CVM GFI #217 Evaluating the Effectiveness of Anticoccidial Drugs in Food-Producing Animals

Administrative guidance, not enacted legislation. This captured publisher edition preserves the listed document and attachments; later developments are not independently consolidated.

Contains Nonbinding Recommendations
Page 1
# 217
Guidance for Industry
Evaluating the Effectiveness of Anticoccidial
Drugs in Food-Producing Animals
(This guidance #217 will supersede the CVM draft Guidance for Industry #40, entitled “Draft
Guideline for the Evaluation of The Efficacy of Anticoccidial Drugs and Anticoccidial Drug
Combinations in Poultry,” dated April 1992.)
Submit comments on this guidance at any time. Submit written comments to the Division of
Dockets Management (HFA-305), Food and Drug Administration, 5630 Fishers Lane, Room
1061, Rockville, MD 20852. Submit electronic comments on the guidance at
http://www.regulations.gov. All written comments should be identified with the Docket No.
FDA-2011-D-0784.
For further information regarding this document, contact Emily R. Smith, (HFV-135), Center
for Veterinary Medicine, Food and Drug Administration, 7500 Standish Place, Rockville,
MD 20855, 240-276-8344, e-mail: emily.smith2@fda.hhs.gov.
Additional copies of this guidance document may be requested from the Communications
Staff (HFV-12), Center for Veterinary Medicine, Food and Drug Administration, 7519
Standish Place, Rockville, MD 20855 and may be viewed on the Internet at either
http://www.fda.gov/AnimalVeterinary/GuidanceComplianceEnforcement/GuidanceforIndust
ry/default.htm or http://www.regulations.gov.
U.S. Department of Health and Human Services
Food and Drug Administration
Center for Veterinary Medicine
November 2012

Contains Nonbinding Recommendations
Page 2
Table of Contents
Introduction ............................................................................................................................. 4
I. 	General Considerations .............................................................................................. 5
A. 	Protocol Development .................................................................................................. 5
B. 	Animal Welfare Considerations .................................................................................... 6
C. 	Substantial Evidence of Effectiveness .......................................................................... 6
D. 	Masking ......................................................................................................................... 7
E. 	Personnel Training and Experience .............................................................................. 7
F. 	Record Keeping ............................................................................................................ 7
G. 	Nutritional Content and Preparation of Experimental Diets ......................................... 7
1. 	Feed Form ............................................................................................................... 8
2. 	Demonstration of Nutritional Adequacy ................................................................. 9
3. 	Nutrient Assays Falling Outside the Specification Limits ...................................... 9
H. 	Drug Assays .................................................................................................................. 9
1. 	Assay Limits for Drugs.......................................................................................... 10
2. 	Feed Sampling for Drug Assays ........................................................................... 10
I. 	Combination Approvals ............................................................................................... 10
II. 	Study Considerations for Poultry ............................................................................ 11
A. 	General Recommendations ......................................................................................... 11
1. 	Study Animals........................................................................................................ 11
2. 	Concomitant Drug Therapy and Vaccination ....................................................... 11
3. 	Bird Placement/Tracking ...................................................................................... 11
4. 	Housing/Management ........................................................................................... 12
5. 	Weighing equipment.............................................................................................. 12
B. 	Battery Studies ............................................................................................................ 12
1. 	Test animals .......................................................................................................... 12
2. 	Housing ................................................................................................................. 13
3. 	Feed management ................................................................................................. 13
4. 	Inoculum ............................................................................................................... 13
Table 1. Maximum acceptable age of commonly recognized field isolates of
Eimeria spp. in chickens ........................................................................................... 14
Table 2. Maximum acceptable age of commonly recognized field isolates of
Eimeria spp. in turkeys ............................................................................................. 14
5. 	Predicting the virulence of the inoculum .............................................................. 14
6. 	Battery Study Design............................................................................................. 15
7. 	Variables for Evaluation of Efficacy..................................................................... 16
8. 	Statistical Analysis ................................................................................................ 20
9. 	Basis of Study Conclusion..................................................................................... 21
10. Anticoccidial Sensitivity Tests............................................................................... 21
C. 	Commercial Field Study.............................................................................................. 21
1. 	Housing ................................................................................................................. 22
2. 	Test Animals .......................................................................................................... 22
3. 	Bird accountability................................................................................................ 22
4. 	Feed accountability............................................................................................... 22
5. 	Experimental Design ............................................................................................. 22

Contains Nonbinding Recommendations
Page 3
6. 	Variables for Evaluation of Effectiveness ............................................................. 23
7. Descriptive Statistics .............................................................................................. 23
III. 	Effectiveness of Anticoccidial Drugs for Minor Uses and for Minor Species ..... 24
IV. 	Effectiveness of Anticoccidial Drugs in Food-Producing Mammals .................... 25
A. 	Study requirements ..................................................................................................... 25
B. 	Method of Infection ..................................................................................................... 25
C. 	Experimental Design ................................................................................................... 25
D. 	Variables for Evaluation of Effectiveness .................................................................. 26
GLOSSARY........................................................................................................................... 27
APPENDIX 1 ......................................................................................................................... 28
APPENDIX 2 ......................................................................................................................... 30
REFERENCES ...................................................................................................................... 31

Contains Nonbinding Recommendations
Page 4
EVALUATING THE EFFECTIVENESS OF ANTICOCCIDIAL DRUGS
IN FOOD-PRODUCING ANIMALS
This guidance represents the Food and Drug Administration’s (FDA’s) current thinking on
this topic. It does not create or confer any rights for or on any person and does not operate
to bind FDA or the public. You can use an alternative approach if the approach satisfies the
requirements of the applicable statutes and regulations. If you want to discuss an alternative
approach, contact the appropriate FDA staff. If you cannot identify the appropriate FDA
staff, call the appropriate number listed on the title page of this guidance.
Introduction
This guidance provides recommendations to industry relating to study design and describes
criteria that the Center for Veterinary Medicine (CVM) thinks are the most appropriate for
the evaluation of the effectiveness of anticoccidial drugs in poultry and other food-producing
animals.
Section I discusses general considerations regarding the development of protocols, study
conduct, animal welfare, substantial evidence of effectiveness, feed preparation, nutritional
content of experimental diets, and the assessment of drug concentrations in experimental
diets.
Section II discusses the studies CVM recommends to substantiate effectiveness of an
anticoccidial drug in poultry.
Sections III and IV provide information regarding the development of studies to demonstrate
the effectiveness of anticoccidials for minor uses and minor species and the effectiveness of
anticoccidials in food-producing mammals, respectively.
The guidance is not a comprehensive source of information on conducting clinical
effectiveness studies. Alternative study designs for providing substantial evidence of
effectiveness may be acceptable. Sponsors should contact CVM to discuss their
development plan prior to initiating any studies. Sponsors and clinical investigators should
consult the Code of Federal Regulations (21 CFR Parts 511 and 514) for information on the
proper shipment, use, and disposition of investigational new animal drugs, as well as
submission of the results of clinical investigations.
FDA’s guidance documents, including this guidance, do not establish legally
enforceable responsibilities. Instead, guidances describe the Agency’s current thinking
on a topic and should be viewed only as recommendations, unless specific regulatory or
statutory requirements are cited. The use of the word "should" in Agency guidances
means that something is suggested or recommended, but not required.

