Risk Screening for ADHD in a College Population

Abstract

Objective:The present study examines the relationship between self-reported levels of ADHD and academic outcomes, as well as aptitude. Method: A total of 523 college students took the Adult Self-Report Scale–Version 1.1 (ASRS-V1.1), and their scores were compared with course performance and ACT (American College Test) composite scores. Results: The measure identified 70 students (13.4%) as being in the “highly likely” category for an ADHD diagnosis. Course exam and ACT scores for the 70 “highly likely” students were statistically identical to the remaining 453 students in the sample and the 77 students identified as “highly unlikely” as well. Only 4 of the “highly likely” 70 students were registered with the university’s Office of Student Disability Services as having been diagnosed with ADHD. Conclusions: The ASRS-V1.1 failed to discriminate academic performance and aptitude differences between ADHD “highly likely” and “highly unlikely” individuals. The use of self-report screeners of ADHD is questioned in contexts relating ADHD to academic performance.

Keywords

ASRS-V1.1 ADHD academic performance ADHD screeners learning disabilities

Over the past 20 years, there has been an increasing awareness among researchers and practitioners that adults can manifest ADHD. In fact, it is estimated that 50% to 70% of children accurately diagnosed with ADHD will continue to present symptoms as young adults (Barkley, 2002), and there may be as many as 11 million diagnosed and undiagnosed adults in the United States who suffer from the disorder (Barkley, Murphy, & Fischer, 2008). It is estimated that somewhere between 4% and 5% of adults exhibit severe enough symptoms to place them into a diagnosable category of ADHD (or ADHD-PI—attention deficit disorder—without the hyperactivity; American Psychiatric Association, 2000; Ramsay & Rostain, 2007).

Although the symptom severity associated with ADHD (inattention, hyperactivity, and impulsivity) occurs on a continuum throughout the population (Gordon & Murphy, 1998; Levy, Hay, McStephen, Wood, & Waldman, 1997), the condition at diagnosable levels is described as leading to problems with executive functioning such as distractibility, attention difficulties, heightened motor activity, social inappropriateness, reduced ability to organize and sequence activities, and general inability to concentrate. Downey, Stelson, Pomerleau, and Giordani (1997) found that the most commonly reported symptoms among adults included mental restlessness, maintaining attention, and impatience (Shaw & Giambra, 1993).

The symptoms associated with ADHD may become problematic in several facets of life, but it is generally agreed that they lead to the greatest deficits when those with the disorder are asked to perform in academic situations. That is, without help, those with ADHD tend to perform more poorly in school, even though there is evidence that only slight differences exist in measured intelligence of ADHD versus non-ADHD individuals (Bridgett & Walker, 2006). For a discussion of the impact of ADHD on academic achievement, see Heiligenstein, Guenther, Levy, Savino, and Fulwiler (1999).

The negative impact of ADHD in learning situations has attracted only limited attention among researchers concerned with its influence in college settings, and the need for more research on ADHD in this population has been documented in the literature (Frazier, Youngstrom, Glutting, & Watkins, 2007; Glutting, Monaghan, Adams, & Sheslow, 2002; Wierzbicki, 2005). In one study, DuPaul et al. (2001) reported the ADHD rate among American college students to be 3.9% for men and 2.9% for women. In another study, Weyandt, Linterman, and Rice (1995) found that 7% to 9% of college students reported clinically significant levels of ADHD symptomology.

In addition, Lewandowski, Lovett, Codding, and Gordon (2008) compared the rate of academic complaints and student self-report scores on a Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; American Psychiatric Association, 1994) derived measure of ADHD. Their sample included 496 students without ADHD and 38 with a previous ADHD diagnosis. By setting a median cutoff score on the measure to create a binary decision (clinically symptomatic or not), they found that both the ADHD and non-ADHD students reported significant rates of ADHD symptoms. Consequently, these authors concluded that their self-report measure of ADHD lacked discriminatory power.

