validity scale with 5 items.6 The P3 was developed using

data from pain patients and community samples by Tollison et al.6 The administration of the tool is

simple, but the scoring algorithms are complex and requires computerized software to accurately score.6

The SCL-90-R is a 90-item self-report checklist and is much like the P3 in its administration and scoring

algorithm, which also requires computerized scoring for accuracy.7 This tool is designed to capture

psychological problems and genuine pain patterns.7 Derogatis used data from adult non-patients, adult

psychiatric outpatients, adult psychiatric inpatients, and adolescent non-patients to identify ‘normal’ and

abnormal’ pain patterns.7 The MSPQ is a 13-item self-report tool that measures somatic complaints in

patients.8 It was developed using data from a sample of patients with chronic backache and community

samples.8 The tool can be easily administered and scored by hand.8

Upon examination of the current literature, deficiencies were identified in the methods of

validation employed for the P3. The choice of a cut-score of 11 points to indicate that a patient is

exaggerating symptoms appears to have been an arbitrary decision. The research indicates that this score

may lead to at least 9% false-positives9. Respondents that score below 11 on the validity scale may be

intentionally exaggerating pain symptoms, indicating that the validity scale of the P3 may not be reliable

in capturing all types of malingering (Table 2).9,10

Similar issues to the P3 were encountered when examining the MSPQ and SCL-90-R as they

have been studied primarily under the assumption that “true” patients with pain exhibit a particular profile

and patients who malinger will deviate from this profile.11,12,13 However, these assumptions have not been

verified in well-designed research studies. Since the profiles themselves may not be accurate, this can

lead to problems in subsequent studies that use these profiles to compare patients with chronic pain to

simulated malingerers.14 14

MMPI-2

August 2011

6

The MMPI-2 is a 567-item self-report measure for adult personality and psychopathology.15 The

original MMPI was developed by Hathaway and McKinley,15 and later revised by Butcher to become the

MMPI-2.16 Normative and psychiatric inpatient group data of various psychiatric disorders has been

collected since the 1940’s in the assessment of response styles and detection of malingered psychological

symptoms. This information is used to assess the degree to which the respondent’s score resembles the

typical scores represented in the normative data for that population.17 Administrating the test is time-

consuming due to the high number of itemsand scoring can be done by hand or usingcomputer software.

Among the 126 scales available in the MMPI-2, 9 validity scales are used to assist the clinician in

identifying intentional exaggeration of symptoms, including the F, FB, F(p) and Fake Bad Scale (FBS).1718

The F or infrequency scale is a 64 item validity scale used to detect attempts at “faking good” or “faking

bad”.15 Individuals who score high on this test are thought to be exaggerating their responses by either

trying to appear better or worse than they actually are.19 The questions within this scale are designed to

determine whether respondents are contradicting themselves in their responses. The FB or Back F scale is

another validity scale containing 40 items used to detect inconsistent responses.18 This scale is analogous

with the F scale except these items are placed later in the test booklet, where deviant responses are

assessed. The F(p) or the infrequency-psychopathology scale is a 27-item validity scale used to detect

malingering of psychopathological symptoms.17,20 This tool was developed to take into account the

elevated rates of psychopathology among psychiatric inpatients. Elevated scores on this scale suggest that

the test-taker is feigning psychopathology as compared to what is commonly found amongst individuals

in inpatient psychiatric facilities.20 Given the significant comorbidity of psychiatric illness to patients

with chronic pain conditions, significant elevations on these scales are thought to point towards a marked

distortion in self-report.4 The FBSis a symptom validity scale containing 43 items that was developed

from a subset of MMPI-2 items.4 This scale is used to identify potentially exaggerated claims of disability

or exaggeration of illness. Depending on the cut-off score used, an elevated score suggests intentionally

exaggerated pain.

August 2011

7

Current literature shows that the MMPI’s infrequency scales (F, F(p), FB) can accurately identify

psychiatric patients who are either generally exaggerating existing psychopathology or feigning a

particular psychiatric condition, such as depression or schizophrenia;4 however, detecting pain

malingering has been less promising. This is due to the heavy reliance on self-reporting and the lack of

scales available on the MMPI-2 to assess the exaggeration of somatic complaints4 aside from the FBS4.

