ORIGINAL ARTICLES

Acne severity grading: Determining essential clinicalcomponents and features using a Delphi consensus

Jerry Tan, MD,a Barat Wolfe, MA,b Jonathan Weiss, MD,c Linda Stein-Gold, MD,d Joseph Bikowski, MD,e

James Del Rosso, MD,f Guy F. Webster, MD,g Anne Lucky, MD,h Diane Thiboutot, MD,i Jonathan Wilkin, MD,j

James Leyden, MD,k and Mary-Margaret Chren, MDl

Windsor, Ontario, Canada; Atlanta, Georgia; Detroit, Michigan; Columbus and Cincinnati, Ohio;

Las Vegas, Nevada; Philadelphia and Hershey, Pennsylvania; and San Francisco, California

From

D

D

A

to

H

M

og

og

H

of

of

Fund

Background: There are multiple global scales for acne severity grading but no singular standard.

Objective: Our objective was to determine the essential clinical components (content items) and features(property-related items) for an acne global grading scale for use in research and clinical practice using aniterative method, the Delphi process.

Methods: Ten acne experts were invited to participate in a Web-based Delphi survey comprising 3 iterativerounds of questions.

Results: In round 1, the experts identified the following clinical components (primary acne lesions,number of lesions, extent, regional involvement, secondary lesions, and patient experiences) and features(clinimetric properties, ease of use, categorization of severity based on photographs or text, andacceptance by all stakeholders). In round 2, consensus for inclusion in the scale was established forprimary lesions, number, sites, and extent; as well as clinimetric properties and ease of use. In round 3,consensus for inclusion was further established for categorization and acceptance. Patient experienceswere excluded and no consensus was achieved for secondary lesions.

Limitations: The Delphi panel consisted solely of the United States (U.S.)ebased acne experts.

Conclusion: Using an established method for achieving consensus, experts in acne vulgaris concluded thatan ideal acne global grading scale would comprise the essential clinical components of primary acnelesions, their quantity, extent, and facial and extrafacial sites of involvement; with features of clinimetricproperties, categorization, efficiency, and acceptance. ( J Am Acad Dermatol 2012;67:187-93.)

Key words: acne; consensus; Delphi; global severity; grading.

Abbreviations used:

CASS: Comprehensive Acne Severity ScaleECLA: Echelle de Cotation des L�esions d’Acn�eFDA: Food and Drug AdministrationGAGS: Global Acne Grading SystemIQR: interquartile range

INTRODUCTIONSeverity is the most important clinical feature

upon which clinicians establish recommendationsfor acne treatment.1 Unlike research settings, thepreferred method for acne severity assessment in

the Department of Medicine, University of Western Ontarioa;

epartment of Psychology, University of Windsorb; Gwinnett

ermatology PC and Gwinnett Clinical Research Center, Inc,

tlantac; Dermatology, Henry Ford Hospital, Detroitd; Derma-

logy, Ohio State University Medical Center, Columbuse; Valley

ospital Medical Center, Las Vegasf; Dermatology, Jefferson

edical College, Philadelphiag, Division of Pediatric Dermatol-

y, Cincinnati Children’s Hospitalh; Department of Dermatol-

y, The Pennsylvania State University College of Medicine,

ersheyi; unaffiliatedj; Department of Dermatology, University

Pennsylvaniak; and Department of Dermatology, University

California San Francisco.l

ing sources: None.

Conflicts of interest: A full list of disclosures for all authors may be

found at the end of the article.

Presented in part as a poster at the 22nd World Congress of

Dermatology, Seoul, Korea, May 24-29, 2011.

Accepted for publication September 1, 2011.

Reprints not available from the authors.

Correspondence to: Jerry Tan, MD, FRCPC, 2224 Walker Rd, Suite

300, Windsor, Ontario, Canada N8W5L7. E-mail: jerrytan@

bellnet.ca.

Published online October 31, 2011.

