|
Veterinary Epidemiology & Evidence-based
MedicinePapers - VM 585P 1998
EARLY DRAFT
Initial draft August 5, 1998
Contents:
Introduction:
This is a list of basic veterinary epidemiology resources that have been published in
journals, serial publications or in books that are out of print. "VMR"
indicates that the material is located in the reserve section of the
Veterinary Library. If you find papers that aren't listed but should be, find
papers listed that shouldn't be (e.g., too out of date), or find mistakes in
citations, please let me know.
Three unique methods to identify current papers are Free MEDLINE: PubMed and Internet
Grateful Med (The National Library of Medicine
literature database, including abstracts, selected from most but not all veterinary
journals, published since the late 1960's) and the Cornell Consultant: A Diagnostic
Support System for Veterinary Medicine. Because Consultant is strongly
disease-intervention oriented, it does not include many relevant publications. For
publications not indexed by MedLine, such as the AABP Bovine Practitioner and Annual
Meeting Proceedings, see the University
of Montreal Veterinary Library Veterinary Table of Contents and Veterinary Conference
Proceedings for issue-by-issue table of contents since 1994.
[Return to Contents List]
"How to
Do" or "How to Use" Papers:
Bonnett, B, R Reid-Smith (1996). Critical appraisal meets clinical reality:
Evaluating evidence in the literature using canine hemangiosarcoma as an example.
Vet Clin North Am Small Anim Pract 26(1):39-61. (VMR)
The many gaps in our understanding of cancer are a great frustration for veterinarians
and their clients. Unfortunately, what has been done has not always addressed the
questions clinicians need answered. How can clinicians deal more effectively with the
literature? Can missing information be identified, coped with clinically, and addressed by
researchers?
Boothe DM, Slater MR (1995). Standards for veterinary clinical trials.
Adv Vet Sci Comp Med 39:191-252.
Budsberg SC (1991). The randomized clinical trial. Vet Surg
20(5):326-328.
Randomized clinical trials were developed to eliminate biases inherent in clinical
medicine. Six types of bias that can occur during patient selection and treatment
allocation and in the collection and analysis of the data have been identified. A
variation of the prospective randomized clinical trial, the nonrandomized surgeon design,
overcomes biases specific to surgical investigations.
Crow SE (1996). Clinical trials in veterinary oncology: a clinician's viewpoint.
Vet Clin North Am Small Anim Pract 26(1):29-37.
Review of the veterinary oncology literature is important for keeping up-to-date, but
it can be time-consuming and sometimes frustrating. Results of clinical trials are
published in a variety of formats, including case descriptions, review articles,
retrospective studies, and prospective trials. They may be found as abstracts, which
usually present preliminary data and interpretation, as interesting items or commentary in
a newsletter, or as articles in referred journals. In this "clinician's
viewpoint," a practical approach to sifting through the mountains of literature is
offered.
Dargatz, DA, MD Salman (1993). Application of epidemiologic principles and
methods to investigating and controlling equine infectious diseases. Vet Clin
North Am Equine Pract 9(2):247-255.
Epidemiology is a tool much like any other diagnostic tool that the practitioner uses
as an aid in the complete physical examination of the group of animals. It can be used to
generate a specific differential diagnosis list. Epidemiology is especially useful to
design a disease control and prevention program by helping to elucidate the key
determinants of disease occurrence.
Dohoo, IR (1984). Decision analysis in bovine practice. Bovine
Practitioner 19:193-196.
Dohoo, IR, D Waltner-Toews (1985). Interpreting clinical research:
- I. General considerations. Comp Cont Educ Pract Vet
7:S473-S477.
II. Descriptive and experimental studies. Comp Cont Educ Pract Vet
7:S513-S516, S518.
III. Observational studies and interpretation of results. Comp Cont
Educ Pract Vet 7:S605-S610, S612.
Elbers AR, Schukken YH (1995). Critical features of veterinary field trials.
Vet Rec136(8):187-192.
