ASTM E1207-14(2022)

This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles for the
Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.
Designation: E1207 − 14 (Reapproved 2022)
Standard Guide for
Sensory Evaluation of Axillary Deodorancy
This standard is issued under the fixed designation E1207; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision.Anumber in parentheses indicates the year of last reapproval.A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.
1. Scope sirable component of axillary odor is caused by degradation of
apocrine sweat by particular bacteria normally found in the
1.1 This guide provides procedures which may be used in
axillary vault.
the design and analysis of studies to quantitatively assess the
intensity of human axillary odor for the purpose of substanti- 1.4 Personal care products are sold and used primarily for
ating deodorant efficacy of personal care products. their ability to reduce the perception of body odor not only by
the individual using the product but also by individuals within
1.2 This guide includes protocols for the selection and
thescopeofcontact.Deodorantprotectionmaybeachievedby
training of assessors, selection of subjects, experimental
these products through various modes of action. Antiperspi-
design, and statistical analyses. This practice is limited to
rants achieve their primary efficacy by means of the action of
assessment of axillary odor by trained assessors. Self-
inorganic salts on the eccrine gland production of sweat.
evaluation protocols are valid for selected sensory tasks but
Antimicrobial agents achieve deodorancy by inhibiting the
may be less sensitive.
growth and activity of the microflora in the axillary vault thus
1.3 With respect to the source of axillary odor, three groups
reducing the microbial decomposition of sweat and the conse-
of secretory glands are present in the axillae which participate
quent production of body odor. Absorbents function either by
to a greater or lesser extent in its production—eccrine,
“binding” available moisture or malodorous substances. Fra-
apocrine, and sebaceous. Axillary odor has been primarily
grances are effective by altering the perception of malodor and
ascribed to the apocrine gland secretion (1). Body odor
increasing the degree of “pleasantness.” Other modes of
intensity has been correlated with the volume of the secretory
control become important from time to time, representing
portion of the apocrine gland (2) and the density of the glands.
changes in the state-of-the-art in product development.
1.3.1 Apocrine glands are found primarily in the axillary
1.5 The studies discussed herein are interpreted through the
vault in conjunction with axillary hairs (3). Pure apocrine
use of statistical tests of hypotheses. These hypotheses are
sweat is sterile and odorless and axillary odor results from
usually of the form:
degradation of apocrine sweat by resident skin bacteria (4).
The Deodorant Efficacy of Treatment A
High bacterial populations are found in moist regions of the
= The Deodorant Efficacy of Treatment B
body, especially in the axillae, providing the appropriate
1.5.1 It should be noted that failure to reject this hypothesis
environment for growth (5).
at a specified level of significance does not prove the
1.3.2 Eccrine glands keep the axillae moist through ther-
hypothesis,butmerelythattheweightofevidenceprovidedby
mally and emotionally induced secretions (6).
the experiment is not sufficient to reject the hypothesis. This
1.3.3 Thesebaceousglandsexcretehighermolecularweight
could occur because either: a) The hypothesis is close to truth
lipid materials which absorb and retain the volatile materials
and great experimental power would be required to reject it, or
resulting from bacterial action (7). The aerobic diphtheroids
b) The experiment by design was low in power and, therefore,
are able to produce the typical acrid axillary odor and the
incapable of rejecting the hypothesis; even when it is far from
micrococcaceae produce an isovaleric acid-like odor when
true.Thiscanoccurduetodesignstructureorlowsamplesize.
incubated with apocrine sweat (8). Therefore, the most unde-
These facts must be taken into consideration when interpreting
study results.
This guide is under the jurisdiction of ASTM Committee E18 on Sensory
1.6 This international standard was developed in accor-
Evaluation and is the direct responsibility of Subcommittee E18.07 on Personal
dance with internationally recognized principles on standard-
Care and Household Evaluation.
Current edition approved Oct. 1, 2022. Published October 2022. Originally
ization established in the Decision on Principles for the
approved in 1987. Last previous edition approved in 2014 as E1207–14. DOI:
Development of International Standards, Guides and Recom-
10.1520/E1207-14R22.
2 mendations issued by the World Trade Organization Technical
Theboldfacenumbersinparenthesesrefertothelistofreferencesattheendof
this standard. Barriers to Trade (TBT) Committee.
Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States
E1207 − 14 (2022)
2. Referenced Documents 3.1.11 trigeminal response—a sensation caused by stimula-
3 tion of the trigeminal nerve.The sensation is that of a physical
2.1 ASTM Standards:
feeling, such as burning and tingling.
E253Terminology Relating to Sensory Evaluation of Mate-
rials and Products
4. Summary of Guide
E1697Test Method for Unipolar Magnitude Estimation of
4.1 The protocols described provide for the designation of
Sensory Attributes
panels of individuals suitably selected and trained to perform
3. Terminology
the functions of assessors and subjects for the purpose of
assessing deodorant efficacy. Details of specific procedures are
3.1 Definitions of Terms Specific to This Standard:
given in Appendix X1 – Appendix X3. Deodorant products
3.1.1 Fordefinitionsoftermsrelatingtosensoryevaluation,
should be tested in a manner which maximizes test sensitivity
see Terminology E253.
while still reflecting normal consumer-use conditions. Ex-
3.1.2 5-alpha-androst-16-en-3-one (delta (5-alpha)
amples are provided to assist the investigator in the design and
androsten-3-one) C H O—CAS No. 18339-17-7—
19 28
performance of test protocols.
component of axillary odor which has a “urinous” character
and results from the action of certain skin bacteria on apocrine
5. Significance and Use
secretion (9).