Currently, there is a paucity of research investigating the relation between ADHD symptom severity and academic performance in college. In one contribution, measuring one of the three symptoms of ADHD, Spinella and Miley (2003) found a significant negative correlation between scores on an impulsiveness scale and exam performance of 27 undergraduates enrolled in a physiological psychology class. In another study, Frazier et al. (2007) found that the subsequent academic probation versus nonprobation status of 380 college students showed a significant .17 correlation with the Inattentiveness subscale of an ADHD symptomology inventory completed by both students and their parents. Based on their results, the authors suggest that students entering postsecondary education should be routinely screened for ADHD.

To investigate the incidence of adult ADHD symptom severity in general populations, a few researchers have developed self-report screening instruments that have received some research attention (Brown, 1996; Conners et al., 1999; Kessler et al., 2005). Most recently, Kessler et al. (2005) published a scale intended to screen for ADHD in adults. Developed under the auspices of the World Health Organization, and known as the Adult Self-Report Scale–Version 1.1 (ASRS-V1.1), it is a short six-item screening instrument, the questions in which were extracted, using stepwise logistic regression, from a larger survey of 18 questions comprising the Adult Self-Report Survey that taps the 18 specific “Criterion A” symptoms defining the disorder in DSM-IV. These six items are presented below:

How often do you have trouble wrapping up the final details of a project, once the challenging parts have been done?

How often do you have difficulty getting things in order when you have to do a task that requires organization?

How often do you have problems remembering appointments or obligations?

When you have a task that requires a lot of thought, how often do you avoid or delay getting started?

How often do you fidget or squirm with your hands or feet when you have to sit down for a long time?

How often do you feel overly active and compelled to do things, like you were driven by a motor?

The 5-point Likert-type scale ranges from “0” (never) to “4” (very often). Thus, the possible range of scores on the ASRS-V1.1 is 0 to 24, with higher scores indicating more ADHD symptomology. In their validation study, Kessler et al. (2005) administered the screener to 154 respondents who previously participated in the U.S. National Comorbidity Survey Replication. It should be noted that for the purposes of their study, the measure was initially administered to a general population and not specifically to individuals who reported having symptoms of ADHD. Their results indicated that the ASRS-V1.1 had a sensitivity (the proportion of persons with the disorder who are correctly identified by a screening test) rating of 68.7%, a specificity (the proportion of persons without the disorder who are correctly identified by the test) of 99.5%, and a classification accuracy (resistance to false positives and false negatives) of 97.9%. The authors recommended the ASRS-V1.1 for a variety of uses, from broad community screening to private clinical practice.

In a subsequent validity study using the ASRS-V1.1 (Kessler et al., 2007), the authors suggested a scoring strategy that involves summing responses from the six items and dividing them into four distinct categories. A respondent scoring 0 to 9 represents a “highly unlikely” candidate for ADHD, 10 to 13 represents the “unlikely” category, 14 to 16 represents the “likely” category, and score from 17 to 24 represents a “highly likely” category. These four diagnostic strata were generated from general adult population estimates derived from the Kessler et al. (2006) study on national comorbidity rates. In their sample of 686 health care subscribers, the six-item screener demonstrated a strong ability to discriminate between ADHD and non-ADHD individuals (determined by computing and comparing area under the curve statistics between those who were administered the screener only and those who took the screener and were also tested for ADHD in traditional clinical methods). They reported an internal consistency for the ASRS-V1.1 of .72.

