This reference tome covers the broad field of the behavioural neurology of dementia in 25 multi-authored chapters. Apart from a chapter by Hodges et al, it is written from a predominantly North American perspective, which will make some of the recommendations for practice seem odd to those of us working in the UK National Health Service – for example, that all patients with suspected dementia should have an MRI scan!

The book is rather idiosyncratically divided into sections of ‘Introduction’, ‘Cognitive impairment, not demented’, ‘Slowly progressive dementias’ and ‘Rapidly progressive dementias’, and not all chapters are where they might be expected – ‘dementia treatment’ can be found in the ‘Cognitive impairment, not demented’ section. Given the variety of voices, it comes as no surprise that there is some repetition of basic information, particularly epidemiology. There is also variation in the referencing style used in different chapters, which is unhelpful when chasing references.

Many of the ‘big names’ in dementia research have contributed to this publication. This means that it is generally well informed and readable. On the other hand, it also results in some concepts which are surrounded by debate being inaccurately portrayed as uncontroversial. For example, mild cognitive impairment is hypothesised to represent the prodromal phase of dementia but has an extremely heterogeneous prognosis, probably due to its lack of conceptual clarity. The chapter by Kelley and Petersen cites evidence that 5–16% of patients with mild cognitive impairment progress to dementia each year (though estimates range from 1–29%) compared with 1–2% of the ‘normal’ population. However, they downplay the fact that a significant proportion of those labelled with mild cognitive impairment return to normal cognitive function. They suggest that this is the case for approximately 5% per year but estimates of the proportion returning to normal have been as high as 44% (this and previous estimates from K. Ritchie, Dialogues in Clinical Neuroscience 2004; 6: 401–8).

More concerningly, the book minimises the well-accepted effect of increased mortality associated with antipsychotic medications in older adults with dementia. In the light of guidance from numerous bodies, and recent reports that these drugs have limited efficacy and are too often used as the first response to behavioural and psychological symptoms of dementia, the book’s coverage of this topic seems inadequate. However, the authors do make appropriate recommendations about the judicious use of these medications in a small proportion of highly symptomatic patients where other approaches have failed.

This book frustrated me slightly and left me crying out for a stronger editorial hand. Although it remains an interesting and valuable collection of rather disparate, varied chapters on this fascinating subject, it definitely remains one for reference in the library rather than your personal collection.