5.1 The procedures recommended in this practice can be
3.1.3 5-alpha-androst-16-en-3-alpha-ol (delta (5-alpha)
usedtoclinicallyassessaxillarydeodorantefficacyofpersonal
androsten-3-alpha-ol) C H O—CAS No. 14152-27-3—
19 30
care products.
componentofaxillaryodorwhichhasa“musky”characterand
results from the action of certain skin bacteria on apocrine
5.2 This practice is applicable to the product categories
secretion (9).
whichincludedeodorantandtoiletsoapbars,liquidbathsoaps
and gels, deodorant sticks, antiperspirants, creams and lotions,
3.1.4 apocrine gland—a highly coiled tubular system found
body talcs, and aerosol and pump delivery deodorants,
primarily in axillary epidermis. These glands continuously
antiperspirants, and body colognes.
produce and store apocrine sweat for later excretion onto the
skin surface via hair follicles. The excretion is activated by
5.3 Procedures of the type described herein may be used to
androgenic sympathetic stimuli such as pain or fear (1).
aid in the communication of efficacy within and between
manufacturers and to the consumer through the various public
3.1.5 deodorant effıcacy—the effectiveness or treatment, or
both, of a product in reducing axillary malodor. communications media. Guidelines are suggested due to the
need to determine the relative or absolute performance of
3.1.6 eccrine gland—a simple unbranched tube with a
experimental materials or of commercial products.
terminal coil.These glands are found in the epidermis over the
entire body surface. The glands are controlled by the auto- 5.4 These procedures may be used by persons who have
nomic nervous system and serve as an evaporative cooling familiarized themselves with these procedures and have had
mechanism. Although heat is the primary stimulus, localized previous experience with sensory evaluation.
eccrine sweating can also occur as a result of emotional stress
5.5 This practice provides suggested procedures and is not
and other physiological stimuli (3).
meant to exclude alternate procedures which may be effec-
3.1.7 IVA, isovaleric acid (3-methylbutanoic acid)
tively used to provide the same clinical result.
C H O;(CH ) CHCH COOH. CAS No. 503-74-2—
5 10 2 3 2 2
6. Subject Selection and Restrictions
component of axillary odor which has a “sweaty, acid”
characterandresultsfromtheactionofcertainskinbacteriaon
6.1 Criteria for Selection—The population should be de-
apocrine secretion.
fined and subjects selected from this population in a random,
and unbiased manner according to the experimental design
3.1.8 right-left imbalance—a condition of some subjects
who have one axilla with notably more intense odor than the considerationsdefinedin8.11.Ifatestisbeingperformedwith
the product directed at a subset of the consuming population,
other axilla as determined from the control odor evaluation.
the subjects should be selected from a population representa-
3.1.9 sebaceous gland—a gland closely related to the hair
tive of the subset.
follicle which produces sebum which combines with apocrine
6.1.1 The subjects should have a recognizable body odor
secretionatthebaseofthefollicle.Sebaceousglandsareunder
level when evaluated under the procedures given in this
androgen control (6).
practice.
3.1.10 sequential analysis—a statistical technique which
6.1.2 In situations where it is desirable to enhance test
may be used to screen potential assessors for sensory acuity to
sensitivity, the following criteria may be adopted:
aspecificstimulus.Theassessorisrepeatedlytesteduntilheor
6.1.2.1 Based on the control odor scores (see 8.3), subjects
she passes or fails the test at a specified level of significance
who have low or extremely high odor should not be selected
(10, 11).
for the test. Subjects may be considered as having a “high”
odor relative to a normal population if they develop an odor
For referenced ASTM standards, visit the ASTM website, www.astm.org, or
score in excess of 7.0 on a 0- to 10-point scale or 3.5 on a 0-
contact ASTM Customer Service at service@astm.org. For Annual Book of ASTM
to 5-point scale. Likewise, subjects may be considered as
Standards volume information, refer to the standard’s Document Summary page on
the ASTM website. having a “low” odor relative to a normal population if they
E1207 − 14 (2022)
developanodorscorebelow3.0ona0-to10-pointscaleor1.5 6.2.2.1 Subjects should use no perfumed substances on the
on a 0- to 5-point scale. A selection process which excludes body such as perfume, after shave, lotions, bath oils, and
“low”odorsubjectsor“extremelyhigh”odorsubjects,orboth, hairspray.
mustbespecifiedforeachtestanddependsuponthenumberof 6.2.2.2 Pre-laundered wearing apparel (see 8.6) may be
subjects required for the test and the relative odor scores of worn by each subject at the option of the test sponsor. Shirts
these subjects. should be collected and laundered in accordance with a
uniform laboratory procedure.