Contains Nonbinding Recommendations
Page 5
I. 	GENERAL CONSIDERATIONS
We consider the indication, “For the control of coccidiosis caused by (specific Eimeria spp.)”
appropriate for anticoccidial drugs; however, a treatment indication may be considered. You
should design your protocol to support your intended indication.
A. 	Protocol Development
A protocol should be developed to specifically describe the plan for conducting an
effectiveness study. The protocol must include a clear statement of the study objective(s),
state the hypothesis, describe the experimental design in detail, and include success, entrance,
and exclusion criteria (21 CFR §§ 514.117(b)(2), (4)-(6)). Your protocol should be based
upon sound scientific principles and processes. The characteristics of an adequate and well-
controlled study are described in 21 CFR § 514.117. Sponsors should follow the format for
writing protocols that is recommended in CVM Guidance for Industry No. 85: Good Clinical
Practices: VICH GL9, Final Guidance, 05/09/01, section 6
(http://www.fda.gov/downloads/AnimalVeterinary/GuidanceComplianceEnforcement/Guida
nceforIndustry/UCM052417.pdf ).
CVM recommends that sponsors submit protocols for review and obtain a Protocol
Concurrence Letter from CVM before beginning essential studies1. CVM’s concurrence
with a study protocol represents a fundamental agreement with the design, execution, and
analyses proposed in the protocol. CVM concurrence represents a commitment that we will
not later alter our perspectives on these issues unless public or animal health concerns appear
that we did not recognize at the time of the protocol assessment. Because this concurrence
does not extend to any subsequent changes a sponsor may make to the protocol, sponsors
may want to seek concurrence on the revised protocol if they make changes. Protocol
concurrence does not guarantee that the results of the study will support a particular finding
or approval of the new animal drug. 2
Sponsors may choose to submit either a master or a site-specific protocol. Sponsors should
identify which type of protocol they are submitting. A master protocol provides general
information on principles that apply to all study sites. Master protocols allow any clinical
investigator to have all the details needed to conduct the entire study, including, but not
limited to the following: test and control article specifications, blocking and randomization
schemes, a description of the animal model, inclusion/exclusion criteria, variables of interest,
statistical analysis, treatment groups, schedule, and success criteria. Additionally, a master
protocol may be used as the basis from which more detailed, site-specific protocols are
developed and written.
A site-specific protocol should contain information present in a master protocol plus any
1 See Guidance for Industry #215: Target Animal Safety and Effectiveness Protocol Development and
Submission, Final Guidance
2 Animal Drug User Fee Act Performance Goals and Procedures,
http://www.fda.gov/ForIndustry/UserFees/AnimalDrugUserFeeActADUFA/ucm042936.htm

Contains Nonbinding Recommendations
Page 6
detailed information pertaining to the study site, including, but not limited to the following:
location of the study/studies, personnel involved, diet(s) to be fed, detailed facility diagrams
showing cage or house locations, location of feeders and waters, environmental conditions,
and standard operating procedures.
If you are submitting a protocol to CVM, we suggest planning your schedule so that there is
sufficient time for us to review it before the proposed start date of any studies.
B. 	Animal Welfare Considerations
All studies using live animals that are conducted in the United States must conform to the
requirements of the Animal Welfare Act (AWA), which is administered by the United States
Department of Agriculture (USDA). The USDA has issued policies and regulations on how
to comply with the requirements of the AWA. In addition, many research institutions that
conduct research studies using live animals are also accredited by the Association for
Assessment and Accreditation of Laboratory Animal Care (AAALAC).
Farm animals used for biomedical research (such as drug studies) fall under the purview of
the AWA and USDA regulations. Two publications: the "Guide for the Care and Use of
Agricultural Animals in Agricultural Research and Teaching," published by the Federation of
Animal Science Societies (FASS), and the “Guide for the Care and Use of Laboratory
Animals,” published by the Institute for Laboratory Animal Research (ILAR), may assist
researchers in the implementation of the USDA regulations. These guides provide
information on the appropriate handling, housing, care, treatment, and transportation of farm
animals for nonagricultural purposes and should be referenced when designing studies to
demonstrate the effectiveness of anticoccidial drugs in food-producing animals.
C. 	Substantial Evidence of Effectiveness
Effectiveness must be demonstrated by substantial evidence consisting of one or more
adequate and well-controlled studies [21 U.S.C. §§ 360b(d)(1)(E) and (d)(3);
21 CFR § 514.1(b)(8)(ii); 21 CFR § 514.4(a); and 21 CFR § 514.117].
CVM recommends battery studies and a multi-location commercial-scale field study to
establish the effectiveness of anticoccidial drugs in poultry. Battery studies are conducted to
evaluate the efficacy of investigational new animal drugs against individual and mixed
Eimeria species in the target animal under controlled laboratory conditions. Commercial
field studies are necessary to confirm effectiveness under conditions of use. You should
conduct a minimum of two battery studies for each Eimeria species being tested and at least
two battery studies with the mixture of Eimeria species. Power calculations can help you
determine the number of experimental units needed at a site. For independent substantiation,
studies should be conducted by at least two different investigators using different recent
United States field isolates. Your protocol should specify the number of individual species
and mixed species studies that you will conduct to demonstrate the efficacy of your
anticoccidial drug. A multi-location commercial field study should be conducted in the
appropriate number of sites to ensure that the two principle objectives for the demonstration
of effectiveness, independent substantiation and inferential value, are met. You should

Contains Nonbinding Recommendations
Page 7
specify in the protocol the number of study sites that you will use. See Sections III and IV
for information on the recommended studies to establish the effectiveness of anticoccidials
for minor uses and minor species and the effectiveness of anticoccidials in food-producing
mammals, respectively.
You are required to demonstrate by substantial evidence that the new animal drug is effective
at the dose or over the dose range selected (see 21 U.S.C § 360b(d)(1)(E)). For the purposes
of dose characterization, you should submit sufficient information to characterize the critical
aspects of the dose-response relationship. You may derive dosage characterization from dose
titration studies, pilot studies, foreign studies, or scientific literature.
D. 	Masking
In order to minimize bias, you should mask all personnel responsible for day-to-day
management of the animals, including those making and recording observations. You must
describe masking procedures in the protocol and final study report (21 CFR § 514.117(b)(7)).
Masking is an important design technique for avoiding bias in clinical trials (see CVM
Guidance for Industry (GFI) #85: Good Clinical Practices: VICH GL9, Final Guidance,
05/09/01).
E. 	Personnel Training and Experience
As specified in 21 CFR § 511.1(b)(7)(i), all personnel involved with the investigation must
have adequate scientific training and experience with the target animal species and disease
models used in the studies.
F. 	Record Keeping
Good record keeping is a critical component in determining the effectiveness of a drug when
it is being tested. The integrity and accuracy of the data collected are critical to the
acceptance of a study as substantial evidence of effectiveness. CVM’s recommended
standard for record keeping and data management is contained in CVM GFI #85: Good
Clinical Practices: VICH GL9, Final Guidance, 05/09/01.
G. 	Nutritional Content and Preparation of Experimental Diets
You should feed animals nutritionally adequate diets so that observed responses can be
attributed to the drug, rather than a possible nutritional effect. You should formulate diets to
meet predominant commercial practices for the species and class of animal being fed, and for
the geographic region where you conduct the study. You may use agricultural survey data,
e.g., Agrimetrics, Agri-Tech, etc., to support information on predominant commercial
practices.
Nutrient recommendations for different food-producing animals, published by the National
Research Council (NRC), may serve as a reference for formulating diets. You should specify
feed additives, such as antioxidants, pellet binders, copper sulfate, etc., in the diet
formulation and ensure that any feed additives do not confound the effects of the test article.

