Understanding the presentation of NPS-AD has generally followed inductive reasoning. Colloquially known as the ‘duck test’ approach, similarities between two ideas or objects or, in this case, conditions are made and conflated or used interchangeably.
That is, ‘if it walks like a duck, swims like a duck, quacks like a duck, it must be a duck’. These are essentially analogs of symptoms from idiopathic psychiatric disorders (that is, disorders of emotion, thought, or behavior for which there is no definitive pathophysiology).
For example, if an individual with AD were to exhibit lack of reactivity or report low mood, poor appetite, or disordered sleep, a diagnosis of depression in AD may be given. Similarly, if an individual exhibits hallucinations or delusions or psychomotor activation, a diagnosis of psychosis may be offered to explain the phenomenon.
Generally then, the descriptions of these phenomena, based on first-person experience (direct and primary observation), generate diagnoses, which direct treatment decisions.
Detection of NPS-AD has relied primarily on caregiver-reported scales. The most commonly used scale is the Neuropsychiatric Inventory (NPI), which has been considered the standard assessment for neuropsychiatric symptoms for the past two decades by assessing symptoms in 12 domains.
Recently, a revised version of the NPI, the Neuropsychiatric Inventory Clinician reported (NPI-C) , has been developed to address several shortcomings of the NPI and other rating scales and is a well-validated and commonly used scale.
A major improvement of the NPI-C was the addition of a clinician rating methodology, which mitigates the reliance on caregiver-provided information and in so doing systematizes the way in which clinical observation and caregiver report are integrated.
There are alternate versions of the NPI, including (a) NPI-NH , designed for use in the nursing home setting, and (b) the NPI-Q , a brief caregiver report questionnaire that can be completed in 5 minutes or less and has good test-retest reliability and convergent validity correlating with the full NPI at 0.91.