Pediatric Attention Disorders Diagnostic System:

1
Pediatric Attention Disorders Diagnostic
System Clinical Utility and Psychometric
acceptability   Presented at the American Academy
of Pediatrics 21st Century Symposium
Incorporating Mental Health Screening in Primary
Care Settings OCTOBER 7th, 2005 Thomas K. Pedigo
Ed. D. Pediatric Adolescent Psychology
P.C. Savannah, GA 31406
Vann B. Scott, Jr., Ph.D.
Armstrong Atlantic State University
Department of Psychology, Savannah, GA
31419 Diane R.
Savage-Pedigo, M.D. Pediatric
Associates of Savannah P.C. Savannah,
GA 31401
2
PEDIATRIC ATTENTION DISORDERS DIAGNOSTIC
SYSTEM(PADDS OVERVIEW)

• PADDS is a system with proven clinical
  reliability and validity in the screening and of
  attention disorders. This process merges three
  short and enjoyable computer administered tasks
  (Target Subtests) with a computer
  administered/scored diagnostic interview.
• The padds cognitive tests are designed to
  closely re-create the basic demands of the
  typical classroom setting by tapping greater
  aspects of executive functioning, i.e. planning,
  organization, working memory, attention to detail
  and changes in stimuli.
• The Computer Assisted Diagnostic Interview or
  CADI covers the major areas of comorbidity needed
  to reliably screen ADHD. Both processes can be
  completed in total in under 45 minutes. All
  information from both components are maintained
  in a data base for collection and comparison over
  time. Since this system can be effectively
  administered by any clinician or assistant, the
  physicians time can be more appropriately spent
  evaluating the results, collected along multiple
  lines of evidence, as well as face to face with
  the patient and family. This standardized
  evidence-based approach is directly in line with
  the current emphasis of best practices call for
  by prominent healthcare agencies.

