As Brain Injury Awareness Month drew to a close it was fitting that we caught up with Dr Gerald (‘Jerry’) Burgess, author of our test battery for the assessment of acquired brain injury and other neurological disorders, the SPANS. We sat down to discuss the test’s development and utility, the upcoming norms extension project, and current research studies utilising the measure. You can read the interview in full below:
You developed the SPANS ‘on the job’ – what led you to design the test?
It was a new job for me, working as a clinical psychologist across two acquired brain injury (ABI) neurorehabilitation inpatient wards in Leicester. I was referred patients who needed first of all to understand what consequence(s) they suffered from an ABI, to assess cognitive skills and understand their new profile of strengths and weaknesses. This information was needed for the patients and their families to understand and so the neurorehabilitation team could advise them, and also in order to devise a rehabilitation plan, make placement decisions, understand the recovery trajectory and predicted level of independence we may expect, and often to make a clinical judgment on their mental capacity to make certain types of decisions.
I wished to be thorough in my assessments, and cover the multitude of abilities I was discovering that can go wrong in a frighteningly wide range of patient’ ABI’s. There was not in existence a comprehensive, yet brief, measure with norms to match my patients’ age range, which was late adolescence to old age. Before I developed the SPANS, in order to get an equally comprehensive battery, I had to borrow subtests from different batteries, with the associated problems of comparing subtest performances that used different metrics and norm groups. I started designing the SPANS because I needed a comprehensive but brief assessment with subtests co-normed side-by-side with each other.
What sort of information does the SPANS provide?
The SPANS now has 30 brief subtests that spread over seven index scores, namely, orientation, attention/concentration, language, memory/learning, visuo-motor performance, efficiency, and conceptual flexibility. Each of these indexes contains two-to-eight subtests. The reason for so many subtests is that when there are multiple subtests measuring the same cognitive skill, one gets more reliable and trustworthy scores in that index score, that it raises the confidence that the label of the index score is intended to measure is indeed what gets measured.
Also there are neuro-anatomically different skills that are independent, or doubly-dissociate, that come under the same broad umbrella label. For example, ‘language’ is multi-faceted and includes expression, naming, comprehension, repetition, reading, and writing. All must be assessed for a thorough ‘language’ assessment to have taken place, or for any one cognitive domain for that matter. Thus as in this example, the SPANS screens many independent skills inherent within its subtests, and then cumulatively produces a very reliable index score in the broader cognitive domains. As a result, it is also possible to screen for neurological syndromes using the SPANS, including aphasia and type, unilateral neglect and spatial impairment, object agnosia, agraphia, acalculia, alexia/dyslexia, and apraxia.
How does this compare to other assessments intended for acquired brain injury?
The SPANS is the only test that considered a wide range of ABI’s and after effects in its design features and item selection, and therefore is like no other. Also, in many, many cases, I see no reason not to make the SPANS the assessment-of-choice when screening or doing a brief sensitive and specific comprehensive assessment with someone with ABI. The SPANS offers the most reliable index scores due to what I explained earlier, but also the most reliable re-test available with its alternative version, and the best test-retest reliability coefficients in its class of tests. The SPANS has more subtests than any other test in its class, and more, and more reliable index scores. The SPANS thus provides a lot of opportunities for norm-referenced observations of behaviour, but in administrating and scoring combined, it does not take longer than other tests in its class. One reason for its comprehensiveness, yet brevity, is that the SPANS was designed to be sensitive and specific with people with normal/average IQ – meaning that at a high percentage it should discriminate between those with no presence or history of ABI or neurological condition to those with even very mild-to-moderate impairment from some condition that affects cognitive abilities. ROC analysis showed that the SPANS is very successful with this in all of its index scores, particularly visuo-motor performance and efficiency. The SPANS was designed to not have to do more testing than is necessary on any one skill or subtest, and still provide good, accurate information if a particular skill or broader domain is impaired or not, and a means to interpret this in real depth. Some tests in this class over-test, require that the patient do more in regard to a single skill or subtest, but to come to the same finding or conclusion. With this, the SPANS can assess more skills but in an equal amount of time – and the patients tend to enjoy it and participate better too as a consequence of not being over-worked, and doing a variety of tasks.
