Download Validity of Denver Developmental Screening Test in Rural and Semirural Canadian Population and more Lecture notes Public Health in PDF only on Docsity!
The Denver Developmental
Screening Test
A Normative Study
KEITH E. BARNES
ANNETTE STARK
The developmental norms of preschoolers from an urban
population, collected for the Denver Developmental
Screening Test, were compared with norms of
preschoolers from a Canadian rural and^ semirural
population.
Introduction
In 1967 Frankenburg and Dodds introduced a new
developmental screening test for infants and preschoolers
between the ages of^ 2 weeks and^ 6.4^ years.' The^ major
purpose of this new screening instrument was to^ detect
slow development in four functional areas of development:
gross motor,^ fine^ motor,^ language,^ and^ personal-social.^ The
test was devised as a clinical tool^ simple enough^ in^ fornat
that it could be administered and scored by people who had
no prior training in psychological testing. Essentially, it was
to be^ used in^ order to^ alert^ professional^ personnel^ to^ the
possibility of developmental delay and the need for special
professional diagnostic investigation. A delay in any one of
the four subtest areas was indicated when the child was not
able to pass an item which more than 90 per cent of
children his^ chronological age were^ able^ to^ do.*
The test was standardized "on^ 1,036 presumably
normal children (2 weeks to 6 years of age) whose families
reflect the occupational and ethnic characteristics^ of^ the
population of^ Denver."'
Dr. Barnes is with the^ South Okanagan Mental Health
Centre, and Ms. Stark is with the^ South Okanagan Health Unit, Kelowna, British Columbia, Canada. Requests for reprints should be sent to^ Dr. Barnes.
- (^) Since the present study was carried out the (^) scoring method has been revised2 ; however, the criteria^ for^ scoring
"delay" have remained^ unchanged.
The South Okanagan Health Unit has^ long^ been
concerned with the problem of screening preschoolers^ in a
variety of functional areas (vision, hearing, general personal
development). When the Denver Developmental Screening
Test (DDST) became available in Canada, the professional
staff at the Health Unit quickly recognized that it could
provide an important addition to their screening program.
However, as the DDST had been standardized on an urban
population and as much of the South Okanagan Health
Unit covers a rural or semirural population which is
significantly different (occupationally) from the Denver
population (^) (X2 = 53.40, df =^ 4, p < 0.001), it was necessary
to look at the need for local norms in order to determine
the validity of using the DDST in the South Okanagan
region. The need for standardization has been^ emphasized
further by the recent study of Bryant et al.,3 who reported
on age/performance differences found between Denver
children and Newcastle (England) children in the subtest
areas of gross motor and language development.
Method
Subjects
Subjects consisted^ of^122 male^ and^104 female,
caucasian preschoolers ranging in^ age from^ 2 weeks^ to^ 6.
years. There were 25 age categoriest all^ together; subjects
t Age categories were taken from the Frankenburg and Dodds study.'
DENVER DEVELOPMENTAL SCREENING^ TEST^363
TABLE 1-intercorrelations^ between^ the Denver^ Norms of^ the DDST and Norms Collected in the South Okanagan Health Unit Region (N = 226)
Denver Region
South Okanagan Health Unit Gross Fine Personal- Region Motor Motor Language Social
Gross Motor 0.996* Fine Motor 0.998* Language 0.996* Personal-Social 0.996*
- (^) Significant at the p < 0.005 level.
were selected on the basis of a stratified random sample of
the general population residing in the South Okanagan
Health Unit region. All those children who were considered
to be of high risk developmentally (premature, breech
deliveries, etc.) were excluded from the study.
The sample population was stratified according to the
occupational status of the head of the family as indicated in
the 1966 Canadian census. The occupational breakdown
was as follows: professional, 8.00 per cent; managerial,
16.44 per cent; sales, 7.56 per cent; craftsman, 28.89 per
cent; and semi- and unskilled, 39.11 per cent. A representa-
tive sample for each of these occupational groups was
selected from each of the^ six^ school^ district^ areas^ in^ the
South Okanagan region.
Procedure
Each subject was assessed on^ the DDST by a^ trained
and experienced examiner. Inter-rater reliability for the five
examiners varied between 0.93 and 0.95.
The norms were computed by determining the^ mean
scores for each age category in the sample population and
then comparing these scores with the mean scores for the
same age categories collected by Frankenburg and Dodds in
the original standardization study.
