Review - A Casebook of Ethical Challenges in Neuropsychology
by Shane S. Bush (Editor)
Psychology Press, 2005
Review by Roy Sugarman, Ph.D.
Mar 8th 2006 (Volume 10, Issue 10)

Perhaps in essence, as Kevin Walsh indicated, there is no such thing, strictly speaking, as a neuropsychological test: there are just psychological tests, the inferences one draws from the results are of course informed by a clinical neuropsychological training and approach.  Testing is of course not assessment either as Joseph D Matarazzo pointed out in his capacity of President of the APA, just a part of it.

Inference becomes opinion when a report is written, and the environment in which it is written, and the purpose for which it is written lead the words and the circumstances in the provision of quandary for the writer.

Such quandaries are the substance of this book.  Each invited author puts forward a seemingly part factual part fictitious scenario for discussion.  In each, a neuropsychologist or like professional is asked to carry out an assessment on some person under some circumstances. The resultant mayhem can emerge from just about any situation, but of course the medicolegal arena is the most likely to present challenges, as are issues of informed consent, as well as privacy, confidentiality and so on. Each scenario includes some narrative around resolution of the case or issues, as a guideline.

Most, if not all, ethical codes common to all professional organisations that act as either guilds or gatekeepers or protectors of the public (it varies) have similar terms in common, such as non-malfeasance, confidentiality, dual relationships and roles, and other related terms and concepts.  Acting within the boundaries of one's profession, one's training and experience, above all else doing no harm and staying within the boundaries of consent, all should be easily recognized and observed.  'Should be', but life is complex and so it is not that simple.  Vexing scenarios can easily be imagined especially when culture intervenes as a complicating factor or conflicting interests especially when third party referrers are involved.

Issues of construct validity and what like professionals would consider valid or proper are more easily complex and vexing.

And so Bush puts together a volume composed of what others have experience in these and other regards.

For instance, an assessment of competency is done faultlessly, but later on, after conviction, the expert returns to discuss punishment, out of the range of the original informed consent or body of data.  Experts are used to deny entitlements, swimming unwittingly or wittingly in the wrong end of the pool of bias versus advocated opinion.  Experts leave the test room, and come back to find the test anonymity violated by theft and unauthorized approval. A client gives his consent for assessment under duress.

These are everyday events for most of us, trying to do what is correct and proper.  So many times, assessment is not within the culture of the doctor-patient relationship, but complex and with conflicting interests and roles being major pitfalls.  Can a patient give consent, and then when the testing is over, withdraw consent? If the report is written? If the report is not yet written or served?  Or written but not served?  Like rape, you can say no before, no during, but not say no afterward?  Or can you?

What then of normal clinical practice, let's say in medical settings?  One scenario is of the patient about to undergo surgery, taped to all kinds of machines, with all kinds of metabolic activity taking place.  Another is of colleagues who want to use the tests themselves to make inferences about cognitive functioning, who want to be taught to screen without the necessary knowledge of how inference is made.  Conflicts of interest may frequently arise when multiple roles are inhabited, or in arenas where there are multiple interests and groups represented.  Nonmaleficence is hard for anyone to action as a philosophy, and my spell checker finds it as hard and unusual a concept as any layperson.  The line between right and wrong is always grey at the edges.

In an assessment, when a client demands the neuropsychologist hear most troubling facts related to her future health, and at the same time demands confidentiality, what does the professional do? 

Psychiatric settings within medicine come with their own risks and the book reserves special space for this.  A patient is interviewed for her competency to make a decision about refusing medical treatment.  There is no neuropsychological deficit, and she can choose to die, can't she?  A second assessment by another expert agrees she is competent from that point of view, but finds her in the grip of a depression.  This depression makes her unfit to be taken at her word, so her choice to die is coloured and amiss, and the inference that she is neuropsychologically intact is right, but the inference that she is competent is thus wrong.

One essential point that emerges at this stage is the desirable faculty of training neuropsychologists initially as clinical psychologists before their speciality in the neural domain.  This is shown to be the common opinion among at least American neuropsychologists even though several western countries train their neuropsychologists with little such input or above average expertise in the issues germane to clinical psychology vs. clinical neuropsychology. 

Psychiatric settings are one thing, rehabilitation settings (such as mine) also provide fertile ground for examining principles such as fidelity, confidentiality, beneficence (which my spell checker recognizes, as does the general public).  Another strong issue is the need to perhaps modify, and thus depart from, strict administration rules during the administration of commonly used tools, perhaps invalidating the norms while still providing what one author here refers to as meaningful clinical information (page 98).  Here, the mirroring is of the abovementioned Matarazzo exhortation to yield diagnostically meaningful and thus useful information, when inference becomes opinion.  The interested reader will find the arguments made here revealing, for as Victor Nell has noted elsewhere, norms without constructs are just numbers, and that is precisely the dilemma here.  How does one make meaningful and informed observations when the norms have been rendered meaningless by a transgression of the rules that created the norm table that has reference?  Then again, norm tables were created on specifically defined populations, and drawing on a client from a different population makes the norms similarly into meaningless numbers. Linking this particular chapter by John DeLuca with those by authors such as Joseph D Matarazzo and Victor Nell would make compelling teaching material for interns.  A minor irritation in this chapter was the miss-citation of Cushman and Scherer's book, being the Psychological Assessment in Medical Rehabilitation, not Medical Practice.

