PEDIATRICS Vol. 121 No. 6 June 2008, pp. 1299-1300 (doi:10.1542/peds.2008-0615)
LETTER TO THE EDITOR |
Should We Screen for Language Delay in Toddlers?
C.P.B. van der Ploeg, PhDC.I. Lanting, MD, PhD
P.H. Verkerk, MD, PhD
Department of Prevention and Health
TNO Quality of Life
2301 CE, Leiden, Netherlands
To the Editor.—
van Agt et al1 evaluated the effects of screening at toddler age on (language) performance at age 8 with a large-scale randomized, controlled trial. Overall, the design of the study was strong, and the results were presented clearly. However, we have serious problems with the interpretation of the results and the conclusion that "nationwide implementation of the intervention as part of a routine monitoring of children's general development can be recommended." In our view, this study provides only weak evidence for some positive effects: for children who were completely screened, only 2 of 7 primary outcome variables showed significant positive effects (according to 1- and 2-sided tests). Five effects were not significant (see Table 41). For those children assigned to the experimental condition (intention-to-treat analysis; see Table 31), 3 primary outcome variables showed significant positive effects (these were not significant when tested 2-sided); however, only 1 of these was the same as above. Four effects were not significant. Thus, we conclude that the results suggest consistent positive effects for only 1 primary outcome measure (special school attendance). Furthermore, for several outcome variables the mean of the experimental group was higher than the mean of the control group and, hence, not in the expected direction. Although these effects are nonsignificant, they raise doubt on the 1-sided test procedure. Design of a study for 1-sided testing (as is done in this study) is allowed only if all effects are in the expected direction.
Large-scale implementation of the screening program is only sensible if benefits of early detection outbalance material and immaterial costs that inevitably arise from population screening. Whether a break-even point can be reached remains questionable, because the sensitivity of the screening instrument is 25% to 50%.1 In other words, 50% to 75% of patients remain undetected, with their parents falsely reassured about their child's language development. Also, causality from screening to beneficial effects needs to be solidly proven. Because the overall response rate was low (19%–47% depending on outcome measure), bias cannot be excluded. Furthermore, only children who screen positively, comply with referral for diagnosis, and receive effective early treatment can possibly benefit from screening. Of the screened children, 2.4% were screen-positive,2,3 63% of which complied with referral for diagnosis.3 Of these referrals, 66% were recognized with a language delay.3 An estimated 40% of children with language delays at the age of 2 to 3 years benefit from treatment; the remaining delays resolve spontaneously.4 Thus, the proportion of screened children who might benefit from screening equals (2.4%*63%*66%*40% =) 0.4%. In practice, even smaller beneficial effects will be expected, because language delay can also be detected among unscreened children via routine child health care. Apparently, the larger reduction in the only consistent effect reported in this study (a 1.5% difference in special school attendance between screened and unscreened children)1 is caused mainly by factors other than the screening alone.
Therefore, the US Preventive Services Task Force statement that there is not enough evidence to justify large-scale screening activities on language delay, in our opinion, remains valid.5
REFERENCES
- van Agt HM, van der Stege HA, de Ridder-Sluiter H, Verhoeven LT, de Koning HJ. A cluster-randomized trial of screening for language delay in toddlers: effects on school performance and language development at age 8.
Pediatrics. 2007;120
(6):1317
–1325
[Abstract/Free Full Text] - de Koning HJ, de Ridder-Sluiter JG, van Agt HME, et al. A cluster-randomised trial of screening for language delay in toddlers.
J Med Screen. 2004;11
(3):109
–116
[Abstract/Free Full Text] - de Koning HJ, de Ridder-Sluiter JG, van Agt HME, et al. Vroegtijdige Onderkenning van Taalontwikkelingsstoornissen 0–3 Jaar. Rotterdam, Netherlands: Erasmus Medisch Centrum; 2000
- Law J, Boyle J, Harris F, Harkness A, Nye C. Screening for speech and language delay: a systematic review of the literature. Health Technol Assess. 1998;2 (9):1 –184[Medline]
- Nelson HD, Nygren P, Walker M, Panoscha R. Screening for speech and language delay in preschool children: systematic evidence review for the US Preventive Services Task Force [published correction appears in Pediatrics. 2006;117(6):2336–2337]. Pediatrics. 2006;117 (2). Available at: www.pediatrics.org/cgi/content/full/117/2/e298
PEDIATRICS (ISSN 1098-4275). ©2008 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