Contains Nonbinding Recommendations
Page 8
You should prepare experimental feeds from a uniform basal diet. A uniform basal diet is
consistent with regard to nutrient densities and ingredients. If you derive the uniform basal
diet from several mixer batches, you should ensure that the diet corresponding to each
treatment group is composed of equal aliquots from each mixer batch. In addition, you
should describe the procedures used to prepare a uniform basal diet and collect feed samples
in your protocol. We realize that there may be alternative methods for preparing a uniform
basal diet. Before initiating any studies, we suggest that you discuss and obtain protocol
concurrence with us regarding the procedure you wish to use.
You should divide each batch equally among all treatment groups. When a batch of feed
runs out for one or more treatment groups, you should discontinue feeding that batch of feed
to all remaining treatment groups and begin feeding the new batch of feed to all treatment
groups at the same time. Feed from the previous batch should be weighed and properly
accounted for in the final study report.
A NOTE ABOUT MIXING FEEDS
Properly mixed feed is critical for the success of your study. Proper mixing of feedstuffs and
feed additives will ensure the uniform dispersal of nutrients and drug(s) in finished feeds.
Personnel responsible for mixing feed should be aware of the performance and capabilities of
the feed mixer(s) that will be used to prepare experimental diets. Mix medicated feeds
according to 21 CFR Part 225 - Current Good Manufacturing Practice for Medicated Feeds,
to ensure adequate homogeneity. All standard operating procedures used for feed mixing
should be consistent with applicable GMP regulations for the manufacture of medicated
feeds. Your protocol should state that the medicated feed will be mixed at a commercial feed
mill and shipped to the study site, or that the medicated feed will be mixed using a validated
mixer. Your final report should include all relevant records.
1. 	Feed Form
The feed form of the experimental diet should reflect predominant commercial practices and
take the following considerations into account: the species and class of animal being fed, the
geographic region where the study is conducted, and the recommendations of the supplier of
the genetic stock.
For example, for poultry battery studies, feed is typically mixed and fed as a mash, and for
poultry commercial field studies the feed is mixed into a mash, pelleted, and crumbled. You
should be aware that the milling process may affect the effectiveness of the drug(s). For
example, poultry feed that is pelleted is typically subjected to temperatures ranging from 170
to 200 °F. Exposure of the feed to high temperatures may affect the stability of the drug(s).

Contains Nonbinding Recommendations
Page 9
2. 	Demonstration of Nutritional Adequacy
Test animals should be fed a nutritionally adequate diet in order to minimize confounding
effects of nutrition on the study results. To demonstrate the nutritional adequacy of the
experimental diets, you should include in your protocol a complete and detailed list of the
amounts of all feed ingredients and vitamin/mineral premixes that will be used in the uniform
basal diet. You should use ingredients that are representative of feedstuffs commonly used in
each geographic location. You should state the standard to which the nutrient levels are
being compared (e.g., NRC, literature, broiler management guide). For a uniform basal
poultry diet, you should report the calculated nutrient levels of crude protein, methionine,
cystine, lysine, calcium, phosphorus (total and available), and metabolizable energy.
To ensure that experimental animals receive proper nutrient densities in the diet, you should
conduct proximate and chemical analyses on a composite sample from the uniform basal diet.
You should include a description of the analyses that you intend to conduct in your protocol
and specify the number of assay replicates. You should indicate whether the analyses are
reported on an as-fed or dry-matter basis.
The purpose of nutrient analysis is to provide additional assurance that the diets are within
the calculated values specified in the protocol. To determine the nutritional adequacy of the
diet, we recommend using a 90% specification limit (SL) approach to calculate upper and
lower specification limits for nutrient concentrations. If you re-assay the composite sample,
you should adjust your specification limits (See Appendix 1); however, the specification
limits should be within those found in predominant commercial practice.
3. 	Nutrient Assays Falling Outside the Specification Limits
Nutrient concentrations that fall outside the SL are not necessarily deemed unsatisfactory.
We will consider the magnitude of the deviation from the SL, the nutritional circumstances,
and the clinical consequences, when the adequacy of the diet is determined. When you
document a deviation, you should provide a written justification of the reason, and possible
clinical relevance of the deviation.
H. 	Drug Assays
The purpose of conducting drug assays is to verify that the drug mixed in the feed and used
in the study is present at the appropriate concentration. Before approval of a new animal
drug, feed assay method transfer studies are conducted to determine the percent coefficient of
variation3 (CV%) for fortified feed samples assayed by the analytical method. The CV% is
then used to calculate the permissible analytical variation (PAV), which accounts for the
inherent variability in the analytical method, as well as in the medicated feed manufacturing
and sampling procedures, e.g. [(2*CV%) + 5%]. Assay limits are determined by applying
the ± PAV to the average percent recovery value of fortified feed samples, giving a range of
assay values that would be acceptable for a medicated feed manufactured at the intended
3 The CV% is defined as the standard deviation divided by the mean, multiplied by 100.

Contains Nonbinding Recommendations
Page 10
concentration. For example, if the CV% was 7.5%, and the PAV for this feed was
(2*7.5%) + 5% = 20%, then the assay limits (assuming 100% recovery) would be 80 to
120%.
1. 	Assay Limits for Drugs
Assay limits for approved drugs in experimental feeds are codified in 21 CFR Part 558.4.
Concentrations of approved drugs in experimental feeds should fall within these assay limits
regardless of where and when a feed sample is collected.
For investigational new animal drugs for which assay limits have not been codified, levels
should conform to the investigational assay limits derived through the feed assay method
transfer studies. If the assay limits are not established before conducting studies, the assay
method used to determine drug concentration(s) should be no more variable than the method
that will be subjected to the method validation process.
Note that a PAV relates to a single application of the assay method. Under the proper mixing
conditions using the correct quantity of the drug, a single assay of a feed sample should fall
within the assay limits. If a drug assay(s) falls outside of the assay limits, the feed should not
be used in the study and the reason for the out-of-assay-limit result(s) should be investigated
and discussed in your final study report.
2. 	Feed Sampling for Drug Assays
In your protocol, you should propose a feed sampling method for the study drug assay. The
method used should provide for a representative sample for the drug assay. The most
relevant feed sampling point is where and when feed is offered to the animals, with the
representative composite samples collected within 24 hours of feeding the animals. If you
choose to also confirm the drug concentration at the feed mill, please include the assay
results with your raw study data.
In addition, you should assay every batch of medicated and control feed for the presence of
the last non-study drug (based on your mixing records), unless all of the equipment used in
the study is dedicated and isolated to avoid cross-contamination.
I. 	Combination Approvals
The Animal Drug Availability Act of 1996 (ADAA) amended the Federal Food, Drug, and
Cosmetic Act (the Act) and changed the requirements for the approval of certain
combinations of new animal drugs that have been previously separately approved. There are
important criteria to consider before you submit a combination new animal drug application.
For example, under section 512(d)(4) of the Act, a sponsor seeking approval of a
combination of two or more previously approved new animal drugs may not need to conduct
additional effectiveness studies, if each new animal drug in the combination has at least one
unique non-overlapping indication. However, among other things, all of the following
criteria must be met: the new animal drugs must provide for appropriate concurrent use,