0190-9622/$36.00

� 2011 by the American Academy of Dermatology, Inc.

doi:10.1016/j.jaad.2011.09.005

187

J AM ACAD DERMATOL

AUGUST 2012188 Tan et al

clinical practice must be practical and time-efficient.Consequently, severity measures are typically basedon qualitative methods of severity grading ratherthan lesion counting. While there are currently morethan 25 acne grading systems in existence, there isneither a gold standard nor a standardized systemconsistently used in clinical practice.2

CAPSULE SUMMARY

d There are multiple scales for acneseverity grading but no singularstandard for use in research and clinicalpractice.

d American experts concluded that anideal acne scale should include primaryacne lesions, quantity, extent, and facialand extrafacial sites as clinicalcomponents; and clinimetric properties,categorization, efficiency, andacceptance as features.

d This is the initial phase in identifying/developing a standard for acne severitygrading to facilitate translation ofresearch findings to clinical practice.

The development of acnegrading systems and the rele-vant features of those inmorecommon use have been re-viewed elsewhere.1,3,4

Overall acne severity is de-pendent on lesion size, den-sity and type; and on thedistribution and intensity ofinvolvement at affected sites.5

This range of dimensionsleads to inherent complexityin developing a uniform,standardized, global severityclassification.6 Nevertheless,clinician-based global acneassessments are of particularvalue in clinical practice,where lesion counts are im-practical because of timeconstraints; and in clinical in-vestigations, where global

qualitative evaluation is considered to be a pivotaloutcome measure in clinical trials of acne treatment.7

While development of a standard for comprehen-sive acne severity evaluation has been a recognizedimperative in acne research for over a decade,1,2 thisgoal has not been achieved. Indeed, a recent sys-tematic review of clinician-based outcome measuresfor acne concluded that the current disparate arrayprohibits secondary trial data analysis, complicatesinterpretation of study results, and may compromisepatient care.8 Clearly, a standard for acne severityevaluation that is accepted by acne experts andclinicians would facilitate consistency in clinicalpractice, clinical trials, and epidemiologicalinvestigations.

The Delphi process is a group communicationmethod that provides group members equalopportunity to contribute to decision making whilereducing the negative effects of group interactions—including normative pressure from vocal or opinion-ated group members.9 Key features of this processinclude anonymity, iteration, controlled feedback,and statistical aggregation of group responses.10 Thisprocess imposes few constraints on the size or com-position of the panel11 and, when applied online, canbe used for geographically dispersed members in a

timely and cost-efficient manner. In dermatology, theDelphi process has recently been applied to establishthe core outcome domains for controlled trials andclinical recordkeeping in atopic eczema.12

On the basis of these considerations, our goal wasto determine the essential clinical components (ie,content-related items) and features (ie, scale-related

properties) of an acne sever-ity global grading scale thatmay be feasible in bothclinical practice and researchapplications by using a Web-based Delphi process forconsensus building.

METHODSThis study was a 3-phase

Delphi process facilitatedby an online survey metho-dology (Perseus SurveySolutions). Panelist re-sponses (in the form of anon-ymous judgments) regardingclinical and measurementitems essential for a globalacne severity scale relevantto practice and clinical trialswere collected at specific

time points (rounds) and disseminated back to thepanel using statistical measures of central tendencyfor further amplification and feedback. This processis outlined in Fig 1.

A panel of dermatologists with clinical and/orresearch interest in acne vulgaris were invited toparticipate. A steering panel of 4 dermatologists(J. T., J. W., J. B., and L. S. G.) with an interest inacne outcome measures was convened and re-quested to invite an additional 7 dermatologistsacknowledged as experts in acne research to partic-ipate as panel-members for this study.