3
Relevance/Application to the Primary Care
Setting   According to the American Academy of
Pediatrics, ADHD is the most commonly diagnosed
childhood psychiatric disorder affecting school
age children. Epidemiological studies have shown
a prevalence rate ranging from 3 percent to 6
percent of school age children. Concern has been
expressed for the these large numbers coupled
with reportedly wide variations in clinical
practice and research approaches all point to
the need to develop pragmatic assessment tools
and approaches for use in the major systems of
service entry. Specifically of importance are
assessment approaches that can be used within
primary care settings, schools and clinics as
well as within the private sector.   Reference
Chan, E., Hopkins, M., Perrin, J. M., Herrerias,
C., Homer, C. J., (2002) VARIATIONS IN
DIAGNOSTIC PRACTICES FOR ATTENTION DEFICIT
HYPERACTIVITY DISORDER A NATIONAL SURVEY OF
PRIMARY CARE PHYSICIANS Homer Division of General
Pediatrics, Children's Hospital, Boston, MA
American Academy of Pediatrics, Elk Grove
Village, IL Center for Child and Adolescent
Health Policy, MassGeneral Hospital for Children,
Boston, MA National Initiative for Children's
Healthcare Quality, Institute for Healthcare
Improvement, Boston, MA. (2002) Pediatric
Academic Societies Abstract.
4
During the 1998 NIH Consensus Development
Conference it was determined that Development and
Validation of Diagnostic Tools Grounded in the
Basic Sciences was warranted.   Key areas of
interest to the NIH   The Development and
Validation of Diagnostic Tools Grounded in the
Basic Sciences   Consequently, there is a
continued need to develop more objective
assessment tools, rating scales and/or diagnostic
interviews that map onto basic underlying
processes as well as a need to supplement
behavioral assessment tools with improved
cognitive and/or neuropsychological measures.
The Development of Strategies for Assessing,
Monitoring and Administering Treatment in
Primary Care Settings   Many of the currently
utilized assessment measures and treatments for
ADHD are incompatible with the primary care
setting. There is also a dearth of practical
decision-making tools for medication monitoring,
differential diagnosis, and the distinction of
referral service needs based upon impairment
severity. Consequently, there is a great need
for the development of practical, reliable and
valid procedures to be used in primary care
settings to identify and manage ADHD symptoms, as
well as to distinguish appropriate referral
needs.
5
Diagnostic Challenges/Comorbidity with ADHD  
Other Comorbid conditions often occur with ADHD.
These conditions may include but are not limited
to Mood Disorders, Anxiety Disorders, Disruptive
Behavior Disorders and Learning Disorders.
Bipolar Disorder is becoming increasingly
recognized by some professionals within
adolescent populations. The importance of
considering other conditions that may mimic or
exacerbate the presence of ADHD is essential to
successful intervention.
6
The following listing of ranges for ADHD and
Comorbid conditions was adapted from the
following source Pliszka, S. R., Carlson, C.
L., Swanson, J. M., (1999). ADHD with Comorbid
Disorders Clinical assessment and Management.
New York, N.Y. The Guilford Press.    Primary
Diagnosis / Secondary Diagnosis Range of
Prevalence Page Number ADHD/ODD-CD 15
to 61 90 ODD-CD/ADHD 35 to 87 90  
ADHD/Depression 0 to 38 127
Depression/ADHD 0 to 57 127  
ADHD/Anxiety 23 to 30 151 Anxiety/ADHD
9 to 35 151
7
Primary Diagnosis / Secondary Diagnosis Range of
Prevalence Page Number ADHD/LD 7 to
60 192 (Across- Reading, Spelling,
Math)   ADHD/OCD 6 to 33 214   Other
related conditions needing assessment/
consideration include   Neurological Impairment
  Developmental disabilities   PDD/Autistic
spectrum disorders
8
Executive Functions and Diagnosis of ADHD
Recent developments within the field of ADHD
have increasingly pointed to the need to evaluate
the various executive operations and working
memory of children suspected of Attention
Disorders. (Brown, T.E., 2002, 2000,1999
Barkley, R.A. 1997,1998 Denckla M, 1996.)
Generally, executive functions are defined as
controls that allow one to perform complex
behaviors that require among other things
planning, attending, organizing input, storing
and retrieving information, modulating emotions
and sustaining effort.
9
While the identification of significantly
hyperactive children can be simple, the
evaluation of children who only display
difficulty in learning or in completing more
complex activities is where the greatest need for
improvement lies. Difficulties in these Executive
Processes (planning, attending, organizing input,
storing and retrieving information, modulating
emotions and sustaining effort) exemplify the
complaints of teachers and parents. Situations
that require an orchestration of these abilities
are often most problematic for AD/HD students.
Parents will often report confusion at their
child's ability to play video games, watch
television or engage in favorite activities.
However, on closer inspection, these activities
often do not produce the same demands as found
within the classroom. These favorite activities
are often overlearned, fast pace, and allow the
child to move freely in and out of the activity.
Changing the structure of these activities
(implementing learning demands) can quickly
produce frustration in AD/HD children.
10
Basic Demands of the Classroom   Attending to
instruction Assimilating information Accommodating
information Organizing, sequencing, manipulating
information Monitoring emotional
activity Formulating a plan of action Implementing
the plan   Other Factors  Working under time
pressure Avoiding distraction Being adequately
prepared THE PADDS TARGET SUBTESTS WERE DESIGNED
TO PRODUCE WORK DEMANDS SIMILAR TO THOSE OUTLINED
ABOVE
11
PADDS
The Clinicians ADHD Toolbox

• Computer based system to collect and compare
  multiple lines of evidence for ADHD diagnosis
• Patient Information Database and Reporting
• Comprehensive Parent and Teacher Interviews in a
  self-running or Clinician input format
• A battery of newly developed cognitive tests
  presented in a challenging, enjoyable format
• Automatic Report Generator with domain specific
  alerts and recommendations, and follow up
  comparisons of Treatments and Progress