6.1.2.2 Thereshouldbenomorethanasmallright-leftodor
imbalance between axillae of each subject. On the basis of a 6.2.2.3 If specified by the test sponsor, laundry additives
category, or interval scale, the consensus of the task group was such as bleach, fabric softeners, etc., may be used on subjects’
that the control odor score differential should not be greater outer clothing.
than 20% of the overall scale (that is, 2.0 points on a 10-point
6.2.2.4 Subjects should minimize physical exertion such as
scale or 1.0 points on a 5-point scale). tennis and jogging.
6.1.2.3 Appendix X1contains additional information on the 6.2.2.5 Subjectsshouldrefrainfromtheuseofbreathmints,
acceptance/rejection history of experimental subject popula- toothpaste, mouth rinses and sprays, chewing gum, and from
tions.Aselection process which excludes approximately 20% drinking coffee or tea at least 1 h prior to each evaluation.
of the lowest odor intensity individuals of a normal population Smoking should be restricted 2 h prior to each evaluation and
is generally recognized as appropriate. alcoholic beverages 8 h before an evaluation.
6.1.3 Chronic medications such as antibiotics, steroids, etc., 6.2.2.6 Subjects should not wash the axillae at home for the
which may affect the test, should be restricted during all test duration of the test. Axillae should only be washed at the test
phases as deemed appropriate by the sponsor. site in accordance with a supervised wash procedure. Care
6.1.4 In addition to the above restrictions it should be should be taken not to get the axillae wet during bathing or
showering at home.
recognized that other factors which contribute to protocol
operating efficiency should be emphasized, including interest,
cooperation, commitment, and punctuality of the subjects. 7. Assessor Selection and Training
6.2 Subject Restrictions—In order to achieve appropriate 7.1 General—Theselectionprocessshouldincludetheprin-
experimental control, the following restrictions should be ciples embodied in Ref (12). The assessor’s task is to detect
imposed upon all subjects during the conditioning and test differences and rate the intensity of perceived axillary odor.
phases.
7.2 Assessors employed for assessing body odor intensity
6.2.1 Conditioning Phase—This period is often referred to
should be screened for the following attributes:
as the “washout” period and is that portion of the protocol
7.2.1 Interest and availability;
preceding the actual test phase. The duration of the condition-
7.2.2 Qualitative and quantitative olfactory discrimination
ing phase should be a minimum of 7 days. The conditioning
ability;
phase for antiperspirants shall be 17 days as defined by the
7.2.3 Ability to carry out basic sensory tasks, and compe-
FDA monograph on antiperspirants (11).
tency with the scale used, and
6.2.1.1 Subjects should use no antiperspirants, deodorants,
7.2.4 Specificanosmias.Whileitisdesirabletoidentifyany
antibiotic creams, antibacterial ointments, or any other cos-
olfactory deficit which an assessor may have, there is experi-
metic products on the axillae. No antibacterial products,
ence which indicates that specific anosmias may not detract
including deodorant and medicated shampoos should be used.
from accurate odor judgments. (See X2.6.3.)
Careshouldbetakennottoexposetheaxillaetoanymedicated
7.3 Recommended procedures are presented in Appendix
product or product containing alcohol.
X2 for the screening and selection of in vivo deodorancy
6.2.1.2 Subjects should use only the control cleansing
assessors.
agent(s) provided by the sponsor as instructed for personal
hygiene.
7.4 Assessor Training—In addition to the following points,
6.2.1.3 Swimmingshouldbestoppedatleast7dayspriorto
therecommendedproceduresaregiveninAppendixX3forthe
the test phase and during the entire test phase.
training of in vivo deodorancy assessors.
6.2.1.4 Subjects who normally shave their axillae should
7.4.1 Assessors should be exposed to the complete range of
shave using the control cleansing agent no less than 24 h prior
quantitative and qualitative malodor stimuli which they will
to the control evaluation and abstain from shaving for the later be asked to rate. This establishes the context in which
duration of the test.
ratings are to be assigned.
6.2.1.5 Spicy foods, including garlic and onions should be
7.4.2 Assessor Training for Category Scales:
restricted24hbeforethecontrolevaluationandduringthetest
7.4.2.1 Afterbeingintroducedtotheratingscaleprocedure,
phase.
assessors should assign ratings to the stimuli in an open
6.2.1.6 It is acceptable to use smokers as subjects, but they discussion to obtain a consensus rating for each stimulus.
are required to refrain from smoking for 2 h before all 7.4.2.2 Assessors should be drilled until the ratings they
evaluations.
independently assign match those obtained by consensus as
6.2.2 Test Phase—In addition to the conditions detailed for closely as possible.Assessors whose ratings disagree with the
the subjects during the conditioning phase, the following consensus rating much more often than those of most other
restrictions are required of the subjects during the test phase: assessors should be eliminated. The criteria for rejection of
E1207 − 14 (2022)
individual assessors must be developed in each laboratory. For anticipated claim substantiation documentation required. Fre-
example, the responses for each assessor can be graphed to quently used post-treatment evaluation intervals are 5, 8, 12,
determine if they fall within a specified range across time. and 24 h.