Because of its brevity and ease of administration and scoring (an advantage for large-scale screening), the ASRS-V1.1 was selected for the current study. To the extent that the ASRS-V1.1 is, indeed, tapping a tendency toward ADHD, it would be expected from the studies reported previously that college students who score higher on the scale and are categorized as “highly likely” should perform more poorly in traditional instructor-led classroom settings in which they are exposed to lectures and textbooks and are graded by their performance on hour exams (Heiligenstein et al., 1999). Unlike the prior studies, a measure of the students’ academic aptitude (ACT scores) was also considered in the analysis. The current study extends previous research on ADHD symptomology in college students (DuPaul et al., 2001; Frazier et al., 2007; Lewandowski et al., 2008; Spinella & Miley, 2003) by using a predictive validity design that includes in the analyses a measure of academic performance as well as a measure of scholastic aptitude (ACT) that research suggests is a good index of general mental ability (Frey & Detterman, 2004). The purpose of the current research was to investigate the efficacy of the ASRS-V1.1 for determining the incidence of ADHD symptoms in a large college student sample and their possible relationship with a measure of actual academic performance and aptitude.

Method

Participants

The study was conducted in a large, urban public university with a generous admission strategy (minimum ACT score of 16 and 2.2 high school grade point average), a student body characterized by a wide range of academic aptitude (ACT scores), and an average 6-year graduation rate of 39%. According to Hess, Schneider, Carey, and Kelly (2009), this graduation rate is similar to the 39.6% average graduation rate of the “less competitive” universities across the nation. A total of 523 student volunteers (73% females) from three large freshman classes in introductory psychology participated in the study for extra course credit. The course is structured around a lecture/PowerPoint format, with course grade determined by performance on three 1-hr exams. The mean age of the participants was 20.6. The mean ACT composite score (taken from university records) of this group was 21.38 (SD = 3.72), with a range of 12 to 32.

Measures and Procedure

An online version of the ASRS-V1.1 was developed for the study. In addition to the six survey items, the website contained the survey instructions, an implied consent form, and some initial demographic questions. The survey took about ten minutes to complete. Based on the sample of 523, the internal consistency of the ARSR-V1.1 yielded a Cronbach’s alpha reliability score of .81.

Final course grades (based on three multiple-choice hour exams) were subsequently available for all participants. The students’ ACT scores were also included in our analysis. As indicated previously, there is research to indicate that ACT composite scores represent good estimates of general cognitive ability (g; Frey & Detterman, 2004), where the authors found that they correlate .77 with measures of g extracted from the Armed Services Vocational Aptitude Battery. The primary outcome variable, course performance, was determined by the grades on the three 1-hr exams (50 multiple-choice questions, each).

Results

The responses of the 523 students were scored using the ASRS-V1.1 scoring protocol for placing respondents in the four categories mentioned previously. As can be seen in Figure 1, the scores are symmetrically distributed and exhibit a continuum of reported symptom severity in this college sample (Gordon & Murphy, 1998). To check for the possibility of gender effects, correlations for both men and women were calculated using the following variables: ASRS-V1.1 scores, exam performance scores, and ACT composite scores. Weak, yet significant, results were only found in the relationship between women’s ACT scores and ASRS-V1.1 scores, r(380) = .12, p < .05. A similar correlation between the same variables was found for men, yet was not significant due to sample size, r(145) = .11, p = .27.

Figure 1.

Distribution of scores on the ASRS-V1.1 Screener obtained from 523 college students

Not surprisingly, a Pearson product–moment correlation calculated between ACT composite score and exam performance for the entire sample resulted in a substantial positive correlation coefficient, r(521) = .49, p <.001. The correlation between exam performance and ASRS-V1.1 scores resulted in a nonsignificant relationship between the two, r(521) = .07, p = .10. Another interesting question is whether ASRS-V1.1 scores were correlated with ACT scores. That analysis yielded a very weak yet significant positive relationship, r(521) = .13, p < .01. A regression analysis was conducted to determine if the ASRS-V1.1 scores contributed any incremental validity for predicting exam scores, once ACT scores were taken into account. Results indicated that the screener scores did not add a significant source of variance to the model and actually lowered its ability to predict the students’ test scores—ACT alone model: R² = .24, F(1, 522) = 140.58, p < .001; ACT and ASRS-V1.1 model: R² = .24, F(2, 521) = 70.27, p < .001.

