Since a perfect study examining the whole population of animals you are interested in will rarely exist (especially in veterinary medicine!), it is up to you to decide if the evidence you have found is pertinent to your individual clinical question. Studies are often conducted on a number of subject animals, and may therefore only be truly representative of a particular subset of a particular population of animals.
Some pertinent questions to ask may be:
- Does the population of animals in the study represent the animals that you see (e.g. animals seen at referral practices versus first opinion practices)?
- Does the evidence focus on animals with single morbidities (as opposed to animals with comorbidities)?
- Does the evidence in the study focus on using one therapy versus combinations of therapies?
Thinking about how to apply the evidence from published studies to the individual animal, or group of animals, you are working with raises four different questions, as outlined by Del Mar et al. (2008):
- What are the potential effects of treatment, both beneficial and harmful?
- Are there differences in the effects of treatments on different sub-groups of animals?
- Are there differences in levels of risk between different groups/sub-groups of animals?
- How do the benefits and harms relate to the individual animal or group of animals you have in front of you?
We will now illustrate these four questions using a clinical scenario about the use of analgesic products for calf dehorning.
1. What are the potential effects of treatment, both beneficial and harmful?
When making clinical decisions, it is important for veterinarians to weigh up the best evidence on both the benefits and harms of any interventions proposed.
In some countries, long-acting analgesic products are not approved for pain relief in livestock. Some of these countries allow veterinarians to use medicines in an extra-label fashion, when the health of the animal is threatened and when the veterinarian determines that particular drug is indicated. Extra-label drug usage, however, is not permitted if it results in violating food residue legislation.
Conversely, in many countries, both long-acting and short-acting products are approved as therapy to provide pain relief.
Because you have recently begun working at a practice that has not historically used analgesia for long-term pain relief post-dehorning in cattle, you wonder if you should propose using a non-steroidal anti-inflammatory drug (NSAID) for calves being dehorned, and if so, which NSAID would be preferable, parenteral flunixin or meloxicam? You search the literature and find two references which appear to be particularly relevant to this question:
- Heinrich A, Duffield TF, Lissemore KD, Millman ST. The effect of meloxicam on behavior and pain sensitivity of dairy calves following cautery dehorning with a local anesthetic. J Dairy Sci. 2010 Jun;93(6):2450-7. doi: 10.3168/jds.2009-2813. Accessed at: http://www.sciencedirect.com/science/article/pii/S0022030210002481
- Fraccaro E, Coetzee JF, Odore R, Edwards-Callaway LN, Kukanich B, Badino P, Bertolotti L, Glynn H, Dockweiler J, Allen K, Bergamasco L. A study to compare circulating flunixin, meloxicam and gabapentin concentrations with prostaglandin E₂ levels in calves undergoing dehorning. Res Vet Sci. 2013 Aug;95(1):204-11. doi: 10.1016/j.rvsc.2013.01.018. Epub 2013 Feb 19. Accessed at: https://aperto.unito.it/retrieve/handle/2318/131393/20567/Fraccaro%20et%20al.%202013.pdf
In the first study (Heinrich et al., 2010), you note that meloxicam (0.5 mg/kg) was shown to significantly prevent the relapse of pain after the effect of a cornual block had worn off in calves undergoing dehorning with cautery when compared to placebo-treated calves. Pain was measured by reduced sensitivity to pressure, ear-flicking and head-shaking, and meloxicam-treated calves were significantly different from calves receiving a placebo (p<0.05). This research was done on Holstein heifer calves that were six weeks of age.
The second study (Fraccaro et al., 2013) described significantly lower blood prostaglandin E2 concentrations in the flunixin-treated (2.2 mg/kg) group compared to the placebo group after surgical/cautery dehorning; the difference between concentrations in the meloxicam group and the placebo group was not statistically significant. However, meloxicam had a 2½ times longer half-life than flunixin, suggesting that its effect should last longer. This research was done on Holstein steer calves that were six months of age.
When appraising the quality of this evidence, you consider a number of things:
- Both studies were performed in cattle.
- Heinrich’s study was carried out in a dairy production system, while Fraccaro’s study was in a beef production system. You consider this, but decide that it is unlikely that the production system would make a difference in interpreting these study results.
- Heinrich’s study was carried out in Ontario, Canada, while Fraccaro’s study was done in Kansas, USA. Again, you think it is unlikely that the region would make a difference in interpreting the results of these particular studies.
- Both studies utilised research animals, although, again, you consider it unlikely that the source of animals would make a difference in interpreting the study results.
- In both studies, animals were randomly allocated to the treatment groups, minimising biases associated with group allocation.
- Heinrich’s study involved two groups of 30 calves, whilst Fraccaro’s study had much smaller groups (seven calves in each group). It is possible, therefore, that because of the smaller group sizes, the effect of the individual variation of animals within Fraccaro’s study might be more likely to account for some or most of the differences between the groups.
2. Are there differences in the effects of treatments on different sub-groups of animals?
Certain sub-groups of animals (e.g. certain age groups) may be more likely to respond either positively or negatively to specific interventions. When thinking about how you might apply the evidence, you will need to consider which sub-groups of animals were utilised in the research being considered.
3. Are there differences in levels of risk between different groups/sub-groups of animals?
By the nature of how and where they are kept, or their innate attributes, different groups or sub-groups of animals may have different levels of immunity and exposure to various pathogens. These differences may lead to different manifestations of disease severity in these different groups.
4. How do the benefits and harms relate to the individual animal or group of animals you have in front of you?
It is down to you as the veterinarian to make a judgement on the applicability of the research findings to your patients, which could involve a number of considerations. You might choose to reflect on your previous experience with similar cases, or to have a more in-depth discussion with the owner. You might also want to discuss the matter with colleagues, or consult an online forum to gain a broader view of the question at hand.
Another pertinent question to ask might be whether or not the findings of the studies are clinically relevant to your case, that is, will they really make a difference to the animal(s) in front of you? Once again, it is up to you to make a judgement about whether or not the outcomes measured would be expected to translate into meaningful clinical benefits to the patient and owner in front of you.
If you do not think the evidence from the papers you are considering is relevant enough to apply to the animal or group of animals you are treating, you can have a discussion with the owner of the animal(s) about the uncertainties around the options available (Legare 2009). Additionally you may choose to:
- Rely on the information in other forms of evidence such as textbooks, and online websites
- Do what you would normally do in these circumstances before you were aware of the published evidence, or
- Rely on your local clinic’s advice or guidelines or advice from colleagues who have handled these types of clinical problems before.