Dear Editor

Thank you for this contribution to the ongoing process of identifying appropriate effect measures for analgesic therapies in companion animals. It is especially important to evaluate the reliability of subjective owner measures given then possibility that these might not accurately measure the pain experienced by their pets.

In light of the recent study by Conzemius and Evans[1] showing significant caregiver placebo effects in canine osteoarthritis studies, I believe there is cause for caution in relying on the results of subjective owner or clinician assessment of lameness and pain in dogs. Despite double-blinding, this study found a highly significant placebo effect when using such measures. For example:

  1. While objective measures were consistent throughout the study, subjective evaluations were inconsistent and tended to show progressive improvement as the study progressed, even in dogs who were not improving objectively.
  2. A placebo effect in the owner evaluations occurred 56.9% of the time.
  3. A placebo effect in veterinarian evaluations occurred between 40% and 45% of the time, depending on which specific method of evaluation was being used.
  4. These placebo effects were highly significant at every point at which the dogs were evaluated (P < .001).

With this context, I wonder about your suggestion that the failure of CBPI and force plate analyses to correlate is due to the fact that both are accurately measuring different manifestations of pain.

If the treatment reduces pain, and pain is causing lameness, then as pain decreases the lameness should decrease. The force plate results appear to show this effect. However, if the CBPI is also measuring the reduction in pain, even if it is not measuring lameness, then the decrease in pain and the improvement in lameness should occur together and the 2 measures should correlate.

I find the analogous situation of evaluating the efficacy of antiemetics in people helpful in thinking about these results. One way of evaluating treatments for nausea is to count the number of episodes of vomiting. If the drug clearly reduces vomiting frequency compared to placebo, then that is an objective, though indirect, measure of nausea (analogous to force plate analysis evaluating pain by measuring lameness). And if we ask people about their nausea, and if they say they feel less nauseated when treated, this is another way to measure nausea and the effect of the treatment (analogous to asking clients about their dogs' pain, though in this case asking a caregiver rather than the patient). If both tests are measuring the same thing (nausea), then the results should correlate. Individual subjects who say they feel less nauseated should vomit less than people who say they don't feel any better.

However, if we give the drug and on average the subjects report less nausea and vomit less, but the individual responses on the survey don't match the frequency of vomiting, then we have results similar to those of this study. People who say the feel better might or might not vomit less. And people might show less vomiting even when they say they don't feel any less nauseated.

So are the survey and the frequency of vomiting both measuring nausea? And which measure is preferable? Do we want people to report feeling better but still vomit just as much, in which case the survey might an appropriate tool? Or is it more important to reduce the amount of vomiting even if some people who vomit less say they don't feel better?

In the case of osteoarthritis pain in dogs, it is unclear if reducing the signs of pain perceptible by the owner and reducing the lameness measured by force plate analysis are equivalent measures of actual pain reduction. This raises the question of when the CBPI would be an appropriate tool to use to measure the efficacy of an analgesic therapy. Is it more important to reduce the signs of pain recognizable and salient to owners or to improve the measurable lameness exhibited by the patient? If animal owners report their pets have less pain but we don't have an objective measure of function that correlates with this, it is clear that the intervention is changing something about the dogs' behavior which the owners are detecting, but it isn't clear that something is pain or that it is what matters to the patient.

Additional research is clearly indicated to evaluate the accuracy and reliability of subjective effect measures for canine analgesia studies and to identify more convenient and affordable objective measures. At present, it still seems prudent to give greater credence to objective assessment and to be cautious in relying entirely on the subjective evaluations of owners and veterinarians.


  1. Top of page
  2. Reference