Developers of the FBS suggest that raw scores above 22 should raise concerns about the validity of self-

reported symptoms, especially with individuals who have been cleared from physical injury or medical

problems.22 Inconsistency in the use of cut scores exists between authors16and 16there is little

psychometric information available on the FBS. Many of the studies that investigated the tool’s properties

were limited by methodological problems including small sample size, unrepresentative samples drawn

by the author, and the lack of cross-validation with more general psychiatric and normative groups.4,22,23

Symptom Validity Tests

In general, aSVT involves presenting an individual with a stimulus and then prompting them to

select the correct answer from a fixed number of options (usually two). The number of items varies

depending on the specific test used. SVTs were intended to detectmalingered sensory-perceptual

deficits24-26 but became more commonly used in testing for feigned memory impairment.27 SVTs were

developed using data from neuropsychological patients by Haughton et al.26 and are easy to administer.

For example, the Portland Digit Recognition Test(PDRT) involves orally presenting patients strings of 5

digits and assessing their ability to recognize them visually.

While commonplace in neuropsychological examinations of patients who have sustained a head injury,

symptom validity tests are not in widespread use among the pain population.31 However, it is argued that

since cognitive impairment is often a component of pain, it is important to address the veracity of

cognitive deficits in pain patients seeking compensation.28,30 Bianchini et al.3 proposed criteria for the

diagnosis of malingered pain-related disability. According to this classification system, a score

statistically below 50% on a forced-choice SVT is regarded as evidence of definite malingering. Several

studies have employed the criteria of Bianchini and colleagues to validate the ability of various SVT’s to

August 2011

8

detect malingering in patients with pain.28,30,32 Greve et al.30 evaluated the accuracy of the PDRT in

detecting malingered pain-related disability. Depending on the cut-off score employed by the PDRT it

was reported to detect between 33% and 60% of “definite malingering” patients while specificity values

ranged from 94% to 97% suggesting a higher number of false negatives but few false positives.30In SVT

validation studies among people with pain, the gold standard used is a statistical score below 50% on

another SVT. For instance, Greve et al categorized patients as definite malingerers if they had a

statistically significant sub-50% score on the Test of Memory Malingering (TOMM).32 Investigations of

the ability of TOMM itself to detect malingered pain-related disability rely on a statistically negative

response bias on the PDRT as the gold standard.32 This circular validation among studies investigating

SVT for malingering diagnosis is common in the literature30,32,33. While these studies demonstrate a strong

correlation between SVT’s, and has established the concurrent validity of these measures, the lack of a

clear gold standard undermines their content validity.

Waddell Signs

Waddell’s Signs are eight physical signs divided into five categories that are intended as a screen

for further psychological evaluation, and predict poor prognosis with treatment.34-36 These signs were

developed by Waddell et al.36 using a sample of Canadian and British patients with chronic back pain,

worker’s compensation claims and a history of failed treatment. Waddell Signs are quick and easy to

assess during a regular physical examination. If an individual scores positive for a single sign, then they

are positive for that category. If three or more of the five categories are positive, then the result is

clinically significant. Isolated positive signs are disregarded.34-35

Two review articles by Fishbain et al.34,37 relating to Waddell’s Signs emerged from the primary

search and each article found no association between malingered pain and Waddell’s Signs. The definition

of malingering proposed by Fishbian et al.37 was generally comprised of the following four factors: 1)

being a patient on worker’s compensation and/or being in active litigation; 2) no improvement of

Waddell’s Signs with treatment; 3) performance on paper-pencil tests (i.e. MMPI) indicating that

performance may be affected by secondary gain issues; and 4) physician dishonesty perception. This

August 2011

9

definition of malingered pain is potentially flawed since the relationship between malingering and the

above concepts has yet to be concretely established. Therefore, the use of a non-validated definition of the

target state makes it difficult to determine whether Waddell’s Signs can detect malingering.

Since the publication of Waddell’s original article in 1980, researchers and clinicians have

misinterpreted and misused Waddell’s Signs to identify malingered pain.35,38 Contrary to popular belief,