12
COMPUTER ASSISTED DIAGNOSTIC INTERVIEW
(CADI) The Computer Assisted Diagnostic
Interview or CADI covers the major areas of
comorbidity needed to reliably screen ADHD. A
review of systems includes
Medical History/Systems Review
Developmental History
Social/Emotional Functioning
Depression/Anxiety
Attention/Hyperactivity Behavioral/School
History The CADI provides two options for
administration. The first is a protocol completed
by the parent and then quickly input by the
examiner or assistant. The second is presented in
the auditory domain via the computer for those
that cannot read or have not completed the
protocol. All information from both components
are maintained in a data base for collection and
comparison over time. A concise report is
generated highlighting parental concerns and/or
comorbid issues that can be cross validated in a
straight forward interview. Parents have
expressed appreciation at the range of
information asked of them as opposed to only
focusing on behavioral scales and ADHD symptoms.
This report will consolidate this important
information by domain so the examiner can
efficiently review and validate concerns or needs
for further referral. BELOW ARE ABBREVIATED
EXAMPLES OF THE CADI INPUT AND REPORT
13
Medical History/Systems Review Developmental
History Social/Emotional Functioning Depression/An
xiety Attention/Hyperactivity Behavioral/School
History
14
(No Transcript)
15
(No Transcript)
16