Literature review, conceptualization of the re-search objective, and identification of survey meth-odology was conceived by a member of the steeringcommittee (J. T.). Accordingly, he was disqualifiedfrom serving on the response panel as his role was toelaborate the initial survey questions, disseminatethem to the panel, review feedback, and then pro-vide further iterations of the Delphi. Thus 10 derma-tologists served as Delphi panelists.

In the first Delphi round, 4 open-ended questionswere posed:a) What components are essential in an acne sever-

ity grading scale? (By components, we meanparts that make up the whole instrument.)

Fig 1. Schematic of the 3 Delphi rounds.

*Green PJ. The content of a college-level outdoor leadership

course. Paper presented at the Conference of the Northwest

District Association for the American Alliance for Health,

Physical education, Recreation, and Dance, Spokane, WA. 1982.

J AM ACAD DERMATOL

VOLUME 67, NUMBER 2Tan et al 189

b) Do any current acne severity grading scalesinclude the components you consider essential?

c) What features are important for an acne severitygrading scale? (By features, we mean qualities orproperties of the scale.)

d) Do any current acne severity grading scalesinclude the features you consider important?

Panel responses underwent content analysis, inwhich they were coded for initial themes reflectingsimilar meaning, and then grouped into conceptualcategories. For example, panelists consistently men-tioned the need for a scale to be valid, which wasplaced under the theme ‘‘validity.’’ Eventually thistheme and similar ones (such as reproducibility)were combined together under the rubric ‘‘clinimet-ric properties.’’ This process was also similar forother categories.

In the second round, a summary of initial re-sponses in addition to an online survey was distrib-uted. The panel was asked to grade the importanceof each component and feature identified in round1 by using a 7-point Likert scale (�3 extremelyunimportant to 13 extremely important). First, pan-elists rated the overall importance of the 6 majorclinical components and 4 major features. Next, theywere asked, if a particular component or feature wasto be included, to specify subcategories for inclusion.For example, for secondary lesions, they were askedto rate the importance of each of scarring, pigmen-tary changes, excoriations, and postinflammatoryerythema for inclusion. Open-ended commentswere also solicited from panelists.

Data were analyzed using SPSS version 17 formeasures of central tendency (ie, mean, mode,median) and level of dispersion (standard deviationand interquartile range [IQR]). Because SPSS doesnot handle Likert data with negative values, thisrange was converted to a 1 (extremely unimportant)to 7 (extremely important) scale for purposes ofanalysis. Cut off ranges for inclusion and exclusionwere based on precedent.13 Using the medians andIQRs, 4 outcomes were possible: consensus forinclusion (median [6, IQR \2), consensus forexclusion (median\2, IQR\2), consensus neutral(median \6 and [2, IQR \2), and no consensus(median of any number, IQR[2).

In the third round, a summary of responses fromthe second round was provided to the panel inaddition to another iteration of the online survey. Ifconsensus for definitive inclusion or exclusion hadbeen achieved for specific items after the secondround, panelists were not questioned further; ifconsensus neutral was achieved after the secondround, panelists were asked to categorically chooseinclusion or exclusion; if no consensus had beenachieved, panelists were asked to rate that item againusing the 7-point Likert scale. For the latter, the samecut off ranges and criteria for inclusion/exclusionfrom the second round were used. For items whererespondents were asked to include or exclude,consensus was defined as 70%.*

RESULTSIn the first round, panelists identified 6 clinical

components and 4 features as essential to the devel-opment of a global acne scale. Clinical componentsidentified by the panel included (a) primary acnelesions (either evaluating both inflammatory andnoninflammatory primary lesions together or sepa-rately); (b) secondary lesions (scarring, pigmentarychanges, excoriations, postinflammatory erythema;either evaluating all types of secondary lesionstogether, or separately); (c) quantity of lesions (eval-uated through counting, numerical ranges such as0-10 or 11-40, verbal descriptors such as few ormany); (d) extrafacial sites of involvement (chest,back, neck, shoulders); (e) extent of involvement(evaluated through proportion descriptors such asone third or entire, percentages at each site, generaldescriptors such as minimal or somewhat); and (f)patient experiences (acne-related quality of life,treatment-related bother). Features identified bythe panel included (a) clinimetric properties (valid-ity, reproducibility, discriminatory capacity, and re-sponsiveness); (b) efficiency (ease of use byclinicians, researchers, nursing staff, and easy toteach); (c) categorization (ability to categorize acneseverity based on photograph examples, descriptive