8.5 Duration of Test Period (Treatment Cycle Duration)—
7.5 Assessor Performance Monitoring—Trained assessors
During the test phase of the study the subjects are treated with
should be tested periodically to confirm their ability to dis-
one or more designated test products and evaluated for odor
criminate(rankings,pairedcomparisons,ratingscanbeusedas
level. Individual product test periods range from 1 to 21 days
appropriate). In order to evaluate rating performance, it is also
depending upon the test objective, the test sensitivity desired,
important to evaluate within- and between-assessor consis-
the product formulation, and the expected end-product use
tency.Onamoreroutinebasis,treatmentsusedforthepurpose
conditions. Generally, 3 to 5 sequential test days will provide
of scale anchors or reference standards can be included in the
sufficient data to document performance claims.
regular testing regimen as “unknowns” to determine if asses-
sors are capable of rating these products consistently. The
8.6 Wearing Apparel—Forstudiesinwhichwearingapparel
procedure for monitoring assessor performance should be
istobecontrolled,shirtsofuniformfibercontent,eithercotton
carried out at least once a year. More frequent monitoring may
or a cotton-polyester blend, but not nylon, should be used.
be required if there is some reason to suspect an assessor’s
ApparelstylemaybeeitherT-shirtsordressshirts.Allwearing
olfactory acuity. (See X3.3.)
apparel should be laundered immediately prior to use using an
unfragranced detergent base. Each subject should be issued a
8. Test Design
fresh shirt after each product application to be worn at least
through the first evaluation point. If successive evaluations are
8.1 Subject Enrollment—A sufficient number of subjects
made between applications, the test sponsor should determine
should be enrolled for the conditioning phase so that the
ifthesameshirtistobeworn,afreshonetobeissued,orifthe
required number of subjects complete the study. The number
subjects are to be allowed to assume normal clothing habits.
enrolled will depend upon the history of the laboratory and the
8.7 Product Assignment—Testproductsshouldberandomly
specificselectioncriteriaforthetest.Ingeneral,itissuggested
assigned to right and left axilla such that each product is
that at least 20% more subjects be recruited than will be
applied to an equal number of right and left axillae. Specific
needed.Eachsubjectshouldbeinformedoftheresponsibilities
experimental designs are given in 8.11.
and obligations of the subjects, provided with a copy of the
restrictions and advised of any regulations and consent appli-
8.8 Test Product Application:
cable under the proposed good clinical practices and any
8.8.1 For deodorant sticks, gels, creams and lotions, body
applicable regulations covering the obligations of sponsors/
talcs, aerosol and pump delivery deodorants and body
investigators.
colognes, the axillae should be cleansed prior to treatment
using a control cleansing agent. It should be determined that
8.2 Conditioning Phase—Each subject should adhere to the
such treatment does not impart a residual odor or produce a
restrictions given in 6.2.1. Each subject should be provided
false treatment effect. Deodorant and toilet soaps and liquid
with the appropriate control cleansing products for personal
bath soaps and gels provide for normal axillary cleansing
hygiene at home during this phase which are to substitute for
during the application process.
products normally used, such as liquid soap, bar soap, and
8.8.2 All axillary treatments during the test phase should be
shampoo, or all three. These products should contain no
monitored by a test supervisor. The level of supervision
antimicrobial ingredients and a minimum level of perfume or
depends upon the experience and number of subjects involved
no perfume.
and the product tested.
8.3 Control Odor Scores—This evaluation is conducted to
8.8.3 Specific recommendations for each product category
determine baseline axillary odor scores for each subject fol-
application condition are given in Appendix X4.
lowing a supervised control wash using the control cleansing
8.9 Test Product Evaluation:
product. The purpose is to uniformly condition the subjects’
8.9.1 This is an example of one specific method of evalua-
axillae prior to the control evaluation. Subjects may then be
tion. Odor assessors are positioned in isolated evaluation
screened from the test if they have unacceptably low or high
stations in the odor evaluation room. Subjects (equal to the
odor or have an accentuated right-left imbalance (6.1.2.2).The
numberofassessors)entertheroomandrandomlyreporttothe
time interval between the control wash and the control evalu-
assessors’ stations so that each assessor has a subject to
ation should be the same as the longest time interval between
evaluate.The subjects stand in front of the designated assessor
test product application and axillary odor evaluation.The soap
with their arms held at their sides for 1 min.At the completion
used for the control wash should be the same as the one used
of the 1-min interval, a signal is given and the assessors
by the subject during the conditioning phase. The specified
evaluate the subjects in front of them, right arm first followed
numberofsubjectswillbeselectedontheircontrolodorscores
bytheleftarm(procedureofrightthenleftisheldconstantfor
in accordance with the selection criteria detailed in 6.1.