Cognitive Tests Executive Functioning
17
Target Recognition presents five large colored
squares with smaller squares inside them. Through
153 presentations some number of the large
squares will have smaller squares of the same
color and some number will be different colors.
The child is taught a strategy to read from left
to right and count only the number of squares
with matching colors. This task requires
suppression of information, attention to detail,
formulation of a response to changes in stimuli,
modulation of emotions and persistence.
18
Target Sequencing presents five large colored
circles. In each of 39 trials a small colored
square appears and then disappears in each
circle, in varied sequences. The child is taught
to only attend to circles with a matching colored
square. At the end of the trial the child is
required to click on each matching circle in the
order observed ( first match first,
second match second and last match last). Target
Sequencing requires the ability to avoid
distraction, attention to detail, organization
and sequencing during input of information,
planning and organization of a response,
modulation of emotion and sustained effort.
19
Target Tracking presents four colored shapes at
the top and bottom of the computer screen. The
computer moves two or three shapes from the top
to the bottom shapes. The child is required to
remember the order of these moves and to recreate
them once all shapes have returned to the top of
the screen. Target Tracking requires the ability
to organize two and three step instructions, and
to recreate these instructions in the order
presented while modulating emotions and
sustaining effort across 20 trials.
20
Generate Reports Cognitive Tests Report
21
SUBJECT SELECTION, METHODOLOGY AND
RESULTS Clinical pilot testing of the PADDS
consisted of 200 children age 6 to 12 years. 95
children diagnosed with either ADD or ADHD and
105 typical peers age 6 to 12 years. Data
collection consisted of two phases with subject
sizes of 113 and 87 respectfully. Phase 1
reviewed the performance of Target Recognition
and Target Sequencing. Phase Two also included
review of the original two subtests along with a
third new subtest Target Tracking. All subjects
lived in small to moderate size cities within
Georgia. Consent forms were sent home to parents
explaining the study and all children with
parents consenting were tested and received a
10.00 dollar gift certificate to TOYS R US for
participating. Children not meeting the screening
criteria outlined below were excluded from the
study. As ADD/ADHD crosses age, gender, race and
socioeconomic status, no data regarding these
demographics were considered appropriate to
report. The ADD/ADHD children were drawn from a
clinical pediatric psychology practice where
comprehensive evaluation and testing had
previously been completed and a diagnosis
rendered. The ADD/ADHD children included in the
study had recently or previously received a
diagnostic assessment and testing process
including background, developmental, medical and
family histories, reports from home and school,
any available school testing and grades,
intelligence testing, cognitive testing for short
term auditory and visual memory, performance on a
continuous performance tests and a review of
criteria met for diagnosis based on the DSM-IV
Criteria for ADD/ADHD.
22
SUBJECT SELECTION, METHODOLOGY AND RESULTS
CONTINUED Prior to clinical testing with the
PADDS Target Subtests each of the ADD/ADHD
children had been prescribed medication with
reports of improvement from home and school.
While taking the PADDS all ADD/ADHD children were
taken off their psychostimulant medication.
Within the ADD/ADHD sample, all psychological
testing data was reviewed to ensure there were no
significant emotional disorders or clinically
significant signs of depression or anxiety
evident in any of the subjects. Only ADD/ADHD
children receiving psychostimulants were included
in the clinical study. All ADD/ADHD children were
tested with the PADDS during the morning hours to
limit diurnal effects on test performance. The
typical or non ADD/ADHD children were drawn from
different elementary schools or Boy Scout Troops
within a moderate size city from the same state
and region. The typical children were also tested
in the morning and were screened with the Conners
Rating Scale-Teacher Version prior to inclusion
in the study. All subjects with a significant
Conners Rating Scale were compared to their PADDS
performance to assess agreement between measures
however, their results were not included in the
clinical study of the PADDS. Initial review of
the 3 Target Subtests showed they demonstrated
acceptable ability to discriminate between
ADD/ADHD children and their typical peers. Tables
1-3 show the results of the Receiver Operating
Characteristics analysis performed to determine
the best cut points for Sensitivity and
Specificity for true diagnosis.
23
PADDS SUPPORTIVE RESEARCH (TABLE1) FOR TARGET
RECOGNITION PHASE 1 N113 PHASE 2 N87 TOTAL
N200
ADHD N 95 Correctly ID 87 Missed 8 Total
Discrimination 91.7
TYPICAL N 1O5 Correctly ID 88 Missed
17 Total Discrimination 84
TOTAL DISCRIMINATION 87.85 of 200 test
subjects
24
PADDS SUPPORTIVE RESEARCH (TABLE 2) FOR TARGET
SEQUENCING PHASE 1 N113 PHASE 2 N87 TOTAL
N200
ADHD N 95 Correctly ID 80 Missed 15 Total
Discrimination 84.
TYPICAL N 1O5 Correctly ID 89 Missed
16 Total Discrimination 85
TOTAL DISCRIMINATION 84.5 of 200 test subjects
25
PADDS SUPPORTIVE RESEARCH (TABLE 3) FOR TARGET
TRACKING PHASE 2 N87
ADHD N 41 Correctly ID 40 Missed 1 Total
Discrimination 97.6
TYPICAL N 46 Correctly ID 40 Missed 6 Total
Discrimination 87
TOTAL DISCRIMINATION 92. of 87 test subjects
26
PADDS SUPPORTIVE RESEARCH (TABLE 4) TEST /
RETEST 43 SUBJECTS DRAWN FROM BOTH STUDIES
43 ADHD SUBJECTS N 43 Trial 1 40 Miss 3 R
.93 INTERCORRELATIONS OF TARGET SUBTESTS Plt.05
Variable Phase 1 Target R Target S
Target R 1.00 .56
Target S .56 1.00
Variable Phase 2 Target R Target S Target T
Target R 1.00 .61 .42
Target S .61 1.00 .40
Target T .42 .40 1.00
27
PADDS PHASE 1-2 DIVERGENT VALIDITYMARKED
CORRELATIONS ARE SIGNIFICANT AT Plt.0500N54
N41TABLE 5
PHASE 1 AGE DX
AGE 1.00 .47
DX .47 1.00
CMSATT -.25 -.06
CMSVER -.22 -.10
FSIQ -.26 .05
TRECOG .41 .39
TSEQ .51 .51
PHASE 2 AGE DX
AGE 1.00 .27
DX .27 1.00
CMSATT -.25 .00
CMSVER -.22 -.10
FSIQ -.26 .05
TRECOG .41 .39
TSEQ .51 .51
TTRACK .44 .39
28
PHASE 1 2 DESCRIPTIVE STATISTICSTABLE 6