Contains Nonbinding Recommendations
Page 11
must be physically compatible, and not have disparate dosing regimens (21
U.S.C. §§ 360b(d)(4)(C) & (D)). If you intend to submit a combination new animal drug
application, we recommend you contact CVM for specific requirements.
II. 	STUDY CONSIDERATIONS FOR POULTRY
General recommendations for all types of poultry studies are provided first, followed by
specific recommendations for battery studies and the commercial scale field study.
A. General Recommendations
1. 	Study Animals
In support of substantial evidence of effectiveness, studies should be conducted using the
target species and class for which the drug is intended. You should use healthy chicks or
poults from contemporary, commercial genotypes. You should also document the source of
the chicks or poults and describe the age and health status of the breeder flocks.
You should minimize stresses outside the scope of the study and handle birds gently at all
times. You should only remove birds from a study to alleviate suffering due to a clinically
diagnosed disease or injury. Humanely euthanize any birds that you remove. You should
describe the removal criteria in your protocol.
2. 	Concomitant Drug Therapy and Vaccination
You should not use concomitant drug therapy during the study because the use of
concomitant drug therapy can influence the test drug response and/or mask adverse drug
reactions.
You may follow a vaccination program to protect the birds against prevalent infectious
disease(s), other than coccidiosis. The vaccination program should not debilitate the birds or
otherwise compromise your study. If vaccinations are used, please address in your protocol
any possible effects on the outcome of the study. In addition, the study protocol should
provide information concerning any vaccination program, including: 1) method (eye, spray
or water); 2) date of vaccination; 3) age of the bird at vaccination; 4) source; 5) vaccine lot
number and expiration date; 6) type of vaccine (live, modified live, or killed organisms);
7) handling of vaccine; and 8) any other information pertaining to the vaccine(s).
3. 	Bird Placement/Tracking
Your protocol should include a description of the method you intend to use to ensure the
accuracy of the number of birds placed in each cage or house at the start of the study. If
early chick/poult mortality is a concern, you may stock birds more densely, but within the
range of commercial stocking density. You should not replace birds during the study.
Your protocol should include a method that provides for: 1) accounting for the birds;

Contains Nonbinding Recommendations
Page 12
2) verifying the randomization of birds to cages, and treatments to cages or houses;
3) detecting migration or misplacement of birds; and 4) preserving the identification of the
original sex.
4. Housing/Management
To maximize the inferential value of the studies, the available square footage for the housing
of study birds should reflect commercial practices in the geographical location of the study
and take into account variation in seasonal temperatures and the final weight of the birds.
You should use a number of birds per feeder or waterer that is consistent with commercial
practices.
You should observe birds at least twice a day to appropriately manage the study and to
collect dead birds for necropsy. Frequent visits to the facility will allow you to obtain
clinical and management observations, such as feathering characteristics, abnormal excreta,
or behavior. You should adopt sanitation and bio-security programs to prevent the
inadvertent introduction or spread of pathogens.
5. Weighing equipment
All scales/balances used for weighing birds and feed should be calibrated on a regular basis
to ensure accurate and consistent measurements. You should include a description of the
calibration method or a copy of the standard operating procedures in your protocol and
provide the documentation in the final study report.
B. Battery Studies
Battery studies are conducted to evaluate the efficacy of investigational new animal drugs
against individual and mixed coccidial species in the target animal under controlled
laboratory conditions.
1. Test animals
You should use day-old birds (chicks or poults) representative of genetic stock used in
contemporary poultry production. You should also acclimate the birds for 10 to 14 days and
individually identify them (via wing band, leg band, etc.) before randomly assigning them to
cages and provide non-medicated feed ad libitum during this acclimation period. You should
house the birds in a way to prevent inadvertent coccidial infection during the acclimatization
period. In addition, as protocols are generally designed to support anticoccidial control
claims, you should start the birds on the experimental diets at least 48 hours prior to the
inoculation of oocysts.
You should identify the method of determining the sex of birds at hatch (i.e., feather or vent
sexing) in your protocol. On the final weigh day, you should verify the sex of each

Contains Nonbinding Recommendations
Page 13
bird phenotypically. If the sex is in doubt or conflicts with the original sex determination,
you should determine the sex by postmortem examination.
2. Housing
You should give special consideration to the placement of birds within the battery. You
should place non-inoculated, non-medicated control groups in the uppermost battery tier to
prevent coccidial contamination of these birds. We prefer that male and female birds be
separately housed. You should randomly assign males and females to separate cages so that
each cage has the same number of birds and the numbers of male and female cages are the
same. You should use uniform lighting within a block, utilizing typical commercial
photoperiods as recommended by the breeder (e.g., 24 hours lighting, or 23 hours lighting
with one hour off, etc.). In this type of study, a block is defined as a row of cages containing
at least one cage for every treatment by sex combination.
3. Feed management
You should measure feed consumption concurrently with body weight gain to document drug
intake. Each time feed is added, you should record the date, the type of feed, and the feed
weight. You should design and manage feed containers in such a way as to minimize feed
wastage and spillage into other feeders i(Williams, 1996). You should record the weight of
the remaining feed, per cage, at the end of the study.
4. Inoculum
The inoculum is a critical factor in the validation of the disease model used in battery studies.
The inoculum should represent field strains of coccidia that have been exposed to currently
approved anticoccidial drugs. You should use sporulated oocysts from recent field isolates of
commonly recognized poultry coccidial species (Table 1 and Table 2) that have been
exposed to contemporary anticoccidial drugs to propagate the inocula. Recent field isolates
reflect a state of anticoccidial susceptibility likely to be seen once the product is in use. The
isolates should not come from houses using anticoccidial vaccinations. Laboratory strains
are not representative of field strains of coccidia and are not acceptable.

Contains Nonbinding Recommendations
Page 14
Table 1. Maximum acceptable age of commonly recognized field isolates of Eimeria
spp. in chickens
SPECIES MAXIMUM ACCEPTABLE AGE
OF THE ISOLATE (Years)
E. maxima 	3
E. acervulina 	3
E. tenella 	3
E. brunetti 	5
E. mitis 	3
E. necatrix 	5
Table 2. Maximum acceptable age of commonly recognized field isolates of Eimeria
spp. in turkeys
SPECIES MAXIMUM ACCEPTABLE AGE
OF THE ISOLATE (Years)
E. meleagrimitis 	3
E. adenoeides 	3
E. gallopavonis 	3
You should use coccidia species that originate from North America, and are representative of
contemporary poultry production trends. You should report from where and when they were
isolated, the identity of the drug(s) in the feed at the time of the outbreak, and predominant
species involved. Your isolate should be as pure as possible; however, single cell isolations
are not necessary. The isolates should be passaged through susceptible birds, oocysts
collected at appropriate times and sporulated to produce the inoculum. Limit the passages to
preserve the identity of the isolate. The mixed inoculum should be comprised of individual
species in proportions adequate to provide the appropriate virulence. Recombine as
necessary to achieve the desired concentration of each individual species for the mixed
inoculum. You should administer the inoculum directly into the crop using a gavage
technique. The non-inoculated, non-medicated control group should receive an equal volume
of water by the same method as the treated birds.
5. 	Predicting the virulence of the inoculum
Virulence titration studies allow you to predict the number of oocysts that should be given in
the subsequent battery studies in order to induce an acceptable coccidial infection. You
should conduct a virulence titration study(ies) for each Eimeria species. You do not have to
conduct studies for each species concurrently. Virulence titration studies should include
birds in a non-infected, non-medicated control group, and multiple groups given
incrementally increasing numbers of oocysts.