Results
The correlations between the two sets of norms for the
four major functional areas of the DDST are shown in
Table 1.
Table 2 includes the^ intercorrelations^ of the four
subtest areas for the South Okanagan Health Unit^ sample
by age groups. Frankenburg and Dodds' do not^ report the subtest intercorrelations^ nor^ the^ intercorrelations^ between
subtest scores and occupational class, school district, and
sex; hence, it^ was not^ possible^ to^ compare^ the^ two^ sets^ of
norms in these areas.*
* The computer program for the original Denver
sample is no^ longer available^ so^ this^ information^ cannot^ be
recalled. Frankenburg, W.^ K., personal^ communication,
Table 3 includes the means and standard deviations for
the four subtest areas by occupational class.
The distribution of scores according to mean,^ standard
deviations, and range for each subtest area by age groups is
presented graphically in Figures 1 to 4.
Discussion
In the present normative study it has been demon-
strated that the DDST can validly be used on a rural and
semirural Canadian population. The correlations between
the two sets of norns (Denver and South Okanagan
regions) were of very high magnitude in each of the four
subtest areas. Such a finding generally indicates very^ good
separation across age groups; however, as can be seen from
Figures 1 to 4, there is enough overlap^ between^ age^ groups
that the DDST does appear to be serving a useful function
in measuring preschool child development, i.e., the^ test
does seem to be differentiating something over and above
simple chronological age.
TABLE 2-Correlational Matrix by Age and Subtest Area
Subtest Areas
Fine Personal- Ages No. Motor Language Social
months Gross Motor 1-3 28 0.820 0.790 0. 4-6 28 0.465 0.277 0A 7-9 25 0.398 0.436 0. 10-12 26 0.239 OA87 0A 13-16 25 0.607 0.337 0. 17-26 27 0.438 0.535 0. 27-45 31 0.538 0.387 0. 46-78 36 0.376 0.510 0A All ages 226 0.972 0.961 0.
Fine Motor 1-3 0.708 0. 4-6 0.299 0. 7-9 OA34 0. 10-12 0.445 0. 13-16 0.596 0. 17-26 0.534 0. 27-45 0.602 0. 46-78 OA94 0. All ages 0.962 0.
Language 1-3 0. 4-6 0. 7-9 0. 10-12 0. 13-16 0. 17-26 0. 27-45 0. 46-78 0. All (^) ages 0-
364 AJPH APRIL, 1975, Vol.^ 65, No.
Fine M5otor Scores 15 20
_~~~~~~~~~~~~~~~~~
i-LZ4ZH
*-E4Zii
Mean
i~ ~ ~ ~ ~ o (^) I -
lowest -ig^ ieat score score
FIGURE 2 Distribution of Fine^ Motor^ scores^ by age groups.
were some rather surprising results. For example, in all four
subtest areas, children of professional families consistently
obtained the^ lowest^ mean^ scores^ of^ all^ the^ occupational
groups in the study while children of craftsman families
consistently scored the highest. Although this finding^ was
not significant statistically for a number of the occupa-
tional classes, it was true in all four subtest areas between
professional and^ craftsman^ families^ (t^ =^ 2.11,^ df^ =^ 81,p^ <
0.025 for Gross Motor; t^ =^ 1.95, df^ =^ 81,p <^ 0.05^ for^ Fine Motor; t =^ 1.81, df =^ 81, p <^ 0.05 for Language;^ and^ t^ = 1.82, df =^ 81,p < 0.05 for Personal-Social). A similar result
was found in^ the^ personal-social area^ between the^ children
of professional and those of^ semiskilled^ families^ (t^ =^ 2.00, df =^ 86, p < 0.025). These findings suggest that, contrary to popular opinion, children^ of^ working^ class and^ lower^ middle^ class parents may well be^ developmentally equal^ or^ superior^ to children of (^) professional and managerial parents. This
relationship is^ particularly^ surprising^ in the^ language^ area, for most reports on the development of language show a
marked positive relationship between language skills and
middle and upper class occupations.' 1-
These findings also differ substantially from the results reported by^ Sandler^ et^ al.20^ In^ that^ study^ considerable differences were found in the language and fine motor areas for black and Puerto Rican preschoolers when^ compared
with the Denver^ DDST^ norms,^ and the authors^ recommend the use of caution when^ using published^ norms^ for lower class preschoolers located in urban areas. The fact that the present study did not find^ significant differences in either the language or fine motor^ areas between middle and lower class children could be attribut- able to two possible factors: one is that all of the preschoolers were^ caucasian^ (there^ are^ no^ black^ or^ Puerto Rican preschoolers living in^ the^ South^ Okanagan^ region); the other factor is that the South Okanagan Valley^ has^ no urban areas within its boundaries. Hence, the^ relatively poor levels of language and fine motor functioning found^ in the Sandler et al. study may be due primarily to conditions of urban^ living and^ ethnic^ characteristics^ rather^ than^ lower socioeconomic class. One possible explanation of these^ present findings^ may have something to do with the changes in^ occupational conditions and educational levels for the average working class person that have taken place in^ the^ last^20 years. There appears to be little doubt that educational level^ has risen substantially for the average working person in^ the^ last two decades; furthermore, the work^ week^ for^ these^ same people has been reduced substantially, while^ the^ work week for professional and managerial occupations has^ increased substantially.