Fishbain et al.34 found that Waddell’s Signs are in fact an organic phenomenon and cannot be used to

differentiate organic from nonorganic causes. Waddell’s Signs should be more appropriately called “pain

behaviours” since they can be explained by the neurophysiology of pain34,35 .For example, “superficial

tenderness” could be due to a patient’s low pain tolerance or the presence of allodynia, a sensitized central

nervous system when exposed to prolonged pain35. Therefore, it is important for physiotherapists to

recognize that Waddell’s signs should not be used to detect malingering. Rather they are best used to

screen patients requiring further psychological assessment, and they also indicate risk of a poor response

to either conservative treatment or surgery.34, 36, 39

Manual Muscle Testing

MMT has been proposed as a tool for detection of intentional pain exaggeration through the

measure of sincerity or consistency of effort with repeated testing of maximal isometric contraction,

commonly referred to as Coefficient of Variation (CV).2,40-45 The use of CV is based on the assumption

that intentional submaximal effort shows greater variability (higher CV) than maximal effort.46 As

proposed by the motor recruitment model, repeated maximal contractions require the simplest motor

control and are therefore easily reproducible while submaximal efforts require the coordination of higher

order motor programming, proprioceptive feedback, and fine motor corrections.46

Clinicians are first required to measure the force output of each contraction using a strength

testing instrument such as a dynamometer.41 CV is then calculated by dividing the standard deviation of

three or more trials by their mean and multiplying by 100 to obtain a unit-less percentage.44 An

individual with a CV above an established cut-off score, whose efforts are considered inconsistent enough

to be labelled submaximal, is interpreted to be exaggerating their pain behaviours.42-44

August 2011

10

The current literature regarding the effectiveness of Coefficient of Variation (CV) to detect

malingered pain is contradictory at best. In fact, authors of review articles and a recent meta-analysis do

not recommend its use in clinical settings.2,40-45 The controversy in the research is likely due to differences

in methods among studies. Utilization of different strength tasks (i.e. grip strength, elbow flexion, knee

extension, trunk flexion and extension, and lifting) by authors is one difference in methods

observed.44 Little agreement between authors as suggested by the broad spectrum of CV cut-off values

(i.e. 7.5% to 20%) may also explain the wide range of sensitivity and specificity values reported (Table

2).46 However, even when factoring the same parameters and type of strength test, studies have yet to find

a specific confidence interval that produces a combination of sensitivity and specificity values adequate

for clinical practice.41,42,44

Investigators have so far reported low test-retest reliability or stability of CVs that are based on 3-

5 repetition.42 Shectman found test-retest reliability to range from 0.02-0.41 for maximal efforts and

0.03-0.64 for submaximal44 with 5 repetitions being more stable than 3.47 While increasing the number of

strength trials in research may result in greater stability of CVs,42-48 administering more than 5 repetitions

may limit its feasibility in clinical settings given time constraints and the potential effects of fatigue.49

Based on the statistical principle for CV, it can only be theoretically useful if the mean and SD

increases proportionally. Thus, for this mathematical model to apply, a larger mean (maximal effort)

should yield greater absolute variability (SD) than a smaller mean (submaximal effort).49 However, this

principle contradicts the muscle recruitment model where submaximal effort is instead expected to have

greater variability (SD).43 Concerns regarding this issue have been expressed by several authors.

Shectman44,49 demonstrated little difference in SD between effort levels. In addition, Fairfax et al. 51 and

Bohannon52 both found negative correlations between mean strength and CV. These results suggest the

possibility that the increase in relative variability or CV in submaximal efforts is due to a decrease in

mean torque rather than a true increase in absolute variability (SD).44,49 Given the inherent bias of CV,

producing inflated values during submaximal efforts, clinicians and researchers should be cautious when

interpreting results.

August 2011

11

Even if a cut-off value can be established and CV is proved to be a valid tool, inconsistency may

not necessarily suggest malingering. For instance, there are many extraneous variables that could affect

muscle testing that have not been accounted for in the research to date. Robinson et al42 describes factors

such as fear of pain, injury or re-injury, anxiety, depression, anger, work satisfaction, motivation,

medication consumption, and even actual pain itself that may contribute to variability in performance.42

Therefore, true strength may not be accurately captured or appreciated using this tool and the ability to

extrapolate the results from research to clinical settings is questionable. The lack of empirical support for

CV and inconsistent methods raises questions of whether MMT can be used independently to determine

sincerity of effort especially given the potential for psychological or physical harm to the patient if an

inappropriate diagnosis is made.

DISCUSSION

We have reported the results of a scoping survey of existing literature describing tools that have

been investigated for the usefulness in detecting intentionally malingered pain. We have limited the

search to tools that could reasonably be performed in a standard rehabilitation clinic that is not outfitted

with advanced laboratory-based equipment. This is an important consideration considering that many

chronic pain conditions exist within the context of a litigious environment, such as post-motor vehicle

accident or work-related injuries, in which third party funders provide compensation for injured clients.