Contains Nonbinding Recommendations
Page 15
You should conduct virulence titration studies no more than two to three weeks before the
battery study to maintain the integrity of the isolate. You should submit the results of the
virulence titration study as part of the final effectiveness study report. If you want CVM’s
concurrence on your virulence titration protocol, you should include it along with the
effectiveness study protocol.
For chickens, poultry pathologists and parasitologists have developed methodology to
evaluate the virulence of coccidia based on intestinal lesion scores, body weight gain, and
mortality.ii A clinically relevant level of virulence is characterized by 30% lower body
weights and a 3.0 unit increase in lesion scoresiii in the inoculated non-medicated birds when
compared to the non-inoculated non-medicated birds. This level of virulence should result in
lesion scores and body weight differences in the battery study that will provide an adequate
model for assessing drug efficacy. For battery studies using species of coccidia that typically
result in mortality, the mortality in the infected, non-medicated birds should be around 25%
to adequately model a naturally occurring outbreak. We expect that the mortality will be
most severe when E. tenella or E. necatrix are the inoculated species.
Standardized lesion scoring systems for turkeys have not been established; however, the
presence of intestinal lesions can be an indicator of virulence.iv We recommend obtaining
CVM concurrence for any new virulence scoring systems before using them in a battery
study. The most reliable clinical predictor of coccidial virulence in turkeys is body weight.
At the conclusion of a battery study, the body weights of the infected non-medicated birds
should be 30% less than the body weights of non-infected non-medicated birds to represent a
clinically relevant level of virulence.
For some species of coccidia in turkeys, mortality is also important for determining
virulence. In species of coccidia that typically result in mortality, 25% mortality in the
infected, non-medicated birds should be expected if the disease challenge adequately mimics
naturally occurring infection. We expect that the mortality will be most severe when E.
adenoeides is the inoculated species.
6. 	Battery Study Design
We recommend a randomized complete block design for battery studies to account for
heterogeneous environmental effects, such as temperature or humidity, which may exist
within a facility. Efforts should be made to ensure homogeneous distribution of cages within
each block; however, you should place the non-inoculated, non-medicated control groups in
the top cages to prevent cross contamination. The cages used in the study should be identical
in size. Male and female birds should be caged separately.
In battery studies, rows of cages usually form homogeneous blocks. Each cage (replication
of a treatment group) is an experimental unit. You should block the battery horizontally to
avoid the confounding factor of environmental differences among the rows of treated cages.
You should describe the randomization procedures used to assign birds to cages, cages to
blocks, and treatments to cages within a block in the protocol and final study report.

Contains Nonbinding Recommendations
Page 16
To allow for individual bird variation, we recommend replication of cages within treatments.
You should consider the number of replicates per treatment relative to the virulence to ensure
statistically significant differences. To accommodate the number of replicates required
batteries can be connected together. You should use an adequate sample size to ensure that
the study has sufficient power to detect differences among treatments. For determination of
clinical significance, however, we recommend a minimum of three or four cages of each sex
per treatment, with a minimum of ten to twelve birds per cage.
You should use a minimum of three treatment groups:
• 	non-inoculated, non-medicated control (NNC);
• 	inoculated, non-medicated control (INC); and
• 	inoculated, medicated (investigational new animal drug).
7. 	Variables for Evaluation of Efficacy
a. Lesion scores
In chickens, evaluate lesion scores 6 or 7 days post-inoculation in all birds. Only one person
should score lesions at each study location. The lesion scoring details will differ depending
on whether single or mixed Eimeria species are being studied. See Johnson and Reid (1970)
iii for examples of single species and mixed species lesion scoring methodology for chickens.
For deaths attributable to coccidiosis, assign the highest lesion score to the most severely
affected region of the intestine.
Alternative methods of lesion scoring may be acceptable, but we recommend that you discuss
them with us before conducting the battery studies. Regardless of the method chosen, you
should provide a detailed description of the pathology associated with each lesion score
grade in your protocol and final study report.
b. Fecal scores (dropping scores)
For turkeys, visual inspection of infected intestines (lesion scoring) for evaluating
anticoccidial efficacy can prove challenging because the pathology associated with turkey
Eimeria spp. can be subtle and is less conducive to standardized scoring. Therefore, fecal
scoring is an option for use as a primary variable for evaluation of anticoccidial efficacy in
turkeys in lieu of lesion scores.
Fecal scores allow for a clinically relevant qualitative assessment of the severity of the
coccidial infection through an examination of the gross appearance of the feces in each
battery cage dropping tray. You should assign fecal scores to indicate the severity of
diarrhea using a well defined numerical scoring system (e.g., 0, 1, 2, and 3). Your protocol
should include a detailed description for each numerical score to ensure consistent scoring
between investigators and a description of the time post-inoculation when scoring will occur.

Contains Nonbinding Recommendations
Page 17
c. Mortality
Some species of coccidia cause significant mortality in the field. Therefore, it is important to
measure mortality as an indicator of drug efficacy. You should maintain a daily record of
mortality by cage and include the following information in the final report: date, cage
identification, bird identification, sex, body weight, and necropsy diagnosis. You should
necropsy all birds that die or are removed from a study and classify as coccidiosis or non-
coccidiosis related mortalities. Perform wet mount examinations from scrapings of intestinal
lesions to identify the species of coccidia. If circumstances prohibit a timely necropsy of the
bird, you may freeze the bird and necropsy at a later date.
Include only coccidiosis mortalities in the statistical analysis for the evaluation of efficacy.
For deaths attributable to coccidiosis, you should assign the highest lesion score to the most
severely affected region of the intestine.
d. Body weight
Body weight change is used in battery studies as a clinical measure of the short-term effects
of the drug against coccidiosis following inoculation of the target coccidial pathogen.
You should record body weights of individual birds on the following days (see Diagram 1):
1) Day 0, Inoculation (10 to 14 days of age)
2) Day 6 or 7, Lesion scoring (20 or 21 days of age)
Diagram 1. Example of timing of body weight measurements by Study Day
Day -14 to Day -10 	Day -2 to -1 	Day 0 	Day 6 to 7
Day old
chicks or
poults
obtained
Body
Weights and
Inoculation
Randomization
and
Drug Initiation
Body Weights