• PHASE 1 N MEAN MEDIAN SD
• AGE ADHD 54 7.67 7.50 1.54
• AGE TYP 59 8.78 9.00 1.78
• TR-ADHD 54 71.57 75.50 35.43
• TR-TYP 59 122.56 132.00 25.73
• TS-ADHD 54 17.69 19.50 9.69
• TS-TYP 59 32.31 35.00 7..97
• PHASE 2
• AGE ADHD 41 7.95 8.00 1.44
• AGE TYP 46 8.85 9.00 1.94
• TR-ADHD 41 60.88 46.00 31.86
• TR-TYP 46 125.43 132.00 27.37
• TS-ADHD 41 17.83 16.00 9.17
• TS-TYP 46 32.15 33.50 6.91
• TT-ADHD 41 4.15 4.00 1.85
• TT-TYP 46 12.78 13.50 4.18

29
RELIABILITY VALIDITY OVERVIEW The PADDS is
a new measure and as such will continue to under
go psychometric evaluation. As pointed out by
Rosenthal Rosnow (1991), When using a test to
measure differences among individuals,
Reliability refers to the ability of the test to
consistently discriminate individuals at some
point in time and over time. As such the
principal criteria for determining the
reliability of a psychological test is that tests
ability to consistently discriminate between
individuals at one point in time referred to as
internal consistency and the ability to do so
over time which is referred to as test retest
reliability. The criteria used to establish
acceptability of the PADDS's reliability was in
keeping with the guidelines suggested by
Rosenthal Rosnow (1991) of .85 or better for
use of clinical measures. Validity refers to
the ability of a test to measure the traits or
dimensions intended. As stated in the Standards
for Educational and Psychological testing (1999),
It is the interpretations derived from a tests
use that are validated, not the test itself. In
the development of the PADDS special attention
was given to provide evidence of Discriminate ,
Convergent and Divergent validity. Data collected
during the clinical trials of the PADDS were
analyzed using the SPSS software program for
windows by SPSS 1999.
30
RESULTS OF TARGET SUBTESTS PSYCHOMETRIC
PERFORMANCE ROC (receiver operating
characteristic curve) analysis was used to
determine the best cut-point to use for TargetR
and TargetS to predict ADHD and Typical status.
Sensitivity is the probability of a positive test
result, given that the patient truly is positive.
Specificity is the probability of a negative
test result, given that the patient truly is
negative. The ROC curve analysis calculates the
sensitivity and specificity of the Target Tests
for every possible cut point in the sample. An
area of 1 would indicate perfect discrimination
for each group. Review of the ROC shows that the
TargetR, TargetS and TargetT subtests are very
similar with respect to overall accuracy. Review
of Phase 1 data shows that the area under the
curve (95 confidence interval) was 0.896 (0.836,
0.956) versus 0.887 (0.825, 0.949) for TargetR
and TargetS respectively. Review of Phase 2 data
shows the area under the curve (95 confidence
interval) was 0.932 (0.876-0.988) for Target
Recognition, 0.905 (0.839-0.972), for Target
Sequencing and 0.971 (0.937-1) for The new Target
Tracking Subtest. Thus, TargetR , TargetS and
TargetT appear to be clinically significantly
with respect to overall accuracy of prediction.
31
RESULTS PSYCHOMETRIC PROPERTIES
CONTINUED Tables 1- 3 shows the sensitivity and
specificity for the final cut-points selected for
analysis. The cut-off value is included for
positive classification. For example, Target R
was set at a cut-point of gt114 (normal) versus
lt114 (ADHD). When this cut point is used is used,
the sensitivity of the test is estimated to be
0.91 and the specificity is estimated to be
0.84. This results in an overall accuracy of
prediction of 87.85 with 200 subjects. The
Target S cut-point was set at gt28 (normal)
versus lt28 (ADHD). When this cut point is used,
the sensitivity of the test is estimated to be
0.84 and the specificity is estimated to be
0.85. This results in an overall accuracy of
prediction of 84.5 with 200 subjects. The Target
T cut-point was set at gt8 (normal) versus lt8
(ADHD). When this cut point is used, the
sensitivity of the test is estimated to be 0.97
and the specificity is estimated to be 0.87.
This results in an overall accuracy of prediction
of 92. with 87 subjects. Table 4 shows the