Although the above correlations over the entire sample are interesting, the central issue is whether the respondents in the higher end of the ASRS-V1.1 score distribution manifested any deficits in academic performance as measured by their scores on the hour exams. Using the classification strategy recommended by Kessler et al. (2007), a total of 70 (13.38%) of the respondents fell into the “highly likely” category (i.e., 17-24). The average exam score for this group was 65.60% versus 65.22% for the remaining 453 participants.

In addition, an anonymous check with the university’s Office of Student Disability Services indicated that only 4 (5.7%) of the 70 students in the ASRS-V1.1 “highly likely” category were registered with the office. Registration allows ADHD students to take tests and quizzes by themselves in a quiet location and to have 50% more time to take them than is allowed for the other students in the class. Of course, it is likely that some students diagnosed with ADHD would ignore the advice to register with the office.

To determine whether responses on the ASRS-V1.1 would discriminate those who are “highly likely” (n = 70) from those who are “highly unlikely” (n = 77) with respect to their exam performance and/or ACT scores, two independent samples t tests were conducted on these contrasted groups. Results indicated statistical equivalence between the two groups’ mean exam performance, t(145) = −0.09, p = .93, as well as their ACT scores, t(145) = 1.06, p = .29.

Finally, to model data on the national prevalence rates of ADHD (Kessler et al., 2006), we restricted the “highly likely” subset of our sample to reflect the previously reported 4.4% rate of ADHD in the adult population. This resulted in 23 participants in our sample falling within a “very highly likely” group. When compared with the “highly unlikely” group of 77 students, there were no significant differences between the two groups’ test scores or ACT composite scores—test scores: t(116) = .39, p = .69; ACT scores: t(116) = −.09, p = .92).

Discussion

Although the ASRS-V1.1 was found to be reliable, our results for a large college student sample indicate that its ability to discriminate between ADHD likely and non-ADHD likely students did not result in group differences in academic performance. Our findings indicate that the ASRS-V1.1 identified 13.4% of the students as being ADHD “highly likely,” a figure that is substantially higher than that reported by either DuPaul et al. (2001) or Weyandt et al. (1995) for a college population. With respect to the weak positive correlation found between scores on the ASRS-V1.1 and ACT (r = .13), this finding is in contrast with a meta-analysis by Bridgett and Walker (2006), which indicated that, on average, individuals diagnosed with ADHD have slightly lower measures of intelligence. This result was also seen in both genders, yet was only significant in women due to those individuals holding the majority in the sample (ACT vs. ASRS-V1.1 in women: r = .12; ACT vs. ASRS-V1.1 in men: r = .11).

Perhaps most intriguing, however, is the finding that there was no difference in course performance between the “highly likely” group identified by the screener and the rest of the sample. The same was true when the cutoff for the “highly likely” category was increased to match national prevalence rates reported in the literature (Kessler et al., 2006). This finding is not congruent with those of Spinella and Miley (2003) or Frazier et al. (2007) in which performance deficits were associated with ADHD symptomology in college populations. However, there is some research evidence, at least among children, that those diagnosed with ADHD may exhibit above-average academic achievement (Forness, 1992; Sandson, 2000). It is possible that similar dynamics may be in place at the college level.

In attempting to understand why the extent of self-reported ADHD symptom severity bears no relationship with academic performance, one possibility is that the personal awareness of such symptoms is context driven. That is, in contexts (e.g., college classes) where focusing, concentration, and time organizational skills are especially required, ADHD symptoms (i.e., a deficit in these skills) become more apparent to the individual placed in the situation. Thus, one might expect self-reported ADHD symptomology (albeit subclinical) in a young adult college population to be more prevalent than in a comparable noncollege population, thus diluting the screener’s discriminative validity. Of the six questions comprising the ASRS-V1.1 (see previous sections in the article), the first five seem tailor-made for college student complaints. This observation may be partially responsible for the 13.4% of our sample who fell into the “highly likely” category. When responses to only the more generic item number 6 were calculated, only 4.3% of our college sample was in the “highly likely” category (i.e., responding “very often” to the item).