As Milner's2l findings on^ the^ relationship of^ language
366 AJPH APRIL, 1975, Vol.65, No.
5 10 25 30
Age (in nmnths)
1 - 3
4 - 6
7 -^9
10 -^12
13 - 16
17 -^26
27 -^45
46 -^78
Language Scores 10 M 201
-1 Mean^ +i r lowes (^) hghst score score FIGURE 3 Distribution of Language scores by age groups. development and^ length of^ time^ spent with^ parents^ in #### conversation and at meal times are probably as^ valid^ for #### nonblack families as for black ones, and^ since^ Irwin^ S findings indicate that working class mothers can^ be #### educated to help develop their preschool children's lan- #### guage through reading, then the present findings are not necessarily contradictory to the results of studies carried #### out 20^ and 30 years ago. Certainly,^ practically^ all^ mothers are better educated in^ infant^ care^ than mothers^ were^ two #### decades ago, and in this particular geographic region the educational program offered by the^ public health^ depart- #### ment to young mothers is especially comprehensive and intensive. It is indeed unfortunate that the normative data from #### the original DDST study were not retrievable, especially the #### sex and^ occupational class^ variables.^ However,^ in^ an^ internal #### analysis of the^ test's^ ability to^ discriminate^ across^ age^ levels, it stood up exceedingly well. Only one^ subtest^ area^ failed^ to #### discriminate adequately and that was^ in^ the^ fine^ motor^ area for age groups 7 to 9^ versus 10 to^12 months^ (t^ =^ 1.268,^ df #### = 49, N.S.). All other age groupings in all four subtest areas #### were significantly different from each^ other.^ It^ is^ possible that this one nonsignificant finding is^ due^ to^ a^ chance factor only. #### In general, the DDST appears to^ be^ an^ adequately constructed screening device in^ that^ it^ discriminates^ across #### the age groupings in all^ four^ subtest^ areas^ with^ one exception as noted above. That there are essentially no sex differences in any of the four subtest areas and that there #### are some unexpected significant differences between chil- dren of^ different occupational classes^ are^ both^ findings contrary to previous studies on early child^ development. Furthermore, how well the DDST actually predicts^ signifi- cant developmental delay needs further investigation, for #### although Frankenburg^ et^ al.3^ report^ relatively^ high^ levels of agreement between the revised^ method^ of^ scoring^ and the DDST and such external criterion^ measures^ as^ the Bayley Infant Scale for Children and^ the^ Stanford^ Binet Form LM, these comparisons are essentially concurrent validation studies and do not indicate the longitudinal #### predictability of DDST scores, especially where^ the #### abnormal and questionable ones might be compared to #### pediatric and neurological follow-up studies, i.e., to deternine exactly how many false positives and false negatives might actually exist over an extended period of time.* Furthermore,, although the^ comparison between DDST scores and scores on the Bayley Infant Scale for Children constitutes a reasonably appropriate concurrent validity test, the comparison between DDST and Stanford Binet Form LM scores does not, if it is indeed true that * (^) The present authors are currently involved in such a follow-up study and will report on the results at a later date. DENVER DEVELOPMENTAL SCREENING TEST^367 0 Ages (in months) 1 -^3 4 -^6 7 -^9 10 -^12 13 -^16 17 -^26 27 -^45 45 -^78 In Child Development Monographs, No. 15. Teacher's College, Columbia University, New York, 1934. 15. Irwin, 0. C. Infant Speech: The Effect of Family Occupational Status^ and of^ Age^ on^ Use^ of^ Sound Types. J. Speech Hear. Disord. 