results of 43 randomly selected test subjects
that were retested with all three target subtests
to determine Test-Retest reliability. The result
was correct reclassification of 40 out of 43
subjects with a corresponding reliability
coefficient of . 93. Review of the Target Tests
subtests indicates clinically acceptable
reliability to support further clinical
development.
32
PADDS PHASE 12 CONVERGENT AND DIVERGENT
VALIDITY As can be seen in tables 1-3 the
individual subtests all met clinical levels of
acceptability for the discrimination of known
clinical and non-clinical groups. Further Table 4
shows the highly significant inter-correlations
of these subtests between themselves. This would
indicate convergent validity for subtests
designed to correctly screen for the clinical
condition of ADHD. Evaluation of diagnostic
sensitivity of the PADDS subtests was compared
with each other and to that of several measures
unrelated to the clinical diagnosis of ADHD.
Table 5 presents the results of correlations from
dummy coding classification of ADHD status as 1
and Typical or non-clinical status as 0. Thus low
scores on the Target subtests would be more
indicative of ADHD status. The expectation would
be highly significant but negative correlations
with Target subtests as ADHD subjects perform
significantly worse on these measures than do
their typical or non-clinical peers. As
indicated by the highly significant negative
correlations of the ADHD subjects with
performance on the PADDS subtests these subtests
are indeed performing the discriminations at
clinically acceptable levels and well beyond
other measures that have no ADHD diagnostic
purpose. In general this would be indicative of
acceptable divergent validity in the clinical
evaluation of the PADDS subtests as related to
ADHD Screening.
33
DISCUSSION OF THE PADDS PHASES 1-2 PSYCHOMETRIC
PERFORMANCE
The initial performance of the PADDS clinical
subtests included 200 children aged 6 to 12 years
split evenly between known diagnostic
classification of ADHD and NON-ADHD. Across
phases 1nad 3 all Target subtests met the
threshold for reliable and valid screening of
ADHD to warrant further external cross validation
of the acceptability of the Target subtests use
in ADHD screening. Future evaluation will require
external evaluation that includes children from
more diverse geographic regions, convergent
comparison with other tests that have received
sound clinical acceptance with the screening of
ADHD and comparison of sensitivity and
specificity across the individual age groupings
of the research samples.
34
INDEPENDENT/EXTERNAL CROSS VALIDATION OF THE
PEDIATRIC ATTENTION DISORDERS DIAGNOSTIC
SCREENER Vann Scott, Ph.D. Department of
Psychology at Armstrong Atlantic State University

• An independent validation of the Pediatric
  Attention Disorder Diagnostic Screener (PADDS),
  conducted by Vann Scott, Ph.D. of the Department
  of Psychology at Armstrong Atlantic State
  University, was completed in the 2004-2005
  academic year. A synopsis of this work follows.
  The Pediatric Attention Disorder Diagnostic
  Screener (PADDS) is a recently developed
  diagnostic tool designed for use with other
  diagnostic criteria to reduce the over
  identification of children with ADHD. It is well
  recognized in the literature on ADHD that
  multiple sources of information should be
  consulted in the diagnosis of the disorder. While
  multiple instruments exist for diagnosis of ADHD
  (such as rating scales and other psychological
  assessments) the costs of these tools can be
  prohibitive and in many cases unnecessary. The
  ultimate goal of the development of this tool is
  to provide physicians, clinical psychologists,
  and other mental health care workers with a quick
  and enjoyable computer-based assessment that,
  when properly combined with other assessment
  criteria, will provide the clinician with
  appropriate information to more accurately
  determine appropriate referrals and/or treatment.
  It is important to note that PADDS is not
  designed to be a stand-alone diagnostic tool.