Professionals seeking to help college students who may be suffering from ADHD are faced with two challenges: For students who were correctly diagnosed with the disorder as younger children, they must be encouraged to continue to monitor their symptoms because of the strong likelihood that they will persist into adulthood. Without confirming evidence, entry into college should not be the time to declare ADHD “outgrown” and the medications left at home. The second challenge is to pursue Frazier et al.’s (2007) suggestion to develop strategies for identifying those students who may be entering college with undiagnosed ADHD. Yet, even though the items are derived directly from official diagnostic criteria, currently available self-report screeners, such as the ASRS-V1.1, may not be appropriately written to match the context of a college student’s ADHD-type behaviors.

A limitation of the current study is that only one ADHD screener was evaluated, and other more elaborative surveys are available (Brown, 1996; Conners et al., 1999), which may provide a more accurate and holistic perspective on ADHD among college students. Future research is needed to determine the incidence and effect of ADHD among college students and what, if any, self-report screening measures are most appropriate for studying this population. If none of the currently available screening instruments for ADHD prove to be effective, researchers should be encouraged to pursue other screening strategies for this disorder among college students.

Another potential limitation is that the current study used self-report data to assess ADHD symptomology. Although young adults’ ability to assess the extent of their own symptoms could be called into question, Frazier et al.’s (2007) suggestion that all entering college students be screened for ADHD would, from a practical perspective, necessitate such self-reports. This obvious practical constraint was taken into consideration in the design of the current study. Considering the prevalence of the disorder and its influence on the dynamics of learning at the college level, researchers should consider the potential positive impact that could come from effective detection and intervention strategies to reduce ADHD in this population of young adults.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Bios

Jonathan Burlison is a Ph.D. candidate in Industrial/Organizational Psychology at the University of Memphis. His research interests are on the effects of attention disorders on college students and employees, and the positive and negative influences of web-based social networking in employment settings.

William O. Dwyer received his PH.D. in Psychology from Southern Illinois University-Carbondale in 1969 and has been on the faculty at The University of Memphis since his graduation. At Memphis he directs the Psychology Department’s Industrial-Organizational concentration, and his current research focus is on the impact of ADHD in the workplace.

References

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text rev.). Washington, DC: Author.

Barkley

R. A.

(2002). ADHD—Long-term course, adult outcome, and comorbid disorders. In Jensen

P. S.

Cooper

J. R.

(Eds.), Attention deficit hyperactivity disorder: State of the science, best practices (pp. 4-7). Kingston, NJ: Civic Research Institute.

Barkley

R. A.

Murphy

K.R.

Fischer

(1998). ADHD in Adults: What the Science Tells Us. New York: Guilford Press.

Bridgett

Walker

(2006). Intellectual functioning in adults with ADHD: A meta-analytic examination of full scale IQ differences between adults with and without ADHD. Psychological Assessment, 18, 1-14.

Brown

T. E.

(1996). Brown Attention-Deficit Disorder Scales. San Antonio, TX: The Psychological Corporation.

Conners

Erhardt

Epstein

Parker

Sitarenios

Sparrow

(1999). Self-ratings of ADHD symptoms in adults I: Factor structure and normative data. Journal of Attention Disorders, 33, 141-151.

Downey

Stelson

Pomerleau

Giordani

(1997). Adult attention deficit hyperactivity disorder: Psychological test profiles in a clinical population. Journal of Nervous and Mental Disease, 185, 32-38.

DuPaul

G. J.

Schaughency

E. A.

Weyandt

L. L.

Tripp

Kiesner

Ota

Stanish

(2001). Self-report of ADHD symptoms in university students: Cross-gender and cross-national prevalence. Journal of Learning Disabilities, 34, 370-379.

10.

Forness

(1992). Classroom instructional characteristics in attention deficit hyperactivity disorder: Comparison of pure and mixed subgroups. Behavioral Disorders, 17, 115-125.