all subsequent evaluations). During evaluation, subjects raise
8.4 Post-Treatment Evaluation Interval—The post- their right arms and then place their right hands behind their
treatment evaluation interval may range from immediately heads. Each assessor takes a sniffing cup (cone-shaped 5-oz
aftertreatmentto30minto48h,ormore.Thespecificinterval paper cup with the pointed end cut off) and places the larger
will be based upon the expected end-product use and the opening of the cup in the center of the right axilla and then
E1207 − 14 (2022)
NOTE 1—It is generally recognized that assessors find it difficult to
sniffs the circumscribed area through the opening at the back
psychologically accommodate more than 10 or 11 points in a scale. With
endofthecone.Eachassessorrecordsthescoreintotherecord
scales consisting of a greater number of points, assessors may stay in one
form while the subjects lower their arms. This procedure is
portion of the scale without using all points available, thereby reducing
repeated for the left arm. The subjects advance to the next
consistency and adding confusion to the evaluation process. However,
designated assessor and the sniffing process is repeated. Once
scales consisting of a larger number of points reduce discontinuity in the
data. Thus, a scale of approximately 10 intervals offers a good compro-
all the subjects in the first group have been evaluated by each
mise between these two considerations. The problem of having consecu-
assessor, this group of subjects is released from the evaluation
tivescoresrepresentconsistentpsychologicaldifferencesacrosstheentire
area and the next group of subjects is brought into the room.
scale may not be overcome by assessor training. However, in practical
8.9.2 Assessors are given breaks after approximately every
terms, these slight distortions are not viewed to be a serious detriment to
20 evaluations, both arms of 10 subjects. Each judge uses a applying statistical analysis to category scales in deodorancy testing.
Category scales provide a heuristic approach to the evaluation of
new sniffing cup for each evaluation.
deodorancy odor which has stood the test of time, and are widely held to
8.9.3 Environmental conditions should be cool room tem-
be an appropriate response variable to which statistical analysis can be
peratures (68°F) with sufficient air flow but no drafts.
applied.
8.10 Odor Assessment Rating Scale—Category scaling is 8.11 Experimental Design Considerations—Includeuniden-
very commonly used to rate axillary odor intensity but any tified controls within the test design. This will help to check
scale used in sensory evaluation to rate intensity, including assessor performance and may shed light on anomalies within
magnitudeestimation(seeTestMethodE1697),isappropriate. the test.
8.10.1 Category Scaling of Axillary Odor: 8.11.1 Introduction to Relevant Experimental Designs—Let
T , T , . T symbolize t deodorant treatments. These may
8.10.1.1 Introduction—This section describes the use of
1 2 t
include: commercial products, experimental substances, pla-
category scaling as one subjective rating method for axillary
cebo formulations, or a null treatment (an “untreated side”).
malodormeasurement.Categoryscalesaretheoldestandmost
8.11.1.1 The three experimental designs commonly used in
frequently used scaling methods for subjective evaluations.
deodorant clinical tests are the Single Pair (1PR) Design, the
The use of category scales for the measurement of axillary
Each versus Control (EVC) Design and the Round Robin
malodor was reported in 1967 (13). The deodorancy assessors
(RRB) Design. Examples of the treatment assignment for each
for the studies by Whitehouse and Carter useda0to10 point
are shown in Table 1.
scale, with “0” meaning no odor, and “10” meaning extremely
8.11.2 Single Pair (1PR) Design—This design is applicable
strong odor. This section discusses background, applications
whenonlytwotreatmentsarecompared.Eachsubjectreceives
and statistical considerations in using category scales for
either T on the left axilla with T on the right axilla or T on
axillary odor evaluations.
1 2 2
the left with T on the right. The assignment of treatments to
8.10.1.2 Background—Category scales applied to deodor-
axillaeisrandomizedinsuchawaythateachtreatmentappears
ancytestingconsistofaseriesofconsecutivenumbers,eachof
anequalnumberoftimesoneachaxillae(orasneartoanequal
whose values represent a “level of odor.” Two common
number of times as possible).
category scales applied in deodorancy testing are [0, 1, 2, 3, 4,
8.11.3 Each Versus Control (EVC) Design—This design is
5, 6, 7, 8, 9, 10] and [0, 1, 2, 3, 4, 5].
applicable when three or more treatments are to be compared,
8.10.1.3 Considerations which arise in the application of
andoneofthetreatments,symbolizedby T,canbesingledout
category scales to deodorancy testing include the following. t
as the control treatment. Carefully consider the choice of the
Assessors may tend to use only the low end or the high end of
control sample. It may be a different treatment, unfragranced
the scale, and not use the entire scale, thus skewing the
base, treatment with water, or no treatment. The remaining
distribution. There is often an inherent tendency on the part of
treatments, T , T , . T−1, are termed test treatments. Each
some assessors not to use the endpoints of the scale. The 1 2 t
subject receives the control treatment on one axilla and one of
distributionofcategoryscalesisdiscreteinnature,whereoften
the t−1 test treatments on the other axilla. Each test treatment
the distribution assumed by the statistical analyses applied is
is randomly assigned to an approximately equal number of
continuous. The psychological difference between two con-
subjects. The assignment of treatments to the left and right
secutive categories may vary, depending upon their location in
axillae is random, but balanced so that each treatment appears
the scale.
the same number of times on the left as it appears on the right
8.10.1.4 Application—Steps may be taken to diminish some
of the difficulties encountered in the use of category scales.