The field trials of drug and vaccine efficacy published in The Veterinary Record in the
last five years were reviewed. The reviewers showed that in addition to some excellent
studies with clear and concise design and analysis, there was a considerable lack of
detail in reporting the method of allocation of animals to treatments, whether groups of
animals or individual animals were allocated to treatments, whether the trials were blind,
and a lack of formal analysis of the results. A considerable number of studies had
problems with clustering of data. Some critical features of the design and analysis of a
clinical field trial are discussed and improvements are proposed.
Jarp J, A Tverdal (1997). Statistical aspects of fish vaccination trials.
Dev Biol Stand 90:311-320.
Vaccine trials are performed to measure the efficacy of vaccines or vaccination
strategies. The statistical aspects pertain both to design and analysis of such trials. In
experimental trials, the design should ensure that the groups to be compared are equal in
all respects except for the factor to be assessed. This is achieved by proper
randomisation procedures. The diagnosis of cases should as far as possible be carried out
without knowledge of vaccination status (blinding). If the administration of a vaccine per
se is considered to influence the outcome, a similar administration of placebo in the
control group should be considered. The required sample size will depend on a specified
significance level, a specified power, and on the magnitude of the effect one wants to
detect. If the cumulative numbers of cases in vaccinated and unvaccinated groups are
registered at the end of the observation period, the effect can be measured by the risk
ratio, risk difference or the relative percent survival. If the estimation is based on a
fish-time approach and the presumption of constant rate of outcome over time is fulfilled,
the rate ratio or the rate difference can be estimated. If the rate is not constant, the
life table would be a method of choice. In the statistical analysis of the vaccination
effect it is important to consider the assumptions on which the statistical tests are
based. Proper choice of the experimental and statistical units in the trials is crucial.
If a trial is set up with the individual fish as the statistical unit, adjusting for
dependence (cluster effect) between the units may be indicated. If possible, the vaccine
effect should be given by the point estimate and the confidence intervals. The statistical
aspects of fish vaccination trials are based on general principles for controlled clinical
trials, but they are also influenced by characteristics connected to the fish population
and the experimental conditions.
Lessard PR, BD Perry (eds., 1988). Investigation of disease outbreaks and
impaired productivity. Vet Clinics NA: Food Animal Practice 4(1)
The chapters, which are by a range of authors, are:
- Investigation planning and data gathering, 1-15,
The characterization of disease outbreaks, 17-32;
The collection and submission of samples for laboratory
testing, 33-60;
The interpretation of laboratory results, 61-78;
Evaluating risk factors in disease outbreaks, 79-95;
The design and use of supportive epidemiologic studies,
97-108;
Techniques of reporting disease outbreak investigations,
109-125;
The investigation of nutritional disorders, 127-144;
The investigation of outbreaks of toxicologic disease,
145-158;
The investigation of outbreaks of infectious
disease,159-168;
The benefit-cost analysis of disease control programs,
169-181;
Disease outbreak investigation: Three case studies,
183-208.
Lund EM, KM James, JD Neaton (1994). Clinical trial design: Veterinary
perspectives. J Vet Intern Med 8(5):317-322. (VMR)
Tremendous potential exists for the use of the randomized clinical trial (RCT) in
veterinary clinical research. Understanding the fundamentals of RCT design not only
benefits clinical researchers, but it can enhance the ability of practitioners to
interpret published RCT reports. In this article, the essential components of RCT design
and implementation are described using examples from clinical veterinary medicine.
Lund EM, KM James, JD Neaton (1998). Veterinary randomized clinical trial
reporting: A review of the small animal literature. J Vet Intern Med
12(2):57-60.
The randomized clinical trial (RCT) is a valuable research method for the evaluation of
new treatment and prevention regimens in veterinary medicine. Reporting of clinical trials
in other disciplines has not been complete. Without complete information on the conduct
and results of a clinical trial, readers cannot optimize their use of the information
presented. This report represents an objective review of randomized clinical trials in the
veterinary small animal literature from 1986 to 1990. Results indicate that RCT reports in
the small animal veterinary literature are incomplete. The importance of reporting on
particular aspects of RCT research is described.
Mitchell H (1997). The pitfalls of field trials in fish vaccinology. Dev
Biol Stand 90:321-332.
