a proportional alteration in the behaviour. This may not be the case with a physio-
logical measurement, for which there may well be some correlation between the
degree of pain and heart or respiratory rate, or the force applied to a limb. However,
if the behaviour response comprises vocalisation or aggression expressed on palpa-
tion, any correlation will necessarily be imprecise. The VAS developed by Holton
et al. ( 1998 ) was the first to provide an evaluation utilising multiple factors as
components of the assessment. For example, a range of behaviours and interactions
was assigned a score, these scores were added and the arithmetic sum indicated the
level of analgesic intervention required. This VAS was modified to produce an
abridged form, known as the composite measure pain score – short form (or CMPS-
SF) for routine use in a clinical setting (Fig. 1 ). These scales have been researched to
provide a response based on the precise meaning of the descriptors and interactions
with individuals familiar with assessing aspects of pain such as reliability and
reproducibility. However, it is important to note that, even with careful effort to
validate such a VAS, each VAS can be used only for one form of pain in one species.
Thus, the “Glasgow scale” applies only to acute pain in dogs, and other CMPS indices
will have to be developed for other pain states and other species (Welsh et al. 1993 ).
As described in the section on laboratory animal pain and analgesia, the appli-
cation of computer analyses of pain associated behaviours, especially video records,
hasbeenextendedtodogs(HansenLascellesetal. 2007 ) and cats (Lascelles et al.
2008 ). Two systems have been most widely reported, “Observer” as for laboratory
animals and “Ethovision”. These techniques generally assess activity and move-
ment and their application to dogs and cats has been more challenging than to
laboratory animals for several reasons, not the least being the need for a more
complex housing situation and a greater variability in individual animals, both in
colour and size. Another area of advancement in recent years has been the use of
computerised gait analysis techniques. These have become increasingly sophisti-
cated and are used to monitor the beneficial effects of analgesics on both acute and
chronic limb pain in both dogs (Waxman et al. 2008 ) and cats (Romans et al. 2005 ).
There have also been recent advances in the understanding of pain responses
through the use of electroencephalography. This has evolved through the availa-
bility of appropriate computer technology (Murrell and Johnson 2006 ). These
techniques have several advantages, in that they provide a direct measure of higher
CNS responses, and hence a true reflection of pain.
Much research has been centred on dogs as the beneficiaries of companion
animal pain studies and there have been fewer studies in cats. This is not logical,
as humans generally regard their companion cats as highly as their companion dogs.
Indeed limited studies in cats meant that until recently they were often denied
analgesia, on the grounds that they could react in a dysphoric manner to opioids and
perceived difficulties in metabolism of NSAIDs which led to the assumption of an
increased susceptibility to toxicity. Although many recent studies have overturned
these misconceptions, the recognition of pain states in cats has still not received the
same level of attention, in terms of development of VAS or other numeric pain
scales, as have pain states in dogs. Recently, devices to measure thermal and
mechanical thresholds in cats have been developed and these have allowed
Pain and Analgesia in Domestic Animals 169