Training assessors to use the entire scale can reduce problems
TABLE 1 Examples of Treatment Assignment for Three
of skewness and tend to make assessors more consistent with
Deodorant Clinical Study Designs
each other in their evaluations. Having assessors compare
Single Pair Each vs. Control Round Robin
scores during training sessions will also improve consistency.
Subject Left Right Subject Left Right Subject Left Right
As assessors gain experience with a particular scale, they tend 1 T T 1 T T 1 T T
1 2 1 3 3 2
2 T T 2 T T 2 T T
2 1 3 1 3 1
tomentallyanchorthescorestoparticularodorlevels.Another
3 T T 3 T T 3 T T
2 1 3 2 1 2
means of improving consistency is to train assessors using
4 T T 4 T T 4 T T
2 1 1 3 2 3
calibrated samples of odor as reference points for each cat- 5 T T 5 T T 5 T T
1 2 3 2 1 3
6 T T 6 T T 6 T T
1 2 2 3 2 1
egory. To reduce problems of discontinuity, it is advisable to
7 T T
3 1
use several assessors (at least three) and take the averaged
8 T T
2 3
scores as the estimate of odor for a particular axilla.
E1207 − 14 (2022)
or as near to the same number of times as possible.Agroup of comparing only one of those treatments with the control in a
subjects all of whom receive the same pair of treatments single pair design, requires that the experimenter use a sample
(ignoring left/right assignment) is termed a cell. The EVC size equal to the number of test treatments (excluding the
design has t−1 cells. control)multipliedbythenumberofpanelistshewouldusefor
8.11.4 Round Robin (RRB) Design—The RRB design is thesinglepairstudy.Iftheexperimenterwouldliketocompare
applicable when three or more treatments are to be compared each test product with another (as opposed to testing the test
but none of them can be singled out as a control treatment. productwiththesinglecontrol)withthesameprecisionasthat
There are t(t−1)/2 possible pairings of t treatments (for obtained in a single pair study, then he must use two times the
example, the three treatments, T , T , and T , generate the number of test treatments (excluding the control) times the
1 2 3
3(3−1)/2=3 pairs T T , T T , and T T ). In the RRB design number of panelists he would use in the single pair study.
1 2 1 3 2 3
each of the t(t−1)/2 possible pairs is randomly assigned to an (b) Round Robin Design—To obtain the same precision
approximately equal number of subjects. As in the other between all pairs of products in a round robin design that
designs, the assignment of treatments to the left and right would be obtained by testing two of those products in a single
axillae is random but balanced, so that each treatment appears pair design requires that the experimenter use a sample
on the right the same number of times as on the left or as near approximately equal to “(t −1)” times the number of panelists
to the same number of times as possible. Clearly, there are used in the single pair design, where “t” is the total number of
t(t−1)/2 cells in a RRB design. products being compared (see Appendix X5).
8.11.5 Order of Evaluation—The order in which the asses- 8.11.6.4 Determiningsamplesizecanbedifficult,especially
sors evaluate the subjects’ axillae, either left first then right or in cases where no prior information about the products being
rightfirstthenleft,isheldconstantthroughoutanystudy;thus, tested is available. In this case, it is probably better to
the effect of presentation order cannot be estimated indepen- overestimateratherthanunderestimatethesamplesize,thereby
dentlyofleft/righteffects.Onlythesumofthetwoeffectsmay achieving the power required (see Appendix X5).
be estimated.
9. Biasing Effect of Fragrances
8.11.6 Choice of Sample Size:
8.11.6.1 Background—The choice of sample size is an 9.1 Odor assessors are trained to assign ratings to the
important one, directly affecting the power and the cost of a
intensity of axillary malodor ignoring any fragrance or base
study. The greater the sample, the more power achieved, and odoroftheaxillarytreatment(seeX3.2.2).Instudieswhereall
the greater the cost. Below are given some general guidelines axillary treatments have the same fragrance, any effects these
for choice of sample size in deodorancy studies. See Refs fragrances may have upon the ratings of axillary malodor
(14-16) for technical discussions of sample size determina- intensity will be the same for all axillary treatments and,
tions. therefore, will not bias estimates of the differences in deodor-
8.11.6.2 In general, deodorancy studies will involve 30 to ant efficacy of the treatments. In studies where there are
60 subjects per treatment pair, depending upon the analysis noticeable differences in the fragrances of the axillary
used and the power required. Depending upon the application, treatments, the structure of the studies described herein does
one might require as few as 20 panelists for rough not preclude the possibility that estimates of the differences in
approximations, or as many as 100 or more panelists for deodorant efficacy of the products will be biased by the
studies involving many products and requiring high power. If fragrance differences, that is, the assessors can’t be fully
the experimenter, based on past experience, knows that the blinded when the axillary treatments have noticeably different
particular products being tested generally show large differ-
fragrances.
ences in efficacy, then a smaller sample may be more cost 9.1.1 Some of the possible biasing effects are given in
effective.Ontheotherhand,ifhesuspectsthattheproductsare
9.1.1.1 – 9.1.1.3.
quite close in deodorant efficacy, then he will want to increase 9.1.1.1 Recognition Effect—The effect of recognizing the
the sample size to enhance the power of the study so that he
identity of the fragrances as those of commercially available
will be more likely to detect the differences between the products.
products, if in fact meaningful differences exist (see 1.5). A
9.1.1.2 Affective Effect—The effect of differences in the
pilot study may be used to determine sample size needs. pleasantness of the fragrances.
