What does it mean to take an evidence-based approach to swine production medicine?
In a nutshell, says Apley, evidence-based medicine is practiced somewhere in between the following two extremes:
Making decisions based only upon well-documented evidence.
Making decisions based upon selective memory and the most recent technology.
Evidence-based medicine, among other things, requires knowing what the complete body of evidence is, which evidence is sufficiently valid to consider and how the evidence applies to your swine practice situation.
“It requires knowing what the complete body of evidence is, which evidence is sufficiently valid to consider, and how this evidence applies to your practice situation,” says Apley, with
In human medicine, evidence-based medicine (EBM) has been described as “the integration of best research evidence with clinical expertise and patient values.” Sackett et al. (See editor’s note at end of article.)
“In production medicine, we replace patient values with client values,” Apley explains. “However, we also must consider the patient in the context of animal welfare and food quality/safety. Hopefully, these are also a major part of our clients’ values.
“Our goals in including client and animal issues in our decisions then become to identify our clients’ production goals, to increase their awareness of industry issues outside their own system and to provide input based on the best evidence available.”
Apley says it is also important to define practices that do not meet the criteria for evidence‑based medicine. Examples of these would include:
Just asking an “expert.”
Only relying on evidence without including experience.
Not critically appraising the evidence.
Not systematically searching for all of the evidence.
What evidence is acceptable in EBM? To answer this question, it is necessary to define what evidence will be accepted, according to Apley.
He points to the human medicine definition of best research evidence by Sackett et al.: “… clinically relevant research, often from the basic sciences of medicine, but especially from patient‑centered clinical research into the accuracy and precision of diagnostic tests (including clinical examinations), the power of prognostic markers and the efficacy and safety of the therapeutic, rehabilitative and preventive regimens.”
“The principle of clinical relevance is one that practitioners should be very astute in applying,” says Apley. For example, “not many of us would apply a nursery study to consideration of the same technology in a finishing barn,” he says.
“However, would we be tempted by vaccine titers in place of efficacy data? Relevancy not only applies to the age, species and production phase of the animal, it also applies to results in the production system as opposed to substitution variables such as titers. An exception would be where the substitution variable has been well correlated to clinical outcome.”
Apley adds that “relevant research also implies that the research was done in a manner such that we can trust the results. Unfortunately, we often only have the publication with which to judge the integrity of the results and conclusions.”
And veterinarians can sometimes be intimidated by the numerous statistical games that may be played, according to Apley. “However, even in the absence of advanced statistical knowledge, there are basic properties of a report that may be used to assess the integrity of the data and conclusions.”
Apley says that answering the following six questions can help to assess integrity of data and conclusions:
1. Is this a paper that applies to the subject as defined in our educational prescription?
2. Was confounding controlled either by randomization or blocking in the study design or by analysis?
Potential confounders include production site, animal source and management practices. Determine if each confounding factor was handled in such a way as to be equally represented in the treatment groups. “Even then, the confounder must be evaluated statistically before excluding it from the analysis,” Apley says. “In examples such as historical controls, or each treatment housed separately in one barn, it would be impossible to remove the confounding factors of time and barn respectively.”
Besides looking for assurance of randomization in the text, evaluate the equality of the number of animals in each treatment group, says Apley. “Except in situations such as unbalanced control designs (fewer negative controls than treated animals), it is reasonable to assume that effective randomization will result in nearly equal numbers in each treatment group,” he adds.
3. Were subjective evaluators masked (also called blinded) to prevent introduction of bias?
4. Are the case definitions for identification of diseased animals and determination of treatment success and failure reported? Do these variables relate to definitions that would be used in your practice?
5. Were the outcome variables, such as morbidity, performance and mortality, appropriate?
6. Was the statistical analysis appropriate? “Here you may need some help to look for tricks such as pseudo-replication, low statistical power in a study used to show equivalency of products and incorrect test selection. If a study is really pivotal to a large management decision, run it by a statistician. Another effective tactic is to question the author directly.”