Field trials are essential for accurately assessing the worth of a vaccine under actual
conditions of use. Compared with the laboratory, the dynamics of the host, pathogen, and
environment in a production setting can produce both subtle and dramatic differences on
the performance of the vaccine and the immune response. Because of this, field trials are
conducted by manufacturers in vaccine development and are required by many national
regulatory agencies to evaluate safety and/or efficacy before granting vaccine licenses.
Aquaculture producers, veterinarians and fish health professionals can use field trials to
analyse the cost-benefit of a vaccination programme for a facility, or to compare
competitive products. Vaccine field trials are more than merely using the products in the
field. Small efficacy effects can result in considerable cost reductions to the fish
farmer. Proper field trial design, conduct and analysis is critical to detecting these
effects. However, field trials are also fraught with many pitfalls that can result in
failure or misleading conclusions. The discussion regarding possible pitfalls of vaccine
field trials in aquaculture is divided into two parts: 1) the art and 2) the science of
successful field trials. The art of successful field trials involves dealing with the
"people" aspect which is necessary for initial and continuing compliance.
Meticulous planning is essential, including a written protocol to which everyone agrees by
signature. The bottom line to the art of field trials is anticipation and discussion of
all possible eventualities together with constant communication with the farmer and site
supervisor. The science of successful fields trials involves anticipating and realizing
the logistical and statistical difficulties in design and implementation. Problems often
encountered are: lack of stated quantifiable purpose; low power of the test due to
inherent small sample size, large variation and small margin of effect; lack of estimate
for anticipated results in controls and unpredictable challenges; non-blinding design with
bias and unequal treatment; one or few measures of outcome; lack of redundancy in sites
selected; poor statistical design and analysis with violation of assumptions; herd
effects; inability to assess long-term effects through several production cycles; and
manufacturer/producer expense.
Perino, LJ, MD Apley (1998). Clinical trial design in feedlots. Vet
Clin North Am FA Anim Pract 14:343-365.
This article discusses the key steps and considerations of clinical trial design, along
with additional considerations in development of a complete protocol. Three types of
feedloct clinical trial examples are provided and discussed.
Polzin DJ (1996). Importance of clinical trials in evaluating therapy of renal
diseases. Vet Clin North Am Small Anim Pract 26(6):1519-1525 (VMR)
Randomized controlled clinical trials (RCCTs) provide the ultimate test of therapeutic
effectiveness. In addition, results of properly performed RCCTs can provide data
concerning the potential harmful effects and costs associated with treatments. This
article addresses the need for RCCTs in veterinary nephrology and provides a background
for applying the results of such studies to patients.
Reeves MJ, NP Reeves (1995). Epidemiology and the veterinary
oncologist--evaluation and critical appraisal of the scientific oncology literature: How
to read the clinical literature. Vet Clin North Am Small Anim Pract
25(1):1-18. (VMR)
Even in the most specialized areas of veterinary medicine, few clinicians or
researchers can keep up with the volumes of scientific literature currently published.
Faced with this onslaught, most clinicians would benefit from a systematic approach for
selecting articles to read and evaluating their scientific merit.
Schukken YH, Deluyker HA (1995). Design of field trials for the evaluation of
antibacterial products for therapy of bovine clinical mastitis. J Vet
Pharmacol Ther 18:274-283.
In this paper, the design and statistical analysis of field trials for the evaluation
of the efficacy of clinical mastitis therapeutics is covered. First, general issues
underlying the design of clinical trials are reviewed. These include bias and confounding;
randomization and blocking; and study objectives and choice of the corresponding
hypothesis. Specific issues in the design of clinical mastitis trials are also discussed.
Selection of subjects is discussed with regard to choice of experimental units,
identification of reference population and study population, inclusion and exclusion
criteria, and sample size calculation. Next, a section on treatment administration and
evaluation of cure reviews treatment, blinding, choice of response measure, as well as
compliance, withdrawal, and early termination. The statistical analysis section addresses
possible statistical models, treatment of confounding, and fixed vs. random effects. In
conclusion, well-conducted clinical mastitis trials represent an invaluable, albeit
difficult and expensive, effort to evaluate efficacy and tolerance under usual
circumstances of use.