8.11.6.3 If the experimenter is testing more than two 9.1.1.3 Expectation Effect—The effect of learning part way
products, and knows the approximate sample size (for the through the study that some fragrances are usually associated
power required) were he testing only two of these products, withlower(orhigher)malodorsothat,bythelatersubjects,the
using the single pair (1PR) design, the following gives the assessors begin to expect lower (or higher) malodor ratings
correct sample size to use for both the Each versus Control when those fragrances are recognized.
(EVC) and the Round Robin (RRB) design: 9.1.2 The potentially biasing effects of axillary treatment
(a) Each Versus Control Design—To achieve the same fragrances are not precluded by the design of these studies;
precision(standarddeviation)incomparingeachofseveraltest however, there is no known alternative test method for assess-
treatments with a single control that would be obtained by ing axillary deodorant efficacy.
E1207 − 14 (2022)
APPENDIXES
(Nonmandatory Information)
X1. SUBJECT ACCEPTANCE/REJECTION HISTORY
TABLE X1.1 Distribution of Accepted/Rejected Subjects
X1.1 General—In an attempt to demonstrate the historical
acceptance of subjects onto deodorancy tests, 25 prior studies Number of Tests 25
Number of Odor Assessors 4
were reviewed from 1 source and 4 studies from an additional
Total Number of Subjects Screened 1066
source.
Total Number of Subjects Accepted 845
Scoring Scale 0–10
X1.1.1 The data presented are for control odor evaluations
Range of Average Control
thatarecarriedoutpriortoacceptanceontothestudy.Subjects Right Axilla, % Total Left Axilla, % Total
Odor Scores
havebeenthroughseveraldaysofabstinencefromdeodorants,
0 0.0 0.0
0.1–1.0 0.2 0.1
antiperspirants,anddeodorantsoaps.Inaddition,a24-hperiod
1.1–2.0 1.1 1.1
has occurred since the axilla have been washed with a
2.1–3.0 6.6 7.5
non-deodorant soap.
3.1–4.0 14.4 15.5
4.1–5.0 27.4 24.2
X1.1.2 The scoring scales used to rank the axillary odor
5.1–6.0 30.5 29.2
were as follows: 6.1–7.0 17.2 16.9
7.1–8.0 2.6 4.5
X1.1.2.1 25 Studies:
8.1–9.0 0.5 0.3
0=No axillary malodor
9.1–10.0 0.0 0.0
10=Very strong and disagreeable malodor
(1 point units are used to rank the scale from 0
to 10)
X1.3 Subjects may be considered as having a “high” odor
4 Studies:
relativetoanormalpopulationiftheydevelopanodorscorein
0=No axillary malodor
excess of 7.0 on a 0- to 10-point scale or 3.5 on a 0- to 5-point
5=Very strong and disagreeable malodor
scale. Likewise, subjects may be considered as having a “low”
( ⁄2 point units are used to rank the scale from 0
odor relative to a normal population if they develop an odor
to 5)
score below 3.0 on a 0- to 10-point scale or 1.5 on a 0- to
X1.2 The following tables show the distribution of the 5-pointscale.Ingeneral,theright-leftodorimbalancebetween
axillae of each subject should be no more than 30% of the
accepted/rejected subjects. The basic criteria for acceptance
was the highest average scores for those subjects presenting overall scale (that is 3.0 units on a 0- to 10-point scale or 1.5
units on a 0- to 5-point scale).
themselves for the control odor evaluation.
E1207 − 14 (2022)
TABLE X1.2 Distribution of Accepted Subjects
Number of Tests 4
Number of Odor Assessors 3
Total Number of Subjects Screened 310
Total Number of Subjects Accepted 310
Scoring Scale 0–5
Range of Average Control
Right Axilla, % Total Left Axilla, % Total
Odor Scores
0–0.4 0.0 0.2
0.5–0.9 0.3 0.2
1.0–1.4 6.0 5.3
1.5–1.9 10.0 9.0
2.0–2.4 21.6 19.0
2.5–2.9 17.5 14.0
3.0–3.4 14.5 14.0
3.5–3.9 12.6 14.0
4.0–4.4 7.5 13.0
4.5–4.9 7.0 5.3
5.0 3.0 6.0
X2. RECOMMENDED PROCEDURE FOR SCREENING AND SELECTING IN VIVO (LIFETIME) DEODORANCY ASSES-
SORS
X2.1 Purpose—The purpose of this series of tests is to products, deodorant product perfumes and various levels of
screen people who are interested in becoming deodorancy IVA mixed with deodorant product perfume solutions. Sug-
assessors. The screening is for olfactory acuity, specific anos- gested tests include paired difference, triangle, duo-trio and
mia to androstenone and androstenol, and interest and avail- ranking. Procedures for these and other tests may be found in
ability for testing. To accomplish this purpose, the screening Ref (17) or any basic text on sensory evaluation.An adequate
should be divided into two phases, conducted in two different number of tests should be conducted to ensure that the
sessions. olfactory acuity of the panelists is accurately assessed.Appro-
priate samples should be determined for the selected tests.The
X2.2 Panelist Recruitment—An adequate number of panel-
probability levels for each sample set of each test should be
ists should be recruited for Phase I testing based on the
established by pretesting and should be no greater than 0.05.
