Denver Scale

From Wikipedia, the free encyclopedia

Jump to: navigation, search

The Denver Developmental Screening Test (DDST), commonly known as the Denver Scale, is a test for screening cognitive and behavioural problems in preschool children. It was developed by William K. Frankenburg and first introduced by him and J.B. Dobbs in 1967.^[1] The test is currently marketed by Denver Developmental Materials, Inc., in Denver, Colorado, hence the name.

The scale reflects what percentage of a certain age group is able to perform a certain task. In a test to be administered by a pediatrician or other health or social service professional, a subject's performance against the regular age distribution is noted. Tasks are grouped into four categories (social contact, fine motor skill, language, and gross motor skill) and include items such as smiles spontaneously (performed by 90% of three-month-olds), knocks two building blocks against each other (90% of 13-month-olds), speaks three words other than "mom" and "dad" (90% of 21-month-olds), or hops on one leg (90% of 5-year-olds).

According to a study commissioned by the Public Health Agency of Canada, the DDST is the most widely used test for screening developmental problems in children.^[2] While this study acknowledges the test's utility for detecting severe developmental problems, the test has been criticized to be unreliable in predicting less severe or specific problems. The same criticism has been upheld for the currently marketed revised version of the Denver Scale, the Denver-II.^[3]. Frankenburg has replied to such criticism by pointing out that the Denver Scale is not a tool of final diagnosis, but a quick method to process large numbers of children in order to identify those that should be further evaluated.^[4]

This revised definition of the Denver's function remains commensurate with what screening tests are designed to do: sort those who probably have problems from those who probably don't. Thus standards for screening test construction still apply to the Denver. Although the instrument has proven reliability, it was not constructed on a large, current, nationally representative sample. It has not been studied for validity (given alongside diagnostic measures to view their relationship or researched for the kinds of problems it may or may not detect). As a consequence, the measure was not studied by its authors for the most critical attribute of any screen, its accuracy. Studies by other researchers showed it to detect only about 50% of children with disabilities, although its specificity in identifying normally developing children is high (when questionables are grouped with normal scores) and the converse when questionable scores are grouped with abnormal results. Since 1991, researchers have appealed to the author to recall and improve the measure but to no avail. Currently the measure is excluded from lists of recommended tools in several states (e.g., Minnesota Department of Education. For a list of accurate alternatives see The website of the American Academy of Pediatrics' Section on Developmental and Behavioral Pediatrics

References

1. ^ Frankenburg, William K.; Dobbs, J.B. (1967). "The Denver Developmental Screening Test". The Journal of Pediatrics 71 (71): 181–191. doi:10.1016/S0022-3476(67)80070-2.
2. ^ Canadian Task Force on the Periodic Health Examination (1994) The Canadian Guide to Clinical Preventive Health Care. Ottawa: Minister of Supply and Services Canada. Chapter 26 "Preschool Screening for Developmental Problems" [1]
3. ^ Glascoe, Frances Page et al. (1992). "Accuracy of the Denver-II in Developmental Screening". Pediatrics (89): 1221–1225.
4. ^ Frankenburg, William K. (2002). "Developmental Surveillance and Screening of Infants and Young Children". Pediatrics 109 (109): 144–145. doi:10.1542/peds.109.1.144. PMID 11773555