In an era where many people with chronic pain are forced to prove the validity of their symptoms often in

the absence of hard objective data, an understanding of how well clinicians are able to accurately

discriminate between legitimate and exaggerated complaints seems particularly relevant.

Several consistent deficiencies have emerged throughout this review; two important concerns

include the lack of a consensus definition and that of a reliable gold standard. Many studies commonly

use simulated patients as their gold standard,4,,6, 9-12, 14, 20, 42,44 requiring healthy individuals to feign a pain

provoking injury. Extrapolating data from studies that use this type of gold standard may be limited due

to difficulties in capturing inherent influential motivators (ex. monetary gain and compensations) of

individuals who are truly malingering. Unfortunately, resolution of this methodological issue can be

August 2011

12

challenging as known malingerers, assuming they could be identified, may not be willing to participate or

volunteer information to researchers.

The MSPQ, SVT’s and Waddell Signs use cross validation against other tools as their gold

standard, which can be potentially problematic.11,34,37,30,32,34 For example, Waddell’s Signs have been

validated against the MMPI as a gold standard.34,36-37 Another example is the circular validation used in

SVT’s, 30,32-33 where the PDRT has been validated against the TOMM as a gold standard, which itself is

validated against the PDRT as a gold standard. Although these studies may demonstrate a correlation

between tests that identify malingering pain and establish the concurrent validity of these measures, the

lack of a clear gold standard undermines their content validity.

Our review revealed other inconsistencies such as the lack of established normative data for pain

for the P3, SCL-90-R, MMPI-2, and MSPQ; however, many research articles claim that patients who are

malingering pain will deviate from a pattern that such patients would ‘normally’ exhibit.4,11-14 This

‘normative’ data has been established from studies that included mixed samples, with various severities,

locations, and causes of pain. Given the variety of influences on any individual's pain experience,56 it

would not be abnormal to see a patient that deviates from a normal presentation or clinical pattern. This

raises the spectre that clinicians could falsely conclude that a patient is exaggerating their pain symptoms

because they do not fit a poorly defined normal presentation for pain. It is challenging to accept this data

since there is no clearly defined ‘normal’ presentation of a patient in pain as it is a subjective experience

that is difficult to objectify.

Many researchers have attempted to identify cut-off scores in the tools that they used to detect

malingered pain in the P3.4,9,32,44 The cut-off values that have been proposed are problematic as there is

little consistency in the values chosen. Cut scores haveyet to be validated in well-designed studies;

therefore it is unknown if the scores capture an acceptable number of true positives and/or true negatives.

Clinicians and researchers that use these values need to be cautious as the implications for falsely

identifying patient as a malingerer could have adverse consequences for a patient who is truly

experiencing pain.

August 2011

13

There are many variables that affect the results of a physical examination type test such as

Waddell’s Signs or MMT. Patient and therapist factors could play a large role in the outcome of this type

of test. For example, it is possible that the therapist’s own biases, their perception that the patient is

malingering, may affect the results of the physical assessment. In addition, patient factors such as

perceptions of pain, fear-avoidance beliefs, coping strategies and many others can influence the patient’s

response to the test, but may well be legitimate concerns. Therefore, it is important to remember that

any physical examination tool is limited by both patient and examiner factors.

LIMITATIONS

We specifically targeted tools that were deemed to be applicable to the clinical setting through

consensus agreement. The factors that influenced this decision were; cost, space required, licensing fees,

requirements for special training, and testing procedures that fell at least marginally within the scope of

physiotherapy practice. It is possible that there are other tools that exist that have yet to be the subject of

a review, which would not have been captured with our search strategy.

We chose reviews as our primary search as such a strategy provided at least some confidence that

the tool had been evaluated more than once. It is difficult to make recommendations about clinical

practice on the basis of only a single research study. The reviews were also deemed to be the best ways to

gather information on the tools that are available in an efficient manner. Once it became clear that none

of the tests we found were going to be confidently endorsed as a valid tool for identifying malingered

pain, the incentive to score the quality of the reviews was reduced and we opted instead to focus on

describing the tests and the current state of the literature pertaining to them, as more of a 'survey of the

landscape' rather than a formalized review of reviews. Of note, some of the included review articles had

known methodological issues that were not addressed; therefore, readers should keep this in mind when

considering the results of this review. However, we believe this exercise has provided a reasonably

accurate overview of the tools that are available, and we have made several suggestions as to how future

research in this area could be conducted to improve confidence in results.