35
PADDS INDEPENDENT CROSS VALIDATION PROCEDURES

• A sample of 88 typical children (aged M 9.44,
  SD 1.16) from a school district in Idaho were
  assessed for ADHD utilizing the PADDS and the
  Connors Behavior Rating Scale (Teachers
  version). This group of typical children was
  compared to a sample of 99 children from
  Savannah, Georgia (aged M 8.24, SD 1.71) who
  were diagnosed with ADHD and were temporarily
  removed from medication for the purposes of
  testing. The clinical sample was administered the
  PADDS and other diagnostic tools (e.g., the Brown
  rating scale, Conners continuous performance test
  2nd ed,,). Particular consideration was given to
  the degree of specificity and sensitivity of the
  PADDS based on previously established
  cut-points.Also evaluated were convergent
  validity with other established ADHD clinical
  measures and divergent validity in comparison of
  the diagnostic sensitivity of the PADDS Target
  subtests with other non ADHD measures (IQ and
  memory measures). Recommendations for development
  are offered.

36
PADDS INDEPENDENT CROSS VALIDATION RESULTS

• In the overall sample, Table 7 shows the
  sensitivity (probability of a positive result
  when the child has ADHD) was 91.9 whereas
  specificity (probability of a negative result
  when the child does not have ADHD) was 86.4.
  This resulted in an overall hit rate 89.3 (167
  of 187). These results are highly consistent with
  the initial phase 1 and 2 results reported
  earlier in this presentation and suggest
  acceptable reliability was again established.
  Convergent validity was assessed with a sample of
  40 ADHD subjects looking at the rate of
  diagnostic sensitivity for the PADDS, the Brown
  ADHD Rating Scales and the Conners CCPT-II.
  Table 8 shows the rates of sensitivity within and
  between each measure for ADHD classification. As
  can be seen the PADDS discriminated 95 of this
  sample correctly followed by 77.5 for the Brown
  and Conners measures respectfully. Also,
  significant correlation was established between
  the PADDS and each of these measures with respect
  to diagnostic sensitivity. Collectively this
  points to strong convergent validity for ADHD
  diagnosis with two clinically accepted measures
  of ADHD screening. As found in the phase 1 and 2
  samples, the PADDS subtests were significantly
  correlated with poorer performance among the ADHD
  group while unrelated measures like IQ and Memory
  testing were not correlated to diagnosis. This
  further suggests divergent validity for the
  Target subtests with regard to ADHD screening.
  Finally, Age was correlated with performance on
  each of the subscales of the PADDS target
  recognition (r .46, p lt .001), target
  sequencing (r .52, p .001), and target
  tracking (r .56, p lt .001). As a result,
  further research should be conducted to determine
  the levels of specificity and sensitivity of the
  PADDS within various age groups.

37
PADDS INDEPENDENT CROSS VALIDATION SENSITIVITY
AND SPECIFICITY (TABLE 7) TOTAL N187
ADHD N 99 Correctly ID 91 Missed 8 Total
Discrimination 91.9
TYPICAL N 88 Correctly ID 76 Missed
12 Total Discrimination 86.4
TOTAL DISCRIMINATION 89.3 of 187 test subjects
38
SENSITIVITY FOR ADHD CLASSIFICATION FOR PADDS,
BROWN RATING SCALES CONNERS CCPT-11TABLE
8N40

• RATE OF DIAGNOSTIC SENSITIVITY/AGREEMENT
• PADDS BROWN SCALES CONNERS CCPT-II
• 38/40 31/40 31/40
• 95 77.5 77.5
• PADDS/BROWN PADDS/CONNERS
• 29/40 24/40
• .72 .62