11.

Frazier

T. W.

Youngstrom

E. A.

Glutting

J. J.

Watkins

M. W.

(2007). ADHD and achievement: Meta-analysis of the child, adolescent, and adult literatures and a concomitant study with college students. Journal of Learning Disabilities, 40, 49-65.

12.

Frey

Detterman

(2004). Scholastic assessment org: The relationship between the scholastic assessment test and general cognitive ability. Psychological Science, 15, 373-378.

13.

Glutting

J. J.

Monaghan

M. C.

Adams

Sheslow

(2002). Some psychometric properties of a system to measure ADHD among college students: Factor pattern, reliability, and one-year predictive validity. Measurement and Evaluation in Counseling and Development, 34, 194-209.

14.

Gordon

Murphy

K. R.

(1998). Attention deficit/hyperactivity disorder. In Gordon

Keiser

(Eds.), Accommodations in higher education under the Americans With Disabilities Act: A no-nonsense guide for clinicians, educators, administrators, and lawyers (pp. 98-129). New York, NY: Guilford.

15.

Heiligenstein

Guenther

Levy

Savino

Fulwiler

(1999). Psychological and academic functioning in college students with attention deficit hyperactivity disorder. Journal of American College Health, 43, 226-228.

16.

Hess

Schneider

Carey

Kelly

(2009). Diplomas and dropouts: Which colleges actually graduate their students (and which don’t). American Enterprise Institute for Public Policy Research. Retrieved from http://www.aei.org/paper/100019

17.

Kessler

R. C.

Adler

Ames

Demler

Faraone

Hiripi

. . . Walters

E. E.

(2005). The World Health Organization adult ADHD Self-Report Scale (ASRS): A short screening scale for use in the general population. Psychological Medicine, 35, 245-256.

18.

Kessler

R. C.

Adler

Barkley

Biederman

Conners

Demler

. . . Zaslavsky

(2006). The prevalence and correlates of adult ADHD in the United States: Results from the National Comorbidity Survey Replication. American Journal of Psychiatry, 163, 716-723.

19.

Kessler

R. C.

Adler

L. A.

Gruber

M. J.

Sarawate

C. A.

Spencer

Van Brunt

D. L.

(2007). Validity of the World Health Organization Adult ADHD Self-Report Scale (ASRS) screener in a representative sample of health plan members. Journal of Methods in Psychiatric Research, 16, 52-65.

20.

Levy

Hay

D. A.

McStephen

Wood

Waldman

(1997). Attention-deficit hyperactivity disorder: A category or a continuum? Genetic analysis of a large-scale twin study. Journal of the American Academy of Child & Adolescent Psychiatry, 36, 737-744.

21.

Lewandowski

Lovett

Codding

Gordon

(2008). Symptoms of ADHD and academic concerns in college students with and without ADHD diagnoses. Journal of Attention Disorders, 12, 156-161.

22.

Ramsay

R. J.

Rostain

A. L.

(2007). Psychosocial treatments for attention-deficit/hyperactivity disorder in adults: Current evidence for future directions. Professional Psychology: Research and Practice, 38, 338-346.

23.

Sandson

(2000). Right hemisphere dysfunction in ADHD: Visual hemispatial inattention and clinical subtype. Journal of Learning Disabilities, 33, 83-90.

24.

Shaw

Giambra

(1993). Task unrelated thoughts in college students diagnosed as hyperactive in childhood. Developmental Neuropsychology, 9, 17-30.

25.

Spinella

Miley

W. M.

(2003). Impulsivity and academic achievement in college students. College Student Journal, 37, 545-549.

26.

Weyandt

Linterman

Rice

(1995). Reported prevalence of attentional difficulties in a general sample of college students. Journal of Psychopathology and Behavioral Assessment, 17, 293-303.

27.

Wierzbicki

(2005). Reliability and validity of the Wender Utah Rating Scale for college students. Psychological Reports, 96, 833-839.