Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
Grand Valley State UniversityScholarWorks@GVSU
Peer Reviewed Articles Psychology Department
7-2001
Self-Report of ADHD Symptoms in UniversityStudents: Cross-Gender and Cross-NationalPrevalenceGeorge J. DuPaulLehigh University
Elizabeth A. SchaughencyGrand Valley State University
Lisa L. WeyandtCentral Washington University
Gail TrippUniversity of Otago
Jeff KiesnerUniversita di Padova
See next page for additional authors
Follow this and additional works at: https://scholarworks.gvsu.edu/psy_articles
Part of the Psychology Commons
This Article is brought to you for free and open access by the Psychology Department at ScholarWorks@GVSU. It has been accepted for inclusion inPeer Reviewed Articles by an authorized administrator of ScholarWorks@GVSU. For more information, please contact [email protected].
Recommended CitationDuPaul, George J.; Schaughency, Elizabeth A.; Weyandt, Lisa L.; Tripp, Gail; Kiesner, Jeff; Ota, Kenji; and Stanish, Heidy, "Self-Reportof ADHD Symptoms in University Students: Cross-Gender and Cross-National Prevalence" (2001). Peer Reviewed Articles. 14.https://scholarworks.gvsu.edu/psy_articles/14
AuthorsGeorge J. DuPaul, Elizabeth A. Schaughency, Lisa L. Weyandt, Gail Tripp, Jeff Kiesner, Kenji Ota, and HeidyStanish
This article is available at ScholarWorks@GVSU: https://scholarworks.gvsu.edu/psy_articles/14
Self-Report of ADHD Symptomsin University Students:Cross-Gender and Cross-National Prevalence
George J. DuPaul, Elizabeth A. Schaughency, Lisa L. Weyandt, Gail Tripp, Jeff Kiesner,Kenji Ota, and Heidy Stanish
Little research has examined the structure and prevalence of attention-deficit/hyperactivity disorder (ADHD) symptoms in universitystudents, including whether symptom structure conforms to the bidimensional (i.e., inattention and hyperactivity-impulsivity) concep-tualization of the Diagnostic and Statistical Manual of Mental Disorders (DSMV-IV; American Psychiatric Association, 1994) and whetherself-reported symptoms vary across gender and country. A sample of 1,209 university students from three countries (Italy, New Zealand,and the United States) completed a 24-item self-report measure (the Young Adult Rating Scale) tapping ADHD symptomatology. Factoranalyses within the U.S. and New Zealand samples supported a bidimensional symptom structure, whereas weaker support for this con-ceptualization was provided by the Italian sample. Participants did not vary significantly by gender in symptom report; however, Ital-ian students reported significantly more inattention and hyperactivity-impulsivity symptoms than students from the United States, andstudents from New Zealand reported more inattention symptoms than students from the United States. The prevalence of self-reportedADHD symptoms beyond DSM-IV thresholds for diagnosis ranged from 0% (Italian women) to 8.1% (New Zealand men). The implica-tions of these results for the use of DSM-IV criteria in identifying university students with ADHD are discussed.
ttention-deficit/hyperactivity/B disorder (ADHD; see Note 1)A is characterized by develop-
mentally inappropriate levels of inat-tention, impulsivity, and motor activ-ity (American Psychiatric Association,1994). ADHD generally begins early inlife (i.e., before age 7) and is unremit-
ting in most individuals throughouttheir lifespan (Barkley, 1998). Further-more, symptoms of this disorder are fre-quently associated with conduct prob-lems and academic underachievement
(American Psychiatric Association, 1994).In fact, 20% to 30% of children and ado-lescents with ADHD also have one ormore learning disabilities (Semrud-Clikeman et al., 1992). Thus, individu-als with ADHD are more likely thantheir typical counterparts to be diag-nosed with learning disabilities and toreceive services for academic problems.
University students diagnosed withADHD constitute a significant and
growing population eligible for ser-
vices under Section 504 of the Rehabil-
itation Act of 1973 and the Americanswith Disabilities Act (ADA) of 1990(Latham, 1995; Richard, 1995). Further-more, university students with ADHDappear to be at higher-than-averagerisk for academic impairment and un-derachievement relative to their coun-
terparts without ADHD (Heiligenstein,Guenther, Levy, Savino, & Fulwiler,1999). The majority of research onADHD, however, has been conductedwith school-age children, and the stateof knowledge and practice regardinguniversity students with ADHD haslagged significantly behind that withyounger children. At the universitylevel, we lack basic epidemiological in-formation regarding the symptomatol-ogy of ADHD at this developmentalstage as well as empirically validatedassessment methods for diagnosis andtreatment evaluation. Such informa-
tion is important to appropriate servicedelivery for any disorder.ADHD is currently conceptualized
as a developmental disorder for whichit needs to be determined whether theindividual is displaying symptoms ofADHD to a developmentally inappro-priate extent (American Psychiatric As-sociation, 1994). Thus, to accurately di-agnose ADHD in university students,information on the typical parametersof the symptom domains of ADHD(i.e., inattention and impulsivity-hyperactivity) in this population is re-quired before we can assert that an in-dividual displays these behaviors to adevelopmentally inappropriate extent.The criteria for ADHD used by the
Diagnostic and Statistical Manual of Men-tal Disorders (DSM-IV; American Psy-chiatric Association, 1994) were estab-lished via reviews of the existingliterature on the factor analysis of
symptoms and via new factor and di-
at GRAND VALLEY STATE UNIV LIB on July 17, 2013ldx.sagepub.comDownloaded from
371
agnostic utility analyses of data fromthe DSM-IV field trials for the disrup-tive behavior disorders (Lahey et al.,1994). The samples for the DSM-IVfield trials for ADHD consisted of chil-dren and adolescents ages 4 to 17 from11 sites across the United States (Laheyet al., 1994). These sources of informa-tion consistently suggested that twoseparate dimensions of symptoms-inattention and motor hyperactivity-impulsivity-constitute ADHD in chil-dren and adolescents. However, theDSM-IV bidimensional structure ofADHD symptoms in postsecondary stu-dents and adults remains to be empir-ically evaluated.
