Faking pain – Is there a test for it?

One of the weird things about pain is that no-one knows if you’re faking. To date there hasn’t been a test that can tell whether you’re really in pain, or just faking it. Well, that’s about to change according to researchers in Israel and Canada.

While there have been a whole range of approaches to checking out faking such as facial expression, responses to questionnaires, physical testing and physical examinations, none of these have been without serious criticism. And the implications are pretty important to the person being tested – if you’re sincere, but someone says you’re not, how on earth do you prove that you’re really in pain? For clinicians, the problem is very troubling because allegations of faking can strain a working relationship with a person, and hardly lead to a sense of trust. Yet insurance companies routinely ask clinicians to make determinations about fraudulent access to insurance money – and worst of all, clinicians often feel they have little choice other than to participate.

In this study by Kucyi, Sheinman and Defrin, three hypotheses were tested: 1) Whether feigned performance could be detected using warmth and pain threshold measurements; 2) whether there were changes in the statistical properties of performance when participants were faking; and 3) whether an “interference” or distractor presented during testing interferes with the ability to fake and therefore provide a clue to when someone is being sincere or not.

Using university students (I hope they got course credits for participating!) who were not health science students, and were otherwise healthy, the investigators gave very little information about the procedure or hypotheses to minimise expectancy bias. Participants were then tested using a thermal stimulator to obtain “warmth sensation threshold” and “heat-pain thresholds” – this is a form of quantitative sensory testing (QST). TENS was used as a distractor in the experimental case, applied for 2 minutes before measuring the pain threshold, and during the heat pain threshold test. This was repeated with first the threshold test, then TENS. Participants were asked to pretend they were in an insurance office, being tested to establish whether they were experiencing genuine pain, after being told the test would be able to tell whether their pain was real.

What did they find out?

Well in situation one, where both threshold and warmth detection were used, and participants were asked to fake the pain intensity, respondents gave higher warmth detection ratings than normal. Not only this, but the ability to repeat the same response with the same temperature was poorer.  Heat pain threshold was also consistently different between the sincere and faked conditions, with heat pain threshold lower when people were faking (to around 3 degrees).

When the second testing option was carried out (using TENS to distract), heat pain threshold was significant lower when participants were faking, and the variance of the feigned + interference condition was three times that of the sincere condition, and the CV of the feigned + interference condition was twice that of the sincere condition.

What does this mean?

Well first of all, it means there are some consistent effects of faking in response to tests of warmth and heat-pain threshold when a distractor like TENS is used. Increased reports of warmth threshold and reduced heat pain threshold were observed, and where statistically significant. Interestingly, it was only when a distractor was used that the variability of reports were found – these authors suggest that people are pretty skilled at giving consistent reports when they’re not being distracted by an additional sensory stimulus.

Now here’s where I begin to pull this apart from a clinical and practical perspective. The authors, to give them credit, indicate that the research is both new and that it may identify some people who do have pain as malingerers. My concerns are that people with chronic pain may not look at all like healthy young university students.

We know very little about the responses to QST by people with different forms of chronic pain. We already know that people with impaired descending noxious inhibitory control respond differently to some forms of QST. We also know that contextual factors including motivation can influence how nervous systems respond to input. But my concerns are far more about the potential harm to those who are tested and found to be malingering when they’re not.

What do you do if you think a person is faking? How do you deal with this? What good does it do to suggest to someone their pain is not real, or isn’t nearly as bad as they make out? Once the words are out of your mouth (or written in a report) any chance of a therapeutic relationship has just flown right out the door. And you’re still left with a person who says they’re having trouble – but now you have an angry, resentful person who has a strong need to prove that they DO have pain.

You see, I think it might be more fruitful to ask why is this person faking pain? If it’s simply for money, surely there are far easier ways to get money than pretending to be disabled by pain? If it’s the case that a person is off out fishing or playing golf or living it up when “supposed” to be in pain, wouldn’t it make more sense to reframe their response as either recovering well (doing what’s healthy) and therefore get on with returning to work; or use a private investigator to demonstrate that he or she is actually capable of doing more than they indicate?

The presence or absence of pain is not really the problem, IMHO. To me we need to address the degree of disability that’s being attributed to pain and work on that. Maybe a greater focus on reducing disability rather than on expensive procedures to remove pain or otherwise get rid of pain is in order?