August 2011

14

Notably future research needs to focus on creating a testable consensus definition of malingered

pain and its operationalization. Most definitions are vague and thus difficult to test experimentally;

therefore, a clear definition is vital for future research. A gold standard of malingered pain needs to be

recognized for any evaluation of discriminatory validity. However, we recognize the difficultly as patients

who are known malingerers would have to be a) identified and b) willing to volunteer for research.

Finally, researchers need to consider whether further investigation of tools to detect malingered pain is

necessary, as pain itself is a subjective phenomenon and currently cannot be objectively visualized due to

the multitude of factors that can influence a patient’s pain experience.

CONCLUSIONS

Despite previous estimates on the prevalence of malingering, which may or may not be accurate,

the general consensus among researchers, clinicians, and funders is that intentional malingering does

occur and remains an important issue. The reviewed studies suggest that there are currently no reliable or

consistently valid methods to identify malingered pain useable in routine physiotherapy clinical practice.

In light of the paucity of evidence to support a single tool’s ability to detect malingering, we contest that

malingered pain cannot be confidently identified. Therefore, clinicians should be cautious using methods

that claim to be valid indicators of malingered pain. As mentioned by Eisendrath54, unless there is clear

evidence that a person is malingering, the subjective reporting of pain should be regarded as the truth. In

fact, the therapist cannot really be certain that a patient who confesses to malingering is in fact even

truthful themselves. Thus, rather than reducing the clinical decision to a simple dichotomy of malingered

vs nonmalingered, clinicians who believe that intentional exaggeration of pain is interfering with

treatment progression should instead consider all of the factors that surround the patient’s expression of

pain (i.e. perception of pain, coping strategies, motivation for malingering, fear, self-efficacy etc.) and

attempt to understand the individual's motivations for such behaviour. Based on this assessment,

therapists are encouraged to consider a multidisciplinary approach to the treatment of pain that

encompasses the whole person including their values, beliefs and goals.

KEY MESSAGES

August 2011

15

What is already known on this topic

Whilst most experts will agree that intentional exaggeration of pain symptoms for external incentives

occurs, especially within the population with chronic non-malignant pain, little is known regarding the

ability of physiotherapists or other non-psychiatrists to identify such exaggerated behavior. Several

clinical tools intended to screen for intentionally malingered pain currently exist, but their application to

routine clinical practice is unknown.

What this study adds

Through a narrative review and consensus process, we have identified 7 different methods that have been

proposed as useful clinical screening tools for identifying intentionally malingered pain. In a field where

reliability and validity are especially important, consistent deficiencies in the definition and

operationalization of malingering, and questionable approaches to establishing clinimetrics, mean that no

tool can currently be promoted as consistently accurate.

August 2011

16

REFERENCES

1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-

IV, 4th ed. Washington, DC: American Psychiatric Association; 1994.

2. Fishbain DA, Cutler R, RS Rosomoff, HL Rosomoff. Chronic pain disability

exaggeration/malingering and submaximal effort research. Clin J Pain. 1999:15;244-274.

3. Bianchini KJ, Greve KW, Glylarrann G. On the diagnosis of malingered pain-related disability:

lessons from cognitive malingering research. Spine J. 2005: 404-417.

4. Arbisi PA, Butcher JN. Psychometric perspectives on detection of malingering of pain: use of the

Minnesota Multiphasic Personality Inventory-2. Clin J Pain. 2004;20(6):383-391.

5. Rohling ML, Binder LM, Langhinrichsen-Rohling J. A Meta-Analytic Review of the Association

Between Financial Compensation and the Experience and Treatment of Chronic Pain. Health

Psychol. 1995:537-547.

6. Tollison CD, Langley JC. Pearson: Pain Patient Profile. http://www.pearsonassessments.com.

Updated on June 10, 2011. Accessed on June 10, 2011.

7. Derogatis LR. SCL-90-R administration, scoring, and procedures manual II. 1983. Towson, MD:

Clinical Psychometric Research.

8. Main CJ. The Modified Somatic Perception Questionnaire (MSPQ). J Psychosomatic Research.

1983; 27:503-514.

9. McGuire BE, Shores EA. Pain Patient Profile and the Assessment of Malingered Pain. J Clin

Psychol. 2001;57(3):401-409.

10. McGuire BE, Harvey AG, Shores EA. Simulated malingering in pain patients: A study with the

Pain Patient Profile. Br J Clin Psychol. 2001;40:71-79.