assumptionthathalfofthepeoplewillpassPhaseItestingand
X2.5.3 Ranking Test—The following three ranking tests are
move on to Phase II testing, from which assessor trainees will
suggested:
be selected. These people should be interested in becoming
X2.5.3.1 The first test is a series of five standard Isovaleric
deodorancy assessors and should be available during the times
Acid (IVA) concentrations prepared in distilled water:
deodorancy evaluations are conducted.
A
Sample Number IVA Concentration (mL/L)
X2.3 Pre-Screening Questionnaire—A brief questionnaire
736 0.014
should be administered to the panelists to determine sex, age,
951 0.058
and other information needed to confirm their willingness and
458 0.22
602 0.89
availabilitytoparticipateindeodorancyevaluations.Questions
059 3.57
updating the assessor’s general health status and smoking
__________________
habits should be included.
A
The assignment of a three-digit code to each test material will follow a computer
X2.4 Test Location and Scheduling—Testing should be
randomized sequence for all test samples in order to keep the odor assessors from
identifying a number series as a specific sample.
conducted in individual booths with adequate ventilation to
preventtheinfluenceofextraneousodorsandsamplecarryover
X2.5.3.2 The second test is the above series of IVAconcen-
effects on test performance. If sample sets are to be reused, the
trations with the addition of 0.1 mL of degassed aerosol
panelists should wear odor-free plastic gloves to prevent
deodorant (commercially available).
contamination of the sample container. The tests should be
X2.5.3.3 The third test is the standard IVA series with the
scheduled to minimize panelist fatigue and permit the ventila-
addition of 0.1 mL of a 5% solution of a fragranced bar soap
tion of odors from the test area.
(commercially available).
X2.5.3.4 The three tests are administered in sequence to the
X2.5 Phase I:
odorassessor.Theassessorispresentedthefivesamplesandis
X2.5.1 Purpose—Usingsequentialanalysis,potentialasses-
requested to rank the samples in order from highest to lowest
sors will be screened for olfactory acuity using deodorant
intensity. Samples should be presented at room temperature in
products and isovaleric acid (IVA).
a randomized and balanced order using procedure appropriate
X2.5.2 Acuity—Avariety of tests may be used to screen the fortheselectedtest.Eachsamplesetshouldbepresentedtwice
sensitivityofpaneliststofragrancematerialsusedindeodorant to obtain 10 to 20 evaluations from each panelist.
E1207 − 14 (2022)
X2.5.3.5 Assessor performance in the ranking tests can be best on the tests and who are the most interested in and
evaluated according to pre-established criteria. (Transposition available for deodorancy evaluations. Although the relation-
of two adjacent odor concentrations may represent a 1-point shipbetween in vitroand in vivodeodorancytestshasnotbeen
odor scale difference and not be significant. However, random established, the performance of the assessors on these screen-
ranking of more than two concentrations would indicate ing tests should not fall below a predetermined minimum. It is
difficultywiththeoverallperformance.)Potentialassessorsare recommended that an adequate pool of assessors be retained
accepted or rejected on the basis of their performance as for deodorancy testing.
determined by the percent correct responses or sequential
X2.6.6 Assessors who are to use magnitude estimation
analysis (18).
should be screened for competency with the scale by having
X2.5.4 A screening test which may also be suitable for themratelinelengths,someofwhicharesosmallthattheyare
determining smell function and measuring assessor acuity has forced to use values between 0 and 1 and some which are so
been prepared (19). close in length that they are forced to use values between two
integers (for example, 3.25). Studies have shown (21) that
X2.6 Phase II:
magnitude estimation ratings assigned to visual line lengths
X2.6.1 Purpose—The purpose of this series of tests is to will virtually be proportional to their actual lengths.Assessors
whoseratingsdepartmarkedlyfromthisrelationshipshouldbe
further screen the olfactory acuity of the panelists who passed
Phase I testing and to monitor the panelists for specific eliminated,especiallyifthereasonforthedepartureisdiscom-
fort with or inability to use decimal fractions.
anosmias relevant to deodorancy testing such as androstenone,
androstenol, isovaleric acid, selected members of the methyl
X2.7 The following protocol is suggested for pretest laun-
ionone family and synthetic and natural musks.
dry of wearing apparel:
X2.6.2 Test Sample Selection—Test samples should be se-
X2.7.1 When using a U.S. style top-loading machine select
lected accordin
...