Kucyi, A., Sheinman, A., Defrin, R. (in press). Distinguishing feigned from sincere performance in psychophysical pain testing. The Journal of Pain.

“Faking” pain – and inconsistency in presentation

There is a common belief amongst some people involved in pain and pain management that a person with chronic pain should be consistent in their presentation. That is, be consistent in various settings, and consistent across various measures.  This assumes that if inconsistencies are present there must be something going on to cause suspicion about the validity of the person’s presentation.

Returning to yesterday’s post, I discussed the distinction between nociception (activation of receptors in the nervous system), pain (the experience produced by the brain once it has interpreted the nociceptive action), disability (the changes in functional performance attributed to the experience of pain), and impairment (tissue changes).  These are not the same! Today’s post refers mostly to pain and disability.

Pain is, as I keep spouting, a personal, subjective experience – you and I can’t share our pain except through our behaviour.  And even when we describe it, we can’t actually experience what it feels like to have each other’s pain. What this means is that our behaviours, or what we do, are what conveys our experience to others but only to a certain extent.

While our experience of pain remains individual, our behaviour is shaped by all the influences around us.  And just to make things more complex,  the meaning of our pain experience is shaped by what we learn and believe about pain from interactions between ourselves and all those influences that shape our behaviour. Complicated? Oh yes.

So if I grow up in a family that is generally stoic, where my complaints of pain are ignored and I see that others in my family don’t generally cry or stop doing things when they are hurt, it’s likely I’ll also be less demonstrative about my pain than another person in a family where this is not the norm.

We also learn to behave differently in different contexts. It’s common to say a few choice words when we hit our thumb with a hammer in the garage on a Saturday afternoon, but we usually wince and gasp quietly if we thump that same thumb in Church the next day!

What this means in terms of consistency is that people often do different things in different settings when they have pain – maybe when they drive from home to a clinic they’re a bit anxious about what the consultation might mean. They get out of the car and walk quite carefully into the clinic and sit down rather gingerly. They’re not sure whether they’re going to hear good news, or not. They go through the examination and consultation, and get told they’re going to try a new medication that might help. Oddly enough, as they walk back out to the car someone looking out the window might see them moving fluidly and getting into the car quite smoothly.  Suspicious minds might start to wonder if they really had “that much pain”.

Turning to assessment of pain now, it’s been said that pain assessment should incorporate a number of different dimensions. At the very least, assessment needs to consider the location of pain, the intensity and quality of pain, and some sort of measure of the interference pain has with everyday activities.  Some commentators consider pain assessment should also make a diagnosis of the type of pain disorder present (or the cause of the pain), while others also think that psychological aspects of pain need also to be included (such as pain catastrophising, measures of avoidance and measures of mood and anxiety). And of course, assessments of range of movement, strength and reflexes are also often part of a pain assessment.

Should these measures all present a “consistent” picture?

Where we hurt and what it feels like (the quality of pain) don’t directly correlate with the amount of interference a person experiences from their pain.  If I’m a pianist I might be really bothered by tingling and burning fingers, while less bothered about the same pain qualities in my feet.  If I’m really anxious about my future as a nurse, I might be really concerned about my low back pain. If I’ve developed some effective coping strategies for managing my overall body aching, I might not be particularly anxious or have low mood despite having quite intense pain.

Inconsistency between various aspects of a person’s presentation doesn’t mean they’re faking, or that their pain is not real. It does mean we need to generate some hypotheses about the relationship between the various factors that could be influencing the person’s behaviour. To decide, on the basis of our own experience (which is always limited!) that another person’s pain is not real just because we don’t understand why they are doing what they’re doing, is a judgement call we can’t make, in all honesty.  More than that, it doesn’t help work out what to do next to move the person from being disabled (or getting a benefit they don’t “deserve”) to returning to function.

Personally, determining the validity of a person’s claim to have pain is not the province of a health professional – it’s a task for administrators of a compensation scheme.  We can describe what we see, we might even make a diagnosis, but we can’t tell whether someone’s cheating the system. Bring out the investigators and be honest that health professionals can’t do what some insurers want us to.