11. Larrabee GJ. Exaggerated Pain Report in Litigants With Malingered Neurocognitive

Dysfunction. Clin Neuropsychol. 2003;17(3):395-401.

12. Torres X. The Symptom Checklist-Revised (SCL-90-R) is Able to Discriminate Between

Simulation and Fibromyalgia. J Clin Psychol. 2010;66(7):774-790.

13. Wallis BJ, Bogduka N. Faking a profile: can naive subjects simulate whiplash responses? Pain.

1996;66(2-3):223-227.

14. McGuire BE, Shores EA. Simulated Pain on the Symptom Checklist 90-Revised. J Clin Psychol.

2001;57(12):1589-1596.

15. Hathaway SR & McKinley JC. The Minnesota Multiphasic Personality Inventory Manual.

Minneapolis: University of Minnesota Press. 1943.

16. Butcher JN, Arbisi PA, Atlis MM, McNulty JL. The Construct Validity of the Lees-Haley Fake

Bad Scale: Does this Scale Measure Somatic Malingering and Feigned Emotional Distress? Arch

Clin Neuropsychol. 2003;18:473-485

17. Arbisi PA, Ben-Porath YS. An MMPI-2 infrequent Scale for use with Psychopathological

Populations: The Infrequency-Psychopathology Scale, F(p). Psychol Assessment.

1995;7:424-431.

18. Hersen M. Comprehensive Handbook of Psychological Assessment: Personality Assessment.

Volume 2. New Jersey: Wiley and Sons; 2004.

19. McGrath RE, Pogge DL, Stein LAR, Graham JR, Zaccario M, Teresa Piacentini. Development of

an Infrequency-Psychopathology Scale for the MMPI-A: The Fp-A Scale. J Pers Assess.

2000;74(2):282-295.

20. Rothke SE, Friedman AF, Jaffe AM, Greene RL, Wetter MW, Cole P and Baker K. Normative

Data for the F(p) Scale of the MMPI-2: Implications for Clinical and Forensic Assessment of

Malingering. Psychol Assessment. 2000;12(3):335-340.

21. Yossef S, Porath B and Tellegen A. MMPI-2; FBS (Symptom Validity). Retrieved from:

August 2011

17

http://psychcorp.pearsonassessments.com/NR/rdonlyres/A25DB8F8-435F-4066-801B-

B641978A97DA/0/MMPI2FBS.pdf. Updated on June 22, 2011. Accessed on June 22, 2011.

22. Lees-Haley PR. Efficacy of MMPI-2 Validity Scale and MCMI-II Modifier Scales for Detecting

Spurious PTSD claims: F, F-K, Fake Bad Scale, Ego Strength, Subtle–Obvious subscales, DIS,

and DEB. J Clin Psychol. 1992;48(5):681–689.

23. Fox DD, Gerson A, Lees-Haley PR. Interrelationship of MMPI-2 Validity Scales in Personal

Injury Claims. J Clin Psychol. 1995:51(1):42–47.

24. Brady JP, Lind DL. Experimental analysis of hysterical blindness. Arch Gen Psychiatry. 1961;

4:331-9.

25. Grosz HJ, Zimmerman J. Experimental analysis of hysterical blindness. Arch Gen Psychiatry.

1965;13:255-60.

26. Haughton PM, Lewsley A, Wilson M, Williams RG. A forced-choice procedure to detect feigned

or exaggerated hearing loss. Br J Audiol. 1979;13:135-8.

27. Bianchini KB, Mathias CW, Greve KW, Houston RJ, Crouch JA. Classification accuracy of the

Portland Digit Recognition Test in traumatic brain injury. Clin Neuropsychol. 2001;15:461-470.

28. Greve KW, Ord J, Curtis KL, Bianchini KJ, Brennan A. Detecting malingering in

traumatic brain injury and chronic pain: a comparison of three forced-choice validity tests. Clin

Neuropsychol. 2008;22:896-918.

29. Greve KW, Binder LM, Bianchini KJ. Rates of below-changes performance in forced-

choice symptom validity test. Clin Neuropsychol. 2009b;23:534-544.

30. Greve KW, Bianchini KJ, Etherton JL, Ord JS, Curtis KL. Detecting malingered pain-

related disability: classification accuracy of the Portland digit recognition test. Clin

Neuropsychol. 2009a;23:850-869

31. Bush SS, Ruff RM, Troster AI, Barth JT, Koffler SP, Pliskin NH. Symptom validity

assessment: Practice issues and medical necessity NAN policy & planning. 2005.