Table I. Clinical components and subcomponentsachieving consensus*

Clinical components

Clinical component item

(and subcomponents) Consensus

Delphi

round for

consensus

Consensus for inclusionPrimary lesions

(inflammatory andnoninflammatory)

Inclusion 2

Evaluated separately InclusionEvaluated together Exclusion

Quantity of lesions Inclusion 2Lesion counting InclusionNumerical range InclusionVerbal descriptors NeutralOther Neutral

Sites of involvement Inclusion 2Chest InclusionBack InclusionNeck InclusionShoulders InclusionOther No consensus

Extent of involvement Inclusion 2Proportion descriptors InclusionPercentage at each site No consensusGeneral descriptors No consensusOther Neutral

Consensus for exclusionPatient experiences Exclusion 3

No consensusSecondary lesions No consensus eIf secondary lesions were included:Scarring InclusionPigmentary changes InclusionExcoriations ExclusionPostinflammatoryerythema

No consensus

Other NeutralEvaluated separately InclusionEvaluated together Exclusion

*Consensus for inclusion (IN) = median $ 6; interquartile range

(IQR) # 2. Consensus for exclusion (OUT) = median # 2, IQR # 2.

Consensus neutral (NEUTRAL) = median\6 and[2, IQR\2. No

consensus (NONE) = median of any number, IQR $ 2. At round 3,

consensus to include or exclude (IN or OUT) = 70% agreement of

respondents.

Table II. Features and subfeatures for inclusion*

Features

Feature item (subfeatures) Consensus

Delphi

round for

consensus

Consensus for inclusionClinimetric properties Inclusion 2Validity InclusionReproducibility InclusionDiscriminatorycapacity

Inclusion

Responsivity InclusionEfficiency

(ease of use)Inclusion 2

For clinicians InclusionFor researchers InclusionFor nursing staff InclusionEasy to teach InclusionOther No consensus

Categorization ofseverity

Inclusion 3

Descriptive text InclusionPhoto examples InclusionBoth text and photos InclusionOther Exclusion

Acceptance Inclusion 3By researchers InclusionBy clinicians InclusionBy regulators InclusionBy patients ExclusionBy lay public ExclusionOther Exclusion

*Consensus for inclusion (IN) = median $ 6; interquartile range

(IQR) # 2. Consensus for exclusion (OUT) = median # 2, IQR # 2.

Consensus neutral (NEUTRAL) = median\6 and[2, IQR\2. No

consensus (NONE) = median of any number, IQR $ 2. At round 3,

consensus to include or exclude (IN or OUT) = 70% agreement of

respondents.

J AM ACAD DERMATOL

AUGUST 2012190 Tan et al

text, or both); and (d) acceptance (by all stake-holders including clinicians, researchers, regulators,patients, lay public). These 6 clinical components, 4features, and their respective subcategories repre-sent the complete list of items for the second andthird rounds of the Delphi. These items and theirconsensus ratings for each round of the Delphiprocess are summarized in Tables I and II.

Additionally, responses to identification of currentscales that included these clinical components and

features were collated. Two of 5 respondents indi-cated that no current scales included all essentialcomponents. Two scales14,15 were posited to includeboth the clinical components and features by onerespondent each, and one further scale16 was sug-gested by two respondents to include only thefeatures. All 10 panelists responded for the first 2rounds; however, only 9 panelists responded for thethird round. Of 54 total items identified by the panel,consensus was achieved for 31 after round 2. Afterround 3, consensus was achieved for an additional10 items, while consensus neutral was achieved for4, and no consensus was achieved for 8.