Gatchel, R., Kishino, N., & Minotti, D. (2010). The Three Major Components of Behavior Used for Assessing Pain: Problems Faced When There Is Discordance Among the Three Psychological Injury and Law, 3 (3), 212-219 DOI: 10.1007/s12207-010-9081-0

The vexed question of determining whether pain is “real”

Every now and then the old chestnut of whether it’s possible to determine if a person’s pain is “real” arises. Usually it’s in the form of an insurer, or maybe an employer, who wants to know if the pain this person is describing can possibly be as bad as the person says because

• “they don’t look like there is anything wrong”
• “there are these inconsistencies”
• “it’s taken too long for them to recover”
• “they’re not participating in their rehab”

For the avoidance of doubt, I have not been able to identify any readily available test or technology that can reliably determine what another person really experiences.

More than that, I’m not sure that it’s helpful in clinical practice to try to determine whether someone is “faking it”, and here’s why (sit back and get comfortable, this could take a while!).

Pain is not an actual or tangible thing.  We infer that someone has pain on the basis of their behaviour.  As Gatchel, Kishino and Minotti say “It is important to distinguish pain as a neurological event (nociception) from pain perception (which is subjective in nature).”  Nociception is about activation of receptors which pass through pathways to the central nervous system. Pain is the interpretation of this activation by the brain once it determines this information needs to be noticed and acted upon.  The same nociception activity in one person may not be perceived in the same way by another.

Having identified that pain is personal, we then turn to how a person responds to this event.  Gatchel, Kishino and Minotti remind us that pain, disability (the functional limitations, or behavioural expression of our individual experience of pain), and impairment (the physiological or physical findings) are not directly correlated.  While there is overlap, it’s not unusual to find someone with significant changes on x-rays and blood tests, who reports high levels of pain – but doesn’t let that stop them doing things.  At the same time, many painful conditions fail to show on x-rays or blood tests, but the person reports high levels of pain and finds it very difficult to carry out normal activities.

Why might someone’s disability from pain be greater than another persons?

That’s the million dollar question!And why it’s so important to obtain a biopsychosocial assessment of a person who is having trouble recovering because of pain.

We do need to understand the physiological processes involved in an individual’s presentation – knowing the diagnosis can be helpful, as can knowing the structural problems that underpin disability and pain.  At the same time, the science of diagnosis is not 100% reliable, and more importantly, unless there is a clear-cut treatment that can remediate the problem, searching for a diagnosis can play an important role in iatrogenic, or “resulting from the activity of physicians”, disability.

On to why identifying whether someone is faking it is unhelpful.

People with chronic pain have usually seen about 4 or 5 (or more) clinicians by the time they come to get a diagnosis of chronic pain.  Each time they see a new clinician, their history is repeated.  Clinicians are trained to find “what’s wrong” – and fix it. We get flummoxed when the person fails to “get better”. For some clinicians, it can mean repeating a treatment in the vague hope that what didn’t work the first time might work the fifth.  For others this failure can be turned on itself and (albeit inadvertently) the person with pain is seen as having “failed”. Maybe they weren’t motivated enough, maybe they didn’t really try hard enough, maybe they just don’t want to get better, maybe they don’t really have pain.

Patients can also be going through a process of feeling as though they’ve failed, maybe they’re not really motivated – maybe this pain is “all in my head”.

I don’t think many clinicians directly tell patients that they suspect they’re not really motivated, or don’t really have this much pain – but in a situation where the patient is already doubting, is anxious about recovery, and given our human tendency to jump to conclusions, it’s very easy for the person with pain to misinterpret what is going on.

In this situation – what do you think a person is likely to do if confronted by someone as to whether they might be “faking”? Hmmmm – I might just try to demonstrate how bad it is for me.  Chance of reducing disability has just dropped. And in the face of factors that reinforce disability such as family members who might be suggesting it’s risky to do anything “because the last time you did you were sore for ages!”, or coworkers who say “you look terrible, you need to go home”, or disability payments that are greater than unemployment benefits, or payment of a mortgage insurance that will make the house completely paid off – well, it’s probably quite hard to recover.

I’ve worked in pain management for 20 years or so, and I can’t tell whether someone is faking bad. I’ve met hundreds of people who say they have chronic pain, with an enormous range of factors that are working to hold the person in to their currently disabled situation. I have yet to find any medical, psychological or functional test (or combination of tests) that can determine the validity of the person’s pain and disability. And believe me, I’ve looked.