In adulthood, the diagnosis of ADHDrelies heavily on symptom self-reportof the adult client (Dulcan & Work
Group on Quality Issues, 1997; Jackson& Farrugia, 1997). As more individualswith ADHD are able to attend univer-
sity or are identified as needing ac-commodations or services in postsec-ondary institutions, the nature of thisdisorder in the university populationrequires greater scrutiny. In light ofthis, it is important to empirically ex-amine self-reported symptoms of ADHDin university students. As is the casewith other issues in the area of ADHD
in young adulthood, little research has
directly examined this issue.Two recent studies have reported on
the prevalence of self-reported atten-tional difficulties in general samples ofuniversity students (Heiligenstein,Conyers, Berns, & Smith, 1998; Wey-andt, Linterman, & Rice, 1995). Usingsymptom cutoff thresholds based onstandard deviation units, Weyandtet al. (1995) found prevalence rates forself-reported attentional difficulties of7% to 8% with a cut-point of 1.5 stan-dard deviations above the mean of thetotal scores on the measures used and4% with a 2.0 SD cut-point in theirsample of 770 university students inthe Pacific Northwest. Including onlythose items in the analyses that corre-sponded to DSM-IV criteria, results in-dicated that 6.8% and 4.7% of partici-pant ratings were respectively 1.5 SD
and 2 SD above the mean on the mea-sures used (Weyandt et al., 1995).
Heiligenstein et al. (1998) examinedself-reported attentional difficulties
using a DSM-IV-based rating scale ina sample of 448 students at a midwest-ern university. They reported preva-lence rates based on symptom counts,with symptom presence operationallydefined as endorsement of often or veryoften for an item. Using the DSM-IV cri-teria for ADHD, they calculated theprevalence for the three subtypes ofADHD as well as the overall preva-lence of an ADHD diagnosis. Heiligen-stein et al. reported an overall preva-lence of self-reported ADHD of 4%,with the majority (56%) of participantsmeeting criteria for ADHD, predom-inantly inattentive type, and the re-
mainder equally distributed betweenthose meeting criteria for hyperactive-impulsive and combined types (22%).Noting a negative correlation be-
tween age and hyperactivity and totalitem scores, Heileigenstein et al. (1998)further examined the impact of an age-adjusted threshold. They calculatedsymptom counts of 1.5 standard devi-ations above the mean for each of theinattention and hyperactive-impulsivesymptom lists. Based on this calcula-tion, they evaluated a 4-symptom, age-adjusted cutoff. This modified cut-
point yielded an 11 % prevalence rate ofself-reported ADHD, with approxi-mately equal percentages of studentsmeeting the criteria for each of the sub-types of ADHD. Heiligenstein et al.
concluded that the current DSM-IV cri-
teria threshold is too high when ap-plied to university students.The present study expanded on the
preliminary work by Weyandt et al.
(1995) and Heiligenstein et al. (1998) byobtaining samples from a range of uni-versities from across the United States
and abroad. In this way, a sufficientlylarge sample size was obtained to per-mit factor analyses to test the appro-priateness of the DSM-IV conceptual-ization of the bidimensional structure
of ADHD symptomatology for this
population. Moreover, by including
cross-national samples, the extent towhich this factor structure generalizescross-nationally could be examined.The latter is important because con-ceptualization and understanding ofthis disorder can vary across countries
and cultures. U.S. samples were drawnfrom three universities located in three
different regions of the United States(Mid-Atlantic, Midwest, and Pacific
Northwest). The cross-national sam-
ples were drawn from universities inNew Zealand and Italy-nations thatdiffer historically in their professionalviews of ADHD (O’Leary, Vivian, &
Cornoldi, 1984). The practice of psy-chology in general, and the conceptu-alization of ADHD in particular, in NewZealand is influenced by both British(e.g., International Classification of
Diseases hyperkinetic syndrome) andU.S. (e.g., DSM-IV) systems. This is re-flected in a number of publications onADHD by New Zealand authors thathave appeared in U.S. professional jour-nals (e.g., McGee et al., 1990; Reeves,Werry, Elkind, & Zametkin, 1987; ’~’Werry, Elkind, & Reeves, 1987). In Italy,ADHD is discussed much less fre-
quently in professional circles, and
empirical investigation of ADHD inchildren in Italy is a relatively recentphenomenon.The Practice Parameters for the Assess-
ment and Treatment of Children, Adoles-cents, and Adults with Attention-Deficit/Hyperactivity Disorder recommend thatthe DSM-IV criteria for ADHD be usedin the identification of adults with thedisorder (Dulcan & Work Group onQuality Issues, 1997). At this stage, thisassertion remains to be empirically eval-uated. The present study began to ad-dress this issue by evaluating therobustness of the bidimensional con-
ceptualization contained in the DSM-IV criteria with this young adult uni-
versity student sample. Furthermore,the degree to which self-reported inat-tention and hyperactivity-impulsivitysymptoms varied as a function of gen-der and country was examined. Stud-ies investigating parent and teacherratings of ADHD symptoms in the
at GRAND VALLEY STATE UNIV LIB on July 17, 2013ldx.sagepub.comDownloaded from
372
child population consistently obtaingender differences, with boys being re-ported to exhibit more symptoms thangirls (e.g., Bauermeister et al., 1995; Du-Paul et al., 1997; Schaughency, McGee,Raja, Feehan, & Siva, 1994). On theother hand, gender differences werenot found for self-report ratings of
ADHD symptoms in a community-based sample of adolescents in NewZealand (Schaughency et al., 1994).The degree to which a gender differ-ence is present for self-reported symp-toms among university students hasnot been examined. The prevalence ofself-reported ADHD based on DSM-IVcriteria was also investigated as a steptoward examining the typical parame-ters of these symptoms in the generalpopulation of university students anddetermining whether an age-adjustedthreshold is indicated. If the DSM-IV
criteria are applicable to university stu-dents, we would expect factor analysisto support a 2-factor solution tappinginto separate dimensions of inattentionand hyperactivity-impulsivity. Also,we would expect men to report moreADHD symptoms than women withno differences in symptom reportsacross countries. Although we lackprevalence data for ADHD in postsec-ondary settings, the generally acceptedprevalence rate for ADHD is 3% to 5%of the general population (AmericanPsychiatric Association, 1994). Given theassociation between ADHD and aca-demic underachievement, some writersin the field have predicted a somewhatlowered prevalence rate in universitysettings and have estimated that 1 % to3% of the university population hasADHD ( Javorsky & Gussin, 1994; Ri-chard, 1995). Because this was an ex-
ploratory investigation, no specific hy-potheses were postulated.