Regarding clinical components, consensus wasachieved that primary lesions (noninflammatory andinflammatory, evaluated separately), quantity of le-sions (evaluated through counting and numericalrange), extrafacial sites of involvement (specifically

J AM ACAD DERMATOL

VOLUME 67, NUMBER 2Tan et al 191

chest, back, neck, and shoulders), and extent ofinvolvement (evaluated using proportion descriptorssuch as one third or less) should be included.Consensus for exclusion was achieved for patientexperiences whereas no consensus was obtained forsecondary lesions, although a slight majority (5 of 9)opted for exclusion (see Table I).

Regarding features, consensus was achieved forinclusion of clinimetric properties (validity, repro-ducibility, discriminatory capacity, and responsiv-ity), efficiency (ie, easy for clinicians, researchers,and nursing staff to use; and easy to teach), accept-ability (by researchers, clinicians, and regulators),and categorization of severity (through descriptivetext and/or photographic examples) (see Table II).

DISCUSSIONWhile there are a myriad of grading systems for

acne severity, none were developed with priorconsensual determination of essential content do-mains. This study used a structured method ofconsensus building, the Delphi process, to identifythe essential clinical components and features ofacne severity global grading for use both in clinicalpractice and in research. Consensus for inclusionwas achieved for the following clinical components:primary lesions, quantity, sites, and extent of in-volvement; as well as the following features: clini-metric properties, efficiency, categorization, andacceptance. Consensus for exclusion comprised pa-tient experiences. No consensus was achieved forsecondary lesions.

While two panelists identified 3 present scales thatmay contain essential clinical components and/orfeatures, further evaluation revealed shortcomings ineach. The scale by Allen and Smith14 of 8 severitygrades features text descriptions of type, quantity oflesions, and proportions of facial involvement—andincluded a separate severity scale for comedones.However, this system was limited to the face. TheEchelle de Cotation des L�esions d’Acn�e (ECLA) scaleby Dreno et al15 comprised numerical ranges ofprimary acne lesions and included extrafacial sites.However, the scale did not include proportiondescriptors of anatomical sites and was not validatedagainst another construct. The Leeds revised acnegrading system presents pictorial examples of overallseverity of facial and truncal acne; including aseparate scale for facial comedones (but not trun-cal).16 However, there was no complementary textdescribing quantity of lesions or extent of involve-ment. Subsequent analysis has shown that the facialcomponent of the scale is skewed to higher grades offacial acne severity with inadequate representationof milder forms.17 Thus these scales did not

adequately fulfill the required elements for clinicalcomponents and/or features identified in this study.In a recent review of measurement properties ofinvestigator-assessed outcome measures for acne, 9separate assessment methods were identified.8

However, when evaluated against the clinical com-ponents identified as important in this study, nonefulfilled all the criteria.

Acne lesion counting, a pivotal component ofacne clinical trial measurements and one identifiedby the panel as important for inclusion, is largelyimpractical in clinical practice. However, other itemsfor inclusion, such as a numerical range of lesionscorresponding to varying severity categories andcategorization based on descriptive text and photo-graphic standards, may be more feasible options.

Results of the current Delphi support inclusion ofextrafacial sites of involvement. Currently, however,only 3 acne grading systems explicitly include ana-tomical regions other than the face. The Leeds system(a photonumeric scale comprising 16 facial, 8 chest,and 8 back severity categories) has been usedextensively in previous epidemiological and investi-gational studies on acne.16 The Leeds system isunique in providing a separate subcategory fornon-inflammatory acne, a feature identified as es-sential in this study. The Comprehensive AcneSeverity Scale (CASS), inclusive of facial and truncalacne, has been validated with the Leeds system andan acne-specific quality-of-life instrument. This scalecomprises 6 categories and is based on a globalassessment previously used in clinical trials.17

However, this scale combines noninflammatory andinflammatory lesions into aggregate categories. TheGlobal Acne Grading system (GAGS), a quasi-quantitative scheme combining regional evaluationof 6 sites (5 facial and 1 truncal) with scores for lesiontype and number summed to provide an overallglobal score, has been proposed but has not yet beenvalidated.18 Most recently, a 5-category global scalehas been proposed.19 While shown to demonstratehigh rater reliability, this scale is limited to the face.