I’ll write more about this topic shortly – taking a look at psychological measures that have been suggested as ways to detect malingering or symptom magnification. I’ll also look at functional measures that supposedly show whether someone is giving “full effort”. In the meantime, I wonder whether it might be more useful to consider that a person’s self report of pain is what it is, and for clinicians to work at identifying the factors that could be responsible for maintaining disability.

Gatchel, R., Kishino, N., & Minotti, D. (2010). The Three Major Components of Behavior Used for Assessing Pain: Problems Faced When There Is Discordance Among the Three Psychological Injury and Law, 3 (3), 212-219 DOI: 10.1007/s12207-010-9081-0

The two faces of pain assessment

This week’s focus is on assessing or evaluating or getting to understand the experience of pain.  There are many reasons to assess someone’s pain, from trying to diagnose the kind of pain they have to attempting to see if someone is ‘real’ or not!  And shades in between.  I’m going to focus on ‘genuine’ reasons to understand someone’s pain experience rather than policing whether the person is a good actor or not, because as I’ve said before, I don’t think very often about people ‘faking’, and because I think it might be more helpful to try to see the world from the person’s eyes first.

So, given that pain is a multidimensional experience, why do I say there are two faces to pain assessment? Well the two I’m thinking of are:
(1) the experience of pain itself – location, quality, intensity etc, along with the beliefs and risk factors that influence the ‘suffering’ component of the ‘onion ring’ model described by Loeser and others.
(2) the impact of pain on function – disability, sleep loss, effect on relationships, work and so on.

I’m going to leave the first aspect for a while, and spend some time looking at the impact of pain on the individual and his/her family/community, but before I do I want to consider some basic concepts relevant to pain assessment that are essential before even beginning to think about the ‘how’ of assessment!

Firstly, all pain assessment is subjective. What that means is that as outside observers, we cannot determine whether the person is actually experiencing what they say (or indicate) they are experiencing. We can’t reliably establish whether someone is or is not carrying out activities to their full capacity. We can’t accurately determine whether a certain number of pain behaviours is the ‘correct’ amount of behaviour for the amount of nociception present.

Secondly, all behaviour is subject to the effects of learning and reinforcement that are well-established in the psychological literature (and in any two-year-old child wanting sweets!). As a result, it’s difficult to tell whether someone is behaving in a certain way because of years of learning within their family context, or whether it’s more immediate social environment demand characteristics that are influencing how the person is performing. We would all probably do more for someone that was holding a gun to our heads than for someone who looked bored (and we knew the results were going to determine whether we got some money!).

Thirdly, what we will or won’t do is also influenced by what we think we can or can’t do – and what we anticipate may happen afterwards, or what we experienced last time we attempted. A recent study of functional capacity evaluations found that even for people without chronic pain, a majority of participants experienced elevated pain the day after completing an FCE, so my guess is that some people who have chronic pain and who have completed a previous FCE may modify their performance at the next one. Our beliefs, attitudes, predictions, memories and so on determine what we are prepared to do in any context. When we look at the impact of pain on function, we need to remember this and take these factors into account.

The final area to consider today is the effect of context on performance. In many ways this is a summary of the three previous areas I’ve mentioned, but bear with me as I explain.
Many people with chronic pain report that they can carry out an activity in one setting, once, but when asked to carry it out repeatedly, and in a work setting, for example, they struggle. In the context of the real world, with all the social cues, cognitive load and need to persist and maintain activity over time, performance becomes variable. This doesn’t just happen for people with persistent pain mind you – let’s think of those Olympic athletes who may have performed well ‘at home’, but in the context of The Olympic Games in China in Front of All Those People – performance may well be very different.

I’ve only mentioned a few of the confounding factors that influence how pain may impact on a person’s daily activity in life and roles. I haven’t even started to use references – but you can bet if you come back this week, there will be more!

Why can he go surfing but can’t do the vacuuming?

Social contract theory is a theory drawn from evolutionary psychology – a ‘cheater detection’ system if you like. Following on from yesterdays post about detecting faking in pain, this study examines the judgements observers (in this case, friends and relatives of people with pain) in a study where four vignettes were presented. Each vignette varied in terms of four cues: the person continuing or stopping liked tasks, continuing or stopping disliked tasks, the availability of medical evidence, and the pain intensity as rated by the person.