Method
ParticipantsA total of 1,209 university studentsfrom three countries (Italy, NewZealand, and the United States) partic-ipated in the study. The Italian sample
was composed of 197 students (170women and 27 men) who ranged inage from 18 to 35 years old (M = 21.6;SD = 3.0) and who attended the Uni-versita di Padova. The Italian partici-pants were all White, and their median
Hollingshead index (based on paren-tal occupation; Hollingshead, 1975)was 40.0 (i.e., representative of middleclass socioeconomic status). The NewZealand sample was composed of 213students (175 women and 37 men) who
ranged in age from 17 to 51 years old(M = 19.4; SD = 4.0) and who attendedthe University of Otago. The New Zea-land sample was predominantly White(n = 182) but also included participantsof Asian (n = 11), Maori (n = 16), andother (n = 4) ethnic backgrounds. Themedian Hollingshead index for this
sample was 70.0 (i.e., representative ofupper middle class to upper class so-cioeconomic status). Finally, the U.S.sample was composed of 799 students(391 women and 407 men) who rangedin age from 17 to 49 years old (M = 21.3;SD = 4.9) and who attended Central
Washington State University (n = 444),Grand Valley State University (n =189),or Lehigh University (n = 166). The
majority of U.S. students were White(n = 695), but the U.S. sample also in-cluded participants of Hispanic (n = 30),Asian (n = 25), African American (n = 14),Native American (n = 8), and other(n = 22) ethnic backgrounds. The me-dian Hollingshead index was 52.0 (i.e.,representative of middle class socio-economic status).
Separate 2 (Gender) x 3 (Country)analyses of variance revealed signifi-cant main effects of country on age,F(2, 1,198) = 11.2, p < .001, and parentaloccupation, F(2,1,151) = 31.77, p < .001.There were no gender or Gender xCountry interaction effects on these
variables. Students in New Zealand
were significantly younger than stu-dents from the United States and Italy.Also, parental occupation status waslower for students from Italy than forstudents from the other two countries.
The percentages of Whites, X 2(12) =119.5, p < .001, and women, x2(2) _142.6, p < .001, were significantly
higher in the Italian and New Zealandsamples than in the U.S. sample.
Procedure
U.S. participants were recruited fromundergraduate psychology classes ateach university. Participants from NewZealand were volunteers from amongfirst-year psychology students at theUniversity of Otago. Italian partici-pants were volunteer second-year stu-dents from one of three psychologyclasses at the Universita di Padova.
Students from the United States and
New Zealand provided written con-sent to participate prior to completingthe self-report protocol. Italian stu-
dents were asked to complete the ques-tionnaire during a class period, andthey were free to decline participationwithout any penalty (i.e., their partici-pation was considered their consent).Participants were asked to provide de-mographic information about them-selves (i.e., age, ethnicity, gender, aca-demic class, and current grade pointaverage) and their parents (i.e., occu-pation). Next, each participant com-pleted a 24-item questionnaire that
included questions derived from theDSM-IV criteria for ADHD (see Note2). Questionnaires were completed onan independent basis and returned toone of the investigators.
Dependent MeasureA 24-item questionnaire, the YoungAdult Rating Scale (YARS), was con-structed by the investigators and in-cluded 17 items derived directly fromthe DSM-IV ADHD symptom list (i.e.,9 inattention and 8 hyperactivity-impulsivity symptoms). An additional7 items were included to address po-tential difficulties (e.g., problems re-membering what was just read) thatuniversity students could encounterin association with ADHD symptoms.All items were completed on a 4-pointLikert scale ranging from 0 (never orrarely) to 3 (very often). Students wereasked to respond to these items re-
garding their behavior over the past
at GRAND VALLEY STATE UNIV LIB on July 17, 2013ldx.sagepub.comDownloaded from
373
6 months (i.e., in accordance withDSM-IV guidelines). Total scores onthis measure can range between 0 and72. A coefficient alpha of .86 was ob-tained with the present sample, indi-cating adequate internal consistencyfor the YARS. Furthermore, Schaugh-ency et al. (1998) found that YARS rat-
ings were significantly related to gradepoint average in a sample of collegestudents and that students with clini-
cally diagnosed ADHD provided sig-nificantly higher YARS ratings thandid students without this disorder.
Data AnalysesSeveral analyses were conducted sepa-rately for the three participant sam-ples. First, a principal axis factor analy-sis was conducted to determine thefactor structure of the YARS. Scree
plots were examined, and the numberof factors to retain was determined. A
second factor analysis was then con-ducted constraining the solution to thedetermined number of factors usingoblique rotation methods, because it
was assumed that the factors would be
correlated.Another set of analyses was con-
ducted to examine differences in self-
reported ADHD symptomatology be-tween countries and genders usingmultivariate analyses of covariancefollowed by univariate analyses of co-variance. For these analyses, a symp-tom was considered present for itemsscored as 2 (often) or 3 (very often), aswas done by Heiligenstein et al. (1998).The 17 items whose content directlyreflected DSM-IV criteria for ADHD
were used to construct inattention (9items) and hyperactivity-impulsivity(8 items) scores. Age and parent occu-pation were used as covariates owingto differences in the means of thesevariables across countries.The final set of analyses examined
the prevalence of self-reported ADHDsymptomatology by converting itemscores to symptom scores as describedin the preceding paragraph, determin-ing the number of symptoms for inat-tention and hyperactivity-impulsivity
domains, and calculating the per-centage of participants who reportedsymptom totals beyond the DSM-IVthresholds for the inattentive, hyperac-tive-impulsive, and combined types ofADHD. In keeping with analyses con-ducted by Heiligenstein et al. (1998),we also examined prevalence rates
using an age-adjusted symptom cutoffcriterion of three symptoms in the
two ADHD domains (inattention and
hyperactivity-impulsivity).
Results
Factor AnalysesFor the U.S. sample, six factors witheigenvalues > 1.0 were obtained; how-ever, examination of the scree plot andthe percentage of variance accountedfor indicated that two factors should be
retained. Together, these factors ac-
counted for 34.8% of the variance. The
factor analysis was conducted againwith a forced 2-factor solution fol-
lowed by oblique rotation. Factor 1 (la-beled inattention) contained 12 itemswith factor loadings > .35, whereas Fac-
tor 2 (labeled hyperactivity-impulsivity)consisted of 7 items with factor load-
ings > .35 (see Table 1). The two factorswere correlated to a moderate degree(r = .63). Five items did not load oneither factor.