Surprisingly, participants in this Delphi agreed toexclude patient experiences. This finding mayreflect the focus of this expert group on clinician-determined severity, and the ready availability ofquality-of-life scales relevant to skin disease ingeneral and acne in particular. Such patient-centric measures are routinely included as a com-plement to clinician-based global acne severityassessments in clinical trials.20-25 Future researchersmay wish to evaluate whether clinicians who do notconduct research would find the inclusion of pa-tient experiences useful for an ideal acne severityscale.

J AM ACAD DERMATOL

AUGUST 2012192 Tan et al

The lack of consensus for secondary lesions mayreflect ambivalence regarding their significance inoverall severity grading.While ongoing acne scarringhas been suggested to reflect more severe acne,thereby predicating more aggressive manage-ment,5,26 a contrasting view is that secondary lesionsreflect previous, not present, activity. Thus a dynamicmeasure of increasing scarring in the presence ofongoing primary acne activity may be more relevantin this context. Consideration of the dynamic con-ceptualization of secondary lesions will require fur-ther exploration. In particular, the development of anacne scar grading system sensitive to clinically rele-vant changes in scarring would be valuable.

For the remaining items for which consensus wasnot achieved, the majority were of the ‘‘other’’category, that is, were there other means of evaluat-ing quantity of lesions, other means of evaluatingextent of involvement, or other means of categori-zation not yet suggested by the panel? The inclusionof this ‘‘other’’ category was intended to allowrespondents to contribute additional items not orig-inally identified in the first Delphi round. Whilecomments were solicited regarding these open-ended items, few panelists responded with sugges-tions, and even fewer responded that the current listbeing considered was inadequate. This implied thatthe panelists did not have other options to offer forconsideration and that item generation at round oneof the Delphi was fairly exhaustive.

The process for group decision-making can bedivided into 3 different methods: interacting (groupdiscussion with pooling of judgments, followed by amajority voting procedure or consensus decision),nominal (individuals silently and independentlygenerate their own ideas, followed by recordedpresentations of ideas, group discussion, then inde-pendent voting through a ranking or rating proce-dure), and Delphi. Unlike the former two, closephysical proximity ofmembers is not required for thelatter. Both nominal and Delphi processes have beenshown to be effective in problem-solving activitiesinvolving pooled judgment of groups. The relativeadvantages are the potential for rapid response withthe nominal process in contrast to lower costs (travel,accommodations, opportunity cost) and lesser in-convenience of the Delphi.27 For this study—thecombined need to minimize cost, maximize partic-ipation, and provide adequate time for reflection onthe issues—led us to select the Delphi process.

A potential limitation of Web-based Delphi sur-veys is the limited range of responses (ie, littleflexibility for participants to include responses un-related to the questions specifically asked). Toreduce this restriction, we solicited panel comments

and concerns with each item for every round.Additional limitations are that this U.S.-based expertpanel is not globally representative and may notadequately represent the perspective of practi-tioners. With regard to the former, a U.S.-basedacne expert panel will more likely have experiencewith acne grading systems for regulatory approval asthe U.S. Food and Drug Administration has clearlydefined recommendations regarding such outcomemeasures for trials in acne.6 Regarding the latter, ourpurpose was to clarify core dimensions relevant for asingular standard for use both in clinical practice andin research. To this end, our panel was selected onthe basis of experience in both domains.