Can it be used in areas beyond acquired brain injury?
It depends on one’s point of view, of which mine is that yes, the SPANS can be used for any neuropsychological assessment if the age of the individual being assessed coincides with the norms range the SPANS possesses. The opposing opinion is that a test should not be used with people with particular clinical/neurological conditions that have not been in sufficient mass assessed using the tool in question, and these data are available, reported to clinicians.
My view is that the neuropsychological profile of particular conditions is well known, and thus with a measure designed like the SPANS is, to assess neuro-anatomically-dictated cognitive, perceptual, and language skills by conventional means, and to do so using many sensitive and specific subtests and reliable index scores, the purpose of any neuropsychological assessment can be achieved. What was useful about clinically designing and norming the SPANS on ABI is that ABI is non-discriminatory – affects right and left hemispheres, includes primary direct damage and secondary effects, every lobe and mid-brain and brain stem, grey matter and white matter, diffuse or focal. It is then up to the clinician, who may use the SPANS manual which provides age-referenced norms and in-depth empirical and theoretical interpretative information on each subtest and index score to apply clinical knowledge, experience, and judgment to conduct a thorough assessment.
The SPANS is a flexible tool – could you explain how the test can be adapted to suit the needs of both the patient and the clinician?
As I’ve just mentioned, the SPANS is norm-referenced to the sub-test level, so any combination of subtests can be administered and interpreted. The SPANS manual suggests 5-minute and 15-minute screening tests to simply detect the presence or absence of cognitive impairment, which are comprised of the most sensitive and specific subtests. One version of this does not even require the use of the stimulus book, but in this case the patient must not be aphasic or have language impairment. If the patient is language impaired, or visual or motor impaired, or only has the faculty of “yes” or “no” responding, the SPANS manual recommends the most useful course to design an individualised assessment that covers the domains of orientation, attention/concentration, memory/learning, conceptual flexibility, and any aspect of visuo-motor performance or language that may be possible.
You have a normative study for older adults in the pipeline – please could you tell us about that?
We are collecting about 220, mostly healthy control, older adult norms to add to our existing database of individuals above the age of 75. When this is complete, the SPANS will be re-launched and provide norms from which to interpret performances of older adults aged 75 to 89, and include findings from initial studies examining the validity of using the SPANS in this age range specifically, and for differential diagnosis of dementias generally. The validation studies will involve detailed understanding of our participants’ demographic and medical histories, ability for independent living, and co-administration of the SPANS with gold standard measures, known to be very sensitive particularly to Alzheimer’s, semantic, and multi-infarct or vascular dementias. We will examine the most useful norm stratifications in these upper age ranges, clinically, and whether the SPANS co-varies and adds more to an assessment than using traditional gold standard tests.
Are there other studies that you know of currently using the SPANS as a measure?
There are many studies under way at the moment, some with manuscripts near-ready for submission to an academic journal for peer-review, and then hopefully publication. These studies either involve examinations of the validity of using the SPANS with particular groups, or the SPANS being used to detect changes in cognition following some kind of intervention, or as a means of describing prevalence of cognitive impairment. So in addition to the older adult study under way described earlier, other group studies include English as a second language, non-Western cultural influences, children/adolescents, and learning disabilities. Further psychometrics have been carried out since the publication of the SPANS manual, and these will be published, including ROC and exploratory factor analysis. There is a study that compares the SPANS head-to-head with the RBANS for sensitivity and utility in ABI neuro-rehabilitation. The SPANS will also be used to describe the cognitive profile of long-term incarcerated individuals, and as a pre- / post- measure of vitamin treatment for individuals with poor nutrition and high alcohol intake, often homeless individuals.
Thank you Jerry!
You can find further information on the SPANS assessment here. If you would be interested in being involved in the older adults norming study Jerry mentioned, you can express your interest by emailing firstname.lastname@example.org.