Similar results were obtained for the
New Zealand sample (see Table 1).
Specifically, seven factors with eigen-values > 1.0 were obtained; however,the scree plot indicated a sharp dropafter two factors. Thus, we retained 2factors that accounted for 33% of the
variance. The factor analysis was con-ducted again with a forced two-factorsolution followed by oblique rotation.Factor 1 also appeared to represent in-attention and included 15 items with
factor loadings > .35. It should be
noted that in contrast to the U.S. sam-
ple, Factor 1 also included 2 items
related to hyperactivity (i.e., di f ficultystaying in seat and feel restless). Fac-
tor 2 contained 5 items related to
hyperactivity-impulsivity with factorloadings > .35. The correlation between
the two factors was .30. Four items did
not load on either factor.
Factor analytic results for the Italiansample deviated somewhat from find-ings obtained in the United States andNew Zealand. The initial principal axisfactor analysis resulted in nine factorswith eigenvalues > 1.0 with no clear
dropoff on the scree plot after two fac-tors. To facilitate comparison with theother two samples, however, we con-ducted a second factor analysis with aforced 2-factor solution followed byoblique rotation. These two factors ac-counted for only 23.1 % of the vari-ance and were negatively correlated(r = -.16). Factor 1 included 9 items
with factor loadings > .35. As was the
case for the other countries, Factor 1
appeared to reflect attention problems,although 2 items were related to hy-peractivity (i.e., feel restless and di f fi-culty engaging in fun activities quietly).Factor 2 contained only 3 items withloadings > .35 that appeared related toimpulsivity. Twelve items (i.e., 50% ofthe 24 items) did not load on either fac-tor. Furthermore, four factors wouldneed to be retained to account for the
same amount of variance (36.5%) astwo factors in the other two samples.
Group Di f ferences inSel f Reported ADHD SymptomsMeans and standard deviations (unad-
justed) for the number of inattentionand hyperactivity-impulsivity symp-toms reported are presented by genderand country in Table 2. To explorewhether self-reported ADHD symp-toms varied across gender or country,a 2 (Gender) x 3 (Country) multivar-iate analysis of covariance was con-ducted using the number of inatten-tion symptoms and the number of
hyperactivity-impulsivity symptomsas dependent variables. Age and par-ental occupation were used as covari-ates owing to differences across coun-tries on these variables. It should be
noted, however, that the within-cell
relationships between the covariatesand dependent measures were equiv-alent across factors. A significant main
at GRAND VALLEY STATE UNIV LIB on July 17, 2013ldx.sagepub.comDownloaded from
374
TABLE 1
Factor Structure for Self-Report of ADHD Symptoms Across Countries
Note. U.S. = United States; N.Z. = New Zealand. Factor loadings > .35 are in boldface type. Only 23 items are included in this table because one item did not loadon either factor for any group.
TABLE 2
Number of ADHD Symptoms Reported by Gender and Country
effect for country was obtained, Wilks’lambda = .98, F(4, 2,228) = 5.44, p <
.001, multivariate effect size = .01. Themain effect for gender and the Gen-der x Country interaction were bothnonsignificant.
Univariate 2 x 3 analyses of covari-ance were conducted for each of the
dependent measures. A significantmain effect for country was found forinattention symptoms, F(2,1,115) = 9.70,p < .001, 112 =.02, and for hyperactivity-impulsivity symptoms, F(2, 1,115) =4.14, p < .05, r~2 = .007. Scheff6 post hoccomparison tests using adjusted means
at GRAND VALLEY STATE UNIV LIB on July 17, 2013ldx.sagepub.comDownloaded from
375
indicated that Italian students reportedsignificantly more inattention and hy-peractivity-impulsivity symptoms thanstudents from the United States (allps < .05). Furthermore, students fromNew Zealand reported more inatten-tion symptoms than did students fromthe United States (p < .05). There wereno significant differences on either ofthe scores between students from Italyand New Zealand. Also, none of theunivariate analyses revealed a signifi-cant main effect for gender or a Genderx Country interaction.
Prevalence of Self-ReportedADHD SymptomsThe percentages of university studentswho self-reported significant symp-toms of one of the three subtypes ofADHD as a function of country andgender are displayed in Table 3. Forthese analyses, a symptom was consid-ered present for items scored as 2 (often)or 3 (very often). Using DSM-IV criteria,participants were identified as havingthe inattentive subtype if six or moreinattention symptoms (i.e., scored 2 or3) were reported with less than six
hyperactivity-impulsivity symptoms.In similar fashion, participants wereidentified as having the hyperactive-impulsive subtype if six or more
hyperactivity-impulsivity symptomswere reported along with fewer thansix inattention symptoms. Finally, indi-viduals who reported six or more
symptoms for both inattention and hy-
peractivity-impulsivity were identi-
fied as having the combined type ofADHD. Although we did not have suf-ficient data to determine whether indi-vidual students experienced functionalimpairment in association with ADHDsymptoms (as is required by DSM-IVcriteria), it should be noted that the cor-relation between self-reported gradepoint average (GPA) and the total
YARS score was statistically significant(r = -.11, p < .001). This negative corre-lation indicated that higher levels ofself-reported ADHD symptoms wereassociated with lower GPA.There were a number of differences
in subtype prevalence across countriesfor both men and women (see Table 3).A total of 2.9% of men from the U.S.
sample were identified as having oneof the three ADHD subtypes, withmost (2%) of these individuals cate-
gorized in the hyperactive-impulsivesubtype. In contrast, a higher percent-age of men from Italy (7.4%) and NewZealand (8.1 %) reported significantADHD symptoms. Similar to men fromthe United States, the majority of Ital-ian and New Zealand men were iden-tified as hyperactive-impulsive. A verylow percentage of students across allthree countries were classified as hav-
ing the combined type of ADHD. Infact, no men in either Italy or New Zea-land reported significant symptoms ofboth inattention and hyperactivity-impulsivity.