32. Greve KW, Etherton JL, Ord J, Bianchini KJ, Curtis KL. Detecting malingered pain-

related disability: classification accuracy of the test of memory malingering. Clin Neuropsychol.

2009c;23:1250-1271.

33. Greve KW, Bianchini KJ. Classification accuracy of the Portland Digit Recognition Test

in traumatic brain injury: Results of a known-groups analysis. Clin Neuropsychol. 2006; 20:816–

830.

34. Fishbain DA, Cole B, Cutler RB, Lewis J, Rosomoff HL, Rosomoff RS. A Strutured Evidence-

Based Review on the Meaning of Nonorganic Physical Signs: Waddell Signs. Pain Med.

2003;4(2):141-181.

35. Ranney D. A Proposed Neuroanatomical Basis of Waddell’s Nonorganic Signs. Am J Phys Med

Rehabil. 2010;89(12):1036-1042.

36. Waddell G, McCulloch JA, Kummel E, Venner RM. Nonorganic Physical Signs in Low-Back

Pain. Spine. 1980;5(2):117-125.

37. Fishbain DA, Cutler RB, Rosomoff HL, Rosomoff RB. Is There a Relationship Between

Nonorganic Physical Findings (Waddell Signs) and Secondary Gain/Malingering? Clin J Pain.

2004;20(6):399-408.

38. Mendelson G, Mendelson D. Malingering pain in the medicolegal context. Clin J Pain. 2004;20:

423-432.

39. Main CJ, Waddell G. Behavioural Responses to Examination: A Reappraisal of the Interpretation

of “Nonorganic Signs”. Spine. 1998;23(21):2367-2371.

40. Baker JC. Burden of proof in detection of submaximal effort. Work. 1998:10;63-70.

41. Lechner DE, Bradbury SF, Bradley LA. Detecting sincerity of effort: a summary of methods and

approaches. Phys Ther. 1998;78:867–868.

42. Robinson ME, Dannecker, EA. Critical Issues in the Use of Muscle Testing for Determination of

Sincerity of Effort. Clin J Pain. 2004;20(6):392-398.

August 2011

18

43. Shectman O. Using the coefficient of variation to detect sincerity of effort of grip strength: a

literature review. J Hand Ther. 2000;13:25-32.

44. Shechtman O, Anton SD, Kanasky WF, Robinson ME. The use of the coefficient of variation in

detecting sincerity of effort: A meta-analysis. Work. 2006;26:335-341.

45. King PM. Analysis of approaches to detection of sincerity of effort through grip strength

measurement. Work. 1998:10;9-13.

46. Kroemer KHE, Marras WS. Towards an objective assessment of the “maximal voluntary

contraction” component in routine muscle strength measurements. Eur J Appl Physiol Occup

Physiol. 1980;45:1-9.

47. Robinson ME, Sadler IJ, O’Conner PD, et al. Detection of submaximal effort and assessment of

stability of the coefficient of variation. J Occup Rehab. 1997;7:207-215.

48. Robinson ME, Giesser ME, Hanson CS, O’Connor PD. Detecting submaximal efforts in grip

strength testing with coefficient of variation. J Occup Rehab. 1993;3(1):45-50.

49. Shechtman, O. The coefficient of variation as a measure of sincerity of effort of grip strength,

Part I: The statistical principle. J Hand Ther. 2001;14:180-187.

50. Shectman O. Is the coefficient of variation a valid measure for detecting sincerity of effort of grip

strength?. Work. 1999;13(2):163-169.

51. Fairfax AH, Balnave R, Adams RD. Variability of grip strength during isometric contraction.

Ergonomics. 1995;38:1819-30.

52. R.W. Bohannon. Differentiation of maximal from submaximal static elbow flexor efforts by

measurement variability. Am J Phys Med. 1987;66:213-218.

53. Simonsen JC. Coefficient of variation as a measure of subject effort. Arch Phys Med

Rehabil.1995:76:516-520.

54. Eisendrath SJ. Chronic Pain Mechanisms and Management. Neurology. 1995;45(9): S26-S34.

55. Derogatis LR. Pearson: Symptom Cheacklist-90-Revised. http://www.pearsonassessments.com.

Updated on June 10, 2011. Accessed on June 10, 2011.