Many variables have been studied with respect to how accurately observers judge another person’s pain – attractiveness, gender, ethnicity, social class, as well as the context (return to work or post-surgical), and the characteristics of the observer (parent, caregiver, spouse, clinician).

Results from past studies have included: unattractive patients were judged to suffer from more pain than physically attractive patients (Hadjistavropoulos et al.1990, 1996, 2000). Generally, higher reported pain intensity appeared to invite higher estimated pain intensity. When comparing sufferers’ self-reports and observers’ estimates, however, low reported pain intensities were overestimated and high reported pain intensities were underestimated (Chibnall and Tait 1995; Chibnall et al. 1997; Krokosky and Reardon 1989; Tait and Chibnall 1997; Zalon 1993).

For health professionals, some studies found no associations (Dudley and Holm 1984; Everett et al. 1994; Hamers et al. 1997; Oberst 1978; van der Does 1989), others found that less experienced observers gave higher pain estimates than did more experienced observers (Mason 1981; Perry and Heidrich 1982; Lenburg et al. 1970).

In terms of contextual cues, the absence of medical evidence to support the person’s report of pain has been associated with lower estimates of pain.

Yesterday I briefly discussed the idea of malingering being the purposeful faking of health problems in order to gain financial benefit: in this study, the evolutionary value of the ‘social contract’ is used as the theoretical model for evaluating our sensitivity to these responses. Social contracting is a situation where the ‘‘individual is required to pay a cost, or meet a requirement, to an individual (or group) in order to be eligible to receive a benefit from that individual’’ (Cosmides 1989, p.197).

It makes sense that as humans we have some systems developed to determine whether we are being exploited and that the normal ‘contract’ between people is being disturbed. There is empirical support for humans being sensitive to cues for cheating from both human and animal studies (Cosmides 1989; Gigerenzer and Hug 1992;Wilkinson 1990) – but as we saw yesterday, we can be deceived reasonably readily, so it makes sense for us to be particularly sceptical about situations where we may be manipulated.

In a situation where one person is likely to receive benefits (caring or free food, for example) because of their reported pain, it makes sense for the observer to be more aware of cues suggesting that the individual receiving the ‘benefits’ is not actually in pain (Williams 2002). This has been studied before, and it has been found that suspicion of cheating or faking leads to conservatism and underestimation of pain. This can explain why such weight is given to medical evidence by health care professionals even when the relationship between ‘evidence’ from radiology and pain, for example, is fairly weak. This is especially true in cases where time off work, or other ‘special’ treatment is given as a result of confirmation of a ‘real’ problem.

The hypotheses in this paper were:
– the combination of two behaviours (continuing liked and
stopping disliked activities) would be judged as unfair
and lead to lower pain estimates, whatever the level of
pain,
– there would be no effect on pain estimates of the
presence or absence of medical evidence, and
– pain as reported by the patient would affect pain
estimates by relatives

Although the results were from a very small group of respondents in the end (only 23% of the initial group recruited actually responded with complete questionnaires), some interesting findings were obtained.
– people who stopped doing things they liked were interpreted as ‘having to stop’, and this was perceived as fair
– people who stopped things they didn’t like but continued with things they did like, were perceived as being unfair
– behaviour of characters reporting high pain was estimated to be fairer

The authors suggest that perhaps greater pain meant greater recognition of the cost in pain incurred by doing any tasks, or that more leeway was exercised in judgements of behaviour as fair or unfair.

– more pain was attributed to patients who stopped liked tasks
– highest pain levels were assigned to patients having stopped both liked and disliked activities
– lowest pain estimates were assigned to patients who had stopped disliked but continued liked chores, the combination which also received the lowest fairness ratings

The authors comment that these findings ‘support our hypothesis concerning lowering of pain estimates by individuals close to someone with persistent pain if they judge patients to be behaving ‘‘unfairly’’, that is, ‘‘accepting the benefits’’ of having pain—of being permitted not to do some tasks—‘‘while not meeting the requirement’’—of being unable to do other preferred activities.’

– participants’ estimates were not significantly affected by presence or absence of medical findings in this sample

– higher given pain intensities led to higher estimated pain intensities
– there were systematic discrepancies in estimates since low given pain intensities were estimated as higher and high given pain intensities as lower.