Body of evidence
These evaluation principles should be applied to a body of evidence instead of just using one source of data to make a decision, according to Apley.
“A complete search for all relevant and valid data is an integral part of evidence‑based medicine,” he says.
And there is a need for a revolution in data presentation to veterinary practitioners, says Apley. To meet this need, he says, is the responsibility of veterinarians in academia and industry, who need “to develop organized methods for assembling, evaluating and then disseminating the results of critical appraisal of available data.”
Editor’s note: This article is based on a presentation at last fall’s Iowa State University Swine Disease Conference for Swine Practitioners. Some of the ideas have been adapted from evidence-based medicine as taught and practiced in human medicine.
Following are the references for the sources of some of the ideas and information.
Sackett D.L., Straus, S.E.,
, W.S., Richardson , W., & Haynes, R.B. (2000). Evidence‑Based Medicine: How to Practice and Teach EBM. (2nd ed.) Rosenberg : Churchill Livingston. New York
Cockcroft, P., & Holmes, M. (2003). Handbook of Evidence‑Based Veterinary Medicine. (1st ed.)
: Blackwell Publishing Ltd. Oxford
Isaacs, D., & Fitzgerald, D. (1991) Seven alternatives to evidence-based medicine. British Medical Journal 319:1618.
Traditional versus evidence-based medicine
This chart shows some of the main differences between the traditional approach and the evidence-based approach to practicing medicine.
Traditional approach: Clinical experience is a valid way of gaining an understanding about diagnosis, prognosis and treatment.
Evidence-based approach: Personal experience may be misleading.
Traditional approach: Pathophysiological rationale is a valid way of guiding treatments.
Evidence-based approach: Randomized studies are required to validate results because predictions based upon physiology may be wrong.
Traditional approach: Common sense and classical medical training are the only qualities needed to evaluate medical literature.
Evidence-based approach: Reading literature requires more than common sense to evaluation the evidence.
Source: Proceedings of the
How to practice evidence-based medicine
Convert the need for information into an answerable question.
“Background questions involve the need for general knowledge about a disease or disorder,” says Apley. “There are two main components. A question root -- (who, what, where, when, how, why) -- with a verb followed by a disorder or aspect of a disorder. For example, ‘what causes fever?’”
On the other hand, foreground questions involve the need for specific information on managing a disease or disorder. “Sackett et al. give three or possibly four parts to a foreground question: patient and/or problem, intervention, comparative intervention if possible and clinical outcome.
An example of this type of question might be, “How does aspirin in the water affect feed and water intake in pigs with a fever?” Apley says that as progress is made in experience, the questions become predominantly foreground questions.
Track down the evidence to answer the question.
This requires a search strategy that is more than a Google search or selected abstracts provided by the company trying to sell the technology, says Apley. “Authorities in your field of interest often have databases they would be able to provide you.
“Review the data and then discuss your questions with them. You can
access statisticians at your alma mater or nearest land-grant institution. With this approach, you are trusting that this database is complete.”
Critically appraise the evidence.
This, explains Apley, includes the validity of the data (can you trust it?), the impact of applying the results, and applicability of the data (do the results apply to the system in which you are working?).
A critical evidence appraisal also means you must develop evaluation parameters and a set of standards to apply to these parameters. Apley adds that it is also helpful to have a ranking system to help appraise the evidence. (He describes one such ranking system in his paper in the Proceedings of the
Integrate the results of your critical appraisal with your experience and client values/patient requirements to make your decision.
Apley concludes that veterinary medicine lags “well behind human medicine” in the ability to search for evidence appraisal. “Examples of human, peer‑reviewed appraisals may be accessed on the Web at the websites of The Cochrane Collaboration--wwwcochrane.org--and The Bandolier--www.jr2.ox.ac.uklbandolier/.
“A group has been formed to work on bringing such an approach to veterinary medicine. This involves participants at
“This group’s goal is to establish the teaching of EBM in colleges of veterinary medicine and to obtain funding to facilitate the production and dissemination of critically appraised topics to veterinary practitioners.”