Contains Nonbinding Recommendations
Page 18
Body weights (initial and final) of birds that died or were culled for non-coccidiosis reasons
should not be used in the analysis of the change in weight for each cage. You should
calculate the change in weight for live and dead birds (coccidiosis related mortality) from the
time of inoculation until the time of lesion scoring. This measurement assesses the efficacy
of the anticoccidial drug on coccidial infection throughout the study period.
The average weight gain (AWG) per bird in a particular cage is the total cage gain divided by
the number of birds used in calculating the total final cage weight.
AWG per bird in cage = 	Total cage gain
# birds1
Total cage gain = Total final cage weight2 - Total initial cage weight3
1 Number of live birds at the end of the study plus the number of birds that died of coccidiosis
during the study.
2Total final cage weight: Sum of body weights at the end of the study plus the body weights
at death of birds dying from coccidiosis.
3Total initial cage weight: Sum of body weights for birds that start the study minus the body
weights of birds that died of non-coccidiosis related causes during the study.
The average daily weight gain (ADWG) for the treatment group equals the sum of the
ADWG values for each cage divided by the number of cages in the treatment group.
ADWG for treatment group= 	Sum of ADWG per cage for the treatment group
# cages
ADWG per cage = [(Final weight of bird 1 - initial weight of bird 1) + (final weight of bird 2
- initial weight of bird 2)...+ (final weight of bird N - initial weight of bird N)] ÷ [(days
survived for bird 1) + (days survived for bird 2) ...+ (days survived for bird N)].
We ask that you outline the procedures used to minimize bias in the measurement of body
weight in the protocol. For example, weigh treatment groups within a block in a random
order to minimize any potential bias. After Day 0, you should not handle the birds until the
end of the study (except for birds that die during the study).
Secondary (supportive) variables
e. 	Feed efficiency
Coccidiosis causes intestinal pathology that may lead to a reduction in the utilization of
nutrients. Feed efficiency is used as a measure of the utilization of nutrients. Feed efficiency
measured in anticoccidial drug studies is an indirect measure of overall animal health. If you
use feed efficiency as a supportive variable, you should outline the procedures used in the
measurement of feed consumption in your protocol. In order to minimize a potential bias,

Contains Nonbinding Recommendations
Page 19
you should weigh feed from treatment groups within a block in a random order. The removal
of feeders from each cage at equal intervals is one acceptable method to weigh back feed.
Calculate feed efficiency (FE) as the ratio of the total feed consumed (FC) to the change in
weight (CW) of all the birds in the entire study as follows:
FE = FC ÷ CW
Total feed consumed is defined as the sum of all feed additions to a cage minus the sum of all
feed weighed back from that cage. The change in weight (CW) during the study period is
defined as the sum of final weights of all birds, including those that are alive at the end of the
study, those that died of coccidiosis during the study, and those that died of non-coccidiosis
causes during the study, minus the sum of the initial weights of all birds at the start of the
study as follows:
CW during study = (Sum of final weights of all birds) – (Sum of initial weights of all birds)
Calculate the overall feed efficiency (OFE) of a treatment group as the mean feed efficiency
of the cages within a treatment group as follows:
OFE of a treatment group =
FE cage 1 + FE cage 2 + ... FE cage N
# cages
f. 	Oocyst counts4
You may use differences in oocyst counts between treatment groups as a supportive variable.
Oocyst counts can be difficult to interpret for reasons such as the following:
• 	Characteristics unique to a specific Eimeria species may affect intestinal peristalsis;
• 	The day of maximum oocyst shedding may vary between Eimeria species;
• 	Oocyst concentration in the excreta is not a linear function of oocyst concentration
in the inoculum; and
• 	Excreta volume varies with many factors, including the Eimeria species inoculated,
the severity of the resulting infection, and feed intake during the infectious period.
You should describe the methodology used to collect (including timing of collection),
process, and analyze the oocysts in your protocol and final study report.
4 Upon agreement with CVM (through protocol concurrence or a pre-submission conference agreement), oocyst
counts may be considered a primary variable in minor poultry species.

Contains Nonbinding Recommendations
Page 20
8. 	Statistical Analysis
The final report should contain:
• Raw data in its original form;
• Raw data in an electronic format (Excel, SAS or text file);
• Program code to read raw data and create calculated variables;
• Statistical programs with documentation; and
• All statistical output (e.g., analysis results).
You should also provide a document identifying the purpose or content of each file and a
document that describes variable names, abbreviations, formats, and how variables are used
in the analysis.
a. 	Evaluation of an acceptable level of virulence
You should compare the inoculated, non-medicated control group to the non-inoculated, non-
medicated control group to verify an acceptable level of virulence. The differences between
these groups should be clinically relevant and statistically significant at α= 0.05 using a two-
sided test. (See section II.B.5 for a discussion of differences CVM considers clinically
relevant).
b. 	Evaluation of efficacy
You should compare the medicated group(s) to the inoculated, non-medicated control group
to establish efficacy. The differences between these groups should be clinically relevant and
statistically significant at α=0.05, two sided. The non-inoculated, non-medicated control
group should not be included in the statistical analysis comparing the inoculated, non-
medicated control group and the investigational treated group(s).
c. 	Analysis
The protocol should describe the statistical model used to analyze each primary and
secondary variable. The statistical model used to analyze the data should reflect the
experimental design and the level of measurement. The type of model should be based on
the response variable. For example, response variables that are at the interval/ratio level of
measurement can be described using a linear model; for those that are nominal/ordinal, a
generalized linear model is appropriate. If a randomized complete block design is used, as
recommended, then the data should be analyzed using a mixed model analysis of variance
(ANOVA). The model should include block as a random effect, the main effects for
treatment and sex (if sexes are included and separated by cage) and the appropriate
interactions as fixed effects.

Contains Nonbinding Recommendations
Page 21
You should conduct a separate statistical analysis for each area of the intestine that is lesion
scored. For all birds that die or are removed from the study due to coccidiosis, you should
assign the highest lesion score to the most severely affected region of the intestine.
9. 	Basis of Study Conclusion
You should define the basis of your study conclusion in your protocol, prior to the start of the
study using the following success criteria (See Appendix 2).
a. In order for a study to be considered in the overall determination of efficacy, the study
should demonstrate an acceptable level of virulence, as described in Section B.8.a. above.
b. In order to conclude that the drug is effective, the study should meet the following success
criteria:
1) For chickens and turkeys, a statistically significant (α=0.05, two sided) and clinically
relevant difference in body weight gain, and mortality (if appropriate) between the
group(s) treated with the investigational new animal drug and the inoculated, non-
medicated control group, and;
2) For chickens, a statistically significant (α=0.05, two sided) and clinically relevant
difference of at least one lesion score unit between the group(s) treated with the
investigational new animal drug and the inoculated, non-medicated control group.
3) For turkeys, a statistically significant (α=0.05, two sided) and clinically relevant
difference in fecal scores between the group(s) treated with the investigational new
animal drug and the inoculated, non-medicated control group.
If a study does not meet the predefined success criteria, the conclusion will be that the drug is
ineffective. All studies that include an acceptable level of virulence, regardless of the
outcome, will be used in the overall determination of efficacy.
10. 	Anticoccidial Sensitivity Tests
Anticoccidial sensitivity tests (AST) are studies conducted with birds in battery cages to
evaluate the effectiveness of a variety of anticoccidial drugs against a particular field isolate.
Because of their study design and objectives, ASTs do not provide substantial evidence of
effectiveness. However, if conducted, the results should be submitted.
C. COMMERCIAL FIELD STUDY
You should conduct a multi-location commercial-scale field study to confirm that the feeding
level you have selected is effective under commercial production conditions. You should
conduct this study under actual conditions of use in commercial-scale facilities, under a
variety of production systems and conditions. You should use an appropriate number of
geographic locations representing different environmental conditions, using a genetic line of