Tanner JE, AP Morgan (1993). Design and analysis of veterinary vaccine efficacy
trials.Vet Microbiol 37:221-230.
Thurmond, MC (1993). Epidemiologic methods in mastitis treatment and control.
Vet Clin North Am Food Amim Pract 9(3):435-444.
Methods and concepts of epidemiology offer means whereby udder health can be monitored
and evaluated. Prerequisite to a sound epidemiologic approach is development of measures
of mastitis that minimize biases and that account for sensitivity and specificity of
diagnostic tests. Mastitis surveillance offers an ongoing and passive system for
evaluation of udder health, whereas clinical and observational trials offer a more
proactive and developmental approach to improving udder health.
Traub-Dargatz, JL, DA Dargatz (1995). Clinical epidemiology: Application to
laboratory data. Vet Clin North Am Eq Pract 11(3):515-524.
Other:
J Am Vet Med Assoc 1994 Jul 15;205(2):344-345
Survey of statistical methods used in the veterinary medical literature.
Hammer AS, Buffington CA
Articles published in 1992 in 6 veterinary journals were reviewed. In 51% of the articles,
statistical analyses were not performed or only descriptive statistics (eg, mean, median,
standard deviation) were used. The most commonly used statistical tests were ANOVA and
t-tests. Knowledge of 5 categories of statistical methods (ANOVA, t-tests, contingency
tables, nonparametric tests, and simple linear regression) permitted access to 90% of the
veterinary literature surveyed. These data may be useful when modifying the veterinary
curriculum to reflect current statistical usage.
Vet Microbiol 1993 Nov;37(3-4):221-230
Design and analysis of veterinary vaccine efficacy trials.
Tanner JE, Morgan AP
Vaccination-challenge tests that involve all-or-none responses and do not require a direct
comparison between vaccinates and controls can be completely characterized by the binomial
distribution. Consumer and producer risks associated with binomial distribution based
tests can be adjusted by altering the number of animals involved and the criterion for
acceptance. Clinical signs or other outcomes measured on an ordinal or ranking scale
should generally be analyzed by nonparametric statistical procedures. Parametric
statistical tests are the most appropriate for data measured on an interval scale if the
necessary assumptions are met concerning the population sampled. The use of in vitro
potency tests in quality control procedures for inactivated vaccines depends on the
demonstration of a significant dose-response efficacy relationship in the host animal.
J Am Vet Med Assoc 1998 Jul 15;213(2):251-256
Program for surveillance of causes of death of dogs, using the Internet to survey small
animal
veterinarians.
Gobar GM, Case JT, Kass PH
Department of Population Health and Reproduction, School of Veterinary Medicine,
University of California, Davis 95616-8734, USA.
[Medline record in process]
OBJECTIVE: To develop a prototype program for surveillance of causes of death of dogs,
using resources developed for the World Wide Web, to enable
collection of data from veterinarians in small animal practice and dissemination of
results in a timely manner at minimal expense. DESIGN: Epidemiologic survey.
SAMPLE POPULATION: Small animal veterinarians who were members of NOAH, Veterinary
Information Network (VIN), or VetPlus-L. PROCEDURE:
Internet electronic communications and Web pages were used for solicitation and collection
of data, dissemination of results, and follow-up discussions with
participants. Data were stored in a relational database. RESULTS: 25 veterinarians
actively submitted case material. On the basis of analysis by region and school of
veterinary medicine attended, these veterinarians were representative of all small animal
practitioners in the United States. During the 6-month study, 621 case
reports were submitted. Analysis of results included determination of number of dogs, with
proportions calculated for primary reason for death, primary clinical sign,
and breed, as well as creation of a map depicting distribution of the practitioners.
Additional data were obtained for analysis to provide information of interest.
CLINICAL IMPLICATIONS: A national database representative of dogs examined by small animal
practitioners would be a valuable source of information.
Rapidly and easily accessible return of information and results is important for any
surveillance system. The program described here appears to be a successful
method for collecting data from practitioners.
Vet J 1997 Jul;154(1):1-3
Randomized controlled trials--the problem of clinical trials in veterinary science.
Higgins AJ
|