In contrast to the findings for men, agreater percentage (3.9%) of women
TABLE 3
Percentages of Self-Reported ADHD Subtypes by Gender and CountryUsing DSM-/V Criteria
from the United States were classifiedas having ADHD than women from ei-ther Italy (0%) or New Zealand (1.7%;see Table 3). The vast majority (2.3%) ofU.S. women and all of the New Zea-
land women were categorized in thehyperactive-impulsive subtype.
In keeping with analyses conductedby Heiligenstein et al. (1998), we alsoexamined prevalence rates using anage-adjusted symptom cutoff criterionof at least three symptoms in the
two ADHD domains (inattention and
hyperactivity-impulsivity). A patternacross countries was obtained for menthat was similar to the one found usingthe DSM-IV cutoff criteria (see Table 4).
Specifically, the preponderance of menin all three countries were identified as
having the hyperactive-impulsivesubtype, and more men in New Zea-land (29.7%) and Italy (37.3%) wereclassified as having ADHD than in theUnited States (27.4%). In similar fash-ion, using age-adjusted criteria re-
sulted in more women from New Zea-land (37.4%) and Italy (43.1 %) beingidentified as having ADHD than wo-men from the United States (24.6%).This pattern of results for women is the
opposite of what was obtained us-
ing the more stringent DSM-IV criteria.As was the case for men, the majorityof women were identified with the
hyperactive-impulsive subtype.
’~ Discussion
’
A primary purpose of this study wasto explore the appropriateness of
DSM-IV criteria for ADHD with post-secondary students and to determinewhether the bidimensional conceptu-alization of ADHD applies to this pop-ulation. Based on factor analysis of theYARS by country, findings from theUnited States and New Zealand sup-port a bidimensional model, with inat-tention and hyperactivity-impulsivityemerging as two distinct factors. Thefirst factor appears to be measuringdifficulty completing tasks, difficultysustaining attention, forgetfulness, andorganizational skills. The second factor
at GRAND VALLEY STATE UNIV LIB on July 17, 2013ldx.sagepub.comDownloaded from
376
TABLE 4
Percentages of Self-Reported ADHD Subtypes by Gender and CountryUsing Age-Adjusted Diagnostic Criteria
appears to encompass difficulties with
self-control, such as impulsivity, exces-sive talking, and interrupting others.These results are consistent with previ-ous literature (e.g., Bauermeister et al.,1995; DuPaul et al., 1997) and DSM-IV
conceptualizations of ADHD.However, findings from the Italian
sample were somewhat inconsistentwith this model, with fewer items
loading on these factors, and 50% ofthe items not loading on any factor.Given the relative paucity of ADHD re-search in Italy, it is difficult to interpretthese findings from an empirical per-spective ; however, several factors couldhave accounted for these differences.One possibility is that difficulties of hy-peractivity and inattention are not gen-erally discussed in Italy with regard tochildhood developmental problems.This lack of discussion may mean that
questions about ADHD symptomspresented unfamiliar ideas or had dif-ferent connotations for Italian partici-pants. There may also be differences inthe acceptability of these behaviorsacross countries because knowledge ofthis disorder in Italy may be limited.Because of differences in novelty oracceptability, Italian participants mayhave encountered difficulties discrimi-
nating among items, which could havecontributed to the lack of a clear factorstructure. It should be noted, however,that items loading on the first factor(e.g., avoid work that requires effort, andfail to finish work) for the U.S. and New
Zealand samples appear to be veryconcrete and unambiguous.Another possible reason for the dif-
ferences in factor structure across coun-tries is that respondents could have beenusing different comparison groupswhen making judgements about symp-toms, because the demographics of stu-dents who attended universities acrossthe three countries differed markedly.A related issue that could contribute tocross-national differences in factor
structure is that less than 50% of psy-chology students enrolled in this Ital-ian university are expected to finish all5 years of university. This high dropoutrate may be related to different ad-mission criteria of Italian relative to
U.S. and New Zealand universities and,therefore, result in a different popu-lation. However, it is unclear how dif-ferent acceptance criteria could affectthese factor structures, which have
typically been supported across dif-
ferent ages and groups in the UnitedStates. Clearly, more research is neededto explore these possibilities.A secondary purpose of the present
study was to explore the differentialprevalence of self-reported ADHDsymptoms in university students bygender and country. In contrast to ear-lier findings from parent and teacherratings of ADHD symptoms in childrenand adolescents (e.g., Bauermeisteret al.,1995; DuPaul et al.,1997; Schaugh-ency et al., 1994), no significant genderdifferences were obtained for self-
reported inattention or hyperactivity-impulsivity symptoms. Murphy andBarkley (1996) found that men withADHD outnumbered women with thisdisorder on a 1.8:1 to 2.6:1 ratio in a
community-based sample of adults. Itis possible that these gender differ-ences in ADHD symptomatology areabsent among higher achieving stu-dents who are able to attend postsec-ondary institutions. The lack of genderdifferences in our sample of universitystudents may also be due to the use of
self-report data in contrast to the par-ent and teacher ratings of symptomstypically obtained with children andadolescents. The latter explanation issupported in part by the results of
Schaughency et al. (1994) who foundgender differences in parent reports ofADHD symptoms for a community-based sample of adolescents in NewZealand but no differences in self-
reports of symptoms between genders.It is also possible that women in gen-eral are more likely to report ADHDsymptoms, as has been found for self-report of depressive symptoms (Com-pas, 1997).