Contains Nonbinding Recommendations
Page 22
birds commonly used in that region, and using a ration that is representative of local
commercial practices.
1. Housing
You should conduct the commercial-scale field study in housing typical for the species and
class being tested. We recommend using paired houses, preferably identical and facing in the
same direction. If facilities permit appropriate separation, you should divide each house into
equal parts (split house design) to allow for more replicates at a study location.
You should not use artificial infection in the commercial-scale field study. Coccidia present
in or on older litter should provide sufficient challenge to accurately assess the effectiveness
in a commercial situation.
2. Test Animals
You should use at least two different genetic lines of birds to provide independent
substantiation of effectiveness among typically used commercial birds. The birds in each
location should originate from the same breeder flock source. Typically, birds are one day of
age at the beginning of the study. You should rear the birds to an age that represents a
marketable weight (by sex). You should consider the indication and target class when
selecting the type of birds you will use in the studies.
3. Bird accountability
Bird accountability includes the number placed in each house, the number that died or were
culled during the study, and the number of live birds sent to market at study end. The
protocol should describe the timing, location, and procedures associated with bird
accountability.
4. Feed accountability
You should document feed weights each time you add feed, make a diet change, and at the
end of the study. The protocol should describe procedures for weighing the feed and
appropriate procedures to minimize feed wastage and spillage.
5. Experimental Design
Because these studies are conducted under commercial conditions using a large number of
birds, two treatment groups are recommended: 1) the investigational new animal drug and 2)
a group treated with a currently approved anticoccidial drug or vaccine. The latter group also
serves as a standard for comparison. The length of study should be consistent with the
targeted birds’ age and class (e.g., broilers, replacements).

Contains Nonbinding Recommendations
Page 23
6. 	Variables for Evaluation of Effectiveness
a. Mortality
You should report historical mortality data for the facility. You should record the number of
mortalities and birds culled daily. You should diagnose the cause of mortalities whenever
possible. When necessary, perform detailed necropsies to obtain accurate diagnoses. If you
diagnose or suspect coccidiosis, you should select a pre-determined number of birds for
lesion scoring and oocyst speciation.
If circumstances warrant ending the study prematurely (excessive mortality, natural disaster,
etc), you should attempt to contact CVM as soon as possible to discuss the impact these
events might have on the development plan. Adverse reactions associated with the
investigational use of a new animal drug must be reported in accordance with
21 CFR § 511.1(b)(8)(ii).
b. 	Bird weights
You should record the final bird weights as soon after harvest as possible to avoid artifacts
introduced during transit to the processing plant. Alternatively, a random sample of birds,
representative of the flock, may be weighed before any pre-harvest fasting. Pre-harvest
fasting includes the removal of feed and water before catching birds for transport to the
processing plant. If you choose the pre-harvest weight collection method, you should
provide a detailed description of the procedure in the protocol.
c. 	Condemnation numbers
You should collect condemnation data from all birds on each treatment. You should separate
the number of condemned carcasses attributable to the farm (i.e., tuberculosis, leukosis,
septicemia, toxemia, inflammatory processes, synovitis, tumors, bruises, and air sacculitis)
from the number of condemned carcasses attributable to the processing plant (i.e., overscald,
contamination, no viscera, and plant rejects).
d. 	Feed consumption
You should calculate feed consumption by subtracting the sum of the feed weighbacks from
the sum of the feed additions to each house.
7. Descriptive Statistics
The following results may be used to evaluate the effectiveness of the investigational new
animal drug in comparison to the group treated with the approved anticoccidial drug or
vaccine: percent mortality, mean final live weight per bird, feed consumption per bird, and
the feed conversion. If the split house design is used, the calculations should be made on a
per-treatment basis; otherwise, the calculations should be made on a per-house basis since the
entire house receives the same treatment.

Contains Nonbinding Recommendations
Page 24
% Mortality =
100 X (# of birds placed at beginning of study – # of live birds at end of study)
(# of birds placed at beginning of study)
Mean final live weight per bird = 	__Total weight of live birds at end of study _
Total number of live birds at end of study
Feed Consumption = Sum of feed additions - Sum of feed weighbacks
Estimated feed consumption per bird = 	Total feed consumed_______
# of live birds at end of study
Estimated feed conversion = 	Total feed consumed
Total live weight
Means, medians, standard deviations, ranges, minimums, and maximums, where appropriate,
should be calculated for the variables listed above. These should be presented in table format
by treatment and location and by treatment combined across locations, where appropriate.
These results will be used in the evaluation of the clinical relevance of the study. Because of
the small number of experimental units, statistical testing is not appropriate for this design.
In addition, we recommend for descriptive purposes only that 95% confidence intervals on
mortality be calculated assuming the binomial distribution. The binomial confidence
intervals treat bird as the experimental unit and thus fail to capture the dependence among
birds within houses. However, these intervals do provide a useful general summary of the
precision of the estimate of mortality.
III. 	EFFECTIVENESS OF ANTICOCCIDIAL DRUGS FOR MINOR
USES AND FOR MINOR SPECIES
The Minor Use and Minor Species Animal Health Act of 2004 (MUMS Act) amended the
Act to establish new regulatory procedures that provide incentives intended to make more
new animal drugs legally available to veterinarians and animal owners for the treatment of
minor animal species and uncommon diseases in major animal species. Minor animal drug
use includes any new animal drug for use in a minor species or a new animal drug for use in
any animal species for control of an infrequently occurring or geographically limited disease.
Minor species include any species other than cattle, horses, swine, chickens, turkeys, dogs,
and cats (21 U.S.C. §§ 201(nn) and (oo)). Minor avian species include game birds, semi-
domestic waterfowl, and ratites. CVM’s Office of Minor Use and Minor Species determines
whether your application qualifies for the incentives provided under the MUMS provisions of
the Act. For additional information on effectiveness requirements for anticoccidial drugs

Contains Nonbinding Recommendations
Page 25
used in minor species, see CVM’s Guidance for Industry (GFI) #61: FDA Approval of New
Animal Drugs for Minor Uses and for Minor Species, Final Guidance (1999).
IV. 	EFFECTIVENESS OF ANTICOCCIDIAL DRUGS IN FOOD-
PRODUCING MAMMALS
We strongly recommend that sponsors contact CVM early in the development process,
before conducting any studies in food-producing mammals. Study designs are difficult to
standardize because coccidiosis is prevalent only under certain management conditions and
because of differences between species and classes in the prevalence and presentation of
clinical disease. Decisions on the specific study requirements, including the experimental
design, appropriate effectiveness variables, and basis of study conclusions including the
appropriate statistical analysis, made in consultation with CVM should greatly facilitate the
development process.
A. Study requirements
Substantial evidence of effectiveness may be satisfied with dose confirmation studies
analogous to battery studies, without a commercial scale field study. Commercial scale field
studies are generally not necessary in mammals because dose confirmation studies are
already conducted under field conditions. In addition, commercial management practices for
food-producing mammals are more variable than in poultry production systems and would be
difficult to standardize.
Effectiveness should be demonstrated for all Eimeria species included in the label
indications. The species of coccidia used in the studies should be representative of the
coccidial species that cause clinical disease. The test animal should be representative of the
age and class of animal susceptible to coccidiosis. Justification of the coccidia species and
the target animal class should be provided in the protocol. We consider mixed species
infections acceptable.
B. Method of Infection
CVM prefers natural infections because they more closely reflect current field isolate
characteristics. We may consider induced infections if you provide acceptable justification
in your protocol for not using natural infections. If you use induced infections, the recently
isolated and propagated inoculum should represent the exposure of field strains of coccidia to
current anticoccidial drugs. Laboratory strains are not acceptable. You should verify
coccidial infection in all test animals by an accepted method before initiating the study.
C. Experimental Design
You should use a minimum of two treatment groups: 1) infected, non-medicated control, and
2) infected, medicated (investigational new animal drug). You should include enough