39
External Cross Validation AASU Convergent
Divergent Validity TABLE 9 P lt.0001
Clinical diagnosis ADHD Pearson Correlation 1
Clinical diagnosis ADHD N 78
FSIQ Pearson Correlation -.165
FSIQ N 78
CMSVIS Pearson Correlation -.031
CMSVIS N 76
CMSV Pearson Correlation -.066
CMSV N 76
CMSATT Pearson Correlation -.183
CMSATT N 76
TR Pearson Correlation -.423()
TR N 78
TS Pearson Correlation -.446()
TS N 77
TT Pearson Correlation -.496()
TT N 77
AGE Pearson Correlation -.394()
AGE N 78

40
RECOMMENDATIONS FOR FURTHER DEVELOPMENT

• Phases 1-2 pilot testing and subsequent
  independent cross validation of the TARGET
  subtests of the PADDS ADHD screening system have
  demonstrated clinically acceptable reliability
  and validity to discriminate between and
  properly classify ADHD children and their typical
  age peers with samples of 200 and 187
  respectfully.
• Collectively these results are taken to support
  the continued research and development of these
  measures for clinical use. Future study should
  focus at minimum on the following areas
• 1. Subsequent independent cross validation
  utilizing samples from a wider geographical
  representation for both ADHD and Typical
  children.
• 2. Evaluation of sensitivity and specificity
  across each age grouping within the PADDS samples
  (6-12).
• 3. Further review of validity with other
  clinically accepted measures of executive
  functions.
• 4. Given the visual nature of the Target
  Subtests, Evaluation of children with primary
  Reading disabilities would be prudent to
  determine the degree if any that these measures
  significantly identify that group.

41
References Barkley, R. A.(1998).
Attention-Deficit Hyperactivity Disorder A
Clinical Workbook, Second Edition. New York
Guliford Press. Barkley, R. A. (1997). ADHD and
the Nature of Self-Control. New York Guliford
Press. Barkley, R. A Grodzinsky, G DuPaul G.
J. (1992). Frontal lobe functions in attention
deficit disorder with and without hyperactivity
a review and research report. Journal of Abnormal
child Psychology 20 (2), 163-184. Brown, T.E.,
(Ed). (1999). DSM-IV ADHD and Executive Function
Impairments. Advanced Studies in Medicine, 2
(25), 910-914. Brown, T.E., (Ed). (1999).
Attention Deficit Disorders and Comorbidities in
Children, Adolescents and Adults. Washington
D.C. American Psychiatric Press. Chan, E.,
Hopkins, M., Perrin, J. M., Herrerias, C.,
Homer, C. J., (2002) VARIATIONS IN DIAGNOSTIC
PRACTICES FOR ATTENTION DEFICIT HYPERACTIVITY
DISORDER A NATIONAL SURVEY OF PRIMARY CARE
PHYSICIANS Homer Division of General Pediatrics,
Children's Hospital, Boston, MA American Academy
of Pediatrics, Elk Grove Village, IL Center for
Child and Adolescent Health Policy,
MassGeneral Hospital for Children, Boston, MA
National Initiative for Children's Healthcare
Quality, Institute for Healthcare Improvement,
Boston, MA. (2002) Pediatric Academic Societies
Abstract. Cohen, M.J., Riccio, C.A., Gonzalez,
J. J. (1994). Methodological differences in the
diagnosis of Attention Deficit Hyperactivity
Disorder Impact on prevalence. Journal of
Emotional and Behavioral Disorders. 02 (01),
31-38. Denckla M. Theory and model of executive
function a neuropsychological perspective. In
Lyon, Krasnegor, eds. Attention, Memory and
Executive Function. Baltimore, Md Brookes
1996263-278. Pliszka, S.R., Carlson, C.L.,
Swanson, J. M., (1999). ADHD with Comorbid
Disorders Clinical Assessment
and Management. New York, N.Y. The
Guilford Press.