Students from New Zealand and Italyreported significantly more inattentionsymptoms than did U.S. students. Fur-thermore, Italian students reportedsignificantly more hyperactivity-impulsivity symptoms than U.S. stu-dents. These findings were somewhatunexpected given the assumption thatADHD is diagnosed more frequentlyin the United States than in the othertwo countries. Cross-national differ-ences in self-reported symptomatologymay have been due to variations in age(e.g., students in New Zealand weresignificantly younger than students inItaly and the U.S.) or socioeconomicstatus (e.g., parents of Italian studentshad a lower socioeconomic status than
parents of students from the othertwo countries). It should be noted that
although cross-national differenceswere statistically significant, the coun-try variable accounted for minimalvariance (i.e., 0.7% to 2%) in self-
reported inattention and hyperactivity-impulsivity symptoms. Thus, the vari-
at GRAND VALLEY STATE UNIV LIB on July 17, 2013ldx.sagepub.comDownloaded from
377
ation in symptom self-report was pri-marily accounted for by factors otherthan gender and country (e.g., impair-ment in academic or social function-
ing).Despite the minimal variance ac-
counted for by gender and country, thepercentages of participants meetingDSM-IV criteria cut-points for self-
reported ADHD symptoms differed tosome degree across these variables. Forexample, 2.9% of men from the U.S.sample reported significant ADHDsymptoms, whereas 7.4% of men fromItaly and 8.1 % of men from NewZealand reported ADHD symptoms.Women from the United States re-
ported slightly higher rates (3.9%),whereas 1.7% of women from New
Zealand and no women from Italy re-ported significant ADHD symptoms.Overall, the U.S. findings are fairlyconsistent with those of Heiligensteinet al. (1998), Murphy and Barkley(1996), and Weyandt et al. (1995), whofound approximately 4% to 5% preva-lence rates for all ADHD subtypes inadult and university student samples.The prevalence rates from Italy andNew Zealand are higher than the ex-pected 3% to 5% as reported by theDSM-IV, although similarly high rateshave been reported in child and ado-lescent studies. Pineda et al. (1999), for
example, investigated the prevalenceof parent-reported ADHD symptomsin children and adolescents ages 4 to 17
living in Colombia and found that
19.8% of the boys and 12.3% of the girlsmet DSM-IV criteria for ADHD. Simi-
larly high percentages have been re-ported in India by Bhatia, Nigam,Bohr, and Malik (1991), who found that29.2% of children ages 11 to 12 met
DSM-111 criteria for all ADHD sub-
types based on parent reports, and inGermany (e.g., Baumgaertel, Wolraich,& Dierrich, 1995) the percentage ofchildren with DSM-III-R ADHD variedfrom 9.6% (age 5) to 17.8% (age 17)based on teacher reports.The most common of the three sub-
types of ADHD across country andgender was the hyperactive-impulsivetype. Using DSM-IV thresholds, 2%
of men and 2.3% of women from theUnited States, 3.7% of men from Italyand no women, and 5.4% of men and1.7% of women from New Zealand were
categorized as having the hyperactive-impulsive ADHD subtype. Less than1 % of U.S. men and women were cate-
gorized as having the inattentive orcombined ADHD subtypes. Similarly,none of the men and women from Italyor New Zealand met DSM-IV thresh-olds for inattentive or combined sub-
types. The U.S. findings are similar tothose reported by Murphy and Barkley(1996), who found that the hyperactive-impulsive subtype was the most com-mon type of ADHD, followed by theinattentive subtype. Heiligenstein et al.(1998), however, found that the inat-tentive subtype was most prevalentamong university students. With chil-dren and adolescents, Pineda et al.
(1999) found that the hyperactive-impulsive type was more frequent inboth boys and girls, whereas the com-bined type was the least frequent andwas absent in adolescents between 12and 18 years of age. The lower per-centage of individuals with combinedtype ADHD in our university samplesmay be due to the more significant im-pairment typically associated with thissubtype relative to the inattentive andhyperactive-impulsive subtypes (for areview, see Barkley, 1998). Because ofthis more severe impairment, individ-uals with combined type ADHD mayhave difficulty gaining acceptance touniversity, especially in New Zealand,where regulations do not require schoolsto provide accomodations and servicesfor students with this disorder.
It is important to note that the pres-ent study explored the prevalence ofself-reported ADHD symptoms and wasnot diagnostic in nature. Obviously, thepresence of ADHD symptoms alone isinsufficient for a diagnosis of ADHD,and in this study limited informationwas available concerning the level offunctional impairment experienced bythe participants and the developmen-tal history of the participants (as is re-quired by DSM-IV criteria). Althoughgrade point average (GPA) was found
to correlate negatively with YARS rat-ings, the magnitude of this correla-
tion was quite small. Furthermore, GPAalone is an inadequate index of impair-ment, especially because in the pres-ent study it was based on self-report.Murphy and Gordon (1998), for exam-ple, have recommended that academicrecords, test histories, performance re-views, and similar data be used to es-tablish impairment for adults beingdiagnosed with ADHD. Overall, thepresent cross-national findings takentogether with findings from previousresearch (e.g., Bhatia et al., 1991; Pin-eda et al., 1999) underscore the impor-tance of obtaining information from avariety of sources and suggest that re-liance on a single indicator (e.g., rat-ing scale) may result in higher-than-expected percentages of ADHD.Given the literature suggesting that
hyperactivity symptoms decrease withage (e.g., Barkley, 1998; Heiligensteinet al., 1998; Hill & Schroener, 1996;
Schaughency et al., 1994), the percent-age of students meeting DSM-IV thresh-olds for the hyperactive-impulsivesubtype was somewhat surprising.Also unexpected were the relativelyhigh percentages of U.S. and NewZealand women who met the criteriafor the hyperactive-impulsive subtype(2.3% and 1.7%, respectively) and thelow percentage who met the thresh-olds for the inattentive subtype (0.8%and 0%). Previous studies (e.g, Gaub &Carlson, 1997) have consistently re-
ported that the inattentive subtype ismore commonly found in women thanin men. This finding, in conjunctionwith the low percentage of studentsacross all three countries meeting crite-ria for the inattentive and combined
subtypes, appears consistent with
Heiligenstein et al.’s (1998) assertionthat DSM-IV thresholds are too high(at least with respect to inattention
symptoms) when applied to universitystudents. When the thresholds for each
subtype were lowered to three (ratherthan six) symptoms, very high rates ofidentification were obtained for allthree countries. Thus, future researchis needed to document symptomatic
at GRAND VALLEY STATE UNIV LIB on July 17, 2013ldx.sagepub.comDownloaded from
378
thresholds that are developmentallyappropriate, related to functional im-pairment, and not subject to seriousoverestimation of diagnostic preva-lence. Such research would provide abetter understanding of the age-relatedchanges associated with ADHD symp-toms and the relevance of these changesto diagnostic criteria for ADHD in uni-versity students and other adults.
Limitations ’Relative to the literature concerningADHD in children, little informationis available concerning the prevalenceand the nature of this disorder in
young adults such as university stu-dents. The present study takes a criti-cal step in helping to better understandthe presence of ADHD symptoms inthe university student population inthe United States and in other cultures.