In summary, essential clinical components andfeatures for acne severity global grading were iden-tified by a Delphi process of consensus amongclinical and research experts in acne. The clinicalcomponents for inclusion consist of the following:separate severity grading of noninflammatory andinflammatory acne lesions, incorporation of facialand extrafacial sites of involvement, determinationof extent, and categorization by descriptive text andphotographic templates. Furthermore, the expertsagreed that such an instrument should demonstraterobust clinimetric properties and be shown to beefficient in application and to be acceptable tostakeholders. Our findings should be viewed as aninitial step to guide in the identification of such ascale among those in existence, or, in its absence, inthe development of such a new standard.

Disclosures: Dr Tan is an advisory board member,speaker, consultant and/or investigator for Bayer,Cipher, Galderma, Photocure, Roche, Stiefel/GSK.Dr Webster is a consultant for Allergan, Cipher,Galderma, GSK, Medicis, Ortho, Anterios, andValecor. Dr Leyden is a consultant and advisoryboard member for Allergan, Obagi, Medicis,Galderma, Stiefel/GSK. Dr Stein-Gold is an advisoryboard member and/or investigator for Galderma,Medicis, Stiefel/GSK. Dr Thiboutot is a consultantand/or investigator for Allergan, Dow, Galderma,Intendis, Johnson and Johnson/Ortho, Medicis,Stiefel/GSK. Dr Lucky has been an investigator forGalderma, Johnson & Johnson, and Coria. DrBikowski is an advisory board member, speaker,and/or consultant for Allergan (and stockholder),Coria, Galderma, Intendis, Onset, Promius,Quinnova, Stiefel/GSK. Dr Wilkin has been a con-sultant/advisor for Allergan, Bayer, Dow, Galderma,Medicis, Photocure, Warner Chilcott, Pierre Fabre,and Promius. Dr del Rosso has been a consultant/advisor, speaker and/or investigator for Allergan,Coria, Galderma, Intendis, Medicis, OrthoNeutrogena, Ranbaxy, Stiefel, Triax, PharmaDerm,

J AM ACAD DERMATOL

VOLUME 67, NUMBER 2Tan et al 193

and Warner Chilcott. Dr Weiss is a consultant and/orinvestigator for Anterios, Coria, Galderma, OrthoNeutrogena, Promius, Quinnova, and Stiefel/GSK.Ms Wolfe and Professor Chren have no conflicts todeclare.

REFERENCES

1. Lehmann HP, Robinson KA, Andrews JS, Holloway V, Goodman

SN. Acne therapy: a methodologic review. J Am Acad

Dermatol 2002;47:231-40.

2. Stathakis V, Kilkenny M, Marks R. Descriptive epidemiology of

acne vulgaris in the community. Australas J Dermatol 1997;38:

115-23.

3. Witkowski JA, Parish LC. The assessment of acne: an evaluation

of grading and lesion counting in the measurement of acne.

Clin Dermatol 2004;22:394-7.

4. Tan JKL. Current measures for evaluation of acne severity.

Expert Rev Dermatol 2008;3:595-603.

5. Pochi PE, Shalita AR, Strauss JS, et al. Report of the consensus

conference on acne classification. J Am Acad Dermatol 1991;

24:495-500.

6. U.S. Department of Health and Human Services Food and

Drug Administration Center for Drug Evaluation and Research

(CDER). Guidance for Industry; Acne Vulgaris: Developing

Drugs for Treatment (Draft Guidance), 2005; 1-14. Available

at: http://www.fda.gov/downloads/Drugs/GuidanceCompli

anceRegulatoryInformation/Guidances/UCM071292.pdf.

7. Food and Drug Administration. Dermatologic and Ophthalmic

Drugs Advisory Committee Briefing, November 4, 2002. Avail-

able from: URL:http://www.fda.gov/ohrms/dockets/ac/02/

briefing/3904b1.htm. Accessed October 17, 2011.