13:224-226, 1948. 16. Sampson, 0. C. A Study of Speech Development in Children 18-30 Months. Br. J. Educ. Psychol. 26:194-202, 1956. 17. Templin, M. C. Certain Language Skills in Children. In Institute of Child Welfare Monographs, Ser. No. 26. University of Minnesota Press, Minneapolis, 1957. 18. Van Alstyne, D. The Environment of Three Year Old Children: Factors Related to Intelligence and Vocabu- lary Tests. In Contributions to Education, No. 366. Teacher's College, Columbia University, New York, 1929. 19. Young, F. M. An Analysis of Certain Variables in a Developmental Study of Language. Genet. Psychol. Monogr. 23:3-141, 1941. 20. Sandler, L., Van Campen, J., Ratner, C., Stafford, C., and Weismar, R. Responses of Urban Preschool Children to a Developmental Screening Test. J. Pediatr. 77:775-781, 1970. 21. Milner, E. A Study of the Relationship between Reading Readiness in Grade One School Children and Patterns of Parent-Child Interaction. Child Dev. 22:95-112, 1951. 22. Irwin, 0. C. Infant Speech: Effect of Systematic Reading of Stories. J. Speech Hear. Res. 3:187-190, 1960. 23. Frankenburg, W. K., Goldstein, A. D., and Camp. B. W. The Revised Denver Developmental Screening Test: Its Accuracy as a Screening Instrument. J. Pediatr. 79:988-995, 1971. A TRIBUTE TO DR. SCHLESINGER #### The Maternal and Child Health Section Council of^ the American^ Public Health^ Association,^ at^ its #### annual meeting in New Orleans October 22, 1974, prepared a resolution, published below, in^ eulogy #### to their esteemed colleague, Edward R. Schlesinger, MD, MPH. #### A fellow of APHA, Dr. Schlesinger was past Chairman of the Program Area^ Committee^ on^ Child #### Health, and had served as Secretary and Vice-Chairman of the Maternal and Child Section. He^ had #### been a member of APHA^ since^ 1940. #### Born in New^ York^ City^ in^ 1911,^ Dr. Schlesinger^ was^ a^ graduate^ of^ Columbia^ University,^ received #### his MD from that University, and^ his^ MPH from^ the^ Johns^ Hopkins^ School^ of^ Hygiene^ and^ Public Health. At the time of^ his^ death,^ Dr.^ Schlesinger^ was^ Professor^ and^ Head^ of^ Maternal^ and Child #### Health, and Associate Dean for Academic Affairs, at^ the^ University^ of Pittsburgh^ Graduate^ School^ of #### Public Health. The Edward R. Schlesinger Memorial Fund^ has^ been established^ at the^ University^ of #### Pittsburgh. #### RESOLUTION #### At this, the annual meeting of the American Public Health^ Association,^ the Maternal^ and^ Child #### Health Section takes^ formal^ note^ of^ the^ untimely^ loss of^ our^ esteemed colleague, Dr. Edward R. #### Schlesinger. #### Dr. Schlesinger had a profound influence on^ health^ workers in maternal and child health.^ Many #### members of the section and of the APHA have been inspired, encouraged, trained, and^ so^ enabled^ to #### improve the^ public health because^ of their valuable association with^ him. #### Throughout his^ long^ career,^ Dr.^ Schlesinger^ was^ a^ loyal^ and devoted member of the^ section^ and #### added substantially to^ its^ strength. During^ his many^ years^ as^ a^ member he held^ a^ number^ of^ positions #### of responsibility, including that^ of Chairman. He^ was^ instrumental in the^ production^ of^ a^ number^ of #### basic studies and contributed extensively to the literature. He^ was^ the author, together with Dr. #### Pauline Stitt, of The Maternal & Child Health^ Section:^ A^ Political &^ Social^ History,^ which is^ being #### distributed for the first time today. #### Administrator, investigator, author, and^ teacher, Ed^ Schlesinger made^ a^ most^ significant^ impact #### on the health of mothers and children. We shall^ miss^ him. DENVER DEVELOPMENTAL SCREENING^ TEST^369