Preliminary findings suggest that theprevalence rate of ADHD symptoms inuniversity students varies from 2.9% to8.1 %, with the highest ratings reportedin New Zealand. The present studywas based on self-report, however, andfuture research is needed to explore thevalidity of self-reported symptomswithin this population. Although self-report ratings are the primary dataused for diagnosis and evaluation oftreatment response in adults, furtherexamination of symptom prevalenceusing multiple assessment methods iswarranted.Another limitation of this study
concerns the lack of sufficient data toestablish functional impairment (al-though GPA was found to correlatenegatively with self-report ADHDratings) or childhood history of symp-toms (as is required for DSM-IV diag-nosis of ADHD). Furthermore, partici-pants were not asked whether theywere currently coping with any life-disrupting events (e.g., death of aloved one or breakup of a long-termrelationship) that may have led to
higher symptom reports. Clearly, fu-ture studies with university samplesneed to go beyond self-reported symp-tom counts to examine prevalence.
There was also a great deal of vari-
ability across individuals in self-reportratings, thereby possibly deleteriouslyaffecting statistical analyses. It is im-
portant to note, however, that this het-
erogeneity is not atypical for ratings ofADHD symptoms (e.g., DuPaul et al.,1997), and careful replication acrosssamples is required to support the ex-ternal validity of findings.Another possible limitation of this
study is that although cross-nationalsamples were obtained, the data werecollected from only five universities,and the degree to which these samplesare representative of the overall uni-versity student population is question-able. Given that the New Zealand sam-
ple was composed entirely of first-yearuniversity students-some of whommay not continue successfully in uni-versity-the degree to which the datarepresent a typical sample of NewZealand university students is limited.The Italian and New Zealand sam-
ples consisted primarily of women,whereas a more equitable gender ratiowas present for the U.S. sample. Thedegree to which gender imbalanceacross samples affected the results ofthis study cannot be determined, but itis possible that symptom reports formen in these countries were skeweddue to a relatively small sample size.Finally, the results provide tentativesupport for the bidimensional concep-tualization of ADHD among univer-
sity students in the United States andNew Zealand; however, given the find-ings from the Italian sample, the sup-port for this model in other culturesis equivocal.
. = ’ ’!’:
Conclusions
These results provide initial evidencethat a percentage of university stu-
dents in the United States, New Zea-land, and Italy report clinically signifi-cant levels of ADHD symptomatology.Furthermore, the DSM-IV conceptual-ization of this disorder has some basisof support in this population. The over-all findings raise questions concerningthe applicability of DSM-IV thresholds
to the university student population.Future research is needed to further ex-
plore the psychometric properties ofthe YARS, the validity of self-reportand its relation to functional impair-ment, the prevalence of ADHD symp-toms in other cultures and universitystudent populations, and the appro-priateness of DSM-IV criteria for diag-nosing university students with thisdisorder.
ABOUT THE AUTHORS
George J. DuPaul, PhD, is professor and coor-dinator of the school psychology program atLehigh University. His research interests in-
clude the school-based assessment and treat-
ment of students with ADHD. Elizabeth A.Schaughency, PhD, is a professor of psychol-ogy at Grand Valley State University. Her cur-rent interests include empirically supported as-sessment and intervention strategies to addressbehavior disorders across the lifespan. Lisa L.Weyandt, PhD, is a professor of psychology atCentral Washington University. Her primaryresearch interest is the assessment and treat-
ment of individuals with ADHD. Gail Tripp,PhD, is a senior lecturer and director of clinicaltraining in the area of ADHD and related dis-ruptive behavior disorders at the University ofOtago. Je f f Kiesner, PhD, is a research fellow atthe Universita di Padova, and his research fo-cuses on peer influence and social development.Kenji Ota, MEd, is a doctoral student in schoolpsychology at Lehigh University who is inter-ested in computer-assisted instruction for stu-dents with ADHD. Heidy Stanish, BA, is a re-search assistant in psychology at Grand ValleyState University who is interested in longitudi-nal predictors of reading in Spanish-speakingchildren. Address: George J. DuPaul, School Psy-chology Program, Lehigh University, 111 Re-search Drive, Bethlehem, PA 18015.
NOTES
z
1. The terminology and diagnostic criteria thathave been used to define and describe thisgroup of disorders have changed over the years.The most recent Diagnostic and StatisticalManual of Mental Disorders (American
Psychiatric Association, 1994) uses the termattention-deficit/hyperactivity disorderand includes subtypes of the disorder that in-volve attentional difficulties in the absence ofhyperactivity-impulsivity (predominantlyinattentive type) as well as those involvinghyperactivity-impulsivity (predominantly
at GRAND VALLEY STATE UNIV LIB on July 17, 2013ldx.sagepub.comDownloaded from
379
hyperactive-impulsive type and combinedtype). Because current terminology encom-passes this range of attentional disorders, theterm ADHD will be used to refer to this
group of disorders in general. However, be-cause differing diagnostic criteria and termi-nology may identify somewhat differentgroups for research, when reviewing researchthe diagnostic criteria and terminology usedby the study authors will be used.
2. A copy of the questionnaire is available fromthe first author upon request.
REFERENCES
American Psychiatric Association. (1994).
Diagnostic and statistical manual of mentaldisorders (4th ed.). Washington, DC: Au-thor.
Americans with Disabilities Act of 1990,P.L. 101-336, 42 U.S.C. § 12101 et seq.
Barkley, R.A. (1998). Attention deficit hyper-activity disorder: A handbook for diagnosisand treatment (2nd ed.). New York: Guil-ford Press.
Bauermeister, J. J., Bird, H. R., Canino, G.,Rubio-Stipec, M., Bravo, M., & Aleg-ria, M. (1995). Dimensions of attentiondeficit hyperactivity disorder: Findingsfrom teacher and parent reports in a com-
munity sample. Journal of Clinical ChildPsychology, 24, 264-271.
Baumgaertel, A., Wolraich, M. L., & Diet-
rich, M. (1995). Comparison of diagnosticcriteria for attention deficit disorders in a
German elementary school sample. Jour-nal of the American Academy of Child andAdolescent Psychiatry, 34, 629-638.
Bhatia, M. S., Nigam, V R., Bohra, N., andMalik, S. C. (1991). Attention deficit dis-order with hyperactivity among pediatricoutpatients. Journal of Child Psychologyand Psychiatry, 32, 297-306.