8. Barratt H, Hamilton F, Car J, Lyons C, Layton A, Majeed A, et al.

Outcome measures in acne vulgaris: a systematic review. Br J

Dermatol 2009;160:132-6.

9. Gupta UG, Clarke RE. Theory and applications of the Delphi

technique: a bibliography (1975-1994). Tech Forecast Soc

Change 1996;53:185-211.

10. Rowe G, Wright G. The Delphi technique as a forecasting tool:

issues and analysis. Int J Forecast 1999;15:353-75.

11. Graham B, Regehr G, Wright JG. Delphi as a method to

establish consensus for diagnostic criteria. J Clin Epidemiol

2003;56:1150-6.

12. Schmitt J, Langan S, Stamm T, Williams HC. Harmonizing

Outcome Measurements in Eczema (HOME) Delphi panel Core

outcome domains for controlled trials and clinical record-

keeping in eczema: international multiperspective Delphi

consensus process. J Invest Dermatol 2011;131:623-30. Epub

2010 Oct 14.

13. Hsu CC, Sandford BA. The Delphi technique: making sense of

consensus. Pract Assess Res Eval 2007;12:1-8.

14. Allen BS, Smith BG. Various parameters for grading acne

vulgaris. Arch Dermatol 1982;118:23-5.

15. Dreno B, Bodokh I, Chivot M, Daniel F, Humbert P, Poli F, et al.

La grille ECLA: un systeme de cotation de l’acne pour la

pratique quotidienned du dermatologue. Ann Derm Venereol

1999;126:136-41.

16. O’Brien SC, Lewis JB, Cunliffe WJ. The Leeds revised acne

grading system. J Dermatol Treat 1998;9:215-20.

17. Tan JKL, Tang J, Fung K, Gupta AK, Thomas DR, Sapra S, et al.

Development and validation of a comprehensive acne severity

scale (CASS). J Cutan Med Surg 2007;11:211-6.

18. Doshi A, Zaheer A, Stiller M. A comparison of current acne

grading systems and proposal of a novel system. Int J

Dermatol 1997;36:416-8.

19. Dreno B, Poli F, Pawin H, Beylot C, Faure M, Chivot M, et al.

Development and evaluation of a Global Acne Severity Scale

(GEA Scale) suitable for France and Europe. J Eur Acad

Dermatol Venereol 2011;25:43-4.

20. Girman CJ, Hartmaier S, Thiboutot D, Johnson J, Barber B,

DeMuro-Mercon C, et al. Evaluating health-related quality of

life in patients with facial acne: development of a

self-administered questionnaire for clinical trials. Qual Life

Res 1996;5:481-90.

21. Gupta M, Johnson AM, Gupta AK. The development of an acne

quality of life scale: reliability, validity, and relation to subjec-

tive acne severity in mild to moderate acne vulgaris. Acta

Derm Venereol 1998;78:451-6.

22. Layton AM, Seukeran D, Cunliffe WJ. Scarred for life? Derma-

tology 1997;195(Suppl 1):15-21.

23. Motley RJ, Findlay AY. How much disability is caused by acne?

Clin Exp Dermatol 1989;14:194-8.

24. Motley RJ, Finlay AY. Practical use of a disability index in

the routine management of acne. Clin Exp Dermatol 1992;17:

1-3.

25. Tan JKL, Fung KY, Khan S. Condensation and validation of a

4-item index of the Acne QoL. Qual Life Res 2006;15:1203-10.

26. Madden WS, Landells ID, Poulin Y, Searles G, Smith KC, Tan JK,

et al. Treatment of acne vulgaris and prevention of acne

scarring: Canadian Consensus Guidelines. J Cutan Med Surg

2000;4(Suppl 1):S2-13.

27. Van de Ven A, Delbeco AL. Nominal versus Interacting Group

Processes for Committee Decision-Making Effectiveness. Acad-

emy of Management Journal 1971;14(2):203-12.