Some food for thought – this is a first cut study using social contract theory as a framework for determining a priori predictions as to which cues are salient, and in which direction. I look forward to finding out more on this – it may help us help our patients reflect on their behaviour, as well as the ways in which family members interpret and respond to patient’s behaviour.

There are some limitations to this study – in particular the small sample size, and the use of written vignettes rather than video-recorded scenarios. But for making us think? I think it’s great.

If you’ve enjoyed this post, there will be more tomorrow – and don’t forget you can use the RSS feed above to subscribe, or you can simply bookmark this page and come on back! I love comments and always respond, so don’t forget to let me know what you think of the topics I cover (***pssst! I don’t bite!***)

Kappesser, J., C. Williams, A.C. (2008). Pain judgements of patients’ relatives: examining the use of social contract theory as theoretical framework. Journal of Behavioral Medicine DOI: 10.1007/s10865-008-9157-4

Faking and malingering (again!)

One of the most popular posts I’ve written on this blog concerns ‘faking’ or ‘malingering’. I’m curious about this, because even though I have been asked many times whether I have had patients that are ‘faking’, I don’t think about it very often myself. I suppose it is a subject that is dear to the heart of anyone who is concerned about ‘motivation’ – especially where entitlements to payment or other benefits are dependent on the authenticity of the person’s claim to be unable to function.

So, time to unpack the whole topic again I think.

Last time I posted I had trouble finding relevant literature – the main problem being that in academic publications where pain is being discussed, the concept of faking has been superceded by research into areas we can work with clinically. The role of determining whether malingering or faking is not one for clinicians: it properly lives within the realm of private investigators – and even then, findings are up for challenge when a person is observed behaving in one way in one setting and another in a different setting. There are so many things that influence behaviour that it’s simplistic to decide that motivation to intentionally deceive is the primary reason.

If we start with the idea that pain itself is an experience – something we can’t share with each other, in the same way we can’t share whether we see the colour blue in the same way as each other, or the taste of banana! So you and I can’t tell just how much pain any person is in.

What we depend on is behaviour – what the person actually does behaviourally when they experience pain. And you and I know that we do different things in different places even when we experience the same pain! Take the example of stubbing your toe – in your garage you’ll probably swear loudly and jump up and down. In a very quiet church? Well, we probably won’t be able to hear your swearing and your jumping might be replaced with a wince…

In another example: if someone asked you to run 200m, you might take off at a bit of a gallop, but I’ll bet that if you knew there was a large and very angry bull coming after you, you might just move a little faster!

So we know that the social environment influences pain behaviour, and we also know that beliefs about the consequences of action also influence behaviour. We know too that what people understand about the significance of their pain is also a factor that changes behaviour. If you are unaware that the tickling feeling down your leg is actually a large and venomous spider biting you, you may well ignore it, if you think the vaguely tender lump under your armpit is cancer you’ll probably pay good attention to it and ask for a medical opinion. And your reaction to this lump might be even greater if you’ve had a history of cancer in your family.

In the compensation situation, there is no doubt that the security of receiving a weekly payment can be very helpful in the initial stages of recovery from an injury. That security can become problematic when the person is being asked to undertake the relatively risky business of returning to work – What if returning to work fails? What if my pain gets so bad that I can’t keep going? What if I let my employer down?

Some of these very realistic fears can interfere with the readiness for someone to return to work – and yes, compensation means it can easier to remain in status quo than to attempt to return to work. Does this equal malingering? Is the person faking? Or would you and I feel very much the same, and be quite hesitant to risk our health and livelihood too?

Can we as health professionals ‘tell’ whether someone is intentionally faking?

The short answer is no – humans are very poor at telling whether someone is lying. That’s why movies are so successful – even though we know the people are ‘actors’!

Take a look at this clip from You Tube:

And this one:

and the truth?

For another good article on deception – if a bit cognitive – this newly published article review the latest and greatest on the art of detecting deception. It’s just too early to apply this to pain and malingering.

SIP, K., ROEPSTORFF, A., MCGREGOR, W., FRITH, C. (2008). Detecting deception: the scope and limits. Trends in Cognitive Sciences, 12(2), 48-53. DOI: 10.1016/j.tics.2007.11.008