Compas, B. E. (1997). Depression in chil-dren and adolescents. In E. J. Mash &
L. G. Terdal (Eds.), Assessment of childhooddisorders (3rd ed.; pp. 197-229). NewYork: Guilford.
Dulcan, M. K., & Work Group on Quality Is-sues. (1997). Practice parameters for theassessment and treatment of children,adolescents, and adults with attention-
deficit/hyperactivity disorder. Journal ofthe American Academy of Child and Adoles-cent Psychiatry, 36, Supplement, 855-1215.
DuPaul, G. J., Power, T. J., Anastopoulos,A. D., Reid, R., McGoey, K., & Ikeda, M.(1997). Teacher ratings of ADHD symp-toms: Factor structure and normative
data. Psychological Assessment, 9, 436-444.
Gaub, M., & Carlson, C. L. (1997). Genderdifferences in ADHD: A meta-analysisand critical review. Journal of the AmericanAcademy of Child and Adolescent Psychiatry,36, 1036-1045.
Heiligenstein, E., Conyers, L. M., Berns, A. R.,& Smith, M. A. (1998). Preliminary nor-mative data on DSM-IV attention deficitdisorder in college students. Journal ofAmerican College Health, 46, 185-188.
Heiligenstein, E., Guenther, G., Levy, A.,Savino, F., & Fulwiler, J. (1999). Psycho-logical and academic functioning in col-lege students with attention deficit hy-peractivity disorder. College Health, 47,181-185.
Hill, J. C., & Schroener, E. P. (1996). Age-dependent decline of attention deficit hy-peractivity disorder. American Journal ofPsychiatry, 153, 1143-1146.
Hollingshead, A. B. (1975). Four factor indexof social status. New Haven, CT: Yale Uni-versity.
Jackson, B., & Farrugia, D. (1997). Diagno-sis and treatment of adults with attention
deficit hyperactivity disorder. Journal ofCounseling and Development, 75, 312-319.
Javorsky, J., & Gussin, B. (1994) College stu-dents with attention deficit hyperactivitydisorder: An overview and description ofservices. Journal of College Student Devel-opment, 35, 170-177.
Lahey, B. B., Applegate, B., McBurnett, K.,Biederman, J., Greenhill, L., Hynd, G. W.,Barkley, R. A., Newcorn, J., Jensen, P.,Richters, J., Garfinkel, B., Kerdyk, L.,Frick, P. J., Ollendick, T., Perez, D., Hart,E. L., Waldman, I., & Shaffer, D. (1994).DSM-IV field trials for attention deficit/
hyperactivity disorder in children andadolescents. Journal of the American Acad-emy of Child and Adolescent Psychiatry, 151,1673-1685.
Latham, P. H. (1995). Legal issues pertain-ing to the postsecondary student withADD. Journal of Postsecondary Educationand Disability, 11, 53-61.
McGee, R., Feehan, M., Williams, S., Par-
tridge, F., Silva, P. A., & Kelly, J. (1990).DSM-III disorders in a large sample ofadolescents. Journal of the American Acad-emy of Child and Adolescent Psychiatry, 29,611-619.
Murphy, K., & Barkley, R. A. (1996). Preva-lence of DSM-IV symptoms of ADHD inadult licensed drivers: Implications forclinical diagnosis. Journal of Attention Dis-orders, 1, 147-161.
Murphy, K. R., & Gordon, M. (1998). As-sessment of adults with ADHD. In R. A.
Barkley, Attention deficit hyperactivity dis-order: A handbook for diagnosis and treat-ment (2nd ed.; pp. 345-369). New York:Guilford Press.
O’Leary, D. K., Vivian, D., & Cornoldi, C.(1984). Assessment and treatment of "hy-peractivity" in Italy and the United States.Journal of Clinical Child Psychology, 13,56-60.
Pineda, D., Ardila, A., Rosselli, M., Arias,B. E., Henao, G. C., Gomez, L. F., Mejia,S. E., & Miranda, M. L. (1999). Prevalenceof attention-deficit/hyperactivity disor-der symptoms in 4- to 17-year-old chil-dren in the general population. Journal ofAbnormal Child Psychology, 27, 455-462.
Reeves, J. C., Werry, J., Elkind, G. S., & Za-
metkin, A. (1987). Attention deficit, con-duct, oppositional, and anxiety disordersin children: II. Clinical characteristics. Jour-nal of the American Academy of Child andAdolescent Psychiatry, 26, 133-143.
Rehabilitation Act of 1973, P.L. 93-112, 29
U.S.C. § 701 et seq.Richard, M. M. (1995). Pathways to success
for the college student with ADD: Ac-commodations and preferred practices.Journal of Postsecondary Education and Dis-ability, 11, 16-30.
Schaughency, E., McGee, R., Raja, S. N., Fee-han, M., & Siva, P. A. (1994). Self-reportedinattention, impulsivity, and hyperactiv-ity at ages 15 and 18 years in the generalpopulation. Journal of the American Acad-emy of Child and Adolescent Psychiatry, 33,173-183.
Schaughency, E., Weyandt, L., Dunham, A.,Hampstead, W., Paschke, R., & Stan-
ish, H. (1998, August). Evaluating collegestudents’ with ADHD report on their ADHDsymptomology. Paper presented at the an-nual convention of the American Psycho-logical Association, San Francisco, CA.
Semrud-Clikeman, M., Biederman, J., Sprich-Buckminster, S., Lehman, B. K., Faraone,S. V., & Norman, D. (1992). Comorbiditybetween ADDH and learning disability:A review and report in a clinically re-ferred sample. Journal of the American
Academy of Child and Adolescent Psychiatry,31, 439-448.
Werry, J. S., Elkind, G. S., & Reeves, J. S.(1987). Attention deficit, conduct, opposi-tional, and anxiety disorders in children:III. Laboratory differences. Journal of Ab-normal Child Psychology, 15, 409-428.
Weyandt, L. L., Linterman, I., & Rice, J. A.(1995). Reported prevalence of attentionaldifficulties in a general sample of collegestudents. Journal of Psychopathology andBehavioral Assessment, 17, 293-304.
at GRAND VALLEY STATE UNIV LIB on July 17, 2013ldx.sagepub.comDownloaded from