Paper 1: Psychometric evaluation of the canine brief
pain inventory in a Swedish sample of dogs with pain related to osteoarthritis.
A. Essner et al.. 2017.
Patient group: 61 dogs referred for physiotherapy treatment of osteoarthritis. Some
dogs were also receiving ongoing anti-inflammatory treatment. Control group of
21 dogs with no signs of osteoarthritis. Inclusion criteria of: greater than 1
year and 9kg, signs and radiographic evidence of osteoarthritis. Exclusion
criteria of: other diseases interfering with mobility or quality of life, lack
of understanding of Swedish language.
Study type: Observational and cross-sectional study. Canine brief pain inventory
(CBPI) translated into Swedish. Questionnaire completed by owner consisting of
4 questions on pain severity, and 6 questions on pain interference with
function. Sums of each section averaged to give pain severity and interference
Outcomes: Results from 79 adult dogs
analysed as 3 incomplete completions of questionnaires. Analysis of
demographic data on age and body condition score and internal consistency and
construct validity were measured.
Key results: Dogs affected by osteoarthritis had significantly higher CBPI total
sums, compared to clinically sound dogs, indicating good internal consistency.
The CBPI indicated that a single modality anti-inflammatory treatment was
inadequate at controlling pain in a number of dogs with osteoarthritis. There
is some overlap with questions as some seemed to assess both pain interference
Study weaknesses: Small sample size for factor analysis. Floor effect present, especially
in pain severity questions, which could indicate subtle behavioural signs of
pain not being recognised.
Paper 2: Psychometric testing of the Helsinki chronic
pain index by completion of a questionnaire in Finnish by owners of dogs with
chronic signs of pain caused by osteoarthritis. A. Hielm-Björkman et al.. 2009.
Patient group: 61 client owned
dogs with Osteoarthritis. Of these, two smaller groups of dogs were treated
with carprofen (n=17) or placebo (n=17).
type: Cohort Study. 61 dogs with
chronic signs of osteoarthritis were chosen. Owners were asked questions from
the Helsinki Chronic Pain Index (HCPI). The questionnaire was asked five times
at four week intervals. Sensitivity testing was performed with smaller
groups of dogs (17 each) treated with carprofen or placebo.
Outcomes: Owners were asked
11 questions and answers were chosen on a 5-point scale. Answers were later
assigned values 0-4 and summed. Owners also assessed lameness using a visual
analogue score and quality of life.
results: The HCPI correlated well
with changes in quality of life indicating this tool can be used as a
predictive indicator of chronic pain. There were significant pain scoring
differences between the placebo control group and the carprofen group showing
that the test had high sensitivity.
Study weaknesses: Small study sample
with a minimal and short questionnaire. Seven owners believed their dogs had
too much pain before medication and gave their dogs NSAIDs against request from
the authors, which would have reduced signs of pain, possibly leading to a
reduced pain score.
Paper 3: Validation of a client-based clinical
metrology instrument for the evaluation of canine elbow
Osteoarthritis. C. Hercock et al.. 2009.
Patient group: 26 Labradors with chronic osteoarthritis in
the elbow from 24 clients.
Study type: Prospective cohort study. Owners were asked
to fill in a LOAD questionnaire assessing the animals function over the
previous two days before data was collected on the force platform to give
results for PVF
Outcomes: After the initial questionnaire, the method
was repeated two weeks later and repeatability was assessed using intraclass
correlation coefficient. Before the third visit, medication was ceased until
the assessment had taken place to give a placebo. Repeatability of the
questionnaire was good.
Key results: The results showed the validity of the
aggregate score of the LOAD questionnaire compared to the PVF score was poor
after both visits. Visual lameness scores do not have a linear relationship
with PVF. A Placebo effect was detected when owners used the LOAD questionnaire
but no effect was detected with the PVF data.
Study weaknesses: The study only assessed elbow osteoarthritis and small sample size of
only Labradors and clients from Liverpool. The study states that it was clear
that some owners had difficulty answering questions related to toileting and
climbing stairs/furniture, resulting in values being missed on the
questionnaire. The study didn’t achieve proof of validity and Type II error was
Paper 4: Evaluation of
Construct and Criterion Validity for the ‘Liverpool Osteoarthritis in Dogs’
(LOAD) Clinical Metrology Instrument and Comparison to Two Other Instruments.
M. Walton et al.. 2013.
Patient group: 222 dogs with
Study type: Prospective study,
split into cross-sectional (n= 143) and longitudinal cohorts (n= 79).
assessment was undertaken using LOAD, CBPI and HCPI. Results were then
compared between instruments. The results were also compared to the gold
standard, which is cited in this paper as the measurement of ground reaction
forces using force plate analysis.
Key results: The results of the
three methods correlated well, providing construct validity. This is
unsurprising due to the similarity of the tests, as all use subtleties of the
main clinical signs of canine osteoarthritis as their underlying theoretical
basis. LOAD is considered a useful tool for use by owners, as the results
appear reliable whilst remaining simple and comprehensible. Both
LOAD and CBPI were suggested to have good criterion validity due to a
significant, although weak, correlation between their analysis of
osteoarthritis and that produced using force plate analysis. There is no
apparent reason as to why the HCPI did not correlate significantly.
Study weaknesses: The primary purpose
of this study was to determine the validity of LOAD, possibly leading to bias
regarding the other tools. The longitudinal cohort was small leading to limited
statistical analysis. Learned behaviour and gait changes may have masked the
measurable effects of the arthritis.
Paper 5: Validation of a structured questionnaire as
an instrument to measure chronic pain in dogs on the basis of effects on health-related
quality of life. M. Wiseman-Orr et al.. 2006.
Patient group: 108 dogs affected by degenerative joint
disease (DJD), 26 clinically healthy dogs.
Study type: Longitudinal study. Initial questionnaires
were completed by owners of dogs with DJD in the first consultation, and
modified questionnaires were at subsequent consultations. Modified initial and
further modified subsequent questionnaires were completed for clinically
healthy dogs. Questionnaire baseis were a 109 descriptor item and 0-6
Likert-type ranking scale. Where appropriate, questions regarding behavioural
changes and health related quality of life (HRQL) were included. Clinician
questionnaires rating patient pain 1-10, with follow-up Likert-type scale
questionnaires, were also provided.
Outcomes: 90 of the DJD cohort dogs were enrolled and
73 provided a minimum of 1 follow up questionnaire. 26 clinically healthy dogs
were included, with a follow up questionnaire from 16. Factor analysis was
performed on a total of 221 questionnaires with a scree plot suggesting 7
factors accounted for the data variance but Kaiser criterion suggested a 15
Key results: 12 factors accounting for 65% of data
variance were identified as key HRQL domains. Linear discriminant analysis of
these HRQL domains discriminated successfully between no pain and some pain
dogs in 88% cases (95% no pain; 87 % some pain).
Study weaknesses: In depth questionnaire format resulted in
exclusion due to missed questions.