Home > Research > Publications & Outputs > A reappraisal of Kendell and Jablensky's accoun...

Links

Text available via DOI:

View graph of relations

A reappraisal of Kendell and Jablensky's account of validity

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published
<mark>Journal publication date</mark>08/2016
<mark>Journal</mark>Journal of Evaluation in Clinical Practice
Issue number4
Volume22
Number of pages8
Pages (from-to)522-529
Publication StatusPublished
Early online date1/05/16
<mark>Original language</mark>English

Abstract

Kendell and Jablensky argue that validity in psychiatry requires either unique biological characteristics or a zone of rarity, where few symptoms of one syndrome are present in another syndrome. Meeting either of these criteria allows the inference that the syndrome is caused by a specific biological mechanism not present in other syndromes. Failing to meet either of these criteria means the syndrome has been arbitrarily grouped and is invalid. Kendell and Jablensky's account of validity is too restrictive. Scientific phenomena are generally produced by a multiplicity of unstable overlapping causes, the causes for one phenomenon typically also present in other phenomena. Despite this, scientific phenomena are not automatically arbitrary. Science employs idealistic models that can successfully describe phenomena produced by overlapping causes and can gain approximately true knowledge of that phenomenon. The specific biological mechanisms that Kendell and Jablensky see as delivering validity are only specific in an idealized sense. Also, approximate truth means Kendell and Jablensky are mistaken to see validity as invariant and independent of context. An alternative approach to inferring causes is the common cause and unifications. Scientists often see otherwise unrelated phenomena regularly co-occur, and this legitimizes inferring common causes responsible for the phenomena. Applied to psychiatry, I show how syndromes that cover many different unrelated phenomena allow an inference to common causes, and this delivers validity. Zones of rarity can actually decrease validity. Idealized models often produce more information about causes by covering more phenomena, whereas zones of rarity often reduce the number of symptoms covered by a syndrome. Ignoring zones of rarity in favour of syndromes that cover many symptoms can sometimes increase validity. This can also occur when validating syndromes through corroborations with other factors. Increasing corroborations may require reducing the number of symptoms, potentially reducing validity.