Continuing the rehabilitation setting focus, it does not shift from informed consent needing to adequately reflect the referral questions, or an anticipation of what might emerge, such as removal of a privilege such as a driving license.  This would be especially true if another agency used the report for a purpose not outlined, or predicted, or if the tests lacked real application in the assessment of the real-life capacity.  Confidentiality in a team is difficult, as there may be members of that team whose use of the information, let alone access to that information, may exceed the original envisaged scenario when consent was obtained.

The neuropsychology of pain is oddly enough also entertained as a source of ethical quagmires.  Again, the special responses of persons afflicted with pain need to be known at an expert level, and knowledge of all and everything associated with a case is strengthened.  This would include the literature on pain affecting or at least having an impact on the functioning of widespread cortical and subcortical areas, as well as the results of clinical assessment.

If clinical neuropsychology is a rocky road strewn with minefields, then pediatric neuropsychology is strewn with disembowelled practitioners who wished they were dead: literature and testing in children, given developmental variance, is difficult. Issues of informed consent are also liable to cloud the water, as is parental involvement and issues of confidentiality, and statutory reporting of abuse and so on.  The presence of others in the room, either family, or court observers, is also an issue.  Geriatric patients are likewise in a vulnerable position requiring some expertise, as are geriatric patients from other cultures different to the origin of the examining neuropsychologist. Both geriatric and cultural concerns are dealt with here, including in the former a thorough review of the impact of illness and age overall, and the limitations of testing.

Ethical challenges move with the times, and the book now introduces the theme of information technology.  No fewer than 13 standards within ethics are challenged immediately in the scenario where an Internet based assessment using a remote company is contemplated, with few guidelines yet published. Another issue is that of providing details on websites, and imitations of confidentiality when prospective clients, where no contract exists, make contact.

Research design has always attracted a risk of ethical violation, and the book notes concerns such as the use of tests being trialed, the interests of others, ideas that the research tools or protocols constitute a thorough and valid workup, or test results or test protocols are not protected, or the rules are fudged with regard to how the results of a particular assessment are used, when other clinicians are involved.

Research into new instruments also presents hazards for the participants.  The determination of response validity is another, well recognized challenge.  It does of course happen that insurers request the use of certain tests, even if they can be deemed inappropriate, or sometimes even out of date. Some responses may look suspicious, but there may be other reasons for this.

The book finishes with an appendix containing the APA's Ethical Code for Psychologists as a reference.

The book overall seldom provides a scenario that is beyond the average neuropsychologist's experience, or at least that resonates.

Perhaps one from my own stable.  A Supreme Court Judge is hearing a matter.  He finds it hard to accept the plaintiff's contention that she has a brain injury following an accident where she did not lose consciousness, but an experienced clinical neuropsychologist has found her to be so damaged.  The judge interrupts the trial and demands that a second neuropsychologist see her.  Reviewing the results, the second neuropsychologist would tend to agree that the test results showed something was really wrong.  However, on applying his own tests, the neuropsychologist finds no indications, and she performs well.  On perusing her list of complaints given to the first neuropsychologist, he notes she suffers from migraine.  She admits that on the day of the first evaluation she had a bad migraine, but in her culture, one doesn't complain too much, and so she didn't tell the neuropsychologist, who didn't ask if she felt fine on the day. 

I wonder if any of us simply ask the question: Are you feeling fit today? Are you at your best? Are you particularly tired today? Can you do your best today?  Are you on any medication that might affect these results?  Did you have three espressos this morning?

Clinical neuropsychology is after all designed to pick up subtle deficits.  Kevin Walsh, the doyen of the Australian community used to note that if the deficits were so gross that the next door neighbor could elaborate on them, then one would not need a neuropsychologist.  The problem is that when sensitivity is high, then specificity is not, with regard to test profiles, and also, vice versa.

One must therefore always come back to the work of Joseph Matarazzo, and accept that variance on tests is a function of the diversity of human capacity, of the presentation of the average human, not necessarily of the presence of pathology. This is what happens when inference becomes opinion, devoid of context.

It is the context that is explored in this volume, with the resultant impact on professional practice and opinion explored in differing scenarios depending on whom is referring, who they are referring, for what ultimate purpose, and most importantly for whose gain?  Above all else, we have to do no harm, but that is not easy in practice, as harm waits to pounce whenever one human delivers comment on another.

© 2006 Roy Sugarman

Roy Sugarman PhD, Acting Director of Psychology, Royal Rehabilitation Centre, Sydney