Contains Nonbinding Recommendations
Page 26
replicates to make a meaningful statistical comparison. Use the standard attributes of an
adequate and well-controlled study, as described in 21 CFR § 514.117.
D. Variables for Evaluation of Effectiveness
Appropriate primary effectiveness variables may include: fecal oocyst counts, fecal scores,
average daily weight gain, and mortality. Clinical observations such as body condition, fecal
consistency, attitude, and hydration may supplement the primary variables. CVM will
consider a literature-based justification of the clinical relevance of any variables.

Contains Nonbinding Recommendations
Page 27
GLOSSARY
The following definitions are presented for use with this guidance document.
Block: A homogeneous, but extraneous (not of interest in the study) grouping of
experimental units designed to account for variance related to the extraneous grouping, thus
(if successful) reducing the error variance used in hypothesis testing. In a statistical model,
blocks are considered a random effect.
Composite sample: A pooled sample representing samples collected from each batch of
feed. The number of samples from each batch of feed included in the pooled sample should
increase with increasing batch size.
Confidence interval for a population parameter: A range of values, calculated from the
sample observations, which are believed, with a chosen degree of confidence, to contain the
true parameter value. For a 95% confidence interval, one is 95% confident that the true value
of the parameter lies within the given range. However, for a particular interval the true value
of the parameter is either in the given range or not. For example, if one took repeated
samples of size 100, calculated 95% confidence intervals from each sample, one expects that
on the average 95 out of the 100 calculated confidence intervals would contain the true
parameter.
Descriptive statistics: The mathematical calculations used to characterize the data collected
in a study; at this step no statistical inferences are examined. Some common descriptive
statistics are means, medians, standard deviations, ranges, minimums, maximums, and
confidence intervals.
Experimental unit: An entity to which a specific treatment is applied. Each unit has an
equal chance of being selected or assigned to a given treatment and units are treated and
observed independently.
Masking: A procedure to reduce potential experimenter and observer bias, both conscious
and unconscious, in which designated study personnel are kept uninformed of the treatment
assignments.
Randomization: A procedure to reduce potential assignment bias by assigning study
animals to treatment or control groups where each animal has an equally likely chance of
being chosen.
Randomized complete block design: A design in which the experimental units to which the
treatments are applied are separated into homogeneous groups of k experimental units, where
k represents the number of treatments. Within each group of k experimental units, the units
are randomly assigned so that each unit in the group receives a different treatment. The
groups are called blocks and the design is called a randomized complete block design.

Contains Nonbinding Recommendations
Page 28
APPENDIX 1 : CALCULATING UPPER AND LOWER SPECIFICATION
LIMITS FOR NUTRIENTS
You should select a laboratory that will conduct the chemical analyses of your uniform basal
diet. Once a laboratory has been selected, you should obtain the coefficient of variation
(CV)5 associated with each laboratory assay procedure. The CV, its justification, and a
reference to the methodologies for each assay should be stated in the protocol.
The specification limits (lower = LSL, upper = USL) for a diet formulated to contain a
theoretical (calculated) nutrient concentration is calculated as follows:
LSL= X - t(α/2,∞) * (CV * X)
USL= X + t(α/2,∞) * (CV * X)
where,
t(α/2, ∞) = t(.10/2, ∞) = 1.645,
CV = coefficient of variation associated with the chemical assay, and
X = theoretical (calculated) nutrient concentration.
For example, the 90% specification limit for a diet formulated to contain 0.50 % dietary
methionine, given a hypothetical CV of .12, would be calculated as follows:
LSL = 0.50% - 1.645 * (0.12 * 0.50%)
USL = 0.50% + 1.645 * (0.12 * 0.50%),
LSL 	= 0.50% - 0.10%
USL 	= 0.50% + 0.10%
Therefore, the calculated lower and upper specification limits for the nutrient concentrations
for methionine may range from 0.40% to 0.60% .
Because the t is equal to t(.10/2, ∞), the CV used should be based on a large number of samples
(greater than 30 samples). This CV should be obtained from the analytical laboratory that
will conduct the nutrient assays. Analytical laboratories that routinely conduct nutrient
analyses of feedstuffs should be able to provide you with CVs for each assay used.
For products labeled for a specific dietary nutrient concentration, the chemically determined
5 The CV is defined as the standard deviation divided by the mean.

Contains Nonbinding Recommendations
Page 29
nutrient concentration should fall within the specification limits. When a single assay is
conducted, the assay value should fall within the specification limits, as given above.
Because multiple assays of a single feed sample provide additional information on the
nutrient concentration, the assay values should be averaged and the specification limit should
be adjusted. In this situation, you can apply the adjusted specification limit method to the
data. This procedure will verify that the population from which the assay values are
collected is similar with respect to the mean and variance as the population used to establish
the assay coefficient of variation.
If there are multiple assays conducted on a composite sample, the specification limit should
be adjusted as follows:
LSLadj = X - t(α/2,∞) * (CV / √n * X),
USLadj= X + t(α/2,∞) * (CV / √n * X),
where
CV = the coefficient of variation, and
√n 	= the square root of the number of assays conducted on a single feed sample.

Contains Nonbinding Recommendations
Page 30
APPENDIX 2: STUDY CONCLUSION DECISION TREE
Is there an
acceptable level of
virulence?
Study not used in
the determination
of effectivenesss
No
Yes
Does the study meet the
predetermined success
criteria?
Study outcome =
not efficacious*
No
Study outcome =
efficacious*
Yes
*Both outcomes will be evaluated in the
overall determination of effectiveness.
The recommended success criteria are:
1) lesion score (chickens) or
fecal score (turkeys),
2) weight gain, and
3) mortality (if applicable).

Contains Nonbinding Recommendations
Page 31
REFERENCES
i Williams, R.B., 1996. The ratio of water and food consumption of chickens and its significance in the
chemotherapy of coccidiosis. Vet. Res. Commun. 20, 437- 447
ii Waletzky, E., 1970. Laboratory Anticoccidial Evaluation Trials: Review of Designs, Variables, Criteria, and
Predictive Value for Field Use. Experimental Parasitology. 28:42-62.
iii Johnson, J., and W.M. Reid, 1970. Anticoccidial drugs: Lesion Scoring Techniques in Battery and Floor-Pen
Experiments with Chickens. Experimental Parasitology. 28:30-36.
iv Holdsworth PA, Conway DP, McKenzie ME, Dayton AD, Chapman HD, Mathis GF et al. World Association
for the Advancement of Veterinary Parasitology (WAAVP) guidelines for evaluating the efficacy of
anticoccidial drugs in chickens and turkeys. Vet Parasitol 2004; 121(3-4):189-212.