56. Melzack R. From the gate to the neuromatrix. Pain. 1999;Suppl 6:S121-S126.

August 2011

19

APPENDIX

Pain Patient Profile6

Administer To Individuals 17–76 years old

Reading Level 8th grade

Items 44 groups of statements with three statements per group

Formats Paper-and-pencil or computer administration

Report Options Interpretive Report, Progress Report

Scoring Options Q™ Local Software

Hand Scoring

Mail-in Scoring Service

Fax-in Service

PAD (Patient Assessment Device) Hand-held Electronic Device

Scales Somatization, Depression, Anxiety and Validity Index

Norms Pain Patients and Community Samples

Modified Somatic Perception Questionnaire8

Administer To Adults

Items 13 item self-report scale, 0-3 or 5pt scale if add moderate

Formats Paper-and-pencil

Report Options Interpretive Report, Progress Report

Scoring Options Hand Scoring

Scales Somatic complaints

Norms Chronic Backache Patients and Community Samples

Symptom Checklist-90-Revised55

Administer To Individuals 13 years and older

Reading Level 6th grade

Items 90 items, 5-point rating scale

Formats Paper-and-pencil, audiocassette, or computer administration

Report Options Interpretive, Profile, and Progress

Scoring Options Q Local™ Software

Mail-in Scoring Service

Hand Scoring

Scales 9 Primary Symptom Dimensions

3 Global Indices Global Severity Index, GSI; Positive Symptom Total, PST

Norms Adult nonpatients, Adult psychiatric outpatients, Adult psychiatric inpatients, Adolescent

nonpatients

August 2011

20

Minnesota Multiphasic Personality Inventory-2 (MMPI-2) 4

Administer 18 years of age and older

Reading Level 5th Grade

Items 567 True/False items

Format Paper and pencil, CD, or computer

Report Options - Extended Score Report

- The Minnesota Report™: Adult Clinical System-Revised Interpretive

Report

- The Minnesota Report: Revised Personnel System, 3rd Edition

Interpretive Report

- The Minnesota Report: Revised Personnel System, 3rd Edition

Adjustment Rating Report

- The Minnesota Report: Interpretive Reports for Forensic Settings

Scoring Options Q™ Local Software

Mail-in Scoring Service

Hand Scoring

Scales 9 Validity Scales

5 Superlative Self-Presentation Subscales

10 Clinical Scales

9 Restructured Clinical (RC) Scales

15 Content Scales

27 Content Component Scales

20 Supplementary Scales

31 Clinical Subscales (Harris-Lingoes and Social Introversion Subscales)

Various special or setting-specific indices

Norms Nationwide adult community sample consisting of 1,138 males and 1,462 females from

various areas of the United States ranging from 18-80 years of age.

Symptom Validity Tests: Portland Digit Recognition Test (PDRT)30

Purpose The Portland Digit Recognition Test (PDRT) is designed for the neurological assessment of

exaggeration and malingering.

Population The test has been developed to evaluate adult individuals.

Type of

Administration

The test is administered to individual client.

Test Description The test requires 36 cards, each with two five-digit numbers printed on them, one above the

other. During the test administration, four sets of 18 trials are implemented. The client counts

backward from a specified integer for a specified amount of time.

Scoring The scores are of three types, those for the Easy items or trials (Sets 1 and 2), for the Hard

items or trials (Sets 3 and 4), and the total number correct.

August 2011

21

Waddell Signs 34,36

Purpose A quick and easy screen to identify patients who necessitate a more thorough psychosocial

assessment and predict those with poor outcome with conservative and surgical treatment.

Items Standardized set of 8 signs divided into 5 categories of “behavioural responses to

examination”.41

Scoring If an individual scores positive for a single sign, then they are positive for that category. If

three or more of the five categories are positive, then the result is clinically significant.

Isolated positive signs are disregarded.

Nonorganic Signs 1. Tenderness

Superficial

Nonanatomic

2. Simulation Tests

Axial Loading

Rotation

3. Distraction Test

Straight Leg Raise

4. Regional Disturbances

Weakness

Altered sensation

5. Overreaction

Manual Muscle Test 44

Description Measures variability of repeated isometric strength testing using a dynamometer.

Coefficient of

Variation

Calculated by dividing the standard deviation (SD) of three or more consecutive trials by their

mean and multiplying by 100.

CV %= [SD/Mean]*100

Repetitions 3-5 per trial

Interpretation of

results

Submaximal effort is associated with greater variability in performance. A larger CV value

thus equates to greater variability and small consistency between repeated trials. CV is

compared to a cut-off value and determines if efforts are inconsistent enough to be labeled

submaximal and insincere